[fyleow]

The shipment from Percy Audio arrived today and I got an Alps Black Beauty. The test results are interesting.....I am using a Fluke 23 DMM

L R
90700 94000
55700 58200
75400 77800
45000 47300
84800 87600
85700 99100
39300 41500

Those are the resistance values. This is the 100k version. Is it me or are the results kinda bad? Also, the pot feels like a stepped atteunater. It clicks when I turn the knob and it's not smooth like the panasonic pot.

It looks alot like this POT here on Ebay, except I paid $35 for mine....

http://cgi.ebay.com/ws/eBayISAPI.dll...tem=1377992923

 

[fyleow]

Nevermind....I think I was measuring the wrong parts. I tested it again and it's closer matched. If anyone is interested I can make an excel graph with the resistance values.

 

[zhtham]

Yeah, a graph will be great

i'd like to know.

thanks,
Z.

 

[tangent]

I will be doing some similar measurements soon, fyleow, and based on some learned comments by others, I'm thinking that measuring resistance is the wrong thing. The short story is that I now think that measuring attenuation of a real-world signal makes more sense. That is, put a stereo test tone CD in and run each signal through the pot and measure the signal with an accurate AC voltmeter. This tests the behavior of the pot under real conditions -- it's a test for the behavior you're actually interested in instead of testing the mechanism that influences that behavior.

Anyway, that's how I'll be doing my own tests.

 

[morsel]

Measuring dc voltage or resistance ratio should be acceptable. Note that resistance ratio is Ra/Ra+Rb where Ra and Rb are the end to wiper resistances. Ra+Rb=Rtotal, which you get from your final reading with the knob all the way clockwise. I created a spreadsheet which references the final cell to get Rtotal.

Also, if you decide to determine average tracking error by converting to db and taking the difference of the two channels, don't forget to use the absolute value of the difference when generating an average, otherwise negative and positive tracking errors cancel each other out and give a false appearance of better tracking.

PS, I'm not really back, I am still missing in action until October. See you then.

 

[AssafL]

Quote:

That is, put a stereo test tone CD in and run each signal through the pot and measure the signal with an accurate AC voltmeter. This tests the behavior of the pot under real conditions -- it's a test for the behavior you're actually interested in instead of testing the mechanism that influences that behavior.

That is definitely a more accurate approach. But you must design the following stage to have a high enough input impedance not to influence the attenuation (i.e. FET input or the like). Otherwise, don't forget to include the following stage in your test.

 

[tangent]

The following stage will be a modern DMM. I should hope that's got a high enough input impedance.