Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 1, 2011 0:23:51 GMT
Many of you will have seen Mike's often stated preference for non polarised electrolytics at the output location. The attached graphs may be of interest. Alex Attachments:
|
|
|
Post by jeffc on Dec 1, 2011 10:16:01 GMT
Hi Alex, Interesting, I've also read 'somewhere' where 'someone' preferred 2 polarised electrolytic caps wired in series back to back for output DC blocking, if this is what's being referred to here, and actually preferred this to using boutique PP caps. I also read 'something' by 'someone' refurbishing an old NAD3020 where non-polar capacitors were recommended for the pre-amp PSU, can't remember the exact reasoning behind this now, it was a while ago. cheers.. jeffc
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 1, 2011 15:55:18 GMT
It's quite simple really. One should NOT use polarised electros when there is little to no DC on it in feedback NOR in input/output decoupling circuits. Simply because they WILL distort as they were not made for this purpose. This is what's shown above... a cap that has been applied wrongly in a certain position that was replaced by a cap that should have been used in the first place in that part of the circuit.
You really will not believe how many manufacturers still use polarised caps in the wrong way ! Either to save some money or ignorance (don't disgard the last remark lightly).
Polarised caps MUST always have a DC on it for them to work properly.
If there is NO DC present either use 'two of them in antiseries (- to - or + to +) IF you are aiming for lowest DC leakage. Also the recommended method if a considerable DC may be present and the polarity/value is uncertain/unknown/may vary. In case you are looking for lowest distortion and care less about leakage and the DC voltage is always close to zero use 2 polarised caps in anti parallel.
Mind you in series the capacitance halves, in anti-parallel the capacity doubles. If you can fit non-polarised caps in both above described circumstances this is always preferred. Above 'escapes' are only recommended if other solutions are not available. Little magic about that.
How that results in better sound is more magical than technical and there seems to be little or no relation between subjective found differences and measurements.
|
|