KinetisV - Peripheral Module Reference

Kinetis V Series Peripheral Module
Quick Reference
Document Number: KVQRUG

Rev 0, 02/2014
Kinetis V Series Peripheral Module Quick Reference, Rev. 0, 02/2014

2
Freescale Semiconductor, Inc.
Contents
Section number
Title
Page
Chapter 1
General System Setup (Software Considerations)
1.1
Software considerations..................................................................................................................................................9
1.1.1
Overview............................................................................................................................................................9
1.1.2
Code execution...................................................................................................................................................9
1.1.3
Reset and booting...............................................................................................................................................9
1.1.4
1.1.3.1
Device state during reset....................................................................................................................10
1.1.3.2
Device state after reset.......................................................................................................................10
Typical system initialization .............................................................................................................................10

1.1.4.1
Lowest level assembly routines.........................................................................................................10

1.1.4.1.1 Initialize general purpose registers...................................................................................10
1.1.4.1.1.1 Unmask interrupts at ARM core ...............................................................11
1.1.4.1.1.2 Branch to start of C initialization code.........................................................11
1.1.4.2
Startup routines..................................................................................................................................11
1.1.4.2.1 Disable watchdog..............................................................................................................11
1.1.4.2.2 Initialize RAM..................................................................................................................11
1.1.4.2.3 Enable port clocks.............................................................................................................11
1.1.4.2.4 Ramp system clock to selected frequency........................................................................12
1.1.4.2.5 Enable UART for terminal communication......................................................................12
1.1.4.2.6 Jump to start of main function for application..................................................................12
Chapter 2
General System Setup (Hardware Considerations)
2.1
Hardware considerations.................................................................................................................................................13
2.1.1
Overview............................................................................................................................................................13
2.1.2
Floorplan............................................................................................................................................................13
2.1.2.1
2.1.3
Connectors.........................................................................................................................................14
PCB routing considerations...............................................................................................................................14

2.1.3.1
Power supply routing.........................................................................................................................14

Section number
Title
Page
2.1.3.2
Power supply decoupling and filtering..............................................................................................15
2.1.3.3
Oscillators..........................................................................................................................................15
2.1.3.3.1 MCG oscillator..................................................................................................................15
2.1.3.4
General filtering.................................................................................................................................18
2.1.3.4.1 RESET_b and NMI_b.......................................................................................................18
2.1.3.4.2 General purpose I/O..........................................................................................................19
2.1.3.4.3 Analog inputs....................................................................................................................19
2.1.4
PCB layer stack-up.............................................................................................................................................19
2.1.5
Other module hardware considerations..............................................................................................................22

2.1.5.1
Debug interface..................................................................................................................................22
Chapter 3
Nested Vector Interrupt Controller (NVIC)
3.1
NVIC...............................................................................................................................................................................25
3.1.1
3.1.2
Overview............................................................................................................................................................25
3.1.1.1
Introduction .......................................................................................................................................25
3.1.1.2
Features .............................................................................................................................................25
Configuration examples.....................................................................................................................................26
3.1.2.1
Configuring the NVIC.......................................................................................................................26

3.1.2.1.1 Code example and explanation.........................................................................................26
3.1.2.2
Relocating the vector table.................................................................................................................27

3.1.2.2.1 Code example and explanation.........................................................................................28
Chapter 4
Clocking System
4.1
Clocking..........................................................................................................................................................................29
4.1.1
Overview............................................................................................................................................................29
4.1.2
Features..............................................................................................................................................................29
4.1.3
4.1.4
Clocking system device hardware implementation...........................................................................................33
4.1.5
Layout guidelines for general routing and placement........................................................................................34

4
Section number
4.1.6
Title
Page
References..........................................................................................................................................................34
Chapter 5
Power Management Control (PMC/SMC/LLWU/RCM)
5.1
Introduction.....................................................................................................................................................................37
5.2
Using the power management controller........................................................................................................................37
5.3
5.2.1
Overview............................................................................................................................................................37
5.2.2
Using the low voltage detection system.............................................................................................................37

5.2.2.1
POR and LVD features......................................................................................................................37
5.2.2.2
Configuration examples.....................................................................................................................38
5.2.2.3
Interrupt code example and explanation............................................................................................39
5.2.2.4
Hardware implementation..................................................................................................................40
Using the system mode controller...................................................................................................................................42

5.3.1
5.3.2
Overview............................................................................................................................................................42
5.3.1.1
Introduction........................................................................................................................................42
5.3.1.2
Entering and exiting power modes.....................................................................................................43
5.3.2.1
SMC code example and explanation..................................................................................................44

5.3.2.1.1 Mode entry sequence serialization ...................................................................................45
5.4
5.3.2.2
Entering wait mode............................................................................................................................45
5.3.2.3
Exiting low-power modes..................................................................................................................45
Using the low leakage wakeup unit................................................................................................................................46

5.4.1
5.4.2
Overview............................................................................................................................................................46
5.4.1.1
Mode transitions ................................................................................................................................46
5.4.1.2
Wake-up sources ...............................................................................................................................46
LLWU configuration examples.........................................................................................................................47

5.4.2.1
Enabling pins and modules in the LLWU..........................................................................................47
5.4.2.2
Module wake-up................................................................................................................................47
5.4.2.3
Pin wake-up........................................................................................................................................47
5.4.2.4
LLWU port and module interrupts.....................................................................................................48

Section number
5.4.2.5
Title
Page
Wake-up sequence.............................................................................................................................48
5.5
Module operation in low-power modes..........................................................................................................................50
5.6
Mode transition requirements.........................................................................................................................................52
5.7
Source of wake-up, pins, and modules...........................................................................................................................53
Chapter 6
enhanced Direct Memory Access (eDMA) Controller
6.1
6.2
eDMA.............................................................................................................................................................................55
6.1.1
Overview............................................................................................................................................................55
6.1.2
Introduction........................................................................................................................................................55
eDMA trigger..................................................................................................................................................................57
6.2.1
DMA multiplexer...............................................................................................................................................57
6.2.2
Trigger mode......................................................................................................................................................58
6.2.3
Multiple transfer requests...................................................................................................................................58
6.3
Transfer process..............................................................................................................................................................59
6.4
Configuration steps ........................................................................................................................................................60
6.5
ExampleADC scan multiple channels........................................................................................................................60

6.5.1
Module configuration.........................................................................................................................................61
Chapter 7
Universal Asynchronous Receiver and Transmitter (UART) Module
7.1
Overview.........................................................................................................................................................................63
7.2
Features...........................................................................................................................................................................63
7.3
Configuration example....................................................................................................................................................64
7.4
7.3.1
UART initialization example.............................................................................................................................65
7.3.2
UART receive example......................................................................................................................................66
7.3.3
UART transmit example....................................................................................................................................67
7.3.4
UART configuration for interrupts or DMA requests.......................................................................................67
UART RS-232 hardware implementation......................................................................................................................68
Chapter 8
Memory-Mapped Divide and Square Root (MMDVSQ)
8.1
Introduction.....................................................................................................................................................................69
Section number
Title
Page
8.2
Features...........................................................................................................................................................................69
8.3
MMDVSQ in Cortex-M0+ core platform.......................................................................................................................70
8.4
MMDVSQ programming model.....................................................................................................................................70
8.5
Square root using Q notation for fractional data.............................................................................................................72
8.6
Execution time................................................................................................................................................................72
8.7
Tips of usage...................................................................................................................................................................73
8.7.1
Tips for minimizing the interrupt latency..........................................................................................................73
8.7.2
Tips for context save and restore in interrupt....................................................................................................73
Chapter 9
Analog-to-Digital Converter (ADC)
9.1
Overview.........................................................................................................................................................................75
9.2
Introduction.....................................................................................................................................................................75
9.3
Features...........................................................................................................................................................................76
9.4
ADC configuration..........................................................................................................................................................76
9.5
ADC hardware design considerations.............................................................................................................................77
Chapter 10
Programmable Delay Block (PDB)
10.1 Overview.........................................................................................................................................................................79
10.2 Introduction.....................................................................................................................................................................79
10.3 Features...........................................................................................................................................................................80
10.4 PDB pre-trigger...............................................................................................................................................................80
10.5 Ping-Pong operation........................................................................................................................................................82
10.5.1 PDB pre-trigger sample mode............................................................................................................................82
10.5.2 Back-to-back sample mode................................................................................................................................82
10.5.2.1 2-channel ADC input back-to-back sample mode.............................................................................83
10.5.2.2 4-channel ADC input pre-trigger sample mode.................................................................................83
Chapter 11
Using FlexTimer (FTM) via Programmable Delay Block (PDB) to Schedule ADC Conversion
11.1 Overview.........................................................................................................................................................................85
11.2 Introduction.....................................................................................................................................................................85
Section number
Title
Page
11.3 Features...........................................................................................................................................................................86
11.4 Configuration code..........................................................................................................................................................87
11.4.1 FTM trigger 4-channel ADC ping-pong conversion via back-to-back mode of PDB.......................................87
11.4.2 ADC configuration.............................................................................................................................................87
11.4.3 PDB configuration.............................................................................................................................................88
11.4.4 FTM configuration.............................................................................................................................................88
11.4.5 ADC ISR............................................................................................................................................................89
Chapter 12
FlexTimer Module (FTM)
12.1 Overview.........................................................................................................................................................................91
12.2 Introduction.....................................................................................................................................................................91
12.3 Features...........................................................................................................................................................................91
12.3.1 FTM clock..........................................................................................................................................................92
12.3.2 Interrupts and DMA...........................................................................................................................................92
12.3.3 Modes of operation............................................................................................................................................92
12.3.4 Updating MOD and CnV...................................................................................................................................93
12.3.5 FTM period........................................................................................................................................................93
12.3.6 Additional features.............................................................................................................................................94
12.4 Configuration examples..................................................................................................................................................94
12.4.1 Example Edge Aligned PWM and Input Capture Mode.................................................................................94

8
Chapter 1
General System Setup (Software Considerations)
1.1 Software considerations
1.1.1 Overview
This chapter provides a quick look at some of the general characteristics of the Kinetis V
series of MCUs. This is a brief introduction of the operation of the devices and typical
software initialization.
For more information, see the device-specific reference manual and data sheet.
1.1.2 Code execution

The Kinetis V series features embedded Flash and SRAM memory for data storage and
program execution.
1.1.3 Reset and booting

When the processor exits reset, it fetches the initial stack pointer (SP) from vector table
offset 0 and the program counter (PC) from vector table offset 4. The initial vector table
must be located in the flash memory at the base address (0x0000_0000). However, the
vector table can be relocated to SRAM after the boot-up sequence if desired. This device
supports booting from internal flash and RAM. This device supports booting from
internal flash with the reset vectors located at addresses 0x0 (initial SP_main), 0x4 (initial
PC), and RAM with the relocation of the exception vector table to RAM.
After fetching the stack pointer and program counter, the processor branches to the PC
address and begins executing instructions.

Software considerations
For more information, see the "Reset and Boot" chapter of the device-specific reference
manual.
1.1.3.1 Device state during reset

With the exception of the SWD pins, during reset the digital I/O pins go to a disabled
(high impedance) state with internal pullups/pulldowns disabled. Pins with analog
functionality will default to their analog functions.
1.1.3.2 Device state after reset

After reset, the digital I/O pins remain disabled until enabled by software. Also, interrupts
are disabled and the clocks to most of the modules are off. The default clock mode after
reset is FLL Engaged Internal (FEI) mode. In this mode, the system is clocked by the
frequency-locked loop (FLL) using the slow internal reference clock as its reference. The
watchdog timer is active; therefore it will need to be serviced, or disabled if debugging.
The core clock, system clock, and flash clock are enabled after reset to support booting.
Also, the flash memory controller cache and prefetch buffers are enabled.
1.1.4 Typical system initialization

The following is a summary of typical software initialization. The code snippets are taken
from a platinum project written in IAR Embedded Workbench. This project is available
in the Kinetis sample code which accompanies this guide.
1.1.4.1 Lowest level assembly routines

These routines are assembly source code found in the file crt0.s. The address of the start
of this code is placed in the vector table offset 4 (initial program counter) so that it is
executed first when the processor starts up. This is accomplished by labeling this section,
exporting the label, and placing the label in the vector table. The vector table can be
found in vectors.h. In this example the label used is __startup.
1.1.4.1.1
Initialize general purpose registers
As a general rule, it is recommended to initialize the processor general purpose registers

(R0-R7) to zero. One way of doing this is with the LDR instruction.

10
Chapter 1 General System Setup (Software Considerations)

LDR
LDR
LDR
LDR
LDR
LDR
LDR
LDR
1.1.4.1.1.1
CPSIE
r0,=0
r1,=0
r2,=0
r3,=0
r4,=0
r5,=0
r6,=0
r7,=0
Unmask interrupts at ARM core
1.1.4.1.1.2
import start
BL
; Initialize the GPRs
; Unmask interrupts
Branch to start of C initialization code

start
; call the C code
1.1.4.2 Startup routines

These routines are C source code found in the files start.c and sysinit.c. This code
provides general system initialization that may be adapted depending on the application.
1.1.4.2.1
Disable watchdog
For code development and debugging, it is best to disable the watchdog. The watchdog
can be disabled by first unlocking the watchdog followed by clearing the WDOGEN bit.
Users should avoid any breakpoints in between the following code lines during
debugging.
/* Disable the watchdog timer */
/* First unlock the watchdog so that we can write to registers */
DisableInterrupts;
WDOG>UNLOCK = 0xC520;
WDOG>UNLOCK = 0xD928; EnableInterrupts;
/* Clear the WDOG bit to disable the watchdog */
WDOG >STCTRLH & =~WDOG_STCTRLH_WDOGEN_MASK;
1.1.4.2.2
Initialize RAM
Depending on the application, the following steps may be required. First, copy the vector
table from flash to RAM, copy initialized data from flash to RAM, clear the zeroinitialized data section, and copy functions from flash to RAM.
1.1.4.2.3
Enable port clocks
To configure the I/O pin muxing options, the port clocks must first be enabled. This
allows the pin functions to later be changed to the desired function for the application.
11
Software considerations
SIM->SCGC5 |= (SIM_SCGC5_PORTA_MASK
| SIM_SCGC5_PORTB_MASK
| SIM_SCGC5_PORTC_MASK
| SIM_SCGC5_PORTD_MASK
| SIM_SCGC5_PORTE_MASK );
1.1.4.2.4
Ramp system clock to selected frequency
The Multipurpose clock generator (MCG) provides several options for clocking the
system. Configure the MCG mode, reference source, and selected frequency output based
on the needs of the system.
1.1.4.2.5
Enable UART for terminal communication
See the section describing UART in this document for more information.
1.1.4.2.6
Jump to start of main function for application
/* Jump to main process */

main();

12
Chapter 2
General System Setup (Hardware Considerations)
2.1 Hardware considerations
2.1.1 Overview
This chapter will outline the best practices for hardware design when using the Kinetis V
series MCUs. The designer must consider numerous aspects when creating the system so
that performance, cost, and quality meet the end-user expectations. Performance usually
implies high speed digital signalling, but it also applies to accurate sampling of analog
signals. Cost is influenced by component selection, of which the PCB may be the most
expensive element. Quality involves manufacturability, reliability, and conformance to
industry or governmental standards.
Evaluation boards are great for evaluating the operation and performance of the many
features of Freescale MCUs. However, evaluation systems are not ideal examples for
implementation of robust system design techniques. This document will mention some of
the hardware techniques found on the Freescale Tower Systems, and will give
recommendations that are more appropriate to conventional systems that are not required
to implement all of the feature options.
2.1.2 Floorplan
The organization of the printed circuit board (PCB) depends on many factors. Typically,
there are connectors, mechanical components, high speed signals, low speed signals,
switches, and power domains, among others, that need to be considered. While placement
of connectors and some mechanical components (switches, relays, and so on) is critical to
the end products form, there are some basic recommendations that can significantly
affect the electrical performance and electromagnetic compatibility (EMC) of the PCB
assembly.

13
Hardware considerations
2.1.2.1 Connectors
The PCB should be organized so that all of the connectors are along one edge of the
board and away from the MCU. The concept here is to prevent placing the MCU in
between connectors that can become effective radiators when cables are attached. This
also keeps the MCU from being in the path of high energy transients that can shoot across
the board from one connector to another. Connectors may be placed on adjacent edges of
the PCB if necessary, but only when the MCU is not in a direct path between the
connectors.
Connector locations should allow for placement of filter components. Noise must be
suppressed at the connector, before it can propagate onto the PCB. For more information
on this topic, see the input filtering section.
2.1.3 PCB routing considerations

This section covers critical power and filtering aspects of PCB layout.
2.1.3.1 Power supply routing

Routing of power and ground to digital systems is a topic that is discussed and debated in
many textbooks and references. The basic concept is to ensure that the MCU and other
digital components have a low impedance path to the power supply. The typical guidance
that was given for one and two layer PCBs was to use wide traces and few layer
transitions. The recommendations for todays high speed MCUs follow those given for
high speed microprocessor systems specifically, use planes for power and ground. This
may raise the PCB cost, but the benefits of crosstalk reduction, reduction of RF
emissions, and improved transient immunity can be realized with lower overall
production and maintenance costs.
In general, the ground routing should take precedence over any other routing. Ground
planes or traces should never be broken by signals. For packages with leads, like the
LQFP, a ground plane directly below the MCU package is recommended to reduce RF
emissions and improve transient immunity. All of the VSS pins of the MCU should be
tied to a ground plane. Ground traces from a plane should be kept as short as possible as
they are routed to circuitry on signal layers (top and bottom). Power planes may be
broken to supply different voltages. All of the VDD pins of the MCU should be tied to

14
Chapter 2 General System Setup (Hardware Considerations)
the proper power plane. Power traces from the planes should be kept as short as possible
as they are routed to circuitry, such as pullups, filters, other logic and drivers, on the top
and bottom layers. More information is given in the PCB layer stack-up section below.
2.1.3.2 Power supply decoupling and filtering

Bypass capacitors, while also called decoupling capacitors, are the storage elements that
provide the instantaneous energy demanded by the high speed digital circuits.
Power supply bypass capacitors must be placed close to the MCU supply pins. The basic
concept is that the bypass capacitor provides the instantaneous current for every logic
transition within the MCU. Fortunately, each Kinetis MCU has a low voltage internal
regulator for the MCU core logic, therefore the abrupt current demands of the internal
high speed logic are not as critical. However, external signals demand energy from the
power rails when they transition from one logic level to the other. The bypass capacitors
provide the local filtering so that the effects of the external pin transitions are not
reflected back to the power supply, which causes RF emissions.
The basic rule of placing bypass capacitors as close as possible to the MCU is still
appropriate. The idea is to minimize the loop created by the capacitor between the VDD
and VSS pins. The implementation of this rule depends on the number of mounting
layers, how the supplies are routed, and the physical size of the capacitors:
Number of mounting layers PCBs with components mounted only on the top side
will have a significant limitation on how close the bypass caps can be located due to
the number of components that require space. PCBs that have components mounted
on both sides of the PCB allow closer placement of the bypass capacitors.
Supply routing For Quad Flat Pack (QFP) packages, the power supply pins may be
supplied radially to the MCU using traces rather than from planes. Although it is
adequate to place the bypass capacitors close to the VDD and VSS pins on the traces
leading to the MCU, it is better to have the ground side of the bypass capacitor tied to
the ground plane (through a via and short trace) close to the VSS pin and the VDD
side tied to the power plane (through a via and short trace) close to the VDD pin.
2.1.3.3 Oscillators
The Kinetis MCU starts up with an internal digitally controlled oscillator (DCO) to
control the bus clocking, and then software can be used to enable an external oscillator if
desired. The external oscillator for the multipurpose clock generator (MCG) can range
from a 32.768 kHz crystal up to a 32 MHz crystal or ceramic resonator.

