DC on pots - how much is OK ?

Hello,

To avoid noises when rotating a pot we know there should be no DC on a pot. But how much would still be OK ? It'd be a pity to add more electrolytics in the signal path just for a few pot movements (which will be rare for my application).

Like when mixing two 5532 outputs with a 'backwards pan-pot-section' (4 resistors, 1 pot with wiper to ground).
The 5532's are unity gain buffers and will be compensated for seeing equal R-source for each of their input (compensating for Ibias).

The 5532's will both have a different offset but it won't be much. The pan-pot in fact morphs from the first to the second offset, but since it's only a few mV could I get away with this ? (I'll add a drawing if necessary)

I should just test it, but I was wondering what the general idea was w.r.t. how much DC can be tolerated.

Thanks,

Peter

Hi Peter,

I have had serious scratching and clicking with sub-1mV DC voltages, so my guess is that you may need to be significantly below 1mV at pots, maybe a bit more at switches. This is why even very low leakages from electrolytics kills pots.

The quality of pots matters also, conductive-plastic pots are much more forgiving regarding DC than carbon-track ones..

Jakob E.

Sorry to jump on this topic, but...

Will a pot go bad over time with DC or will it happen immediately? And once the pot begins to get scratchy, is the pot ruined? Or if the DC is removed, will the pot stop being clicky/scratchy?

Just curious.

I've heard of pots getting ruined by DC before, but I can't really see how it can happen except if there is significant currents flowing.

The main issue is noise that's generated as you move the pot, which modulates the DC and folds it into the audible band (makes a click as you break out of the friction band of the pot and then a lower frequency swoosh as you continue to turn). A lower fricion wiper to track connection may reduce the scratchyness of pots with DC on them.

Cheers,

Kris

The DC ruins the pot over time. It burns/oxidizes small spots in the carbon surface. Sometimes you can restore the pot by using contact cleaner.

Is this the case for conductive plastic though? What about trimmer pots...are these not designed for DC? In my day job I work with flight test instrumentation packages that use pots for measureing things like control deflections, and for the most part we're talking DC (sub 2 Hz). I haven't seen one of these pots electrically fail yet. They are very high quality though. But I have seen carbon audio pots go bad.

Cheers,

Kris

If a pot has, or can have DC on it, (more than a few microamps)
It should be a wirewound. Even trim pots. Theres nothing wrong with having a high quality cap in your circuit. You might should also consider if the Op-Amp goes to the rail for what ever reason. If there's no load to speak of, a tantalum is the best. Next would be a low leakage electrotyic. Panasonic has several.



Cheers
RonL

A pot will survive DC without actually being ruined, but it wil become scratchy to audio.

For areas where it's only ment to pass DC, this may not be a problem at all (like in the threshold/makeup controls in the ssl compressor).

But if you want to be able to turn it with audio through it without noise, you'd better make sure that it's DC neutral..

Jakob E.

What idf it has the same voltage in both ends, so no current will be passing, but There´s DC aplied?

All the big poly synths like the Propet 5's and the like have front panels filled with pots that only deal with DC, such as supplying a 0-5V line to a multiplexer. I think rafafredd may have something there, as I rarely see problems with them. The current must be really low. Next time I service one I will take some measurements.

Hello & thanks all for the responses.

Didn't expect this to be so critical, but the sub-mV mentioned by Jakob makes me adding a few caps to the circuit. The 5532-offset will sure be more than this.

It does make me wonder how many designs seem to get away with this issue though. Especially SWR-circuits (bass amps); they have TL07X all around and only a few caps in the signal path. Most of the various stages and the I/O are DC-coupled. I'm now not in the neighborhood of a unit, but I don't seem to recall my SWR-gear being scratchy potwise.


FWIW, I'm mostly using the metal shaft Piher pots, they're sturdy and haven't given me much problems. I guess they're carbon though.


from Jakob:

The DC ruins the pot over time. It burns/oxidizes small spots in the carbon surface.

Click to expand...

That's only for large DC-voltages, correct ? Can't imagine a sub-mV voltage can already harm a pot.


from Ron L:

Theres nothing wrong with having a high quality cap in your circuit. You might should also consider if the Op-Amp goes to the rail for what ever reason. If there's no load to speak of, a tantalum is the best. Next would be a low leakage electrotyic. Panasonic has several.

Click to expand...

You're right, it's good to have a cap somewhere in the path for safety. But a tantalium ?


from Rafafredd

What idf it has the same voltage in both ends, so no current will be passing, but There´s DC aplied?

Click to expand...

I'd say it's OK if the whole pot is at the same potential.
If the complete pot is at +1000V or -23nV w.r.t. ground should make no difference w.r.t. scratchyness - how could the pot know ?


Bye,

Peter

You're right, it's good to have a cap somewhere in the path for safety. But a tantalium ?

Click to expand...

Tantaliums can sound as good as anything else as long as you don't
load it very heavy. Say an input to Non Inverting opamp ect.
I would perfer a good high qualityelectrolytic though.

[quote author="rafafredd"]What idf it has the same voltage in both ends, so no current will be passing, but There´s DC aplied?[/quote]

If it has the same voltage at both end then there is no voltage present across the pot and hence no current can flow.

I think we are getting into semantics here, but I can have 250VDC on pins 1 and 3 of a pot and there is no current flow in the pot as long as pin 2 (wiper) is not placed at a lower potential. There is no potential difference in the pot but there is voltage present.

I also do think so! There is tension (voltage) but no current flowing...

[quote author="clintrubber"]Hello,

To avoid noises when rotating a pot we know there should be no DC on a pot. But how much would still be OK ? It'd be a pity to add more electrolytics in the signal path just for a few pot movements (which will be rare for my application).
Peter[/quote]

Somewhere DC current trough contact is better than zero current.
In relays in phones swich centers it was be done.

Electrolytes on small DC are source of failures.
Is bigger hell to insert elyt to eliminate small DC than small DC

xvlk

See next post.

As the Penny & Giles spokesman on this list, I must state that DC voltage on a pot does not damage the pot - if the current is within the ratings of the pot. Many P&G faders are used in DC applications with no degradation due to the DC voltage.

Scratchy noise from a pot has 3 factors. If the interface between the wiper contacts and the resistive track degrades due to wear or dirt, the instantaneous resistance jumps as the wiper moves will produce noise. Clean the pot or replace old pots. Environment (dust, cigarette smoke, etc.) may shorten the useful time between cleanings.

The amount of DC voltage affects the noise. The voltage may increase with time as electrolytic capacitors age and allow more leakage current.

The amount of gain following the pot determines how much any noise will be amplified. If you have a low-level circuit, don't use pots, or if you must, keep the DC to very low values.

A few millivolts of DC in line-level circuits shouldn't be a problem with a reasonable quality pot.

Thanks, Dale!

Jakob E.