LED Audio VU Meter
LED Audio VU Meter
LED Audio VU Meter
Share |
Please Note: PCBs are available for this project. Click the image for details.
Introduction
It is quite true that there are many variations of this circuit already on the Net, but for
the sake of completeness - and because there are PCBs for this version - here is yet
another.
The LED meter is simpler and smaller than its analogue counterpart, and is very
common in audio equipment. This version is based on a National Semiconductor IC,
and uses the logarithmic version. Each LED operates with a 3dB difference from the
previous one, and a jumper is provided to allow dot or bar mode.
Thanks to Uwe
Beis for the
meter display
This project is also an essential part of the expandable analyser to be published
soon (or perhaps "eventually"), and one meter circuit is used for each frequency
band. There are many other uses for a simple LED VU meter. They are ideal as power
meters on ampliers, can be used with mixers (including the high quality mixer
described in the project pages), preamps and any other application where it is
important to know the signal level.
The circuit is completely conventional, and is based on the application notes from National
Semiconductor. The circuit is shown in Figure 1 and as you can see it uses a single IC and a few
discrete components. DC to the LEDs is almost unltered - C1 is included to make sure the IC does
not oscillate, and is not a lter cap. This allows a higher LED current with lower dissipation than
would be the case if the DC were fully smoothed, and full smoothing would also require a much
larger capacitor. This increases the size and cost of the project - especially important if it is to be
used in larger numbers as may be the case with a mixer or analyser.
L1 to L7 will normally be green (normal operating range) and L8 to L10 should be red (indicating
overload). This gives a 9dB overload margin when the unit is calibrated as described below. As
shown, full scale sensitivity (with VR1 at maximum) is 4 Volts peak (approximately 2.8 volts RMS).
This is designed for direct connection to high-level preamps or low power speaker output of an
amplier. Sensitivity is easily changed.
JP1 determines dot or bar mode. With the jumper installed, the unit operates in bar mode, meaning
that LEDs will light in a continuous bar. If the jumper is omitted, then only the LED corresponding to
the current signal level will light. Dot mode uses far less current, but the display is not as visible.
Power comes from a 15V transformer (connected to AC1-AC2). You can generally use the smallest
one available, as average power is quite low. The peak current is about 120mA DC, so a 5VA
transformer will be sufcient to power two meter circuits. One 15V output goes to the terminal
AC1, the other to AC2. The 10 ohm resistor isolates the earth connection to help prevent hum if the
same transformer is used to power a preamp (for example).
Figure 2 - Power Supply Cir cuit (Single 15V AC Winding)
The power supply is very simple, and can easily be hard-wired. A 15-0-15V transformer can also be
used, so the circuit can use the same transformer as a preamp (for example). The supply voltage
must not exceed 25V DC or peak.
If you wish to use a centre-tapped transformer, use the circuit shown in Figure 3. Performance is
identical to that of Figure 2 for all intents and purposes.
Note: The total supply v oltage must be gr eater than the r eference voltage, but the
circuit will work per fectly with supply v oltage down t o 5V for a r eference voltage of 4V
or less. If a low v oltage supply is used, RDC and DC+ ma y simply be joined together. The
use of raw DC is only a r equirement wit h supply voltages abo ve around 12V t o keep the
dissipation of the LM3915 within r atings.
Battery operation is possible, but be wa rned that a 9V batter y won't last very long. Using
the circuit in 'dot' more will prolong batter y life because of lower average current.
The formula for sensitivity is somewhat complex, and is further complicated by the fact that the
same resistors that change the reference voltage also affect the LED current. As shown, LED
current is about 12mA. To save you the (very) tedious calculations, I have prepared a table to use to
set the reference voltage (the reference voltage sets the signal level for the 'all LEDs on' condition).
This always needs to be slightly lower than the voltage to be measured, so that ne adjustments
can be made with VR1. LED current is xed at about 10-13mA for all voltages.