15
2.1.3.3.1
MCG oscillator
The high speed oscillator that can be used to source the MCG module is very versatile.
The component choices for this oscillator are detailed in the device-specific reference
manual. The placement of this crystal or resonator is described here.
The EXTAL and XTAL pins are located on the outside pad ring of the BGA package and
on corner pins of the LQFP/QFN package. This allows room for placement and routing of
the crystal or resonator on the top layer, close to the MCU. The feedback resistor and
load capacitors, if needed, can be placed on the top layer as well. See Figure 2-1, Figure
2-2, and Figure 2-3.
Note that the low power modes of this oscillator do not require a feedback resistor, and
may not require external load capacitors. See the device-specific reference manual for
details. This makes it as simple as possible because only one component has to be placed
and routed. Low power oscillators are more susceptible to interference by system
generated noise, therefore the guidelines for crystal routing are important.
The crystal or resonator must be located close to the MCU. No signals of any kind should
be routed on the layer directly below the crystal. The load capacitors and ground of the
crystal package must be connected to a single ground trace coming from the closest VSS
pin or the recommended ground under the MCU. An unbroken ground plane on the layer
directly below the crystal is recommended. A ground pour must be placed around the
crystal and its load components to protect it from crosstalk from adjacent signals on the
mounting layer.
R2
XTAL
R1
EXTAL
C1
Y1
C2
Figure 2-1. Typical crystal circuit

16
Ground under
MCU package
XTAL
EXTAL
VSS
VDD
C0
VDD
Ground pour
(not ground ring)
R1
VSS/GND
Y1
C1
C2
No signals or
components
under crystal
components!
Figure 2-2. Crystal layout for low power oscillator

17
Ground under
MCU package
XTAL
EXTAL
VSS
VDD
R2
VDD
C0
Ground pour
(not ground ring)
R1
VSS/GND
Y1
C1
No signals or
components
under crystal
components!
C2
Figure 2-3. Crystal layout for high power oscillator
2.1.3.4 General filtering

General purpose I/O pins should have adequate isolation and filtering from transients.
2.1.3.4.1
RESET_b and NMI_b
The RESET_b pin, if enabled, should have a 100 nF capacitor close to the MCU for
transient protection. The NMI_b pin, if enabled, must not have any capacitance
connected to it. Each pin, when enabled as their default function, has a weak internal
pullup, but an external 4.7 k to 10 k pullup is recommended. As with power pin
filtering, it is recommended to minimize the ground loop for the capacitor and the VDD
loop for the pullup resistor for these pins.

18
The RESET_b pin also has a configurable digital filter to reject potential noise on this
input after power-up. The configuration bits are located in the RCM_RPFC register.
While use of this filter may negate the need for the pullup and capacitor mentioned
above, it is still recommended to use external filtering in electrically noisy environments.
2.1.3.4.2
General purpose I/O
General purpose inputs, such as low speed inputs, timer inputs, and signals from offboard should have low pass filters (series resistor and capacitor to ground) to prevent data
corruption due to crosstalk or transients. The filter capacitor should be placed close to the
MCU pin, while the resistor can be placed closer to the source.
Inputs that come from connectors should have low pass filtering at the connector to
prevent noise from propagating onto the PCB. This requires a robust ground structure
around the connector. Series resistors for signals that come from off-board should be
placed as close to the connector as possible. A filter cap closer to the MCU input pin may
be required if the signal trace length is very long and can pick up noise from other
circuits.
Output pins must not have any significant capacitance placed close to the MCU. These
signals can have capacitors at the load or connector to minimize radiated emissions if
necessary.
NOTE
Must ensure that VDD be powered prior to /RESET_B and all
others GPIO pins.
2.1.3.4.3
Analog inputs
Analog inputs should also have low pass filters. The challenge with analog inputs,
especially for high resolution analog-to-digital conversions, is that the filter design needs
to consider the source impedance and sample time rather than a simple cutoff frequency.
This topic cannot be discussed in detail here, but the general concept is that fast sample
times will require smaller capacitor values and source impedances than slow sample
times. Higher resolution inputs may require smaller capacitor values and source
impedances than lower resolution inputs.
In general, capacitor values can range from 10 pF for high speed conversions to 1 uF for
low speed conversions. Series resistors are added for RC filter and a typical value is 100
Ohms.

19
2.1.4 PCB layer stack-up

The Kinetis V series MCUs are high speed integrated circuits. Care must be taken in the
PCB design to ensure that fast signal transitions, such as rise/fall times and continuous
frequencies, do not cause RF emissions. Likewise, transient energy that enters the system
needs to be suppressed before it can affect the system operation (compatibility). The
guidance from high speed PCB designers is to have all signals routed within one
dielectric (core or prepreg) of a return path, which usually is a ground plane on a multilayer PCB and an adjacent ground on a two layer PCB. This allows return currents to
predictably flow back to the source without affecting other circuits, which is the primary
cause of radiated emissions in electronic systems. This approach requires full planes
within the PCB layer stack and partial planes (copper pours) on signal layers where
possible. All ground planes and ground pours must be connected with plenty of vias.
Likewise, all like power planes and power pours must be connected with plenty of vias.
Recommended layer stackups:
4-Layer PCB A:
Layer 1 (top MCU location)Ground plane and pads for top mounted
components, no signals
Layer 2 (inner)Signals and power plane
Thick core
Layer 4 (bottom)Ground plane and pads for bottom mounted components, no
signals
4-Layer PCB B:
Layer 1 (top MCU location)Signals and poured power
Layer 2 (inner)Ground plane
Thick core
Layer 4 (bottom)Signals and poured power
6-Layer PCB A:
Layer 1 (top MCU)Power plane and pads for top mounted components, no
signals
Layer 2 (inner)Signals and ground plane
Layer 3 (inner)Power plane
Layer 6 (bottom)Ground plane and pads for bottom mounted components, no
signals
6-Layer PCB B:
20
Layer 1 (top MCU)Signals and power plane

Layer 6 (bottom)Signals and ground plane
6-Layer PCB C:
8-Layer PCB A:
Layer 1 (top MCU)Signals
Layer 3 (inner)Signals
Layer 8 (bottom)Signals
8-Layer PCB B:
8-Layer PCB C:
Layer 1 (top MCU)Signals and ground plane
Thick core
21
Other module hardware considerations

8-Layer PCB D:
Layer 1 (top MCU)Signals and ground plane
Thick core
In general, avoid placing one signal layer adjacent to another signal layer.
2.1.5 Other module hardware considerations

2.1.5.1 Debug interface
The Kinetis V series MCUs use the Cortex Debug interfaces for debugging and
programming. The 19-pin Cortex Debug interfaces provides connections for Serial Wire
debugging, as well as target power. The 9-pin Cortex Debug interfaces provides
connections for Serial Wire debugging only. Figure 2-4 shows the 20-pin header
implementation with 19 pins populated. Figure 2-5 shows the 10-pin header
implementation with 9 pins populated as used on the TWR system and Freedom boards.

22
Figure 2-4. 20-pin debug interface
Figure 2-5. 10-pin debug interface
The debug signals are multiplexed with general purpose I/O pins, therefore some signals
will require proper biasing to select the operating mode. The SWD_CLK pin has an
internal pull down device and SWD_DIO has an internal pull up device. The connectors

23
Other module hardware considerations
for this interface are keyed dual row 0.050 centered headers. When implementing either
of these headers on a target system, pin 7 must be depopulated to use the 19-pin or 9-pin
adapters from the debug tool. The Samtec part numbers for these connectors are:
FTSH-110-01-L-DV-K 20-pin keyed connector

FTSH-105-01-L-DV-K 10-pin keyed connector
FTSH-110-01-L-DV 20-pin connector, no key
FTSH-105-01-L-DV 10-pin connector, no key
This interface is useful during the development phase of a project. The header may not
need to be populated in the production phase of the project, but the PCB pads should be
kept available for future debugging purposes.

24
Chapter 3
Nested Vector Interrupt Controller (NVIC)
3.1 NVIC
3.1.1 Overview
This chapter shows how the NVIC is integrated into the Kinetis MCUs and how to
configure it and set-up module interrupts. It also demonstrates the steps to set the
interrupts for the desired peripheral and how to locate the vector table from flash to
RAM.
3.1.1.1 Introduction
The NVIC is a standard module on the ARM Cortex-M series. This module is closely
integrated with the core and provides very low latency entering and exiting an interrupt
service routine (ISR). It takes 15 cycles to exit an ISR, unless the exit from the interrupt
is into another pending ISR. In this case, the MCU tail-chains and the exit and re-entry
takes 11 cycles.
The NVIC provides four different interrupt priorities which can be used to control the
order in which interrupts must be serviced. Priorities are 0-3, with 0 receiving the highest
priority. For example, in a motor-control application, if a timer interrupt and UART occur
simultaneously, the timer interrupt that moves the motor is more critical than the UART
interrupt receiving a character. The timer priority must be set higher than the UART.

25
NVIC
3.1.1.2 Features
On Kinetis V1x series MCUs the NVIC provides up to 48 interrupt sources including 16
that are core specific. It also implements up to four priority levels that are fully
programmable. The NVIC uses a vector table to manage the interrupts. This vector table
can be stored in either flash or RAM, depending on the application.
Table 3-1. Core exceptions
Address
Vector
IRQ
Source module
Source description
ARM core
Initial stack pointer
ARM core
Initial program Counter
ARM core
Non-maskable Interrupt (NMI)
ARM core
Hard fault
11
ARM core
SVCall
12
14
ARM core
Pendable request for system service
15
ARM core
System tick timer(SysTick)
ARM Core System Handler Vectors

0x0000_0000
3.1.2 Configuration examples

The NVIC is easy to configure, as demonstrated in the following examples. The first
example shows how to configure the NVIC for a module, using the low power timer
(LPTMR) as a base. The second example shows how to locate the vector table from the
flash to RAM.
3.1.2.1 Configuring the NVIC

Configuring the NVIC for the specific module involves writing three registers: NVIC Set
Enable Register (NVIC->ISERx), NVIC Clear Pending Register (NVIC->ICPR[0]), and
NVIC Interrupt Priority (NVIC->IP[IRQn]). After the NVIC is configured and the
desired peripheral has its interrupts enabled, the NVIC serves any pending request from
that module by going to the module's ISR.
3.1.2.1.1
Code example and explanation
This example shows how to set up the NVIC for a specific module, using the LPTMR.
The steps to configure the NVIC for this module are:

26
Chapter 3 Nested Vector Interrupt Controller (NVIC)
1. Identify the vector number and the IRQ number of the module from the vector table
in the device-specific reference manual in the section Interrupt Channel
Assignments. For the LPTMR the vector is 44 and IRQ is 28
2. Identify which bit to set, perform a modulo operation dividing the IRQ number by
32. This number is used to enable the interrupt on NVIC->ISER[0] and to clear the
pending interrupts from NVIC->ICPR[0].
Example:
LPTMR BIT = 28 mod 32
LPTMR BIT = 28
3. At this point, the interrupt for the LPTMR can be configured:
NVIC->ICPR[0|=(1<<28);
NVIC->ISER[0|=(1<<28);
//Clear any pending interrupts on LPTMR

//Enable interrupts from LPTMR module
4. Next, set the interrupt priority level. This is application dependent. On Kinetis V1x
series MCUs there are four different priority levels. To set the priority, write to the
NVIC->IP[IRQn] register; the "IRQn" represents the IRQ number divided by 4. Note
the most significant nibble is used to set up the priority, the lower nibble is reserved
and reads as zero. The LPTMR example sets the priority to priority 3:
NVIC->IP[_IP_IDX(IRQn)] = (NVIC->IP[_IP_IDX(IRQn)] & ~(0xFF << _BIT_SHIFT(IRQn))) |
(((priority << (8 - __NVIC_PRIO_BITS)) & 0xFF) << _BIT_SHIFT(IRQn)); //Set Priority to
priority 3 to the LPTMR module
5. After the NVIC registers are set up, finish the peripheral configuration that must
enable the interrupt.
6. In the ISR, clear the peripheral interrupt flag and read back the status register to
avoid re-entrance. For this example:
void LPTMR_Isr (void)
{
LPTMR0->CSR|=LPTMR_CSR_TCF_MASK; //Clear LPTMR Compare flag
LPTMR0->CSR = ( LPTMR_CSR_TEN_MASK |
LPTMR_CSR_TIE_MASK |
LPTMR_CSR_TCF_MASK );
/*ISR code goes here*/
}
3.1.2.2 Relocating the vector table

Some applications need the vector table to be located in RAM. For example in an RTOS
implementation, the vector table needs to be in RAM, which allows the Kernel to install
ISRs by modifying the vector table during runtime.

27
NVIC
The NVIC provides a simple way to reallocate the vector table. The user needs to set up
the Vector Table Offset Register (VTOR) with the address offset for the new position.
If you plan to store the vector table in RAM, you must first copy the table from the flash
to RAM. Also note that in some low power modes, a portion of the RAM will not be
powered, which can lead to a vector table corruption. In this case, locate the vector table
in the flash prior to entering a low power mode.
3.1.2.2.1 Code example and explanation
The CM0+ core adds support for a programmable Vector Table Offset Register (VTOR)
to relocate the exception vector table. This device supports booting from internal flash.
The vector table is initially in flash. If the vector table is needed in RAM, move it in this
manner:
1. Copy the entire vector table from flash to RAM. The linker command file labels are
useful in this step. Refer to the following sample code:
/*Address for VECTOR_TABLE and VECTOR_RAM come from the linker file*/
extern uint32 __VECTOR_TABLE[];
extern uint32 __VECTOR_RAM[];
/* Copy the vector table to RAM */
if (__VECTOR_RAM != __VECTOR_TABLE)
{
for (n = 0; n < 0x104; n++)
__VECTOR_RAM[n] = __VECTOR_TABLE[n];
}
2. After the table has been copied, set the proper offset for the VTOR register:
/* Point the VTOR to the new copy of the vector table */
write_vtor((uint32)__VECTOR_RAM);
It is important to follow these steps in order, to ensure that there is always a valid vector
table.

28
Chapter 4
Clocking System
4.1 Clocking
4.1.1 Overview
This chapter will discuss the clocking system and the multipurpose clock generator
(MCG) module. Examples will provide an overview of how to switch between the MCG
modes and specifically how to enable the on-chip FLL for high-speed operation. Clock
selection options will be discussed for the RTC.
4.1.2 Features
An example of the clocking system is summarized in the following figure. Not all clock
sources will be available on specific devices. Refer to the individual device reference
manual for full details of the available clock sources.

29
Clocking
MCG
4 MHz IRC
SIM
IRCS
FCRDIV
MCGIRCLK
CG
32 kHz IRC
FLL
Core clock,
platform clock,
and system clock
fast peripherals
OUTDIV1
CG
OUTDIV4
CG
Bus clock/
Flash clock
OUTDIV5
CG
Alt ADC
clock
CLKS
IREFS
Clock options for

some peripherals
(see note)
MCGOUTCLK
FRDIV
EREFS0
System oscillator
EXTAL0
OSCCLK
XTAL0
OSC
logic
OSC32KCLK
MCGFLLCLK
CG
OSCERCLK
ERCLK32K
Clock options for some

peripherals (see note)
XTAL_CLK
CG Clock gate
Note: See subsequent sections for details on where these clocks are used.
Figure 4-1. Clock distribution diagram
The system level clocks are provided by the MCG. The MCG consists of:
Two individually trimmable internal reference clocks (IRC), a slow IRC with a
frequency of ~32 kHz and a fast IRC with a frequency of ~4 MHz, which can be
reduced by means of the FCRDIV divider
Frequency locked loop (FLL) using the slow IRC or an external source as the
reference clock
Auto trim machine (ATM) to allow both of the IRCs to be trimmed to a custom
frequency using an externally-generated reference clock
The clocks provided by the MCG are summarized as follows:
MCGOUTCLK The main system clock used to generate the core, bus, and memory
clocks. It can be generated from one of the on-chip reference oscillators, the on-chip
crystal/resonator oscillator, an externally generated square wave clock, or the FLL.

30
Chapter 4 Clocking System
MCGFLLCLK The output of the FLL and is available any time the FLL is enabled.
MCGIRCLK The output of the selected IRC. The selected IRC will be enabled
whenever this clock is selected.
In addition to the clocks provided by the MCG, there are three other system level clock
sources available for use by various peripheral modules:
OSCERCLK The clock provided by the system oscillator and is the output of the
oscillator or the external square wave clock source.
ERCLK32K The output of the system oscillator if it is configured in low power
mode at 32 kHz, the external RTC_CLKIN path or the low power oscillator (LPO).
LPO The output of the low power oscillator. It is an on-chip, very low power
oscillator with an output of approximately 1 kHz that is available in all run and low
power modes except VLLS0.

