Is there some collective knowledge about the sound of potentiometers?

[jan.didden]

I can limit my inquiries to those that are compatible.

Thank you.

Indeed. If you have preconceived notions it is always best to ask only those with the same preconceived notions. 😉

Jan

 

[john curl]

Yes Jan, he should focus his questions to me, as I stand behind what I, and others, have heard.

 

[cbdb]

So what?

 

[Richard Ellis]

Ah yes, a new religion is forming....We have the loco inter-connect sect, the Speaker-wire faction, wire, resistor, capacitor ......let's see....And now the Potentiometer group. No doubt, a Lutheran type split from the wire guys.
All bow before the mighty potentiometer, listen carefully, the potentiometer has endless words of wisdom.




__________________________________________________Rick.......

 

[JMFahey]

You piqued my curiosity, so I just tried measuring the THD+N in an Alps 10k linear pot, set to the -6dB potision, with 15Vrms input, using an AP1. I could detect nothing but noise across the full audio range (noise floor hovering around 0.0006% THD+N). I find this reassuring, if a little boring!

Way to go 😀

Well, I didn't receive the information that I was seeking, but I did receive information.

Now that I know the nature of the biases and superstitions of the residents here, I can limit my inquiries to those that are compatible.

Thanks for stating the rules of the game, so we don't waste time trying to explain you what you will never ever accept, no matter what.

 

[Xoc1]

A cheap potentiometer is likely to have poor channel balance. It is also likely to have a very large temperature coefficient.
I think cermet pots make sense as the TC is lower than carbon.
The comment that using law faking resistors put more current through the wiper contact is interesting. The advantage of using a law faking resistor is that it makes it possible to use a linear pot which usually has better channel balance than a comparable part with an audio taper.
Saying all that my current choice of attenuator is a home built switched shunt attenuator that uses a 10K bulk foil resistor in the signal path with a switched pull down that mostly consists of common 1% 50ppm metal film.
I'm not going to make any claims about its performance other than to say that I'm happy with the way that it works, I don't think it compromises my system, and the channel balance is better than the speakers.😉

 

[Destroyer OS]

I've accumulated every piece of information through a deliberate process of refinement. Despite the massive amount of data, there's one notable thing that occurs every time, beyond significant standard deviation. At first this was startling because, well, we've all got concerns about what fits our needs in so many ways. But all said and done one couldn't argue that it's less than a rule as one exception cannot break a rule. Yearn no more. Potentiometer < Stepped Attenuator. On a budget? Check Ebay, they've got small ones made in china, and ones stocked with Dale resistors.

 

[SY]

A cheap potentiometer is likely to have poor channel balance. It is also likely to have a very large temperature coefficient.

Tempco is irrelevant- remember, you're looking at a ratio and the pot is pretty likely to be isothermal.

 

[Xoc1]

Is thermal modulation of the signal relevant?
In an active buffered system I an see that the TC of the resistive attenuator is largely irrelevant.
A Cermet pot is about 150PPM which is probably a closer match to a law faking resistor than a carbon pot would be. Does the TC of the various resistors in a passive system signal path have any effect?

 

[john curl]

Cermet is not advised. Carbon, plastic or wire is.

 

[HeyBill]

The really good stereo pots track together well. I've seen tests showing that some don't. I'm not an expert, but bandwidth is an issue in volume control design.

My volume control has input transformers and a 10K DACT stereo attenuator. The bandwidth is said to be very good with this approach. The DACT pot also tracks very well. It sounds good too IMHO.

Regards, HeyBill

 

[Pano]

Funny things, potentiometers. I have tested several (Alps, Nobel, Panasonic) for microphonics and could find nothing at all. Not a bit.

On the other hand, I've used pots I thought sounded crap. There was some big round metal pots that my friends were all crazy about, don't remember the brand. Awful, scratchy, bad tracking, grungy sounding. I just threw away 3 of them. And the expensive TKD pots? Yeah, they work great. Hate to like them because they cost so much.

Maybe it's just how well the wiper works.

 

[SY]

Is thermal modulation of the signal relevant?

In a potentiometer, that doesn't happen unless you do something pathological like intermittently hitting the lower or upper part of the tract (with respect to the wiper) with freeze spray, or do something really stupid like running high power signals through it, in which case thermal modulation is the least of your problems.

 

[Xoc1]

Cermet is not advised. Carbon, plastic or wire is.

This is an intriguing statement, I wonder what your objection is to cermet?
A internet search shows that some people are enthusiastic about cermet.
NVA Audio for instance.
Looking at comparable specs for Cermet vs conductive plastic I came across the Vishay P11s and P11a
The S version is cermet and the A version conductive plastic
The conductive plastic version is sold specifically for audio use.
Spec sheet here
http://www.vishay.com/docs/51031/p11.pdf
The only spec that stands out as preferring the CP version is the contact resistance variation.
Is there something else that is not immediately obvious, RF diodic effects for instance?
Regards Martin

 

[lcsaszar]

The question is: does the material (resistive layer and wiper) matter at all? If there is an audible difference between signal wires made of copper, OFC, silver, iron, then why there would not be at potentiometers? Do we know at all what are the materials they are made of? What is the composition of metal in wirewound resistors, in cermet, in carbon, in conductive plastic, in metalfilm? Also dissimilar metals may form a thermocouple and perhaps a diode. Could this effect occur at the barrier of the conductive layer and the wiper? Or shouldn't we worry about all this?

 

[SY]

If there is an audible difference between signal wires made of copper, OFC, silver, iron, then why there would not be at potentiometers? Do we know at all what are the materials they are made of?

In order, big "if" (highly unlikely, and if Pano's demonstration that no-one could distinguish copper from a banana or mud is considered...) and yes, very well known.

 

[john curl]

Oshifi, there is often a difference between pots. We didn't think so, 40 years ago, but experience and often, even measurement shows problems in even the 'best' most expensive, pots that we could buy. Often this is because of a defect in the plastic resistive layer at certain positions of the pot setting, creating a 'back to back diode'. I have made these measurements and this forced changing to a different brand, or even type of pot.
Unfortunately, there are still differences that cannot be easily measured. I suspect that this is due to the substrate under the resistive layer. In any case, it has been heard.
You will find that for the best hi fi equipment, we take pots rather seriously. Many of my competitors make their own ladder network, because they find it to sound better.

 

[SY]

Perhaps you can explain to Merlin (post 5) what he's doing wrong.😀

 

[john curl]

Alps makes good value, but often audible pots.

 

[stephend1714]

Pano -- in a post above -- mentions TKD pots . . . I used them [a second one when the first got a bit noisy after several thousand wipes changing gain between my CD player and my bi-amped configuration]. Highly satisfactory --- doing a neat disappearing act, IMHO.

A stepped quality-resistor stereo ladder attenuator was my first choice, but the dB changes in any I could find for sale back in the day (1991 to 2000 or thereabouts) were too big --- and the rather stratospheric price just added to the non-appeal factor.