Ref. Voltage R3 (k) R4 (k) I led (mA)
12 (11.6) 2.2 15 12
10 (9.99) 2.7 15 10.2
8 (8.13) 2.2 10 10.4
6 (5.81) 1.8 5.6 10.5
4 (3.81) 1.2 2.2 12.9
2 (2.20) 1.2 0.82 11.9
Table 1 - Resist or Values For Different Voltages
Now, if the above looks too irksome, or fails to meet your needs, you can download a little
calculator that will do exactly what you want, and can even check what values you will get from
"real world" resistor values. Click here to download LM3915.zip (12,583 bytes), and extract the les
into the directory (folder) of your choice. (Note, the program needs the Visual Basic 4 runtime
libraries.)
The circuit only senses the positive signal (i.e. it is half-wave only). In most cases this is not a
problem, because although audio waveforms are asymmetrical, the overall signal usually balances
out over a period of time. If this is not desirable, a simple rectier circuit using a dual opamp (a
cheap one is quite OK) is shown in Figure 2, and can be added between the signal source and the
input. This is not a precision rectier, so it will introduce a small error into the signal, causing the
sensitivity of low level signals to be reduced. The lowest couple of LEDs will therefore not be
exactly 3dB apart, but for monitoring purposes this error can be completely ignored.
If this is to be used, substitute a xed 100k resistor for VR1 (from Pin 5 to ground) in Figure 1, and
bring the signal into the IC via R1 as shown by the dashed line. VR1 in the signal rectier will be
used to change the gain rather than the meter circuit. R3 and R4 should use the values shown in
Figure 1 for best accuracy.
The signal rectier needs a supply of 15 volts, and the audio signal is fed directly into the 'Audio'
input of the meter circuit. I suggest that the signal level to the rectier be reasonably high (or use
the 'Set Gain' control to increase the gain of the rst stage). This will minimise the errors from the
less-than-perfect rectier. The reason for not using a precision rectier circuit is simply one of gain
- a standard precision rectier circuit doesn't have any gain so you lose the ability to monitor low-
level signals. The speed of the circuit can be adjusted by varying the value of C3. With a high value
(say 10uF), the meter will act more like a peak programme meter, holding the highest peaks for a
relatively long time. The lower the value, the more quickly the meter will respond.
Note - the input t o this circuit must be l ess than 10V RMS at all times. Higher le vels will
be clamped b y the protection diodes (D1, D2), but these cannot be r elied upon for
continuous pr otection against high le vel input signals. Ex cessive levels will destr oy the
opamp's input circuit. For higher v oltage an input attenuator must be used, and an
external limiting r esistor (10k) in series with the input is r ecommended.
The gain of this circuit (as shown) is limited to a maximum of 11. At higher gain values, cheap
opamps (such as the 1458) will be unable to amplify the highest frequencies due to their
bandwidth limitations. This means that the lowest level signal you can have for a full scale reading
will be about 1.3V peak, or about 900mV RMS. The maximum gain I would recommend is obtained
using a 4.7k resistor for R3. This will give a gain of about 22, at which point the response will barely
make it to 20kHz. This equates to a maximum signal sensitivity of a little under 400mV RMS. It is
unlikely that this will ever be needed in practice, as 400mV is far too low to operate any preamp or
mixer and retain respectable noise performance.
Calibration
You have (of course) selected the resistors R3 and R4 to give a reference voltage slightly lower
than the peak voltage to be measured. Now the meter can be calibrated to suit your application.
This could not be simpler. At the maximum level you wish to operate the equipment (as shown on
an audio millivoltmeter or oscilloscope with signal applied), adjust VR1 so that the signal
illuminates all the green LEDs (L1 is the most sensitive, and L10 indicates maximum level, so L1 to
L8 should be lit). If the input is directly from a speaker output, an additional series resistor should
be used at the 'Aud' input terminal to reduce the level. This can be determined by calculation (I
leave this to you) or by experiment. As a guide, for a 50W amplier, the external resistance should
be about 47k ohms.
If you are using the external signal rectier, VR1 should have been omitted from the circuit as
described above. Apply the signal voltage to the input of the signal rectier at the maximum
permitted level. Adjust VR1 (on the rectier) to illuminate LEDs L1 to L8.
If you are calibrating the meter for a power amplier, set the output to a level just below clipping.
Adjust the level control until all LEDs are illuminated. This way, if the last LED (L10) lights when you
are listening to music, you will know that you are very close to clipping, and the volume should be
reduced.
Projects Index
Main Index