The MCG can be configured in one of several modes to provide a flexible means of
providing clocks to the system for a wide range of applications.
After exiting reset, or recovering from a very low leakage state, the MCG will be in FLL
engaged internal (FEI) mode with MCGCLKOUT at 20.97 MHz, assuming a factory
trimmed slow IRC frequency of 32.768 kHz. If a different MCG mode is required, the
MCG can be transitioned to that mode under software control.
It is only possible to transition directly to certain MCG modes. Refer to the individual
device reference manual for details on this. It may be required to transition through
several modes to reach the desired MCG mode. When transitioning from one clock mode
to another, you must ensure that you have fully entered that mode before moving to the
next mode. The mcg.c file within the sample code contains examples of how to perform
all the individual clock mode transitions. The fei_fee function causes the MCG mode
transition from FEI to FEE. The specific MCG register operations will be discussed
below.
In this example, the FLL is configured to use an external 10 MHz from the crystal
oscillator to generate a 75 MHz output frequency which is a typical configuration for
FEE mode.
A code example for transition from FEI to FEE follows:
int fei_fee(int crystal_val, unsigned char hgo_val, unsigned char erefs_val)
{
unsigned char frdiv_val;
unsigned char temp_reg;
// short i;
int mcg_out, fll_ref_freq, i;

31
Clocking
// check if in FEI mode
/*
if (!((((MCG -> S & MCG_S_CLKST_MASK) >> MCG_S_CLKST_SHIFT) == 0x0) && // check CLKS mux
has selcted FLL output
(MCG -> S & MCG_S_IREFST_MASK)))
// check PLLS mux
has selected FLL
{
*/
return 0x1;
// return error code
// check external frequency is less than the maximum frequency

if (crystal_val > 50000000) {return 0x21;}
//check crystal frequency is within spec. if crystal osc is being used
if (erefs_val)
{
if (
((crystal_val > 40000) && (crystal_val < 3000000)) ||
(crystal_val > 32000000)) {return 0x22;} // return error if one of the available
crystal options is not available
}
// make sure HGO will never be greater than 1. Could return an error instead if desired.
if (hgo_val > 0)
{
hgo_val = 1; // force hgo_val to 1 if > 0
}
OSC0 -> CR = OSC_CR_ERCLKEN_MASK
|OSC_CR_EREFSTEN_MASK
|OSC_CR_SC16P_MASK;
// configure the MCG -> C2 register
// the RANGE value is determined by the external frequency. Since the RANGE parameter
affects the FRDIV divide value
// it still needs to be set correctly even if the oscillator is not being used
temp_reg = MCG -> C2;
temp_reg & = ~(MCG_C2_RANGE0_MASK | MCG_C2_HGO0_MASK | MCG_C2_EREFS0_MASK); // clear
fields before writing new values
if (crystal_val 40000)
{
temp_reg |= (MCG_C2_RANGE0(0) | (hgo_val << MCG_C2_HGO0_SHIFT) | (erefs_val <<
MCG_C2_EREFS0_SHIFT));
}
else if (crystal_val 8000000)
{
}
else
{
}
MCG -> C2 = temp_reg;
// if the external oscillator is used need to wait for OSCINIT to set
if (erefs_val)
{
for (i = 0 ; i < 20000000 ; i++)
{
if (MCG -> S & MCG_S_OSCINIT0_MASK) break; // jump out early if OSCINIT sets before
loop finishes
}
if (!(MCG -> S & MCG_S_OSCINIT0_MASK)) return 0x23; // check bit is really set and
return with error if not set
}

32

// determine FRDIV based on reference clock frequency
// since the external frequency has already been checked only the maximum frequency for each
FRDIV value needs to be compared here.
if (crystal_val 1250000) {frdiv_val = 0;}
else if (crystal_val 2500000) {frdiv_val = 1;}
else {frdiv_val = 5;}
// The FLL ref clk divide value depends on FRDIV and the RANGE value
if (((MCG -> C2 & MCG_C2_RANGE0_MASK) >> MCG_C2_RANGE0_SHIFT) > 0)
{
fll_ref_freq = ((crystal_val) / (32 << frdiv_val));
}
else
{
fll_ref_freq = ((crystal_val) / (1 << frdiv_val));
}
if( crystal_val == 32768 )
{
MCG -> C4 |= MCG_C4_DMX32_MASK | MCG_C4_DRST_DRS(1);
}
//KV10 will run up to 75 MHz, core clock = 10 MHz/32*8 * 1920 = 74.9 MHz
MCG -> C4 |= MCG_C4_DRST_DRS(2);
// Check resulting FLL frequency
mcg_out = fll_freq(fll_ref_freq); // FLL reference frequency calculated from ext ref freq
and FRDIV
if (mcg_out < 0x5B) {return mcg_out;} // If error code returned, return the code to
calling function
// Select external oscilator and Reference Divider and clear IREFS to start ext osc
// If IRCLK is required it must be enabled outside of this driver, existing state will be
maintained
// CLKS=0, FRDIV=frdiv_val, IREFS=0, IRCLKEN=0, IREFSTEN=0
temp_reg = MCG -> C1;
temp_reg &= ~(MCG_C1_CLKS_MASK | MCG_C1_FRDIV_MASK | MCG_C1_IREFS_MASK); // Clear values
in these fields
temp_reg |= (MCG_C1_CLKS(0) | MCG_C1_FRDIV(frdiv_val)); // Set the required CLKS and FRDIV
values
MCG -> C1 = temp_reg;
// wait for Reference clock Status bit to clear
for (i = 0 ; i < 20000 ; i++)
{
if (!(MCG -> S & MCG_S_IREFST_MASK)) break; // jump out early if IREFST clears before
loop finishes
}
if (MCG -> S & MCG_S_IREFST_MASK) return 0x11; // check bit is really clear and return
with error if not set
// Now in FBE
// It is recommended that the clock monitor is enabled when using an external clock as the
clock source/reference.
// It is enabled here but can be removed if this is not required.
// MCG -> C6 |= MCG_C6_CME_MASK;
return mcg_out; // MCGOUT frequency equals FLL frequency
} // fei_fee

33
Clocking
4.1.4 Clocking system device hardware implementation

The main system oscillator can be configured in various ways depending on the crystal
frequency and mode being used. Refer to the device-specific reference manual for details.
When the oscillator is configured for low power, an integrated oscillator feedback resistor
is provided. The oscillator also has programmable internal load capacitors when it is
configured as a 32kHz oscillator (RANGE = 0). The internal crystal load capacitors are
selectable in software to provide up to 30 pF, in 2 pF increments, for each of the EXTAL
and XTAL pins. This provides an effective series capacitive load of up to 15 pF. The
parasitic capacitance of the PCB should also be included in the calculation of the total
crystal load. The combination of these two values will often mean that no external load
capacitors are required when using a 32 kHz crystal.
If either of the oscillator pins are not being used, they may be left unconnected in their
default reset configuration or may be used as GPIO.
4.1.5 Layout guidelines for general routing and placement

Use the following general routing and placement guidelines when laying out a new
design. These guidelines will help to minimize electromagnetic compatibility (EMC)
problems:
To minimize parasitic elements, surface mount components must be used where
possible
All components must be placed as close to the MCU as possible.
If external load capacitors are required, they must use a common ground connection
shared in the center
If the crystal, or resonator, has a ground connection, it must be connected to the
common ground of the load capacitors
Where possible:
Keep high-speed IO signals as far from the EXTAL and XTAL signals as
possible
Do not route signals under oscillator components - on same the layer or layer
below
Select the functions of pins close to EXTAL and XTAL to have minimal
switching to reduce injected noise

34
4.1.6 References
The following list of application notes associated with crystal oscillators are available on
the Freescale website at www.freescale.com. They discuss common oscillator
characteristics, potential problems, and troubleshooting guidelines:
AN1706: Microcontroller Oscillator Circuit Design Considerations

AN1783: Determining MCU Oscillator Start-Up Parameters
AN2606: Practical Considerations for Working With Low-Frequency Oscillators
AN3208: Crystal Oscillator Troubleshooting Guide

35
Clocking

36
Chapter 5
Power Management Control (PMC/SMC/LLWU/RCM)
5.1 Introduction
This chapter is a brief description of the power management features of the Kinetis V
series 32-bit MCU.
There are four modules covered in this chapter:
Power Management Controller (PMC)

System Mode Controller (SMC)
Low Leakage Wake-up Unit (LLWU)
Reset Control Module(RCM)
5.2 Using the power management controller

5.2.1 Overview
This section will demonstrate how to use the power management controller (PMC) to
protect an MCU from unexpected low VDD events. References to other protection options
will also be made.
5.2.2 Using the low voltage detection system

5.2.2.1 POR and LVD features
The POR and LVD functions allow protection of memory contents from brown out
conditions and the operation of the MCU below the specified VDD levels. As noted in
the module operation in the low power modes section, the LVD circuit is available only

37
Using the low voltage detection system
in RUN, WAIT, and STOP modes. POR circuitry is on in all modes and can be optionally
disabled in VLLS0. The user has control over whether LVD is used and whether the POR
is enabled in the lowest power mode (VLLS0). When using the low voltage detect
features, the user has full control over the LVD and LVW trip voltages. The LVW is a
warning detect circuit and the LVD is reset detect circuit.
As voltage falls below the warning level, the LVW circuit flags the warning event and
can cause an interrupt. If the voltage continues to fall, the LVD circuit flags the detect
event and can either cause a reset or an interrupt. The user can choose what action to take
in the interrupt service routine. If a detect is selected to drive reset, the LVD circuit holds
the MCU in reset until the supply voltage rises above the detect threshold.
The POR circuit for the MCU will hold the MCU in reset based upon the VDD voltage.
Before entering the VLLS0 low power mode, the user can choose to disable the POR
circuit. Because the MCU is switched off in VLLS0, the POR protection is not really
needed and can be disabled. This saves a few hundred nano amps of power while the
MCU is in this mode.
If the POR circuit is disabled in VLLS0, the MCU will continue to hold the state of the
pins until the VDD levels are much lower than the POR trip voltage levels.
Exiting VLLS0 follows the reset mode. The POR circuit is reenabled protecting the MCU
operation during the recovery.
5.2.2.2 Configuration examples

LVD and LVW initialization code is given below: Notice the comments describing the
chosen settings. You should select the parameter options for your application. The NVIC
vector flag may be set and is cleared if interrupts are enabled. The Interrupt is enabled in
the NVIC in this initialization with the call to function enable_irq(LVD_irq_no):
/***************************************************************/
/* LVD and LVD initialzation routine.
* sets up the LVD and LVW control registers
*
* This function can be used to set up the low voltage detect
* and warning. While the device is in the very low power or low
* leakage modes, the LVD system is disabled regardless of LVDSC1
* settings. To protect systems that must have LVD always on,
* configure the SMC's power mode protection register (PMPROT)
* to disallow any very low power or low leakage modes from
* being enabled.
*
* Parameters:
* lvd_select = 0x00 Low trip point selected (V LVD = V LVDL )
*
= 0x01 High trip point selected (V LVD = V LVDH )
*
= 0x10 Reserved
*
= 0x11 Reserved
* lvd_reset_enable = 0x00 LVDF does not generate hardware resets
*
= 0x10 Force an MCU reset when LVDF = 1
* lvd_int_enable = 0x00 Hardware interrupt disabled

38
Chapter 5 Power Management Control (PMC/SMC/LLWU/RCM)

*
= 0x20 Request a hardware interrupt if LVDF = 1
* lvw_select = 0x00 Low trip point selected (VLVW = VLVW1)
*
= 0x01 Mid 1 trip point selected (VLVW = VLVW2)
*
= 0x10 Mid 2 trip point selected (VLVW = VLVW3)
*
= 0x11 High trip point selected (VLVW = VLVW4)
* lvw_int_enable = 0x00 Hardware interrupt disabled
*
= 0x20 Request a hardware interrupt if LVWF = 1
*/
void LVD_Initalize(unsigned char lvd_select,
unsigned char lvd_reset_enable,
unsigned char lvd_int_enable,
unsigned char lvw_select,
unsigned char lvw_int_enable){
/*enable LVD Reset ? LVD Interrupt.select high or low LVD */
PMC -> LVDSC1 = PMC_LVDSC1_LVDACK_MASK |
(lvd_reset_enable) |
lvd_int_enable
|
PMC_LVDSC1_LVDV(lvd_select);
/* select LVW level 1,2,3 or 4 */
PMC -> LVDSC2 = PMC_LVDSC2_LVWACK_MASK |
(lvw_int_enable) |
//LVW interrupt?
PMC_LVDSC2_LVWV(lvw_select);
/* if interrupts requested
clear pending flags in NVIC and enable interrupts */
if (((PMC -> LVDSC1 & PMC_LVDSC1_LVDIE_MASK)
>>PMC_LVDSC1_LVDIE_SHIFT) |
((PMC -> LVDSC2 & PMC_LVDSC2_LVWIE_MASK)
>>PMC_LVDSC2_LVWIE_SHIFT))
{
enable_irq(LVD_irq_no); // ready for this interrupt.
}
}
5.2.2.3 Interrupt code example and explanation

The LVD circuitry can be programmed to cause an interrupt. You should create a service
routine to clear the flags and react appropriately. An example of such an interrupt service
routine is given.
To clear a warning or detect interrupt flag two things must happen:
1. The VDD voltage must return to a nominal voltage above the threshold.
2. A write to the LVDACK bit must be done to clear the LVDF indicator or a write to
the LVWACK bit must be done to clear the LVWF indicator.
If the ACK bit is written and the voltage does not go back above the threshold, the
interrupt flag will not clear and the interrupt routine will be reentered.
void pmc_lvd_isr(void)
{
if (PMC -> LVDSC1 &PMC_LVDSC1_LVDF_MASK){
printf("[LVD_isr]LV DETECT interrupt occurred");
}
if (PMC -> LVDSC2 &PMC_LVDSC2_LVWF_MASK){
printf("[LVD_isr]LV WARNING interrupt occurred");
}

39
Using the low voltage detection system

// ack to clear initial flags
PMC -> LVDSC1 |= PMC_LVDSC1_LVDACK_MASK;
PMC -> LVDSC2 |= PMC_LVDSC2_LVWACK_MASK;
}
5.2.2.4 Hardware implementation

RESET PIN: The reset pin is an open drain and has an internal pullup device. The pin is
driven out if the internal circuitry detects a reset. This is true for all resets, except when
there is a recovery from the VLLSx modes.
Although the wake-up recovery from VLLSx modes is through the reset flow, the reset
pin is not driven out of the MCU. The reason for this is that I/O is held in the pre-lowpower mode entry state so the internal reset action is blocked from being driven out.
If reset is driven low for longer than minimum pulse width to pass the analog filter and
the digital filter settings, the MCU will wake from any low-power mode and the PIN bit
in the RCM_SRS0 register will be set.
The reset pin can be disabled by clearing the RESET_PIN_CFG bit in the Flash Option
Register (FOPT).
VDD: The VDD supply pins can be driven between 1.71 V and 3.6 V DC.
The following diagram shows a representative timing diagram during POR.
Figure 5-1. Representative timing diagram during POR

40
The following diagram shows the observed behavior of the reset pin during a ramp of
VDD. In both of the following diagrams RESET asserts initially as the POR circuit is
powered up. Next RESET is released when untrimmed LVD level is reached. Next
RESET is asserted when LVD trim register is loaded, followed by RESET being released
when trimmed LVD level is reached.
Figure 5-2. Observed RESET pin behavior during normal slow VDD rise
The following diagram shows the observed behavior of the reset pin during a fast ramp of
VDD.
Figure 5-3. Observed RESET pin behavior during normal fast VDD rise

41
Using the system mode controller
5.3 Using the system mode controller

5.3.1 Overview
This section will demonstrate how to use the system mode controller (SMC). The SMC is
responsible for controlling the entry and exit from all of the run, wait, and stop modes of
the MCU. This module works in conjunction with the RCM, PMC, and the LLWU to
wake-up the MCU and move between power modes.
5.3.1.1 Introduction
There are 10 power modes and some new clocking options. These modes and options are
described below.
1. Run Default Operation of the MCU out of Reset, On-chip voltage regulator is On,
full capability.
2. Very Low Power Run (VLPR) On-chip voltage regulator is in a mode that
supplies only enough power to run the MCU in a reduced frequency. Core clock up
to 4 MHz and Bus frequency up to 1 MHz.
3. Wait ARM core enters Sleep mode, NVIC remains sensitive to interrupts,
Peripherals Continue to be clocked.
4. Stop ARM core enters DeepSleep mode, NVIC is disabled, WIC is used to wake
up from interrupt, peripheral clocks are stopped.
5. Very Low Power Wait (VLPW) ARM core enters Sleep mode, NVIC remains
sensitive to interrupts (FCLK = ON), On-chip voltage regulator is in a mode that
supplies only enough power to run the MCU at a reduced frequency.
6. Very Low Power Stop (VLPS) ARM core enters DeepSleep mode, NVIC is
disabled (FCLK = OFF), WIC is used to wake up from interrupt, peripheral clocks
are stopped, On-chip voltage regulator is in a mode that supplies only enough power
to run the MCU at a reduced frequency, all SRAMs are operating (content retained
and I/O states held).
7. Very Low Leakage Stop3 (VLLS3) ARM core enters SleepDeep mode, NVIC is
disabled, LLWU is used to wake up, peripheral clocks are stopped, all SRAMs are
operating (content retained and I/O states held), and most modules are disabled.
8. Very Low Leakage Stop 1 (VLLS1) ARM core enters SleepDeep mode, NVIC is
disabled, LLWU is used to wake up, peripheral clocks are stopped, all SRAMs are
powered down, and I/O states held. Most modules are disabled.
9. Very Low Leakage Stop 0 (VLLS0) Lowest Power Mode ARM core enters
SleepDeep mode, NVIC is disabled, LLWU is used to wake up, peripheral clocks are
42
stopped, All SRAMs are powered down, and I/O states held. Most modules are
disabled, LPO shut down, optional POR brown-out detection.
The modules available in each of the power modes are described in a table. Please see
Module operation in low-power modes for the details of the module operations in each of
the low-power modes.
5.3.1.2 Entering and exiting power modes

SMC controls entry into and exit from each of the power modes. The WFI instruction
invokes wait and stop modes for the chip. The processor exits the low-power mode via an
interrupt. For VLLSx modes, the wake-up sources are limited to LLWU generated wakeups, NMI pin, or RESET pin assertions. When the NMI pin or RESET pin have been
disabled through associated FOPT settings, then these pins are ignored as wake-up
sources. The wake-up flow from VLLSx is always through reset.
NOTE
The WFE instruction can have the side effect of entering a lowpower mode, but that is not its intended usage. See ARM
documentation for more on the WFE instruction.
On VLLSx recoveries, the I/O pins continue to be held in a static state after code
execution begins, allowing software to reconfigure the system before unlocking the I/O.
RAM is retained in VLLS3 only.

How you decide which modes to use in your solution is an exercise in matching the
requirements of your system, and selecting which modules are needed during each mode
of the operation for your application. The best way to explain would be to work through
an example.
For example, consider the case of a battery-operated human interface device that requires
an LPTMR timebase. It will wake up periodically and check the conditions of several
sensors. Then it will take action based upon the state and, when requested, perform high
levels of computation to control the operation of a device. After reviewing the power
modes table in Module operation in low-power modes, you should be able to identify
which of the modules are functioning in each of the low-power modes.

