[Danster.dk]

Hi guys.

My first post here!
Im building my first diy headphone amp, and I got a stereo pot meter, which (of cause?) is made of two single turn potentiometers. But the two arent giving out the same resitance. It varies a lot, and I was jut wondering, should I buy another one, or dont you think I can hear the difference?
How exact should a potmeter be?

I've made some measuring:
Pot a: Pot b:
1,19k 1,21k
6,48K 6,86K
18,6K 18,07K
44,2K 43,2K

Thanks!

/Danster.dk
Denmark

 

[balou]

Most potentiometers really have high differences. yours doesn't look all too bad, and the human ear isn't so good in hearing little volume differences. if you still aren't satisfied, try out the alps rk27

if you want a high quality volume control, use a stepped attenuator. but they are way more expensive than potentiometers

 

[Danster.dk]

Thanks balou!

 

[Yen]

I have used some bad quality pots and they totally ruin your listening experience. Not even caring about the sound quality, bad pots have so poor balance that it is unbearable. E.g. once I had one in a cmoy, but it is panning from louder left to louder right and back to left. There is only one spot that is balanced, but that's too low.

However, you dont necessarily need a RK27. Cheaper RK09 is enough, unless you are building something more than a cmoy.

 

[Pars]

This is what is meant when people discuss the term "tracking" when referring to pots. Yours actually looks pretty good, so I wouldn't worry about it. Most pots do not track well particularly on the low end (volume turned down) of the scale. Yours appears to have an error of ~20 ohms out of 1.2K (lowest values you measured), so 1.6% on a pot that has a tolerance of probably 10% but maybe 20%. Measure it turned all the way down.. values should be 10 ohms or less.

 

[tangent]

Remember that the division ratio is more important than the resistance difference. That is, if you measure 10K and 40K at a certain position on one channel of a 50K pot and the other channel measures a total of 48K (well within 10% tolerance) and splits to 9K and 39K, this is almost exactly the same division value despite the resistance differences. You probably wouldn't hear any difference in this particular situation.

This is why I did my potentiometer tests using a voltage measurement instead of a resistance measurement.

 

[Danster.dk]

Im building a Kevin Gilmore class A. So I want a decent pot. I can tell you the numbers on the back of it: 50k(ohm)AX2 8632(A).
Im guessing AX2, means log right?

Pars: Im not sure when you write: "values should be 10 ohms or less", is that when every meassuring Im making below 1,2K, the differenes of the pots should maximum be 10ohms?

Measuring:

Pot A: Pot B:
1,2 1,0
24 33
61 77
113 128
258 263
404 374
539 534
694 658
920 892
1276 1205
1551 1559

So this is not okay?

Thanks!

 

[Pars]

No I would think that would be fine. I meant at the lowest volume that you should be seeing ideally 0 ohms, but of course in the real world, that isn't going to happen. Your values of 1.2 and 1.0 ohms, respectively are fine.

Tangent brings up a very good point: the absolute values themselves aren't that important, but the ratio of the values from wiper to ground vs. wiper to input are what really matters, as a pot is nothing more than a variable voltage divider.

 

[Danster.dk]

Thanks guys for your help. I will try the pot I got now, or else buy an alps or some similar. Im going to make the boards tomorrow, and hopefully get the last parts soon, so Im excited