Bipolar CV and signal level/peak LED circuits

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, luketeaford, Joe.

Post Reply
rutgerv
Common Wiggler
Posts: 69
Joined: Mon Dec 06, 2010 7:25 pm
Location: Elst, Netherlands
Contact:

Bipolar CV and signal level/peak LED circuits

Post by rutgerv » Thu Feb 11, 2016 9:32 am

Hi everyone,

I always love those threads where small circuit ideas are shared. In software you would call them "code snippets", so I thought I'd share a few myself.

I was inspired by this lovely thread on LED drivers: http://electro-music.com/forum/topic-49449.html.

Image

The bipolar CV viewer is a simple circuit. It takes a -5 to +5V input signal and visualizes the positive half with one color LED, and the negative half with a different color LED. You can either use two LEDs or a single bicolor LED. If you use superbright LEDs you may want to reduce the current through the LEDs by taking something larger for the 1K resistor. The 3V3 zener diodes are there to overcome the typical dead-spot of many LED circuits (where current flows, but not enough to light the LED). At the same time it makes sure both LEDs are off when no signal is present (it maintains a very narrow 0 point). The graph below shows currents through both LEDs as a function of input signal.

Image

Image

The Level/Peak indicator is meant to visualize audio signals with a red-green bicolor LED (or two separate LEDs):

*green*: signal is present (between 0 and 5V)
*orange*: signal is hot, beyond standard Eurorack specs (between 5V and 10V)
*red*: signal is getting close to clipping level of typical Euro circuits (above 10V)

Although there are simpler circuits, this one has a couple of special advantages:
* the BAT42s make a rectifier, such that both positive and negative peaks are detected
* the 3V3 zeners help to overcome the dead-zone on the green 'signal present' indicator

Also note: C23 and R14 determine the decay-time of the peak detection. If you want peaks to be held longer or shorter, change their values.

Regards,

Rutger

User avatar
roglok
Super Deluxe Wiggler
Posts: 2560
Joined: Fri Dec 30, 2011 3:49 pm
Location: germany

Post by roglok » Thu Feb 11, 2016 2:04 pm

very useful. the back 2 back zeners seem like a cool trick. thanks for sharing!

User avatar
abelovesfun
Super Deluxe Wiggler
Posts: 1051
Joined: Thu Dec 13, 2012 1:19 pm
Location: Portland, OR
Contact:

Post by abelovesfun » Thu Feb 11, 2016 2:36 pm

Is there an eagle or other PCB file? This is something I could use a lot of.
Super Easy, Awesome DIY kits @ http://www.aisynthesis.com
modular ambient and composery music @ http://www.forcedamage.com/
https://www.facebook.com/forcedamagemusic

User avatar
executiveBlaster
Veteran Wiggler
Posts: 513
Joined: Wed Oct 12, 2011 12:59 pm

Post by executiveBlaster » Thu Feb 11, 2016 4:39 pm

sweet! thanks for sharing rutger

I use this guy all the time:


Image


No dead spot. The resistor value can be tweaked to taste. Using WP937EGW LEDs, 1k keeps it really off at 0V, and indicates signal level nicely.

Your Level / Peak indicator looks awesome, I'll have to try that out!

User avatar
latigid on
Super Deluxe Wiggler
Posts: 1024
Joined: Fri Oct 25, 2013 11:05 am

Re: Bipolar CV and signal level/peak LED circuits

Post by latigid on » Fri Feb 12, 2016 4:19 am

rutgerv wrote:Hi everyone,

I always love those threads where small circuit ideas are shared. In software you would call them "code snippets", so I thought I'd share a few myself.