43
Using the system mode controller
At this point, notice that the LPTMR, and optionally, the brown-out detection, are the
only modules that are fully functional in all of the lowest power modules. Notice also the
modules that allow wake-up in the low-power modes such as the GPIO, LLWU, or the
comparator.
In this example system, the MCU spends most of the time in one of the lower power
modes waking up periodically to check sensor signals, plus other house-keeping tasks.
The MCU also wakes up from a user input. This can be hitting a button, a touch of a
capacitive sensor, the rise or fall of an analog signal from a sensor feeding the
comparator. To enable these sources please refer to the LLWU section 3 in the devicespecific reference manual for configuration details.
The example drivers code for SMC are available from the Freescale Web site
www.freescale.com.
Please refer to AN4503 for power management ideas and explanations as well as mode
entry and exit drivers.
5.3.2.1 SMC code example and explanation

There are four registers in the system mode controller:
PMPROT: Power Management Protection
PMCTRL: controls entry into low-power run and stop modes
STOPCTRL: provides various control bits enabling the user to fine tune power
consumption during the stop mode
PMSTAT: a read-only register that indicates the current power mode of the system
PMPROT is a write once register after a reset. This means that when written, all
subsequent writes are ignored. In our example system above, our two basic modes of
operation are Run mode and VLPS mode. If we do not want the MCU to be in any other
low-power mode, we would want to write the AVLP bit in the PMPROT register.
SMC -> PMPROT = SMC_PMPROT_AVLP_MASK;
This write allows the MCU to enter WAIT, Normal STOP or VLPS only. It is then no
longer possible to enter any other low-power mode.
After the PMPROT register has been written, the write to PMCTRL and STOPCTRL
determines the mode that will be entered. For our example, entry into VLPS mode would
be enabled with this write sequence.
SMC -> PMCTRL &= ~SMC_PMCTRL_STOPM_MASK;
SMC -> PMCTRL |= SMC_PMCTRL_STOPM(0x02);

44
5.3.2.1.1
Mode entry sequence serialization
To ensure that the correct mode is entered there are some serialization considerations.
This means that the write to PMCTRL takes 6-7 cycles to complete and if the write to
PMCTRL is done immediately before the WFI instruction is executed, see the bad code
below, then the MCU may try to enter the mode that was defined before the write was
made.
//BAD CODE//
/* Set the SLEEPDEEP bit to enable deep sleep mode (STOP) */
SCB -> SCR |= SCB_SCR_SLEEPDEEP_MASK;
/* if PMCTRL register was previously a 0x00 before the write,
the low power mode entered with the execution of next
instruction WFI may be NORMAL STOP not VLPS */
asm("WFI");
If you need to change the mode control value of STOPM right before entering the lowpower mode, then it is best to do a read back of the PMCTRL register before executing
the WFI. This insures the write has completed before the core starts the low-power mode
entry. See the updated serialized code sequence below. To make sure the read does not
get optimized out by the compiler define as volatile.
// BETTER CODE //
volatile unsigned int dummyread;
/* Set the SLEEPDEEP bit to enable deep sleep mode (STOP) */
SCB -> SCR |= SCB_SCR_SLEEPDEEP_MASK;
dummyread = SMC -> PMCTRL;
asm("WFI");
5.3.2.2 Entering wait mode

If you want to use WAIT mode, then the SLEEPDEEP bit needs to be cleared before
executing the WFI instruction.
SCB -> SCR &= ~SCB_SCR_SLEEPDEEP_MASK;
5.3.2.3 Exiting low-power modes

Each of the power modes has a specific list of exit methods. Mode exit from low power
modes WAIT, VLPW, STOP, and VLPS are initiated by an interrupt. Mode exit from
VLLSx is from the LLWU enabled wake-up source. These exit methods are discussed
later.

45
Using the low leakage wakeup unit
Recovery from VLLSx is through the wake-up reset event. The MCU will wake from
VLLSx by means of reset, an enabled pin, or an enabled module. See Table 5-3 in the
LLWU configuration section for a list of the sources. The wake-up flow from VLLS0,
VLLS1, and VLLS3 is through reset. The wake-up bit in the SRS registers is set,
indicating that the MCU is recovering from a low power mode. Code execution begins
but the I/O are held in the pre-low-power mode entry state and the oscillator is disabled
even if EREFSTEN had been set before entering VLLSx. The user is required to clear
this hold by writing to PMC_REGSC[ACKISO].
Prior to releasing the hold the user must reinitialize the I/O to the pre-low-power mode
entry state, so that unwanted transitions on the I/O do not occur when the hold is released.
5.4 Using the low leakage wakeup unit

5.4.1 Overview
This section will demonstrate how to use the Low Leakage Wake-up Unit (LLWU). The
LLWU is responsible for selecting and enabling the sources of exit from VLLS3, VLLS1,
and VLLS0 low-power modes of the MCU. This module works in conjunction with the
PMC and the MCU to wake-up the MCU.
5.4.1.1 Mode transitions

There are particular requirements for exiting form each of the power modes. Please see
Mode transition requirements for a table of the transition requirements for each of the
modes of operation.
5.4.1.2 Wake-up sources

There are a possible 16 pin sources and up to 8 modules available as sources of wake-up.
Please see Source of wake-up, pins, and modules for a table of external pin wake-up and
module wake-up sources.

46
5.4.2 LLWU configuration examples

There are 8-bit wake-up source enable registers for the pin and module source selection.
There are 8-bit wake-up flag registers to indicate which wake-up source was triggered, 8bit flag register and 8-bit filter control register to control the digital filter enable for up to
two external pins.
5.4.2.1 Enabling pins and modules in the LLWU

With Kinetis V series devices the LLWU wakeup pins are identified and numbered to be
compatible with the K series MCUs. LLWU pins range from LLWU_P3 to LLWU_P15
but are not contiguous.
5.4.2.2 Module wake-up

To configure a module to wake-up the MCU from one of the low-power modes requires a
study in the control and function of each of the modules capable of waking the MCU.
Because the LPTMR can be on in all low-power mode, we can configure the LPTMR to
wake up the system when its interrupt flag is set. To do this we need to enable the
LPTMR module to cause an interrupt and then allow that interrupt to cause a wake-up.
To enable the LPTMR to cause a wake-up the corresponding module wake-up bit must be
set.
LLWU -> ME = LLWU_ME_WUME0_MASK;
// enable the LPTMR overflow to wake up from low power modes
Other modules have to be enabled in the same way.
5.4.2.3 Pin wake-up

To configure a pin to wake-up the MCU from the low-power modes requires a study of
the port configuration register controls and the GPIO functionality.
The PCR registers select the multiplex selection, the pull enable function, and the
interrupt edge selection. If we want to initialize the first wake-up pin, PTE1, as an LLWU
wake-up enabled pin we need to:
1. Initialize the PCR for PTD6.
2. Make sure the pin is an input.
3. Enable PTD6 as a valid wake-up source in the LLWU.
47
Using the low leakage wakeup unit
The code below configures a pin as a GPIO input pin. It can be a LLWU wake-up as long
as it is selected as a digital input pin.
/* Enable Port D6 to be a digital pin. */
SIM -> SCGC5 = SIM_SCGC5_PORTD_MASK;
PORTD -> PCR6 = (PORT_PCR_ISF_MASK |
// clear Flag if there
PORT_PCR_MUX(01) |
// GPIO
PORT_PCR_IRQC(0x0A) | // falling= A Rising = 9
PORT_PCR_PE_MASK |
// Pull enable
PORT_PCR_PS_MASK);
// pull up/down enable
GPIOD -> PDDR &= 0xFFFFFFBF;
// set Port D6 as input
/* Set the LLWU pin enable bits to enable the PORTD6 input
* to be a wakeup source.
* WUPE15 in the LLWU_PE4 register is used in this case
* since it is associated with PTD6. */
LLWU -> PE4 = LLWU_PE4_WUPE15(2); //falling edge detection
This needs to be done for each of the pins you want to enable as an interrupt and low
leakage mode wake-up source.
5.4.2.4 LLWU port and module interrupts

In low-power modes the ARM core is off, the NVIC is off some of the time, and the WIC
is kept alive allowing an interrupt from the pin or module to propagate to the mode
controller to indicate a wake-up request. To enable the LLWU interrupt we would
initialize the LLWU vector in the interrupt vector table with the appropriate LLWU
interrupt handler with the following sequence:
/* example code - in the isr.h file */
#undef VECTOR_023
#define VECTOR_023 LLWU_Isr
#undef VECTOR_047
#define VECTOR_047 PortBCDE_Isr
For example, to allow the processing of the pin PTE1, add the following initialization
code:
/* example code in the interrupt vectors initialization code */
NVIC_EnableIRQ(LLWU_IRQn);
// ready for this interrupt.
NVIC_EnableIRQ(PORTBCDE_IRQn); // ready for this interrupt.
Then, there is a need for interrupt service routines for the two enabled interrupt sources,
the LLWU interrupt, and the port BCDE interrupt.
5.4.2.5 Wake-up sequence

The wake-up sequence is not obvious for some of the modes. For most of the wait and
stop modes code execution follows a predictable flow.
48
For VLLS0, VLLS1, or VLLS3, the exit is always through the reset vector and then
through the interrupt vector of the LLWU immediately after the LLWU interrupt is
enabled in the NVIC with the "NVIC_EnableIRQ(LLWU_IRQn)" function call. There is
a WAKEUP bit in the RCM_SRS0 register that allows the user to tell if the reset was due
to an LLWU wake-up event.
An example of wake-up test code is shown below:
if (RCM -> SRS0 & RCM_SRS0_WAKEUP_MASK){
printf("\nWakeup bit set from low power mode ");
if (((SMC -> PMCTRL & SMC_PMCTRL_STOPM_MASK)== 4) &&
((SMC -> STOPCTRL & SMC_STOPCTRL_VLLSM_MASK)== 0))
printf("VLLS0 exit ") ;
If the LPTMR is the wake-up source and the LLWU interrupt is enabled in sequence
before the LPTMR interrupts you must clear the source of the module interrupt or else
the code execution will never leave the LLWU interrupt service routine. An example is
given in the code snippet below:
if (LLWU -> F3 & LLWU_F3_MWUF0_MASK) {
SIM -> SCGC5 |= SIM_SCGC5_LPTMR_MASK;
LPTMR0 -> CSR |= LPTMR_CSR_TCF_MASK;
// write 1 to TCF to clear the LPT timer compare flag
LPTMR0 -> CSR = ( LPTMR_CSR_TEN_MASK
| LPTMR_CSR_TIE_MASK
| LPTMR_CSR_TCF_MASK );
// write one to clear the flag
LLWU -> F3 |= LLWU_F3_MWUF0_MASK;
}
The I/O states and the oscillator setup are held if the wake-up event is from VLLS0,
VLLS1, or VLLS3. The user is required to clear this hold by writing to the ACKISO bit
in the PMC_REGSC register. Prior to releasing the hold the user must reinitialize the I/O
to the pre-low-power mode entry state, so that unwanted transitions on the I/O do not
occur when the hold is released.
if (PMC -> REGSC & PMC_REGSC_ACKISO_MASK)
PMC -> REGSC |= PMC_REGSC_ACKISO_MASK;

49
Module operation in low-power modes
5.5 Module operation in low-power modes
Table 5-1. Module operation in low power modes
Modules
VLPR
VLPW
Stop
VLPS
VLLSx
static
static
OFF
Core modules
NVIC
FF
FF
System modules
Mode Controller
FF
FF
FF
FF
FF
static
static
static
static
FF2
low power
low power
ON
low power
low power in
VLLS3, OFF in
VLLS0/1
disabled
disabled
ON
disabled
disabled
Brown-out
Detection
ON
ON
ON
ON
ON in VLLS1/3,
optionally disabled
in VLLS03
eDMA
FF
FF
Async operation
Async operation
OFF
FF
FF
FF
static
static
static
OFF
ON
ON in VLLS1/3,
OFF in VLLS0
LLWU1
Regulator
LVD
Async operation in
CPO
Watchdog
FF
static in CPO
EWM
FF
FF in PSTOP2
Clocks
1kHz LPO
System oscillator
(OSC)
ON
ON
OSCERCLK max of OSCERCLK max of

16 MHz crystal
16 MHz crystal
ON
OSCERCLK
optional
OSCERCLK max of
limited to low
16 MHz crystal
range/low power in
VLLS1/3, OFF in
VLLS0
MCG
4 MHz IRC
4 MHz IRC
static - MCGIRCLK static - MCGIRCLK

optional
optional
OFF
Core clock
4 MHz max
OFF
OFF
OFF
OFF
Platform clock
4 MHz max
4 MHz max
OFF
OFF
OFF
System clock
4 MHz max
4 MHz max
OFF
OFF
OFF
1 MHz max
OFF
OFF
OFF
OFF in CPO
Bus clock
1 MHz max
OFF in CPO
25 MHz max in
PSTOP2 from RUN
1 MHz max in
PSTOP2 from
VLPR
Memory and memory interfaces
Table continues on the next page...

50
Table 5-1. Module operation in low power modes (continued)

Modules
Flash
VLPR
VLPW
Stop
VLPS
VLLSx
1 MHz max access

- no program
low power
low power
low power
OFF
low power
low power
low power
low power in
VLLS3, OFF in
VLLS0/1
No register access
in CPO
SRAM
low power
Communication interfaces
UART0
1 Mbit/s
1 Mbit/s
static
static
OFF
1 Mbit/s
Async operation
Async operation
OFF
static
static
OFF
static, address
match wakeup
static, address
match wakeup
OFF
Async operation in
CPO
UART1
1 Mbit/s
Async operation in
CPO
SPI0
FF in PSTOP2
master mode 500

kbit/s
master mode 500

kbit/s
slave mode 250

kbit/s
slave mode 250

kbit/s
static, slave mode

receive in CPO
I2C0
50 kbit/s
50 kbit/s
static, address
match wakeup in
CPO
FF in PSTOP2
Timers
FTM0, FTM1,
FTM2
PDB
FF
FF
static
static
OFF
FF
static
static
OFF
FF
Async operation
Async operation
Async operation4
ADC internal clock

only
OFF
HS or LS compare
LS compare in
VLLS1/3, OFF in
VLLS0
static
static, OFF in
VLLS0
static
static
Async operation in
CPO
FF
Async operation in
CPO
LPTMR
FF
FF in PSTOP2
Analog
16-bit ADC
FF
FF
ADC internal clock

only in CPO
CMP5
FF
FF in PSTOP2
FF
HS or LS compare
in CPO
6-bit DAC
FF
FF
HS or LS compare
FF in PSTOP2
FF
static in CPO
12-bit DAC
ADC internal clock

only
static
FF in PSTOP2
FF
static

51
Mode transition requirements
Table 5-1. Module operation in low power modes (continued)

Modules
VLPR
VLPW
static in CPO
Stop
VLPS
VLLSx
static output,
wakeup input
OFF, pins latched
FF in PSTOP2
Human-machine interfaces
GPIO
FF
FF
IOPORT write only

in CPO
static output,
wakeup input
FF in PSTOP2
1. Using the LLWU module, the external pins available for this chip do not require the associated peripheral function to be
enabled. It only requires the function controlling the pin (GPIO or peripheral) to be configured as an input to allow a
transition to occur to the LLWU.
2. Since LPO clock source is disabled, filters will be bypassed during VLLS0
3. The STOPCTRL[PORPO] bit in the SMC module controls this option.
4. LPO clock source is not available in VLLS0. To use system OSC in VLLS0 it must be configured for bypass (external
clock) operation. Pulse counting is available in all modes.
5. CMP in stop or VLPS supports high speed or low speed external pin-to-pin or external pin-to-DAC compares. CMP in
VLLSx only supports low speed external pin-to-pin or external pin-to-DAC compares. Windowed, sampled & filtered modes
of operation are not available while in stop, VLPS, or VLLSx modes.
5.6 Mode transition requirements

The following table defines triggers for the various state transitions:
Table 5-2. Power mode transition triggers
Transition #
From
To
RUN
WAIT
Sleep-now or sleep-on-exit modes entered with SLEEPDEEP

clear, controlled in System Control Register in ARM core.
WAIT
RUN
Interrupt or Reset
RUN
STOP

set, which is controlled in System Control Register in ARM
core.
STOP
RUN
Interrupt or Reset
RUN
VLPR
The core, system, bus and flash clock frequencies are

restricted in this mode.
Trigger conditions
Set PMPROT[AVLP]=1, PMCTRL[RUNM]=10

VLPR
RUN
Set PMCTRL[RUNM]=00 or
Reset.
VLPR
VLPW

clear, which is controlled in System Control Register in ARM
core.
VLPW
VLPR
Interrupt
VLPW
RUN
Reset.

52
Table 5-2. Power mode transition triggers (continued)

Transition #
From
To
VLPR
VLPS
Trigger conditions
PMCTRL[STOPM]=000 If PMCTRL[STOPM]=000 and
STOPCTRL[PSTOPO]=00, then VLPS mode is entered
instead of STOP. If PMCTRL[STOPM]=000 and
STOPCTRL[PSTOPO]=01 or 10, then only a Partial Stop
mode is entered instead of VLPS or 010,
core.
VLPS
VLPR
Interrupt
NOTE: If VLPS was entered directly from RUN, hardware
will not allow this transition and will force exit back to
RUN
RUN
VLPS
PMPROT[AVLP]=1, PMCTRL[STOPM]=010,
core.
VLPS
RUN
Interrupt and set PMCTRL[RUNM]=00, transitioning to RUN

before entering VLPS or
Reset.
RUN
VLLSx
VLLSx
RUN
VLPR
VLLSx
PMPROT[AVLLS]=1, PMCTRL[STOPM]=100,
STOPCTRL[VLLSM]=x (VLLSx), Sleep-now or sleep-on-exit
modes entered with SLEEPDEEP set, which is controlled in
System Control Register in ARM core.
Wake up from enabled LLWU LPTMR input source or RESET
pin
PMPROT[AVLLS]=1, PMCTRL[STOPM]=100,
STOPCTRL[VLLSM]=x (VLLSx), Sleep-now or sleep-on-exit
modes entered with SLEEPDEEP set, which is controlled in
System Control Register in ARM core.
5.7 Source of wake-up, pins, and modules
Table 5-3. Source of wake-up, pins, and modules
LLWU pin
Module source or pin name
LLWU_P3
PTA4
LLWU_P5
PTB0
LLWU_P6
PTC1
LLWU_P7
PTC3
LLWU_P8
PTC4
LLWU_P9
PTC5
LLWU_P10
PTC6
LLWU_P12
PTD0

53
Source of wake-up, pins, and modules
Table 5-3. Source of wake-up, pins, and modules (continued)

LLWU pin
Module source or pin name
LLWU_P13
PTD2
LLWU_P14
PTD4
LLWU_P15
PTD6
LLWU_M0IF
LPTMR0
LLWU_M1IF
CMP0
LLWU_M2IF
CMP1
LLWU_M3IF
Reserved
LLWU_M4IF
Reserved
LLWU_M5IF
Reserved
LLWU_M6IF
Reserved
LLWU_M7IF
Reserved

54
Chapter 6
enhanced Direct Memory Access (eDMA) Controller
6.1 eDMA
6.1.1 Overview
This chapter is a compilation of code examples and quick reference materials created to
shorten the development time of your applications that use the eDMA controller of the
Kinetis V series. Consult the device-specific reference manual for specific part
information.
This chapter demonstrates how to configure and use the eDMA controller to create data
movement between different memory and peripheral spaces without CPU intervention.
6.1.2 Introduction
The eDMA controller provides the ability to move data from one memory mapped
location to another. After it is configured and initiated, the eDMA controller operates in
parallel to the core, performing data transfers that would otherwise have been handled by
the CPU. This results in reduced CPU loading and a corresponding increase in system
performance. Figure 6-1 illustrates the functionality provided by the eDMA controller.

