Sanity check PCB layout for stereo VU meter (using Audio Guru's pre-stage)

Status
Not open for further replies.

danrogers

Member
Hi all, its been a long time since I decided to make some VU meters. I remember Audio Guru suggesting to use his pre-processing circuit to smooth the peaks etc before running into the LM3915 IC.

I have just finished a basic PCB design, incorporating the pre-processing stage suggested and the LM3915 circuit. I have also used a 12v linear regulator to power the LM3915's and a 9v regulator to power the op amp and the pre-stage.

I'd like any suggestions on how to improve the circuit and to pick up if I have made any silly errors! The circuits are mirrored, I have used the component numbering from the pre-stage schematics and added the letter 'B' for the second channel.

I will attach the board layout obviously and the two schematics I have used.

Thanks in advance
 

Attachments

  • VU meters.png
    0 bytes · Views: 19
  • peak detector. (1).PNG
    16.2 KB · Views: 828
  • lm3915 circuit.jpg
    0 bytes · Views: 20
Last edited:
Several things:
1-It should not be necessary to have seperate V regulators. The opamp (part# not given) should be OK with 12 V
2-Common and + traces could be wider
3-I did not see level set controls
I did not check the layout for functionality. E
 
Hi, thanks for the reply.

1- the Op amp is listed on the schematic for the pre-stage. Its a MC33171. I did look at the datasheet and it should be fine as you say but I notice the 12v is also feeding the transistor that switches too the input pin of the LM3915 so I'm not sure how much 12v would effect it?
2- Thanks I will address that
3- I just copied the schematic in the LM3915 datasheet, but also thought about how to set the level, I guess I would use a pot in place of the resistor divider that feeds the REF ADJUST input?

I have just noticed a flaw in that I have used through hole and SMD parts so the regulator will be reserved on the underside of the board...DOH

Thanks
 
danrogers said:
Hi all, its been a long time since I decided to make some VU meters. I remember Audio Guru suggesting to use his pre-processing circuit to smooth the peaks etc before running into the LM3915 IC.
Click to expand...
No.
The circuit is a "peak detector". Your vision sees normal brightness when a light is turned on for 30ms or longer. Shorter durations are seen as dimmed. The peak detector holds the peak voltage for a moment that is 30ms in duration so you can see it brightly. Then it decays fairly quickly when the signal is gone.

Simple VU meters without a peak detector have the lower voltage LEDs lighted almost all the time but the higher voltage LEDs can barely be seen because they are very dim.

I used the MC3317x opamps because their slew rate allows a max frequency of 35kHz (an LM324 or LM358 is similar but has a low slew rate which allows a max frequency of only 2kHz).

I have just finished a basic PCB design, incorporating the pre-processing stage suggested and the LM3915 circuit. I have also used a 12v linear regulator to power the LM3915's and a 9v regulator to power the op amp and the pre-stage.
Click to expand...
I agree that the LM3915 does not need a voltage regulator since it already has a regulated reference voltage.
Most opamp circuits including mine do not need a voltage regulator.
 
Last edited:
Hi Guru.

Thanks for explaining the purpose. How do you adjust the gain?

What voltage do you suggest running the LM3915 at? Is there a way to run your peak detector circuit and the LM3915 from the same source?

Thanks
 
danrogers said:
How do you adjust the gain?
Click to expand...
The gain can be adjusted by the signal source. The peak detector is fed from a preamp opamp that also can have variable gain or have an input volume control.

In my Sound Level Indicator project I have a mic preamp (gain is 101) driving the peak detector (gain is 1.8) and have an automatic gain adjustment that reduces the gain 10 times when loud sounds are detected. It shows a pin dropped on the floor in the next room or shows very loud sounds from the stereo or TV.

What voltage do you suggest running the LM3915 at? Is there a way to run your peak detector circuit and the LM3915 from the same source?
Click to expand...
The LM3915 gets too hot when the LEDs have a fairly high supply voltage, when they are bright and when all LEDs are lighted in the BAR mode. The datasheet shows that the LED supply voltage can be reduced or be fed through a power resistor that shares the heat with the LM3915 which is what my circuit does.

My project is powered by a "9V" (actually 8.4V) Ni-MH battery that is always trickle charged to 10V. The electret mic and bias for the preamp is powered from a 5V low-dropout voltage regulator IC that also provides a DC reference voltage to the automatic gain circuit.
Each output of the LM3915 has two 1.8V red LEDs in series and their current is set to about 26mA. I connected a 10 ohm/1W resistor in series with all the LEDs so it dissipates 0.68W when all LEDs are lighted.
 

Attachments

  • Sound Level Indicator 002.jpg
    47.1 KB · Views: 331
  • Sound Level Indicator 004.jpg
    28.4 KB · Views: 457
  • Sound Level Indicator schematic rev1.PNG
    0 bytes · Views: 12
Thanks Guru. So it should be OK to run the LM3915 from a 9V source then?

What would be the preferred method to adjust the gain then? I'm not sure how it auto adjusts, is that the job of the transistor? I'm unsure how to adjust the gain of the Op-amp.

Just to check, the audio line signal goes into C4?

Many thanks
 
 
thanks Guru, that really helps to understand a bit more about it.

Just been searching through more LM3915 posts on here. I have found the circuit attached that I think you posted somewhere. I looks quite a bit more simple, would you mind explaining the differences between it and the circuit you have been helping me with?

 
Last edited:
danrogers said:
I have found the circuit attached that I think you posted somewhere. I looks quite a bit more simple, would you mind explaining the differences between it and the circuit you have been helping me with?
Click to expand...
The transistor charges the timing capacitor 10 times to 20 times faster so you see every short duration sound displayed on the LEDs. The diode circuit does not display short duration sounds.
 
Status
Not open for further replies.

Similar threads

T
Could use checking of my schematic for Teensy Audio Shield pcb
Replies
17
Views
3K
TonyAme
T
T
Need help checking over schematic for Arduino Nano Audio Sequencer PCB
Replies
5
Views
2K
TonyAme
T
F
how should you design milliohm meter using dvm display
2 3 4
Replies
73
Views
12K
Flashdrive34
F
T
  • Question
Potentiometer board schematic check for boo-boos
Replies
6
Views
2K
TonyAme
T
B
  • Question
Schematic PCB review - Capacitive Touch Sensor AT42QT1012
Replies
4
Views
2K
Tony Stewart
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…