/snip
Thanks for sharing, they look great! Do they noticeably influence the power rails/crosstalk when running, or do the op amps provide sufficient buffering? Have you tested them with e.g. a 1 volt/oct CV signal? Adding LED bling has a few risks.

rutgerv
Common Wiggler
Posts: 69
Joined: Mon Dec 06, 2010 7:25 pm
Location: Elst, Netherlands
Contact:

Post by rutgerv » Fri Feb 12, 2016 11:15 am

abelovesfun wrote:Is there an eagle or other PCB file? This is something I could use a lot of.
Not at the moment, but if there's a lot of interested, I could have a set of boards made. I was actually thinking of a set of circuits like these to go into 1u tiles.
latigid on wrote:Thanks for sharing, they look great! Do they noticeably influence the power rails/crosstalk when running, or do the op amps provide sufficient buffering? Have you tested them with e.g. a 1 volt/oct CV signal? Adding LED bling has a few risks.
As far as I've been able to test, the opamps do a pretty good job at isolating/smearing out current loading on the power rails. Keep the ground clean is more of a concern, so that's why I'm returning the current to the DGND (dirty/digital ground) after passing through the led. I actually use this as a separate trace on the module towards the power connector (I wish I could go down separate to the PSU, but that's not within Euro specs).

Rutger

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1801
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Post by neil.johnson » Fri Feb 12, 2016 3:23 pm

executiveBlaster wrote:sweet! thanks for sharing rutger

I use this guy all the time:


Image


No dead spot. The resistor value can be tweaked to taste. Using WP937EGW LEDs, 1k keeps it really off at 0V, and indicates signal level nicely.
Yes this is a well-proven circuit I've used many times. LEDs are current devices, so putting them in the feedback loop of the op-amp ensures that the op-amp's output will do whatever is needed to drive a nice linear current through the LEDs. And it saves two resistors and two zener diodes.

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

rutgerv
Common Wiggler
Posts: 69
Joined: Mon Dec 06, 2010 7:25 pm
Location: Elst, Netherlands
Contact:

Post by rutgerv » Mon Feb 15, 2016 6:26 am

neil.johnson wrote:
executiveBlaster wrote:sweet! thanks for sharing rutger

I use this guy all the time:


Image


No dead spot. The resistor value can be tweaked to taste. Using WP937EGW LEDs, 1k keeps it really off at 0V, and indicates signal level nicely.
Yes this is a well-proven circuit I've used many times. LEDs are current devices, so putting them in the feedback loop of the op-amp ensures that the op-amp's output will do whatever is needed to drive a nice linear current through the LEDs. And it saves two resistors and two zener diodes.

Neil
It's simple alright, but I assume you're aware it doesn't behave so well with smaller signals. Even though the current through the LED is linear, due to the opamp's action. Just breadboard it, feed it an LFO and turn back the input level and you'll see what I mean.

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1801
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Post by neil.johnson » Mon Feb 15, 2016 3:53 pm

rutgerv wrote:It's simple alright, but I assume you're aware it doesn't behave so well with smaller signals. Even though the current through the LED is linear, due to the opamp's action. Just breadboard it, feed it an LFO and turn back the input level and you'll see what I mean.
Yes I was aware, since that was what the circuit as designed for. The trouble with handling low level signals is that the SNR drops and you start having to discriminate between signal and noise. In your circuit the zeners increase the loop gain at low signal levels, the downside being that noise can start to affect the display.

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

rutgerv
Common Wiggler
Posts: 69
Joined: Mon Dec 06, 2010 7:25 pm
Location: Elst, Netherlands
Contact:

Post by rutgerv » Tue Feb 16, 2016 3:49 am

neil.johnson wrote:
rutgerv wrote:It's simple alright, but I assume you're aware it doesn't behave so well with smaller signals. Even though the current through the LED is linear, due to the opamp's action. Just breadboard it, feed it an LFO and turn back the input level and you'll see what I mean.
Yes I was aware, since that was what the circuit as designed for. The trouble with handling low level signals is that the SNR drops and you start having to discriminate between signal and noise. In your circuit the zeners increase the loop gain at low signal levels, the downside being that noise can start to affect the display.