55
eDMA
Figure 6-1. eDMA operational overview
The eDMA controller contains a 16-bit data buffer as temporary storage. Because the
Kinetis V series is a crossbar based architecture, the CPU is the primary bus master
connected to the M0 master port. The eDMA is connected to the M2 master port of the
crossbar switch. Therefore, the CPU and eDMA can access different slave ports
simultaneously. With this multi-master architecture, the system can maximize the use of
the eDMA features. Figure 6-2 shows the basic architecture of the Kinetis V series. A
specialized device may have differencesrefer to the device-specific reference manual
for details.
GPIO
controller
Crossbar Switch
ARM core
unified bus
M0
S0
Flash
controller
M2
S1
SRAML
BME
SRAMU
S2
DMA
Master Modules
Peripheral
bridge 0
Peripherals
Slave Modules
Figure 6-2. Crossbar switch configuration

56
Chapter 6 enhanced Direct Memory Access (eDMA) Controller
The crossbar switch forms the heart of this multi-master architecture. It links each master
to the required slave device. Both fixed priority and round robin arbitration schemes are
available. If both masters attempt joint access to the same slave, an arbitration scheme
begins and eliminates bus contention.
6.2 eDMA trigger

Each channel of the Kinetis eDMA controller can be configured to begin DMA transfers
from multiple peripheral sources or software. The eDMA controller integrates the DMA
Mux to route a different trigger source to the 16 channels. With the DMA Mux, up to 63
events occurring within other peripheral modules can activate an eDMA transfer. In
many modules, event flags can be asserted as either eDMA or Interrupt requests. These
sources can be selected through DMAMUX_CHCFGn[SOURCE] registers. However,
different devices may have different peripheral source configurations. Refer to the
device-specific reference manual for details.
6.2.1 DMA multiplexer

The eDMA channel multiplexer routes the eDMA trigger source. The 52 peripheral slots
and 10 always-on slots can be routed to the 16 channels. Each channel router can be
assigned to one of the 52 possible peripheral DMA slots or to one of the 10 always-on
slots.

57
eDMA trigger
Figure 6-3. DMA MUX block diagram
6.2.2 Trigger mode

The eDMA multiplexer supports two options to trigger eDMA transfer requests.
Disabled modeNo request signal is routed to the channel and the channel is
disabled. This is the reset state of a channel in DMA MUX. Disabled mode can also
be used to suspend a DMA channel while it is reconfigured or not required.
Normal modeA DMA request is routed directly to the specified eDMA channel.

58
6.2.3 Multiple transfer requests

Only one channel can actively perform a transfer. To manage multiple pending transfer
requests, the eDMA controller offers channel prioritization. Fixed priority or round robin
priority can be selected.
In the fixed priority scheme each channel is assigned a priority level. When multiple
requests are pending, the channel with the highest priority level performs its transfer first.
By default, fixed priority arbitration is implemented with each channel being assigned a
priority level equal to its channel number. Higher priority channels can preempt lower
priority channels. Preemption occurs when a channel is performing a transfer while a
transfer request is asserted to a channel of a higher priority. The lower priority channel
halts its transfer on completion of the current read/write operation and enables the
channel of higher priority to work.
In round robin mode, the eDMA cycles through the channels from the highest to the
lowest, checking for a pending request. When a channel with a pending request is
reached, it is allowed to perform its transfer. After the transfer has been completed, the
eDMA continues to cycle through the channels looking for the next pending request.
6.3 Transfer process

Each channel requires a 32-byte transfer control descriptor (TCD) for defining the desired
data movement operation. The channel descriptors are stored in the eDMA local memory
in sequential order.
Each time a channel is activated and executes, n bytes are transferred from the source to
the destination. This is referred to as a minor transfer loop. A major transfer loop consists
of a number of minor transfer loops, and this number is specified within the TCD. As
iterations of the minor loop are completed, the current iteration (CITER) TCD field is
decremented. When the current iteration field has been exhausted, the channel has
completed a major transfer loop. This figure shows the relationship between major and
minor loops. In this example a channel is configured so that a major loop consists of three
iterations of a minor loop. The minor loop is configured as a transfer of 4 bytes.

59
Configuration steps
Source Data
Transferred
(bytes
- n) =4
DMA Request
Minor Loop
CITER
3
DMA Request
1
2
3
4
CITER
2
DMA Request
1
2
3
4
1
2
3
4
CITER
1
(channel activated)
Time
Major Loop
Figure 6-4. eDMA transfer process-major and minor loops
6.4 Configuration steps

To configure the eDMA, the following initialization steps must be followed:
1. Write the eDMA control register (only necessary when other than the default
configuration is required).
2. Configure channel priority registers in the DCHPRIn (as required).
3. Enable error interrupts using either the DMAEEI or DMASEEI register (as required).
4. Write the transfer control descriptors for the channels that will be used.
5. Configure the appropriate peripheral module and configure the eDMA MUX to route
the activation signal to the appropriate channel.
All transfer attributes for a channel are defined in the unique TCD for the channel. Each
32-bit TCD is stored in the eDMA controller module. Only the DONE, ACTIVE and
STATUS fields are initialized at reset. All other TCD fields are undefined at reset and
must be initialized by the software before the channel is activated. Failure to do this
results in unpredictable behavior. Refer to the device-specific reference manual for the
TCD detail description.
6.5 ExampleADC scan multiple channels

In this example, the eDMA is used to supply the analog-to-digital converter with a
command word and move the result of AD to a location in the internal SRAM. The AD
command word stores all the information that the AD module requires for a conversion,
so by using the eDMA to provide the command words, the module can be instructed to
perform conversions without any CPU intervention. After the result is transferred by the
eDMA to internal SRAM, the application can make further analysis on the data.
60
6.5.1 Module configuration

To implement this example two eDMA channels are required: one to transfer the
command word and the other to transfer the result. The command transfer requests a
software trigger by setting the START bit of channel 1, and an always-on trigger.
Channel 1 is configured to perform this transfer.
Channel 0 is used to transfer the AD result to RAM. This transfer is activated when the
AD result ready flag is asserted, and after transfer is complete. Link to channel0 to
transfer the next command to the ADC SC1 register, until all channels complete
conversion and BITER counter is equal to 0, generate eDMA interrupt request.
The following code provides the eDMA driver and defines the eDMA structure to
complete the configuration.
sDMAConfig.Channel = 1;
sDMAConfig.Slot = DMA_SLOTS_ALWAYS_EN; /* always on */
sDMAConfig.sSetting.bIntEn = TRUE;
sDMAConfig.sSetting.bDebugEn = TRUE;
sDMAConfig.sSetting.bRoundRobinEn = TRUE;
sDMAConfig.sSetting.bReqEn = FALSE;
sTCDInfo.SADDR = (uint32_t)&u32ADCChannel[0];
sTCDInfo.SOFF = sizeof(u32ADCChannel[0]);
sTCDInfo.ATTR = DMA_ATTR_SMOD(0)|
DMA_ATTR_SSIZE(2)|
DMA_ATTR_DSIZE(2)|
DMA_ATTR_DMOD(0)
;
sTCDInfo.NBYTES_MLNO = sizeof(u32ADCChannel[0]);
sTCDInfo.DADDR = (uint32_t)&ADC0->SC1[0];
sTCDInfo.DOFF = 0;
sTCDInfo.BITER_ELINKNO= DMA_BITER_ELINKYES_BITER(ADC_CHANNEL_NUMBERS)
sTCDInfo.CITER_ELINKNO = sTCDInfo.BITER_ELINKNO;
sTCDInfo.SLAST = -(sizeof(u32ADCChannel[0])*ADC_CHANNEL_NUMBERS);
sTCDInfo.CSR = DMA_CSR_INTMAJOR_MASK;
sDMAConfig.sTCDInfo = sTCDInfo;
/* Initialize the eDMA channel1 to transfer channel value to ADC_SC1 */
DMA_Init(DMA0,&sDMAConfig);
sDMAConfig.Channel = 0;
sDMAConfig.Slot = DMA_SLOTS_ADC0; /* ADC0 request */
sDMAConfig.sSetting.bReqEn = TRUE;
sTCDInfo1.SADDR = (uint32_t)&ADC0->R[0];
sTCDInfo1.SOFF = 0;
sTCDInfo1.ATTR = DMA_ATTR_SMOD(0)|
DMA_ATTR_SSIZE(2)|
DMA_ATTR_DSIZE(2)|
DMA_ATTR_DMOD(0)
;
sTCDInfo1.NBYTES_MLNO = sizeof(u32ADCResult[0]);
sTCDInfo1.DADDR = (uint32_t)&u32ADCResult[0];
sTCDInfo1.DOFF = sizeof(u32ADCResult[0]);
sTCDInfo1.BITER_ELINKYES = DMA_BITER_ELINKYES_BITER(ADC_CHANNEL_NUMBERS)
|DMA_BITER_ELINKYES_ELINK_MASK
|DMA_BITER_ELINKYES_LINKCH(1)
;

61
ExampleADC scan multiple channels

sTCDInfo1.CITER_ELINKYES = sTCDInfo1.BITER_ELINKYES;
sTCDInfo1.DLAST_SGA = -(sizeof(u32ADCResult[0])*ADC_CHANNEL_NUMBERS);
sTCDInfo1.CSR = DMA_CSR_INTMAJOR_MASK;
sDMAConfig.sTCDInfo = sTCDInfo1;
/* Initialize the eDMA channel0 to transfer ADC result to buffer */
DMA_Init(DMA0,&sDMAConfig);
DMA_SetCallBack(DMA_Ch0Task,0);
NVIC_EnableIRQ(DMA0_IRQn);
DMA_SetCallBack(DMA_Ch1Task,1);
NVIC_EnableIRQ(DMA1_IRQn);
/* Set eDMA channel 0 "START" bit, launch ADC conversion */
DMA_SetSTARTbit(DMA0,0x1);
Using the above configurations, the required DMA functionality for this example has
been achieved.

62
Chapter 7
Universal Asynchronous Receiver and Transmitter
(UART) Module
7.1 Overview
The UART module on the Kinetis V series devices supports asynchronous, full-duplex
serial communications with peripheral devices or other CPUs. The UART module has
four main modes of operation -- UART, IrDA, and LIN mode.
The following sections will discuss the features and use of the UART in UART mode. In
particular the use of the UART as an RS-232 serial communication port will be
described. For full details on the UART module, including all of its features and modes
of operation, please see the device-specific reference manual.
7.2 Features
The feature set available on the UARTs vary from UART to UART. Basic UART
functionality is available on all UARTs, but the clock source for the module and the
transmit and receive FIFO sizes can vary. The UART intantiation table lists the UART
features that vary based on UART module instantiation.
Table 7-1. UART instantiations on Kinetis
UART instance
LIN supported?
FIFOs
Module clock
UART0
Yes
8 entry TxFIFO, 8 entry

RxFIFO
Core clock
UART1
Yes
8 entry TxFIFO, 8 entry

RxFIFO
Peripheral clock
NOTE
This table describes the UART instantiations on the Kinetis V
series devices available at the time of publication. As new

63
Configuration example
Kinetis devices become available the UART instantiations

might change. See the Chip Configuration chapter of the
device-specific reference manual to verify the UART
instantiation information for your device.
7.3 Configuration example

The examples included here poll UART status flags to determine when receive data is
available or when transmit data can be written into the FIFO. This approach is the most
CPU intensive, but it is often the most practical approach when handling small messages.
As message sizes increase it might be useful to use interrupts or the eDMA to decrease
the CPU loading. However, the overhead required to set up the interrupts or eDMA must
be taken into account. If the additional overhead outweighs the reduction in CPU loading,
then polling is the best approach.
Using the UART interrupts to signal the CPU that data can be read from or written to the
UART will help to decrease the CPU loading. The UART has a number of status and
error interrupt flags that can be used, but for typical receive and transmit operations the
receive data register full flag (UARTx_S1[RDRF]) and transmit data register empty flag
(UARTx_S1[TDRE]) are enabled using the UARTx_C2[TIE, RIE] bits. The names of
these flags are misleading, because they do not always indicate a full or empty condition.
For UARTs that include a FIFO, the full or empty condition is determined based on the
amount of data in the FIFO compared to a programmable watermark. If both the RDRF
and TDRE interrupt requests are enabled, then the UART interrupt handler needs to read
the S1 register to determine which condition is true then read and/or write to the UART
data register (UARTx_D) to clear the flags. Because the CPU is responsible for moving
data there is CPU loading associated with an interrupt-driven software approach.
Using the eDMA to move data minimizes CPU loading even more than using the UART
interrupts. The UART's same RDRF and TDRE flags used for an interrupt driven
software approach can be re-routed to the eDMA controller instead. This is done by
setting the UARTx_C5[TDMAS, RDMAS] bits. Each of these requests are routed to a
different eDMA channel (the specific eDMA channels are selected by programming the
eDMA channel mux). One eDMA channel is responsible for handling receive traffic, so it
reads one or more bytes from the UART for each request. The second eDMA channel is
responsible for handling the transmit traffic, so it writes one or more bytes to the UART
for each request. When the entire transmit or receive eDMA movement is complete the
eDMA can interrupt the core to notify it of the completion. In this approach the CPU has
no loading associated with the actual data movement. All of the CPU loading is the result

64
Chapter 7 Universal Asynchronous Receiver and Transmitter (UART) Module
of the initial configuration of both the UART and eDMA modules and then any
processing of data that is required to prepare it for transmission or interpret it after
reception.
7.3.1 UART initialization example

The initialization code provided in this section can be used to configure the UART for 8N-1 operation (eight data bits, no parity, and one stop bit) with interrupts and hardware
flow-control disabled. The parameters passed into this function are the UART channels to
initialize (uartch), the module clock frequency for the UART in kHz (sysclk), and the
desired baud rate for communication (baud).
NOTE
The UART modules are pinned out in multiple locations, so the
initialization function below doesn't know which UART pins to
enable. The desired UART pins should be enabled before
calling this initialization function.
void UART_Init(UART_Type *pUART, UART_ConfigType *pConfig)
{
uint16_t u16Sbr, u16Brfa;
uint8_t u8Temp;
uint32_t u32SysClk = pConfig u32SysClkHz;
uint32_t u32Baud = pConfig u32Baudrate;
/* Sanity check */
ASSERT((pUART == UART0) || (pUART == UART1));
/* Enable the clock to the selected UART */
if (pUART == UART0)
{
SIM->SCGC4 |= SIM_SCGC4_UART0_MASK;
}
if (pUART == UART1)
{
SIM->SCGC4 |= SIM_SCGC4_UART1_MASK;
}
/* Ensure the transmitter and receiver are disabled when changing settings*/
pUART -> C2 & = ~(UART_C2_TE_MASK | UART_C2_RE_MASK );
/* Configure the UART for 8-bit mode, no parity */
pUART -> C1 = 0;
/* Calculate baud settings */
//u16Sbr = (((u32SysClk) >> 4) + (u32Baud >> 1))/u32Baud;
u16Sbr = (uint16)((((u32SysClk*1000) >> 4))/u32Baud);
/* Save off the current value of the UARTx_BDH except for the SBR field */
u8Temp = pUART -> BDH & ~(UART_BDH_SBR_MASK);
pUART
pUART
->
->
BDH = u8Temp | UART_BDH_SBR(((u16Sbr & 0x1F00) >> 8));

BDL = (uint8_t)(u16Sbr & UART_BDL_SBR_MASK);
/* Determine if a fractional divider is needed to get closer to the baud rate */

u16Brfa = (((u32SysClk*32000)/(u32Baud * 16)) - (u16Sbr * 32));

65
Configuration example
/* Save off the current value of the UARTx_C4 register except for the BRFA field */
u8Temp = pUART -> C4 & ~(UART_C4_BRFA(0x1F));
pUART -> C4 = u8Temp |
UART_C4_BRFA(u16Brfa);
/* Enable receiver and transmitter */

pUART -> C2 |= (UART_C2_TE_MASK | UART_C2_RE_MASK );
The initialization above can be simplified to the following steps:

1. Enable the UART pins by configuring the appropriate PORTx_PCRn registers (not
shown in the code example).
2. Enable the clock to the UART module.
3. Disable the transmitter and receiver. This step is included to ensure that the UART is
not active when it is being configured. This step is not needed if the uart_init
function is always called when the UART is already in a disabled state (the UART is
disabled after reset by default).
4. Configure the UART control registers for the desired format. For 8-N-1 operation no
UART registers need to be configured (the default register settings configure the
UART for 8-N-1 operation).
5. Calculate the baud rate dividers. This includes calculating the 13-bit whole number
baud rate divider, the SBR field stored in the UARTx_BDH and UARTx_BDL
registers, and the 5-bit fractional baud rate divider, the UARTx_C4[BRFA] field.
6. Enable the transmitter and receiver to start the UART.
7.3.2 UART receive example

The function below shows an implementation for a simple polled UART receive
function. The parameter passed into this function is the UART channel to receive a
character (uartch). The function returns the character that is received.
uint8_t UART_GetChar(UART_Type *pUART)
{
/* Sanity check */
ASSERT((pUART == UART0) || (pUART == UART1));
/* Wait until character has been received */
while (!(pUART S1 & UART_S1_RDRF_MASK));
/* Return the 8-bit data from the receiver */

return pUART D;
Because this is a polled implementation, the function will wait until a character is
received. If no character is received, then the code will remain in the while loop
indefinitely. To avoid code getting stuck when no traffic is being received, include a
function to test whether a character is present. The uart_getchar_present function can be
called prior to calling the uart_getchar function in cases when UART receive traffic is not
guaranteed or required before moving on with program execution.
66
Chapter 7 Universal Asynchronous Receiver and Transmitter (UART) Module

int UART_GetChar_Present (UART_Type *pUART)
{
return (pUART S1 & UART_S1_RDRF_MASK);
}
7.3.3 UART transmit example

This function is an implementation for a simple polled UART transmit function. The
parameters passed into this function are the UART channel that will be used to transmit
(uartch) and the character to be sent (ch).
void UART_PutChar(UART_Type *pUART, uint8_t u8Char)
{
/* Wait until space is available in the FIFO */
while (!(pUART S1 & UART_S1_TDRE_MASK));
/* Send the character */

pUART D = (uint8_t)u8Char;
7.3.4 UART configuration for interrupts or DMA requests

The examples included here poll UART status flags to determine when receive data is
available or when transmit data can be written into the FIFO. This approach is the most
CPU intensive, but it is often the most practical approach when handling small messages.
As message sizes increase it might be useful to use interrupts or the DMA to decrease the
CPU loading. However, the overhead required to set up the interrupts or DMA should be
taken into account. If the additional overhead outweighs the reduction in CPU loading,
then polling is the best approach.
Using the UART interrupts to signal the CPU that data can be read from or written to the
UART will help to decrease the CPU loading. The UART has a number of status and
error interrupt flags that can be used, but for typical receive and transmit operations the
receive data register full flag (UARTx_S1[RDRF]) and transmit data register empty flag
(UARTx_S1[TDRE]) would be enabled using the UARTx_C2[TIE, RIE] bits. The names
of these flags are a bit misleading, since they don't always indicate a full or empty
condition. For UARTs that include a FIFO, the full or empty condition is determined
based on the amount of data in the FIFO compared to a programmable watermark. If both
the RDRF and TDRE interrupt requests are enabled, then the UART interrupt handler
would need to read the S1 register to determine which condition is true then read and/or
write to the UART data register (UARTx_D) to clear the flags. Since the CPU is still
responsible for moving data there is CPU loading associated with an interrupt-driven
software approach.

