Ripple/noise/voltage checker

Status
Not open for further replies.

Mosaic

Well-Known Member
Hi,
I am thinking of making two battery operated portable tools.
1) Tests for constant current ( up to 5A DC) or constant voltage up to 24 VDC and assesses loaded DC ripple/noise p-p, if possible down to 2 mV.

2) Tests audio capacitors for value & A.C. coupling noise/distortion.

For item 1) I suppose Linear Bipolar power transistors on a fan cooled processor heatsink can apply the 5A load.
An instrumentation OPA to amplify the ripple with a peak hold into a uC ADC to measure it with a precision Vref.

For Item 2:
I am not sure what is the best approach. I am wondering if I just rely on hearing to work. So I use a high quality reference capacitor into one side of a headphone and the output of the test cap into the other side with matched parts for volume and amplification from a standard 'source' material. Either a pure sinewave with variable freq. or an mp3 hifi source.
 
A ceramic capacitor can be microphonic where sound causes it to generate or modulate a signal which might cause acoustical feedback howling or distortion.
An electrolytic capacitor might be polarized without a DC voltage then it rectifies audio creating severe distortion. A ceramic or electrolytic capacitor value changes when the signal voltage changes that causes low frequency even-harmonics distortion.

It is easy to make a low distortion sinewave oscillator and distortion analyser.
 
For your distortion you could use a good soundcard in a PC, and use FFT to analyse the output. You could also use the soundcard as the sinewave source.
 
In my entire audio career I never tested any capacitors for noise or distortion because I used film capacitors (for coupling audio) that do not produce noise or distortion.
Maybe you want to use cheap electrolytic capacitors in a passive crossover network for a speaker? Then their wide tolerance and distortion will mess up the sound. Many people can hear 0.1% distortion, and some people can hear less.
 
Status
Not open for further replies.
Cookies are required to use this site. You must accept them to continue using the site. Learn more…