Neil
Indeed, that's the trade-off. Though in practice the noise sensitivity is counteracted by the slowness of the LED a fair amount. I've never had problems with noise, even with a rather noisy LFO output.

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1801
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Post by neil.johnson » Tue Feb 16, 2016 8:15 am

One advantage of your scheme is that you can adjust the relative brightness of the two LED colours by changing the knee voltages of the zener diodes.
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
sicpaul
Wiggling with Experience
Posts: 342
Joined: Thu Dec 22, 2011 9:28 am
Location: Berlin

Post by sicpaul » Mon Mar 07, 2016 10:39 am

Hello,

i think the peak/level indicator might be exact what i want, but after building on portoboard i don't get it working as i thought. So i try to understand the circuit.


1st OP stage (IC10A) is an inverter to make negative signal parts positive for common detection (behind the two BAT42). C23 and R14 slowing down reaction time.

4th OP (IC10D) compares signal level with about 4,9V and lights one LED (i assume the red one?) above.

3rd OP (IC10C) compares signal level with about 9.9V and goes high above.

What does OP2? and what happens around D7?

Will there be a friendly wiggler explaining it to me in easy words without killing me by maths? :help:

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1801
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Post by neil.johnson » Mon Mar 07, 2016 12:13 pm

sicpaul wrote:1st OP stage (IC10A) is an inverter to make negative signal parts positive for common detection (behind the two BAT42). C23 and R14 slowing down reaction time.
Correct.
sicpaul wrote:4th OP (IC10D) compares signal level with about 4,9V and lights one LED (i assume the red one?) above.
Yes, that would be the peak LED, since it is either hard-on or hard-off.
sicpaul wrote:3rd OP (IC10C) compares signal level with about 9.9V and goes high above.

What does OP2? and what happens around D7?
You're almost there. When the output of IC10C goes low it pulls all the current away from D8A. Diode D7 ensure that IC10C cannot drive current into D8A. The difference in thresholds gives you the "orange" zone when both green and red LEDs are on, so the colour progression is:

off -> dim green -> green -> bright green -> orange -> red
sicpaul wrote:Will there be a friendly wiggler explaining it to me in easy words without killing me by maths? :help:
Hope that helps!

Neil
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

User avatar
sicpaul
Wiggling with Experience
Posts: 342
Joined: Thu Dec 22, 2011 9:28 am
Location: Berlin

Post by sicpaul » Mon Mar 07, 2016 2:48 pm

neil.johnson wrote:Hope that helps!
Neil
Yes, thanks Neil, it is good to have my suggestions confirmed and to understand what D7 does.

The only thing left unclear to me:
neil.johnson wrote:
sicpaul wrote:3rd OP (IC10C) compares signal level with about 9.9V and goes high above...
You're almost there. When the output of IC10C goes low it pulls all the current away from D8A. Diode D7 ensure that IC10C cannot drive current into D8A.
From your explanation i would think output of IC10C it should go down to switch off the green light an high input signals.... and looking at the circuit i would assume it goes high over 9.9V of input level instead. Shouldn't the pins 9 and 10 of IC10 be swapped? :hmm:

User avatar
sicpaul
Wiggling with Experience
Posts: 342
Joined: Thu Dec 22, 2011 9:28 am
Location: Berlin

Post by sicpaul » Mon Mar 07, 2016 2:55 pm

:bang: :bang: :bang: Sorry, i don't know why my printing shows the pins swapped (less ink and too much dirt?). So it is my fault. Clearer looking tells me THEY ARE in opposite connection than OP IC10D.
Circuit is clear now. Thanks again and sorry for my mistake :hihi: .

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1801
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Post by neil.johnson » Mon Mar 07, 2016 5:40 pm

No worries :)
Random ramblings :: http://www.njohnson.co.uk
SSI parts and more :: https://www.cesyg.co.uk

Post Reply

Return to “Music Tech DIY”