67
UART RS-232 hardware implementation
Using the DMA to move data can help to decrease the CPU loading even more than using
the UART interrupts. The UART's same RDRF and TDRE flags used for an interruptdriven software approach can be re-routed to the eDMA controller instead. This is done
by setting the UARTx_C5[TDMAS, RDMAS] bits. Each of these requests would be
routed to a different eDMA channel (the specific eDMA channels would be selected by
programming the DMA channel mux). One eDMA channel would be responsible for
handling receive traffic, so it would read one or more bytes from the UART for each
request. The second eDMA channel would be responsible for handling the transmit
traffic, so it would write one or more bytes to the UART for each request. When the
entire transmit or receive DMA movement is complete the eDMA can interrupt the core
to notify it of the completion. In this approach the CPU has no loading associated with
the actual data movement. All of the CPU loading is the result of the initial configuration
of both the UART and eDMA modules and then any processing of data that is required to
prepare it for transmission or interpret it after reception.
7.4 UART RS-232 hardware implementation

This block diagram illustrates the hardware connections for an RS-232 implementation.
The diagram shows the optional hardware flow control signals, but only the RX and TX
data connections are required.
Figure 7-1. UART RS-232 hardware connections block diagram

68
Chapter 8
Memory-Mapped Divide and Square Root (MMDVSQ)
8.1 Introduction
The ARM v6M instruction set architecture (ISA) implemented on the Cortex-M0+ core
does not support integer divide instructions or square root functions. However, these
arithmetic operations are widely used in motor control applications. To maximize such
system performance and minimize the device power dissipation, the co-processor
MMDVSQ is included in Kinetis V series to support execution of the integer divide and
unsigned integer square root operations. The supported integer divide operations include
32/32 signed (SDIV) and unsigned (UDIV) calculation.
8.2 Features
The supported features of the MMDVSQ are as follows:
Lightweight implementation of 32-bit integer divide and square root arithmetic
operations
Supports 32/32 signed and unsigned divide (or remainder) calculations
Supports 32-bit unsigned square root calculations
Simple programming model includes input data and result registers plus a control/
status register
Programming model interface optimized for activation from inline code or software
library call
Fast Start configuration minimizes the memory-mapped register write
overhead
Supports two methods to determine when result is valid, including software
polling
Configurable divide-by-zero response

69
MMDVSQ in Cortex-M0+ core platform
Memory-mapped AHB coprocessor based at 0xF000_4000 address space reserved

for CPU devices
Pipelined design processes 2 bits per cycle with early termination exit for minimum
execution time
8.3 MMDVSQ in Cortex-M0+ core platform

The generic core platform diagram illustrates the location of the MMDVSQ module
within the Cortex-M0+ platform.
Cortex-M0+ Core
Fetch
Dbg
NVIC
AGU
Dec
Rn
LD/ST
SHFT
ALU
MUL
AHB Bus
IO Port
MTB Port
RGPIO
32
PRAM
RAM
Array
MTB
MCM
s1
m0
MMDVSQ
AXBS
-Lite
Alt-Master
DMA_4ch
32
m3
s2
m2
s0
BME
PBRIDGE
32
FMC
Slave
Peripherals
NVM
Array
Figure 8-1. Generic Cortex-M0+ Core Platform Block Diagram
All functionality associated with the MMDVSQ module resides in the core platforms
clock domain. The MMDVSQ is only clocked when responding to bus requests to its
programming model or is busy performing a calculation. This minimizes system power
dissipation.
8.4 MMDVSQ programming model

There are four 32-bit memory mapped registers and one status and control register:
70
Chapter 8 Memory-Mapped Divide and Square Root (MMDVSQ)
MMDVSQ_DEND (dividend)
MMDVSQ_DSOR (divisor))
MMDVSQ_RCND (radicand))
MMDVSQ_RES (result))
MMDVSQ_CSR (control/status))
These registers can only be accessed via word-sized (32-bit) accessed. At any instant in
time, the MMDVSQ can perform either a divider or square root calculation.
For divide, the specific operation to be performed according to the following table:
Table 8-1. Determining Divide Specific Operations
MMDVSQ_CSR[REM]
MMDVSQ_CSR[USGN]
Which divide operation to perform?
Signed divide
MMDVSQ_RES = quotient
(MMDVSQ_DEND/MMDVSQ_DSOR);
Unsigned divide
MMDVSQ_RES = quotient
(MMDVSQ_DEND/MMDVSQ_DSOR);
Signed divide
MMDVSQ_RES = remainder
(MMDVSQ_DEND%MMDVSQ_DSOR);
Unsigned divide
MMDVSQ_RES = remainder
(MMDVSQ_DEND%MMDVSQ_DSOR);
For overflow divide (dividend = 0x80000000, divisor = 0xFFFFFFFF), MMDVSQ

exactly follows the ARM Cortex-Mx definition and return 0x80000000 as quotient with
no indication of the overflow condition. The remainder is 0 in this case.
For divide-by-zero, the MMDVSQ_CSR[DZ] indicator will be set and the quotient or
remainder result is forced to 0. If MMDVSQ_CSR[DZE] = 1, then an attempted read of
the result register is error terminated to signal software of a divide-by-zero condition.
For square root, MMDVSQ_RES = integer (MMDVSQ_RCND); the result is limited to
16-bit value with MMDVSQ_RES[31:16]=0x0000.
There is fast start or normal start of divide operation which is controlled by
MMDVSQ_CSR[DFS]. fast start means that a write to MMDVSQ_DSOR register
starts the divide, whereas normal start means that the divide operation is started by
setting the MMDVSQ_CSR[SRT] bit. The fast start is used by default.
The following code shows how to do signed divide and get quotient in u32Result:
MMDVSQ -> CSR & = ~(MMDVSQ_CSR_REM_MASK | MMDVSQ_CSR_USGN_MASK);

71
Square root using Q notation for fractional data

MMDVSQ -> DEND = i32Numerator;
MMDVSQ -> DSOR = i32Denominator;
u32Result = MMDVSQ -> RES;
The following code shows how to do signed divide and get remainder in u32Result:
MMDVSQ -> CSR = (MMDVSQ CSR & ~(MMDVSQ_CSR_USGN_MASK)) | MMDVSQ_CSR_REM_MASK;
MMDVSQ -> DEND = i32Numerator;
MMDVSQ -> DSOR = i32Denominator;
i32Result = MMDVSQ -> RES;
The following code shows how to get the square root of u32Data in u16Result:
MMDVSQ -> RCND = u32Data;
u16Result = MMDVSQ -> RES;
8.5 Square root using Q notation for fractional data

For the unsigned fractional data using u(nsigned)Qm.n notation, it requires m+n bits (m
+n = 32) for the input radicand. The uQm.n format produces a uQ(m/2).(n/2) square root.
For example, in uQ02.30 format is 0xC90FDAA0. To obtain , MMDVSQ_RCND
= 0xC90FDAA0, and the result of its square root is 0x0000E2DF which is 1.7724304199
when interpreted as uQ01.15 format. The error of result is -0.001%.
8.6 Execution time

The MMDVSQ module includes early termination logic to finish both divide and square
root calculations as quickly as possible, based on the magnitude of the input operand.
Therefore, the execution time for the calculations is data dependent on
MMDVSQ_DEND. The execution time is defined from the register write to initiate the
calculation until the result register has been updated and available to read (i.e., time with
MMDVSQ_CSR[BUSY]=1).
For both divide and square root operation, the shortest execution time is 1 core cycle, and
the longest is 17 cycles. For example, if MMDVSQ_DEND = 0, then the execution time
is 1 core cycle; if MMDVSQ_DEND = 01xx xxxx xxxx xxxx xxxx xxxx xxxx xxxx in
binary format, then the execution time is 17 core cycles.
For more detail information on the execution time, refer to the reference manual.

72
Chapter 8 Memory-Mapped Divide and Square Root (MMDVSQ)
8.7 Tips of usage

8.7.1 Tips for minimizing the interrupt latency
A memory read o MMDVSQ_RES register while the calculation is still being performed
causes the access to be stalled via the insertion of bus wait states until the new result is
loaded into this register. A stalled bus cycle cannot be interrupted. So if system interrupt
latency is a concern, a simple wait loop code shall be added to poll
MMDVSQ_CRS[BUSY] before reading the result register. Because the polling code is
fully interruptible, the interrupt latency is minimized.
8.7.2 Tips for context save and restore in interrupt

If the last calculation was a zero divide and the divide-by-zero enable is set, then a read
of the MMDVSQ_RES register is error terminated. To avoid such error termination
during a context save, follow the steps below:
1. Read MMDVSQ_DEND, DSOR, and CSR register and save their values
2. Clear MMDVSQ_CSR[DZE] bit
3. Read MMDVSQ_RES register and save its value
When restoring the context, do not initiate another divide calculation by disabling the
fast start of divide before reloading the MMDVSQ_DEND, DSOR registers. Follow
the steps below:
1. Write 0x00000020 to MMDVSQ_CSR to disable the fast start mechanism
2. Reload MMDVSQ_DEND, DSOR, and RES register with their previous saved
values

73
Tips of usage

74
Chapter 9
Analog-to-Digital Converter (ADC)
9.1 Overview
Each sub-family of the Kinetis V series may have a different ADC module. This chapter
describes the ADC module used in KV1x sub-family. The ADC module is a SAR type
with up to 16-bit resolution. It has two independent ADCs, ADC0 and ADC1. The ADC
module supports up to 4 pairs of differential inputs and up to 24 single-ended inputs and
features a self-calibration mode, automatic compare, and hardware average. The ADCx
(x=0, 1) has two conversion result registers, ADCx_RA and ADCx_RB, and completes
4-channel samples using a trigger event.
9.2 Introduction
The ADC module can be configured to single-end mode or differential mode. The Output
is right-justified unsigned format for single-ended modes, and it is 2's complement 16-bit
sign extended for differential modes. The single mode has 16-bit, 12-bit, 10-bit and 8-bit.
The differential mode includes 16-bit, 13-bit, 11-bit and 9-bit. The MSB is symbol
placement, 1 is negative, and 0 is positive.
Each ADC has 12 external channels from ADCx_SE0 to ADC_SE11 in single-end mode,
and ADC0 can be configured 2 pairs differential mode ADC0_DP0 and ADC0_DM0
and ADC0_DP1 ADC0_DM1, and ADC1 can be configured 2 pairs differential mode
ADC1_DP1 and ADC1_DM1 and ADC1_DP2 ADC0_DM2. There is internal
temperature sensor, Bandgap, VREFH and VREFL channel, it is convenient to test
junction or ambient temperature using integral temperature sensor and it can remove
temperature drift error in reference via internal reference channel at any time.
ADC has self-calibration function, the user has to calibrate ADC at beginning of the
conversion. Calibration function can remove the offset effectively.

75
Features
ADC has hardware average function, it can be set to 4, 8,16 and 32 samples average.
Automatic compare function has less-than, greater-than or equal-to, within range, or outof-range, programmable value.
ADC has low power mode for lower noise application, and long sample selection is good
choice for low speed and high accuracy operation. ADC also can be configured to high
speed for fast conversion like PMSM motor control application.
ADC has hardware trigger and software trigger mode, ADC can be configured to
continuous conversion via only one trigger signal.
ADC supports DMA operations, the conversion flag can trigger DMA transfer when the
DMA function is enabled.
9.3 Features
16-bit resolution maximum.

Four pairs of differential and 24 single-ended external analog inputs
Single-ended 16-bit, 12-bit, 10-bit, and 8-bit modes. Differential 16-bit, 13-bit, 11bit, and 9-bit modes
Long sample, high speed and low power selection
Conversion complete/hardware average complete flag and interrupt
Asynchronous clock source for lower noise operation with option to output the clock
Hardware or software trigger function
Automatic compare with interrupt for less-than, greater-than or equal-to, within
range, or out-of-range, programmable value
Temperature sensor
Hardware average function
Selectable voltage reference: VREFH/VREFL or VDDA
Self-Calibration mode
Transfers of DMA triggers using COCO flag
9.4 ADC configuration

An initialization procedure is necessary when the user want to get the valid data from
conversion result register, the typical sequence as follows:
1. 1. Enable ADC clock gate.
2. ADC calibration, the user can find the details information of calibration on reference
manual 32.4.7 Calibration function.

76
Chapter 9 Analog-to-Digital Converter (ADC)
3. Configure ADCx_CFG1 and ADCx_CFG2. Select input clock source, divider, ADC
conversion mode, long sample time configuration. Once configure ADCx_CFG1
register, the user has to update ADCx_CFG2 register according to ADCx_CFG1
configuration.
4. Configure ADCx_SC2 and ADCx_SC3. Select trig mode, compare function and,
reference voltage and average function, etc. if compare function is enable,
ADCx_CVn register should be correct value.
5. Configure ADCx_SC1n, set up differential or interrupt control bit.
6. Write channel number in ADCx_SC1n. If software trigger is enable, it will kick off
ADC conversion.
7. Check COCO conversion complete flag.
8. Read ADCx_Rn.
The hardware trigger is from PDB module, so the user has to configure PDB module if
hardware is enable. The ADC configuration should be prior to PDB and peripheral. For
example, if the users want to use FTM channel flag to trig ADC, the user needs to
configure ADC firstly, secondly configure PDB, FTM is configured at last. It is order to
do not neglect the trigger when ADC or PDB is not ready.
9.5 ADC hardware design considerations

There are some tips for ADC sample circuit design on PCB. Analog inputs should also
have low pass filters. The challenge with analog inputs, especially for high resolution
analog-to-digital conversions, is that the filter design needs to consider the source
impedance and sample time rather than a simple cutoff frequency.
This topic cannot be discussed in detail here, but the general concept is that fast sample
times will require smaller capacitor values and source impedances than slow sample
times. Higher resolution inputs may require smaller capacitor values and source
impedances than lower resolution inputs.
In general, capacitor values can range from 10 pF for high speed conversions to 1 uF for
low speed conversions. Series resistors can range from a few hundred Ohms to 10 k.
Use the following general routing and placement guidelines when laying out a new
design. These guidelines will help to minimize signal quality problems. The ADC
validation efforts focused on providing very stable voltage reference planes and ground
planes.

77
ADC hardware design considerations
Use high quality RC components for the anti-aliasing filter. Place this RC filter as
close to the ADC input pins as possible where it can remove the most noise.
Provide very stable analog ground and voltage planes, both for analog power and
voltage references if full accuracy of the ADC is required.
Note
When building an external voltage divider on the ADC input
pin to sample high voltage, choose the voltage divider with the
lowest equivalent resistor value, because the ADC input
resistance is only 5K .

