[sdiy] Running a Bipolar Comparator..

Tom Wiltshire tom at electricdruid.net
Sat Mar 25 13:32:23 CET 2017

On 25 Mar 2017, at 07:36, Busby Bergson <busby.bergson at gmail.com> wrote:

> Hi everyone -
> I'm currently trying to run an LM339 on a bipolar power supply. The
> application notes suggest this is possible, as long as the input
> signals don't go negative. I'm feeding it with a fully-rectified input
> and a variable threshold from 0-12v.
> Attached to it is a pretty standard Attack/Release circuit, feeding a buffer.
> There are two issues:
> 1. The "Release" circuit shorts at the far end of its range.
> 2. Diodes are getting blown out.
> Does anyone have any idea why this is happening?
> Here's a drawing of the circuit:
> http://imgur.com/a/U0F8e

Given that the input is positive, and the threshold voltage is positive, and the output across the cap is positive…I'm not really seeing why you're using a bipolar supply for this. It looks like Ground and +V would be sufficient, except for the final output buffer op-amp, which probably wouldn't go fully to ground unless given a bipolar supply.

You need to make sure that the current that flows when the pots are at zero is low enough that the rest of the circuit can cope. That 100R resistor next to the cap is there to limit this current. The downside is it also limits minimum attack/release times. Thus it needs to be as small as safely possible, but no smaller. Some juggling of timing cap/pot values may be required to get a good compromise, though 1uF/1M is a combination I've seen before.

If the LM339 is on a bipolar supply, does its output get pulled down all the way to the negative rail? That'd really screw things up…cap suddenly turned upside down, etc.


More information about the Synth-diy mailing list