A small comparison about different capacitors

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

MuteRecord

New member
Joined
Aug 3, 2018
Messages
3
Hey,

before I start, I wanted to say thank you to all you nice people in this group; because your discussions always helped me with my little diy-projects :)


So here is what I wanted to share with you all:
For my bachelor-thesis I made a small comparison about different capacitors. I built an Amek 2500-PreAmp with switchable capacitors and measured the frequency response, the phase, the distortion and the noise of the PreAmp. After that I recorded some audio-samples and put them on a Google-Drive-link, where people can listen to them and tell their thoughts about that.

So if you are interested in my little “study” feel free to visit the link below and listen to the different samples. And if you leave a comment in the table-document I put on there, you would also help me with my bachelor-thesis :)

https://drive.google.com/open?id=1Q9wxaacplPf17jlkKckWbS7L940xA8jO

Thanks a lot and lots of greeting from Germany
Xaver
 
There are three capacitors I didn´t change. A 2.200µF and a 150pF because of their values and 10µF which is connected to a -16V supply. All the others were changed every time I recorded and messured them.

greetings
Xaver
 
Well good for you for devising an experiment. You're on the right track. But your methodology is a little off. If you want to study the characteristics of a particular component, you should create a test circuit specifically designed to exercise those characteristics. If you just change all of the caps on a board like that, if something sounds different, it is more likely be because the tolerance of capacitance is wide enough that you can hear the difference in a filter part of the circuit. Or it sounds different / better because you worked on it. I must confess I enjoy using stuff that I worked on, fixed or build from scratch. It makes it sound better even though there's no way it can be.

Capacitors do not sound different IMO. And if they did, many of the caps on that board are not even in the signal path. Also, were the caps replaced old?

Recapping is done largely to reduce noise and maybe fix up filtering w/ electrolytics.
 
Of course he carefully selected the capacitors to be exactly the same capacitance.  ;D

If he measured the amplitude response, phase response, linearity (distortion), and noise he can surely correlate any audible differences heard with objective differences measured.

Listening tests can require a lot of work to discern subtle differences with statistical significance.

JR
 
squarewave said:
Well good for you for devising an experiment. You're on the right track. But your methodology is a little off. If you want to study the characteristics of a particular component, you should create a test circuit specifically designed to exercise those characteristics. If you just change all of the caps on a board like that, if something sounds different, it is more likely be because the tolerance of capacitance is wide enough that you can hear the difference in a filter part of the circuit. Or it sounds different / better because you worked on it. I must confess I enjoy using stuff that I worked on, fixed or build from scratch. It makes it sound better even though there's no way it can be.

Capacitors do not sound different IMO. And if they did, many of the caps on that board are not even in the signal path. Also, were the caps replaced old?

Recapping is done largely to reduce noise and maybe fix up filtering w/ electrolytics.

You´re totaly right if you say, that the methodology would be wrong for the characteristics of a particular component. But, what I probably forgot to mention is, that the idea was to analyze differences in an audio-path. My thesis is something like "there are measurable and hearable differences between different capacitors in mic-pres". So if I had only measured different capacitors in a specific test circuit, it would have been against my thesis. But it will probably be something interesting to do for the future :)

And yes, the capacitors were all new. I built the PreAmp from scratch, so that I could switch the capacitors easily.
 
MuteRecord said:
You´re totaly right if you say, that the methodology would be wrong for the characteristics of a particular component. But, what I probably forgot to mention is, that the idea was to analyze differences in an audio-path. My thesis is something like "there are measurable and hearable differences between different capacitors in mic-pres". So if I had only measured different capacitors in a specific test circuit, it would have been against my thesis. But it will probably be something interesting to do for the future :)

And yes, the capacitors were all new. I built the PreAmp from scratch, so that I could switch the capacitors easily.
The age of the capacitors within reason is not as important as the exact capacitance. More or less capacitance can alter the pole frequency of simple filters and therefore affect the amplitude response.  If your bench measurements are precise enough they should reflect this.

Good luck.

JR

 
Back
Top