78
Chapter 10
Programmable Delay Block (PDB)
10.1 Overview
The Programmable Delay Block (PDB) is interconnected with the ADC, CMP, and
FTMs. The PDB has a pre-trigger function for ADC and provides an interval interrupt as
a 12-bit DAC hardware trigger. Individual PDB pulse-out signals are connected to each
CMP block and used for the sample window. The PDB channel 1 pre-trigger can also be
used as an FTM synchronous input.
10.2 Introduction
The PDB provides the programmable delay time for trigger input source, the PDB
interval trigger provides the hardware trigger for 12-bit DAC, and the PDB pulse out for
CMP module. The PDB has pre-trigger and back-to-back operations. When pre-trigger is
not enabled, the PDB works under bypass mode.
There are two channels in the PDB. Each channel is associated with one ADC, for
example, channel 0 is assigned to ADC0, and channel 1 is assigned to ADC1. The PDB
can be configured to pre-trigger mode and bypass mode. One PDB channel can be
configured as two pre-triggers, pre-trigger0 and pre-trigger1, when the input trigger event
is coming, the PDB starts to run, when the PDB counter (the value in PCBx_CNT)
matches the channel delay (the value in PDB_CHnDLY0 or PDB_CHnDLY1), the
channel flag will be set, and the trigger can be used as ADC hardware trigger. If the pretrigger is disabled, it works in bypass mode, the input trigger source of PDB can be
connected to ADC directly.
The PDB pre-trigger back-to-back mode configures the PDB pre-triggers as a single
chain or several chains. PDB back-to-back operation acknowledgment connections are
implemented as follows:
PDB channel 0 pre-trigger 0 acknowledgement input: ADC1SC1B_COCO

79
Features
PDB channel 0 pre-trigger1 acknowledgement input: ADC0SC1A_COCO

PDB channel 1 pre-trigger0 acknowledgement input: ADC0SC1B_COCO
DAC interval trigger can from external hardware trigger: ADC0_SC1A[COCO] flag, or
DAC interval trigger, the 16-bit interval time can be set in PDBx_DACINTn[INT]. The
interval trigger is available once the DAC interval counter matches
PDBx_DACINTn[INT] value.
10.3 Features
The programmable delay block features the following:
Input hardware trigger sources and one of software trigger source, the user can find
PDB input trigger source from chip configuration on KV10 reference manual.
KV10 has two configurable PDB channels for each ADC.
One PDB channel is associated with one ADC, one trigger output for ADC hardware
trigger and has two pre-trigger outputs for ADC trigger select per PDB channel.
16-bit delay register per pre-trigger output Trigger outputs, it can be enabled or
disabled independently, and it also support bypass of the delay registers of the pretrigger outputs, the input trigger event is served to ADC directly.
PDB can works under One-Shot or Continuous modes.
Optional back-to-back mode operation, which enables the ADC conversions
complete to trigger the next PDB channel, there are channel delay interrupt and
channel sequence error interrupt and its flag.
One 16-bit resolution interval trigger output per DAC, optional bypass of the delay
interval trigger registers, DAC also has external trigger from ADC0_SC1A[COCO].
2 channel pulse outputs for CMP0 and CMP1. Pulse-out's can be enabled or disabled
independently, and the pulse width is programmable via PDBx_POnDLY register.
DMA support.
10.4 PDB pre-trigger

Each PDB channel can produce two pre-trigger outputs, pre-trigger0 and pre-trigger1.
The PDB clock source is the bus clock. The pre-trigger delay time (bus clock cycles) is
calculated using one of the following:
Pre-trigger0 delay time:

80
Chapter 10 Programmable Delay Block (PDB)
PDBx_SC[PRESCALER] * PDBx_SC[MULT] * PDBx_CHnDLY0 + 2 bus clock

cycles. Furthermore, add one additional bus clock cycle to determine the time at which
the channel trigger output change.
Pre-trigger1 delay time:
PDBx_SC[PRESCALER] * PDBx_SC[MULT] * PDBx_CHnDLY1 + 2 bus clock
cycles. Furthermore, add one additional bus clock cycle to determine the time at which
the channel trigger output change.
The pre-trigger delay time must be set carefully. The delay time setup primarily depends
on the ADC conversion time.
Back-to-back function
The back-to-back mode function can be set in the PDBx_CHnC1[BB] bits which
configures the PDB pre-triggers as a single chain or several chains. The PDB back-toback operation acknowledgment connections signal are from the ADC COCO flag:
PDB channel 0 pre-trigger 0 acknowledgement input: ADC1SC1B_COCO

PDB channel 1 pre-trigger0 acknowledgement input: ADC0SC1B_COCO
Figure 10-1. Back-to-back mode
PDBx_CHnC1[BB] bits is the back-to-back mode control bit, when PDBx_CHnC1[BB]

bit is set to 1; the pre-trigger is produced by the COCO flag. For example,
PDBx_CH0C1[BB] =0x02, thats to say, channel 0 pre-trigger1 is set to 1, this pretrigger should be produced by ADC0_SC1A COCO flag. When PDBx_CHnC1[BB] bit is
set to 0, the corresponding pre-trigger is produced by the channel delay trigger,
CHnDLY0 or CHnDLY1.
When all the corresponding back-to-back operations are enabled, all of the pre-trigger
output is produced by ADC COCO flag, which creates a closed loop, and the trigger
input invalid. The user must begin one of the ADC conversions to produce the COCO
flag.
81
Ping-Pong operation
10.5 Ping-Pong operation

10.5.1 PDB pre-trigger sample mode
The user can implement an ADC ping-pong operation with the PDB module. The PDB
has a single pre-trigger mode and a back-to back mode.
The user can only use the pre-trigger out signal to realize a ping-pong operation. For
example, there are two channel input signals, the user can choose PDB channel 0 or
channel 1. Two pre-trigger signals are sufficient because of the PDB channel delay time,
CHnDLY0 and CHnDLY1, and also because enable of the channel pre-trigger and its
output. The ping-pong operation will be finished by PDB pre-trigger output source.
Likewise, a 4-channel input signal sample can be finished by ADC0 and ADC1 via PDB
channel 0 and channel 1. The PDB module in KV10 can produce a total of four pretrigger output signals.
Figure 10-2. 2-Channel ADC Pre-Trigger Sample
10.5.2 Back-to-back sample mode

Back-to-back mode is a useful function similar to ADC high-speed conversionthat is,
the pre-trigger output is produced by ADC acknowledgement. You must set the pretrigger delay time correctly according to the ADC conversion speed, however, back-toback mode is very compact, therefore decreasing unnecessary pre-trigger delay time.
82
Chapter 10 Programmable Delay Block (PDB)
For example, there are two channel input signals. You can choose between PDB channel
0 or channel 1. To begin, set the channel delay time, CHnDLY0, and then enable the
channel pre-trigger and its output. The ping-pong operation will be completed by the
PDB pre-trigger output source.
10.5.2.1 2-channel ADC input back-to-back sample mode

The next figure provides a sample of the 2-channel ADC back-to-back timing.
Figure 10-3. 2-channel ADC back-to-back sample
10.5.2.2 4-channel ADC input pre-trigger sample mode

A 4-channnel ADC input can be sampled by one input trigger. To begin, the input trigger
starts the PDB and the PDB channel 0 produces pre-trigger0. When CH0DLY0 matches
the PDB counter, then ADC0_SC1A begins the ADC conversion; ADC0_SC1A COCO
triggers ADC0_SC1B; ADC0_SC1B COCO triggers ADC1_SC1A; ADC1_SC1A
COCO triggers ADC1_SC1B. When ADC1_SC1B conversion ends, the new input
trigger begins the next 4-channel ADC conversion.

83
Ping-Pong operation
Figure 10-4. 4-channel ADC back-to-back sample

84
Chapter 11
Using FlexTimer (FTM) via Programmable Delay
Block (PDB) to Schedule ADC Conversion
11.1 Overview
This chapter demonstrates how to use the PDB module to schedule and perform ADC
conversions via FTM period trigger input, the ADCs works under 16-bit single-end
mode, ADC0 samples the voltage on the ADC0_SE1(PTE16), and channel
ADC0_SE5(PTE17), and ADC1 samples the voltage ADC1_SE1(PTE18) and
ADC1_SE7(PTE18) on TWR-KV10Z32 board. And print the buffered 256 set
conversion result via CDC virtual serial port of OpenSDA.
The code example shows how to:
Make a low-level driver for the ADC, PDB and FTM

Show how to calibrate the ADC before ADC conversion
Configure the ADC for averaging a single-ended voltage conversion
Schedule ADC conversions with the PDB via FTM channel interrupt
11.2 Introduction
Timing of ADC conversions relative to system events is a key to applications, such as
motor control, and metering, requiring timing of ADC conversions for the best time to get
a noise-reduced reading. When the Kinetis V MCU is acting as a controller, it will output
control changes from time to time. Scheduling ADC conversions around these changes,
which may make transient disturbances in the system, is key. Scheduling the ADC
conversions at a time after the transient effects of the last control change has been made
can enable smooth operation of control loops. The PDB allows simple scheduling of one
or both of the ADC peripherals conversions. In this example, both ADCs will be
scheduled.

85
Features
11.3 Features
The ADC features demonstrated by the PWM_PDB_ADC_Lab example code include:
Simple calibration of the ADC:
A simple driver for the ADC, which facilitates using both ADCs and their calibration
with minimal software, is included in the adc_demo example code. Prior to taking
the first measurement, during the initialization of the demo project the ADC will be
calibrated. The use of the driver of the ADC will simplify this. While the ADC can
be used prior to calibration for conversions, the calibration of the ADC enables it to
meet its specifications.
Averaging by 1, 4, 8, 16, or 32:
The ADCs ability to average up to thirty-two conversion values prior to ending the
conversion process and generating a result will be demonstrated. This feature reduces
CPU load; it also reduces the effect of a noise spike on any readings. It is a simple
arithmetic averaging of thirty-two (or less if so configured) ADC conversions. These
conversions are taken upon the PDB triggering the ADC.
The ADCs interrupts:
The interrupt feature of the ADC is also used in the example. The ADC conversion
result is read from ADCx_Rn registers in the Interrupt Service of ADC0 and ADC1.
Hardware triggering of the ADC with the FTM via PDB:
The ADC works with the FTM to trigger the ADCs conversions. Determining which
ADC trigger to convert is based on your configuration choices. In this case the ADC
will be configured to be triggered only by the FTM channel flag via PDB. The pretrigger works under back-to-back mode.
16-bit resolution:
The conversion results in this example are 16 bit unsigned.
Differential or single-ended:
Single-ended mode is illustrated in this example.

86
Chapter 11 Using FlexTimer (FTM) via Programmable Delay Block (PDB) to Schedule ADC Conversion
11.4 Configuration code

In this case the ADCs are configured with four channels in ping-pong fashion trigged by
the back-to-back mode of the PDB. The input trigger source is from the periodic FTM
channel flag. ADC inputs are not connected to anything of interest for this demo.
ADC0_SE1(PTE16) pin is written to ADC0_SC1A[ADCH] and channel
ADC0_SE5(PTE17) pin is written to ADC0_SC1B[ADCH]. ADC1_SE1(PTE18) pin is
filled in ADC1_SC1A[ADCH] and ADC1_SE7(PTE19) pin is filled in
ADC1_SC1B[ADCH].
11.4.1 FTM trigger 4-channel ADC ping-pong conversion via

back-to-back mode of PDB
There are several steps taken in the course of the execution of this demo, involving
setting up the peripherals. These steps are further detailed with code from the
PWM_PDB_ADC_Lab project and explained in the sections that follow, numbered after
the manner of the steps:
1. ADC calibration.
2. ADC initialization. Configure ADC clock source and set ADC clock to 18.75MHz,
select ADC 12-bit single mode, enable ADC interrupt, and setup ADC ISR. Write 4
channel ADC external input to ADC0 and ADC1.
3. Configure PDB module, configure PDB trigger source and write pre-trigger delay
time. PDB works under back-to-back mode, only channel 0 pre-trigger0 is produced
by CH0DLY0 register. The rest pre-triggers are produced by ADC conversion
COCO flag.
4. Configure the FTM, set FTM0 Combine mode, and set duty cycle, dead time. Enable
FTM clock. Select FTM0 Channel 0 flag as FTM0 trigger source, then enable FTM
clock source and start FTM.
5. When FTM0 channel0 flag is set, it is PDB input trigger, and PDB start to run, when
PDB channel 0 pre-trigger0 is set, it trigger ADC conversion, the COCO flag trigger
ADC0_SC1B, ADC1_SC1A and ADC1_SC1B conversion one by one.
6. When the 4-channel ADC conversion ends, the ADC will be triggered by next input
trigger of FTM0.
7. The ADC conversion result is read in ADC ISR, if 256 set ADC conversion result is
available, MCU will send them via serial port.

87
Configuration code
11.4.2 ADC configuration

The code provided here enables ADC configuration.
SIM->CLKDIV1 |= SIM_CLKDIV1_OUTDIV5EN_MASK | SIM_CLKDIV1_OUTDIV5(0x03);
/*!<out5 divider is 4, clock source is 75/4MHz*/
pADC_Config->bClocklSelect
= 0x02; /*!<clock source is alternative clock*/
pADC_Config->bConversionMode
= 0x01; /*!<12-bit single-end mode*/
pADC_Config->bHardwareTriggerEn = TRUE; /*!<enble ADC hardware trigger*/
pADC_Config->bIntEn
= TRUE; /*!<ADC0 and ADC1 interrupt enable*/
ADC_Cal(ADC0);
/*!<ADC0 calibration*/
ADC_Cal(ADC1);
/*!<ADC0 calibration*/
ADC_Init(ADC0, pADC_Config);
/*!<inital ADC0*/
ADC_Init(ADC1, pADC_Config);
/*!<inital ADC1*/
ADC_SetCallBack(0, ADC0_Task);
ADC_SetCallBack(1, ADC1_Task);
/* write ADC0 channel and ADC1 channel */
ADC0->SC1[0] = (ADC0->SC1[0]&(~0x1F))|0x01;
ADC0->SC1[1] = (ADC0->SC1[1]&(~0x1F))|0x05;
ADC1->SC1[0] = (ADC1->SC1[0]&(~0x1F))|0x01;
ADC1->SC1[1] = (ADC1->SC1[1]&(~0x1F))|0x07;
/*!<write
/*!<write
/*!<write
/*!<write
ADC0
ADC0
ADC1
ADC1
channel
channel
channel
channel
1*/
5*/
1*/
7*/
11.4.3 PDB configuration

The following code shows how to configure the PDB.
/* PDB module configuration */
pPDB_Config->bPDBEn
=
TRUE; /*!<enable PDB*/
pPDB_Config->u16ModulusValue
=
1500; /*!<set modulo value*/
pPDB_Config->u16InterruptDelay =
1500; /*!<set interrupt delay value, it is not
necessary if disable interrupt*/
pPDB_Config->u16Ch0Delay0
=
200; /*!<set channel 0 pretrigger 0 delay time*/
=
=
=
1000; /*!<set channel 1 pretrigger 1 delay
time*/
pPDB_Config->u8Ch0PretriggerEn =
0x03; /*!<enable channel 0 pretrigger */
pPDB_Config->u8Ch1PretriggerEn =
0x03; /*!<enable channel 1 pretrigger */
pPDB_Config->u8Ch0PretriggerOut =
0x03; /*!<enable channel 0 pretrigger output */
pPDB_Config->u8Ch1PretriggerOut =
0x03; /*!<enable channel 1 pretrigger output */
/*channel delay is not necessary back-to-back is enable, the pretrigger is produced by
ADC conversion ack*/
pPDB_Config->u8Ch0Back2BackEn
=
0x02; /*!<enable channel 0 back-to-back mode */
pPDB_Config->u8Ch1Back2BackEn
=
0x02; /*!<enable channel 1 back-to-back mode
*/
pPDB_Config->bTriggerSource
=
0x08; /*!<trigger source is from FTM0 */
PDB_Init(pPDB_Config);
/*!<inital PDB module */
11.4.4 FTM configuration

The following code demonstrates how to configure the FTM.
/* FTM module configuration */
FTM_PinConfig(); /*!<FTM ouput pin configuration*/
FTM_PWMInit(FTM0, FTM_PWMMODE_COMBINE, FTM_PWM_LOWTRUEPULSE); /*!<inital FTM to combine

88
Chapter 11 Using FlexTimer (FTM) via Programmable Delay Block (PDB) to Schedule ADC Conversion
mode*/
/* set MOD value */
FTM_SetModValue(FTM0, 4678); /*!<set FTM period is 16KHz, the modulo value is equal
(75000/16)-1 */
FTM_PWMDeadtimeSet(FTM0, 2, 0x13);/*!<set channln and channel n-1 dead time, 1us*/
/* set clock source and start the counter */
FTM_ClockSet(FTM0, FTM_CLOCK_SYSTEMCLOCK, FTM_CLOCK_PS_DIV1);
/* set the duty cycle, note: only fit for combine mode */
FTM_SetDutyCycleCombine(FTM0, FTM_CHANNEL_CHANNEL1, 50);
FTM0->EXTTRIG = 0x10;
/*!<trigger source is from FTM0 channel 0 trigger */
/* read ADC conversion u8Result */
while(u8CycleTimes<255);
/*!< wait for u8Result0A[i],u8Result0B[i],u8Result1A[i] and
u8Result1B[i] is full
*/
for(i=0;i<256;i++)
/*!< print adc conversion result. */
{
printf("%d, %d, %d, %d\n",u8Result0A[i],u8Result0B[i],u8Result1A[i],u8Result1B[i]);
}
printf("Please enter any character which will echo...\n");
/* echo chars received from terminal */
while(1)
{
u8Ch = UART_GetChar(TERM_PORT);
UART_PutChar(TERM_PORT, u8Ch);
}
}
11.4.5 ADC ISR

The following code shows the ADC interrupt service routine which reads ADC results
into buffers.
/*****************************************************************************//*!
*
* @brief ADC0 module task
*
* @param none
*
* @return none
*
* @ Pass/ Fail criteria: none
*****************************************************************************/
void ADC0_Task(void)
{
if((ADC0->SC1[0] & ADC_SC1_COCO_MASK)==ADC_SC1_COCO_MASK)
{
u8Result0A[u8CycleTimes] = ADC0->R[0];
PTD->PTOR = (1<<7);
}
{
u8Result0B[u8CycleTimes] = ADC0->R[1];
PTD->PTOR = (1<<6);
}
/*****************************************************************************//*!
*
* @brief ADC1 module task
*

89
Configuration code
* @param none
*
* @return none
*
* @ Pass/ Fail criteria: none
*****************************************************************************/
void ADC1_Task(void)
{
{
u8Result1A[u8CycleTimes] = ADC1->R[0];
}
{
u8Result1B[u8CycleTimes] = ADC1->R[1];
u8CycleTimes++;
}
}

90
Chapter 12
FlexTimer Module (FTM)
12.1 Overview
This chapter will demonstrate the features of the FlexTimer Module (FTM) of Kinetis V
series devices. It also presents examples of how to properly configure the module to
achieve its required operational mode. One of the examples included in this chapter
utilizes two different modes of FTM operation: input capture and PWM mode. Another
example mentions the PWM functionality of FTM working in very low power stop mode
(VLPS).
12.2 Introduction
The FTM is built on the base of TPM module well known from Freescale 8-bit
microcontrollers. The FTM extends the functionality to meet the demands of motor
control, digital lighting solutions, and power conversion, while providing low cost and
backwards compatibility with the TPM module.
12.3 Features
The features of FTM on all Kinetis V series devices can vary in the number of FTMs and
the channels included in each module. The FTM functionality can be summarized into six
basic modes of operation:
Input capture
Output compare
PWM
Deadtime insertion
Quadrature decoder
Fault control
91
Features
The timer can also operate as a free running 16-bits counter.
12.3.1 FTM clock

The FTM source clock is selectable:
System clock
Fixed frequency clock
External clock
Source clock can be the system clock, the fixed frequency clock, or an external clock.
Fixed frequency clock is an additional clock input to allow the selection of an on-chip
clock source other than the system clock. Selecting external clock connects FTM clock to
a chip level input pin therefore allowing to synchronize the FTM counter with an off chip
clock source.
The prescaler divide-by 1, 2, 4, 8, 16, 32, 64, or 128.
12.3.2 Interrupts and DMA

FTM supports three types of interrupt:
Timer overflow interrupt
Channel (n) interrupt
Fault interrupt
The channel generates a DMA transfer request according to DMA and CHnIE bits.
12.3.3 Modes of operation

If no mode is selected by MSBx bits in channel status and control register, the counter is
fully operational but there is no pin control. Therefore edge/level sensitivity selection by
ELSx bits does not play any role.
In an input capture mode, rising, falling, or both edges can be captured. Considering
some restrictions, it allows this module to measure, for example, pulse width or time
period of input signal. Captured edge values can be read from FTMx_CnV. Any writes to
this register are ignored in input capture mode. For some selected FTM channels,
comparator output can also be used for capturing.

