Question about when a bipolar electrolytic is needed ...

[Derringer]

I've done a couple builds now using opamps with a +/-15 volt supply.

Most recently, I boxed up the preamp of the ampeg VH140c
https://ampeg.com/support/files/Schematics/V%20Series/VH-140C/VH-140C%20Schematics.pdf

Take a look at C9. It's near the top of the schematic, between IC4B and IC4A.

First, is it even required since the signal is riding on zero volts dc?

Second, is there any harm in using a polarized 1 uf electrolytic there since the dc on both sides of it would be the same?

[Rob Strand]

First, is it even required since the signal is riding on zero volts dc?

I'd say it is required because the output of the clipper is asymmetrical and any average DC from the output of that stage would upset the clipping points of the next stage.

Second, is there any harm in using a polarized 1 uf electrolytic there since the dc on both sides of it would be the same?

I wouldn't recommend an Electro.  When the output of IC4B is at full swing and the input is at low frequencies you might get too much reverse voltage across the cap.

[R.G.]

Take a look at C9. It's near the top of the schematic, between IC4B and IC4A.

First, is it even required since the signal is riding on zero volts dc?

It's not riding on zero volts DC. It's riding on the offset voltage of the preceding opamp times the gain of that opamp, plus any DC riding into that opamp.  Opamps have errors and offsets. If the gains are always near unity, you get away with direct coupling outputs to inputs for DC purposes. But with effective gains, you generally need to block the buildup of DC offsets with a DC blocking cap so the errors don't accumulate in all the wrong directions.

Second, is there any harm in using a polarized 1 uf electrolytic there since the dc on both sides of it would be the same?

Yes, possibly. Polarized caps really need polarization to survive. You don't want them to have any reverse voltage, even small ones, although they will sometimes survive minor small reverses for a while. Since you don't know what the direction of built-up offsets from preceding stages is, you need to use a NP type, or at least fake that with two polarized types hooked in series minus to minus.

[Prehistoricman]

That schematic is full of non-polarised caps. Bad design maybe? I'm not qualified to say.

You'd have trouble finding some of the larger caps in sensible physical sizes as non-electrolytic. You can substitute a NP cap with two polarised caps back-to-back.

https://electronics.stackexchange.com/questions/21928/can-you-make-a-non-polar-electrolytic-capacitor-out-of-two-regular-electrolytic

[Derringer]

cool

thanks folks!

[PRR]

> That schematic is full of non-polarised caps. Bad design maybe?

NP caps are not expensive (not like film!). In no-polar work they survive lack of bias pretty well (and longer than most warranties). Availability of >=1uFd allows low impedance levels for less hiss. Design is all trade-offs. I don't see "bad" here.

Derringer--- if you are sitting on an empty bin of NP caps, tack in a Polar. Put your voltmeter across and see if the small (probably <0.1V) bias is in the right direction; not? then flip it. That will work a very long time.

FWIW: mike preamps often have a huge (4,700uFd) cap with "unknown" small bias *either* way. NP that size is a hassle. Most of the console makers just use a good polar electrolytic.

[Derringer]

As usual, you know the situation well PRR 

I built it with a couple polar 1uf caps because that's what I had on hand and didn't want to take up too much space running a pair of polar 2.2 caps.

I did see that small bear has what the schem calls for for very cheap .... I'd pay more for the postage, but I'll get them with the next order.

I figure I'll drive the circuit now like it is and maybe learn what a failing electrolytic in the signal path sounds like. Do they short out? Open? Closed?

Point taken about measuring each side to see what's what. That's something I had intended to do while it was opened up, but got all excited when it sounded good, so I didn't bother. I ran them all with the + side towards the output of the previous opamp stage like you would with a virtual ground setup.

The enclosure used to house a power supply for a friend's old trainset.
Getting those pots all attached was a pain ... i.e. ... no rush to get back in there!