92
Chapter 12 FlexTimer Module (FTM)
In an output compare mode, set, clear, or toggle of output on match can be achieved. This
mode also allows the generation of positive or negative pulses on the match event. This
event occurs on compare match of FTMx_CNT and FTMx_CnV.
The most useful feature of the FTM is pulse width modulation (PWM). Edge- (up
counting) and center- (up-down counting) aligned PWM are available in this timer
module via FTMx_SC[CPWMS]. Both modes of operation provide positive (high true) or
negative (low true) PWM pulse generation. The channel pulse width is a proportional part
of modulo value and is defined by FTMx_CnV. Combine mode is not available in this
module.
12.3.4 Updating MOD and CnV

Any writes to FTMx_MOD or FTMx_CnV latch the value into an equivalent register
buffer. The register is updated by its buffer value depending on clock mode selection:
When the FTM is disabled by CMOD then MOD or CnV registers are updated
immediately after they are written.
When the FTM is enabled by CMOD and selected mode is output compare the CnV
register is updated on the next counter increment immediately after CnV register was
written.
In every other case the MOD and CnV registers are updated immediately after the
FTM counter reaches the MOD value.
12.3.5 FTM period

The FTM time period is based on the period of the FTM source clock, prescaler value,
modulo value, and also depends on alignment in the case of PWM mode. Therefore, the
period of FTM can be calculated as:
Equation 1. FTM period calculation
For example, when:

The system clock = 48 MHz is selected as a source clock for counter. The CLKS[1:0]
bits in the SC register are 01
Prescaler factor is selected to divide source clock by 16 (FTMx_SC_PS = 0x4)
The FTM modulo value is set to 3000 (FTMx_MOD = 0x0BB8) and edge aligned
mode is selected

93
Configuration examples
Equation 2. FTM period calculation example
12.3.6 Additional features

The global time base feature can be used for the synchronization of all modules. In this
case, all modules use the same time base. Only one FlexTimer module is used as global
time, as detailed in the device reference manual.
FTM continues to function in debug mode. The behavior of FTM in debug mode is
configured by FTMx_CONF.BDMMODE[1:0].
12.4 Configuration examples

Two basic examples of FTM configuration will be demonstrated in this section. The first
example uses edge-aligned PWM and the second example shows input capture features of
FTM working in normal run mode.
NOTE
In these examples, TWR-KV10Z32 with PKV10Z32 is used.
Optionally, TWR-PROTO, TWR-ELEV, and TWR-SER can be
used.
12.4.1 Example Edge Aligned PWM and Input Capture Mode

This example demonstrates the basic features of the FTM, such as input capture and
edge-aligned PWM mode. In this case, FTM0 is configured to work in edge-aligned
PWM mode. This module uses two channels: channel 1 and channel 2.
Channel 1 is configured to generate positive PWM pulses. Channel 2 is configured to
generate negative PWM pulses (inverse PWM). Module FTM1 is configured to work in
input capture mode. Channel 0 is configured to capture rising edges and in contrast
channel 1 is configured to capture falling edges. The purpose of this configuration is to be
able to measure, for example, the pulse width of the pulses generated by FTM0. This can
be achieved by interconnecting channel 1 of FTM0 and both channels of FTM1, as
shown in the following figure.

94

Edge Aligned PWM
CH1
Positive pulses
CH2
Negative pulses
out
in
out
in
FTM0
Input Capture Mode

CH0
Rising edge
CH1
Falling edge
FTM1
Figure 12-1. Interconnection of FTM modules in example
Before configuring both FTM modules, SIM must configure all required clock options. In
this example, system clock is used as a source clock for FTM. Therefore, it is necessary
to set FTMx_SC.CLKS[1:0] to 1. Next, the clocks for the ports whose pins will be used
must also be enabled. In this example, pin PTC2 and PTC3 are used as channel outputs of
FTM0, and PTD6 and PTD7 are used as channel inputs of FTM1. Therefore, clocks for
PORTC must be enabled. In the end, the clock gates must be enabled for both of the
FTMs used. Follow the next few lines of code with required SIM module configuration.
/* Enable PORTC/D clocks */
SIM->SCGC5 |= SIM_SCGC5_PORTC_MASK | (SIM_SCGC5_PORTD_MASK;
/* Enable FTM0/1 clocks */
SIM->SCGC6 |= SIM_SCGC6_FTM0_MASK;
SIM->SCGC6 |= SIM_SCGC6_FTM1_MASK;
NOTE
If any additional clock settings are required in your application,
they must also be implemented in SIM configuration.
After SIM initialization, required port pins must be configured according to their use. It is
necessary to clear only interrupt status flag, select an alternative pin for the FTM channel,
and for outputs you can enable drive strength.
/* Enable the FTM0 Ch1/2 function on PTC2/3 */
PORTC->PCR[2] = PORT_PCR_MUX(0x4);
PORTC->PCR[3] = PORT_PCR_MUX(0x4);
/* Enable the FTM1 Ch0/1 function on PTD6/7*/
PORTD->PCR[6] = PORT_PCR_MUX(0x5);
PORTD->PCR[7] = PORT_PCR_MUX(0x5);
The next step is the configuration of FlexTime modules. FTM0 configuration is

demonstrated first. In this module, timer overflow interrupt will be enabled. Therefore,
NVIC must be set before module interrupt is enabled.
/* enable FTM2 interrupt in NVIC */
NVIC_EnableIRQ(FTM0_IRQn);
You can also redefine interrupt vector to your interrupt service routine. This must be
done in a different module, such as isr.h.
extern void FTM0_Isr(void);
#undef VECTOR_033
#define VECTOR_033
FTM0_Isr

95
Then FTM0 module can be initialized. FTM0_CONF should stay in its default state. For
best performance, initialize the counter, that is, write to FTM0_CNT, before writing to
the modulo register. Modulo value MOD is set to 4800 to generate a PWM signal with
10 kHz,assuming that the system clock is set to 48 MHz. Then FTM0_SC will enable
timer overflow interrupt, set edge-aligned (up counting) mode, select clock mode to allow
the counter to increment on every clock, and set the prescaler divider to 1.
/* update MOD value */
FTM_SetModValue(FTM0, 4800);
/* set clock source, start counter */
Channels configuration must be done after general FTM0 module configuration. Channel
1 of FTM0 is configured as edge-aligned PWM with high-true pulses (positive PWM
pulses). Channel 2 is configured as edge-aligned PWM with low-true pulses (negative
pulses). Channel interrupt is disabled. Channels value must also be initialized.
pFTM->CONTROLS[1].CnSC
pFTM->CONTROLS[1].CnV
pFTM->CONTROLS[2].CnSC
pFTM->CONTROLS[2].CnV
=
=
=
=
FTM_CnSC_MSB_MASK | FTM_CnSC_ELSB_MASK;
0;
FTM_CnSC_MSB_MASK | FTM_CnSC_ELSA_MASK;
0;
NOTE
Both channels of the FTM are configured in inverse PWM to
demonstrate how one leg of the three phase converter used in
motor control application can be configured. Also note that
dead time control is available in the Kinetis V series FTM.
This module generates interrupt on timer overflow. It is convenient to clear the
corresponding flag in interrupt service routine. Value register is also set.
void FTM0_Isr(void)
{
/* clear the flag */
FTM_ClrOverFlowFlag(FTM0);
/* update the channel value */

FTM_SetChannelValue(FTM0, FTM_CHANNEL_CHANNEL1, u16ChV_new);
NOTE
Setting the compare value in the timer overflow interrupt
service routine of the same FTM is not a best practice. CnV
registers are updated by their buffer value immediately after
timer overflow. Therefore, one period of FTM can be lost in
such a case. Instead, set compare value immediately before
timer overflow.

96
After FTM0 initialization, FTM1 should be initialized. As mentioned previously, this

module is configured to work in input capture mode. This module is also configured to
generate interrupt on timer overflow. It is then required to enable vector interrupt for this
module and set priority as in previous FTM configuration.
NVIC_EnableIRQ(FTM1_IRQn);
In isr.h, redefine interrupt vector.

extern void FTM1_Isr(void);
#undef VECTOR_034
#define VECTOR_034
FTM1_Isr
Because pulse width measurement uses FTM1, it is desirable to trigger start of FTM1
counting by FTP0 overflow. Therefore, FTM0_CONF sets input trigger to TMP0
overflow and enables counter start on trigger. For more information, see Figure 13-2.
Counter is set to stop on overflow and input trigger provides its restart. The rest of the
configuration is similar to FTM0, except modulo value, which can be set higher. In this
example, it was set to 11200, assuming FTM1 period is 4285 Hz.
FTM_EnableChannelInt(FTM1, 1);
Channels of FTM1 are configured to input capture mode. Channel 0 is configured to

capture rising edges and channel 1 is configured to capture falling edges. There is no
interrupt on the generated channel event.
/* configure FTM1_CH0 and FTM1_CH1 to measure the input wave pulse width */
FTM_DualEdgeCaptureInit( FTM1, FTM_CHANNELPAIR0,
FTM_INPUTCAPTURE_DUALEDGE_ONESHOT,
FTM_INPUTCAPTURE_DUALEDGE_RISINGEDGE,
FTM_INPUTCAPTURE_DUALEDGE_FALLInGEDGE );
FTM1 also generates interrupt on time overflow. In the interrupt service routine, the
timer overflow flag is cleared and pulse width is calculated from the rising and falling
edges values of channels. Channel flags must also be cleared in to be able to capture
edges in the next FTM1 period.
void FTM1_Isr(void)
{
FTM_ClrChannelFlag(FTM0, FTM_CHANNELPAIR0);
FTM_ClrChannelFlag(FTM0, FTM_CHANNELPAIR0+1);
u8IntMark = 1;
}
NOTE
When a different input signal will supply FTM1 inputs, then it
is recommended to increase modulo value to 65535 to reach the
highest possible resolution of pulse width measurement. Also
note that if the result of pulse width is negative, it should not be
taken into account. In this case, rising edge (FTM1_C0V) was
captured two periods before and falling edge (FTM1_C1V) was
97
captured only in the previous period. Therefore, the result is

incorrect.
CV1
CV2
FTM0
FTM0
TOF
PM0
FTM0
ISR
FTM0
TOF
FTM0
TOF
FTM0
ISR
FTM0
ISR
FTM0
TOF
FTM0
ISR
FTM0
TOF
FTM0
TOF
FTM0
ISR
FTM0
TOF
FTM0
ISR
FTM0
ISR
FTM0
TOF
FTM0
ISR
CH1
out
CH2
out
CSOO
FTM1
CSOO
CSOO
FE
FE
RE
CSOT RE
RE
FE
CSOT
t
FTM1
TOF
FTM1
ISR
FTM1
TOF
FTM1
TOF
FTM1
ISR
FTM1
ISR
Figure 12-2. Functional description of example (RE rising edge, FE falling edge,
CSOO counter stop on overflow, CSOT counter start on trigger)

98
Appendix A
How to Load QRUG Examples
A.1 Overview
This chapter describes how to load and run the sample code described in other sections of
the Kinetis V series Quick Reference User Guide. It describes the procedures used to
ensure your Tower system or Freedom board is connected properly, and explains how to
load the example projects.
A.2 Software configuration

For compiling, flash downloading, and debug functioning, you will need to install IAR
System's Embedded Workbench for ARM (EWARM) V6.70.3 or later, as well as any
patches and updates available at www.iar.com. EWARM supports Open SDA, the
firmware located on your Kinetis V series Tower board that enables you to flash and
debug code with only a mini-B USB cable.
The projects you will be working with can be found on the Freescale website:
www.freescale.com
Using the EWARM and installing and downloading the project files to the Kinetis V
series Tower board will enable you to use the OPEN SDA interface.
A.3 Hardware configuration

The examples can be run with the Kinetis V series microcontroller module in stand-alone
mode. Alternatively, you can put together your Tower kit for examples using serial a
channel interface without using the Open SDA serial channel.
Connect a USB cable to the mini-USB port on the Kinetis V series board. This will be:
J23 on TWR_KV10Z32
99
Terminal configuration
When you plug in the USB cable to your board, you will see LEDs on the board turn on.
A.4 Terminal configuration

The OPEN SDA feature on the Kinetis V series Tower board will create a serial port that
communicates to your computer over the USB cable, which was connected in the
previous section. This virtual serial port is connected to UART0 on the TWR-KV10Z.
To configure the terminal:
1. Open your computers device manager and look in the COM Ports and read what
COM port number is assigned to the OPEN SDA port.
2. Open the Terminal Utility.
3. Configure the terminal client to use indicated COM port, 115200 baud, 8 data bits, 1
stop bit, and no parity.
4. Then open the Serial Port to start the connection.
A.5 Download sample code

To download sample code:
1. Save locally the latest sample code for your Tower module from:
http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=TWRKV10Z32
2. Run the install and save into any directory.
3. Go to kvxx-drv-lib\build\iar\ to see all of the different projects available.
4. The next section describes running the basic low power demo example, but the same
instructions can also be used with other projects.
A.6 Running the "Blinky_green" project

To run this project:
1.
2.
3.
4.
Open IAR and go to File -> Open -> Workspace in the menu bar.
Open the Blinky_green workspace at kvxx-drv-lib\build\iar\kv10\Blinky_green\.
The workspace that opens contains a Blink_green project for the TWR-KV10Z.
There are two flash combinations available in the demo which this project supports;
FLASH_32KB and FLASH_16KB. Select FLASH_32KB for the TWR_KV10
board.
100
Appendix A How to Load QRUG Examples
5. The selected project will appear in bold font.

6. To ensure a fresh start, clean the project by right-clicking on the project name and
selecting Clean.
7. Compile the project by clicking the Make icon, or right-click on the project and
select Make.
8. In the build dialog box at the bottom, you will see any errors or warnings. If the
compilation was successful and there are no errors, you will see something like the
image below. There may be some warnings depending on the code:

101
9. Download the code to the board and start the debugger by pressing the Download
and Debug button.
10. The code will download into flash. The debugger screen will appear and pause at the
first instruction. Click the Go button to start running.
11. After you select Go, the software will print out information and indicate to input any
character which will echo, meanwhile the green LED (D1,D3,D5) begins to blink.
Running the Blinky_green project.
Please enter any character which will echo...

102
How to Reach Us:

Home Page:
freescale.com
Web Support:
freescale.com/support
Information in this document is provided solely to enable system and

software implementers to use Freescale products. There are no express
or implied copyright licenses granted hereunder to design or fabricate
any integrated circuits based on the information in this document.
Freescale reserves the right to make changes without further notice to
any products herein.
Freescale makes no warranty, representation, or guarantee regarding
the suitability of its products for any particular purpose, nor does
Freescale assume any liability arising out of the application or use of
any product or circuit, and specifically disclaims any and all liability,
including without limitation consequential or incidental damages.
Typical parameters that may be provided in Freescale data sheets
and/or specifications can and do vary in different applications, and
actual performance may vary over time. All operating parameters,
including typicals, must be validated for each customer application by
customer's technical experts. Freescale does not convey any license
under its patent rights nor the rights of others. Freescale sells products
pursuant to standard terms and conditions of sale, which can be found
at the following address: freescale.com/SalesTermsandConditions.
Freescale, the Freescale logo, and Kinetis are trademarks of Freescale
Semiconductor, Inc., Reg. U.S. Pat. & Tm. Off. Tower is a trademark of
Freescale Semiconductor, Inc. ARM, ARM Powered, and Cortex are
registered trademarks of ARM Limited (or its subsidiaries) in the EU
and/or elsewhere. All rights reserved. All other product or service names
are the property of their respective owners.
2014 Freescale Semiconductor, Inc. All rights reserved.
Document Number KVQRUG

Revision 0, 02/2014

KinetisV - Peripheral Module Reference

Uploaded by

Copyright:

Available Formats

KinetisV - Peripheral Module Reference

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

KinetisV - Peripheral Module Reference

Uploaded by

Copyright:

Available Formats

Kinetis V Series Peripheral Module

Document Number: KVQRUG

Kinetis V Series Peripheral Module Quick Reference, Rev. 0, 02/2014

Freescale Semiconductor, Inc.

Reset and booting...............................................................................................................................................9

Device state during reset....................................................................................................................10

Device state after reset.......................................................................................................................10

Typical system initialization .............................................................................................................................10

Lowest level assembly routines.........................................................................................................10

PCB routing considerations...............................................................................................................................14

Power supply routing.........................................................................................................................14

Freescale Semiconductor, Inc.

Power supply decoupling and filtering..............................................................................................15

PCB layer stack-up.............................................................................................................................................19

Other module hardware considerations..............................................................................................................22

Configuring the NVIC.......................................................................................................................26

Relocating the vector table.................................................................................................................27

Clocking system device hardware implementation...........................................................................................33

Layout guidelines for general routing and placement........................................................................................34

Kinetis V Series Peripheral Module Quick Reference, Rev. 0, 02/2014

Freescale Semiconductor, Inc.

Using the power management controller........................................................................................................................37

Using the low voltage detection system.............................................................................................................37

POR and LVD features......................................................................................................................37

Interrupt code example and explanation............................................................................................39

Using the system mode controller...................................................................................................................................42

Entering and exiting power modes.....................................................................................................43

SMC code example and explanation..................................................................................................44

Entering wait mode............................................................................................................................45

Exiting low-power modes..................................................................................................................45

Using the low leakage wakeup unit................................................................................................................................46

Mode transitions ................................................................................................................................46

Wake-up sources ...............................................................................................................................46

LLWU configuration examples.........................................................................................................................47

Enabling pins and modules in the LLWU..........................................................................................47

LLWU port and module interrupts.....................................................................................................48

Freescale Semiconductor, Inc.

Module operation in low-power modes..........................................................................................................................50

Mode transition requirements.........................................................................................................................................52

Source of wake-up, pins, and modules...........................................................................................................................53

Multiple transfer requests...................................................................................................................................58

Configuration steps ........................................................................................................................................................60

ExampleADC scan multiple channels........................................................................................................................60

UART initialization example.............................................................................................................................65

UART receive example......................................................................................................................................66

UART transmit example....................................................................................................................................67

UART configuration for interrupts or DMA requests.......................................................................................67

UART RS-232 hardware implementation......................................................................................................................68

Freescale Semiconductor, Inc.

MMDVSQ in Cortex-M0+ core platform.......................................................................................................................70

MMDVSQ programming model.....................................................................................................................................70

Square root using Q notation for fractional data.............................................................................................................72

Tips for minimizing the interrupt latency..........................................................................................................73

Tips for context save and restore in interrupt....................................................................................................73

ADC hardware design considerations.............................................................................................................................77

Kinetis V Series Peripheral Module Quick Reference, Rev. 0, 02/2014

Freescale Semiconductor, Inc.

1.1.2 Code execution

1.1.3 Reset and booting