Bypassing filter caps with Smaller caps

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi everyone,

I am wondering what results people have with bypassing large filter and other large bypass caps with a small bypass capacitor to provide a better path to ground for high frequencies.

I have heard of this a few times but I am wondering if it is a good thing to conceder? I usually see somewhere around .1 uf recommended.

Thanks
 
I have asked before and didn't get an answer so here we go again. I believe with the proper test you should be able to see what you hear. Where will you see the difference with the PS cap bypassed? Output at HF? Sine Wave shape? Square Wave shape? Or??

You can be an objectivist and attempt to find an appropriate measure to show what you are looking for, or you can insert a high quality film cap onto the last stage of your power supply.

One is easy and you are the judge of any change, the other requires sophisticated high resolution test equipment and a rigorous protocol to judge your results by.

There are a few people on these boards who claim that if they cant measure differences in THD then there will be no differences in the final sound - experience has told me they are wrong. I know which approach I go for in these circumstance.

I am a trained scientist and know full well the limits of measurement techniques in deciding these questions. On this particular point about caps - I was extremely skeptical until I tried it. I was also extremely skeptical about using transformers for input phase splitting and interstage duties (knowing absolutely for a fact that they degrade THD figures) but now I have not built a valve amp without them for about five years.

You takes your choices and builds your own crosses.

Shoog
 
I have asked before and didn't get an answer so here we go again. I believe with the proper test you should be able to see what you hear. Where will you see the difference with the PS cap bypassed? Output at HF? Sine Wave shape? Square Wave shape? Or??

It depends on how it's done. If the smaller cap is placed local to the signal circuit, it can prevent or reduce the likelihood of oscillation. This is particularly true with circuits using high gm RF tubes like ECC88. If the small cap simply goes across the large one several inches away, its effect is negligible-to-zero.
 
It depends on how it's done. If the smaller cap is placed local to the signal circuit, it can prevent or reduce the likelihood of oscillation. This is particularly true with circuits using high gm RF tubes like ECC88. If the small cap simply goes across the large one several inches away, its effect is negligible-to-zero.

And yet people commonly observe it and so it is there, it cannot be so easily dissmissed - and I very much doubt that it is predominantly attributable to oscillations. I have heard it reported that the film cap may induce oscillation when used as a bypass due to inductive- capacitive interactions between the two caps - who am I to say without having seen the particular evidence.

Should you also be aware that measuring different characteristics may be capable of showing the effect of the modifications, rather than relying solely on the subjectivist view that only y/our ears can tell the difference?

They may show differences (or they may not) but often these are grouped variables which are not directly attributable to any one particular property of the circuit. What exactly are you reporting in these circumstances ? Have you gained any better understanding of what is really going on within the circuit ?

I would not advocate anyone taking my subjective opinion of this - I advocate them taking their own subjective opinion. As directdrive pointed out, he observed the difference, didn't like it and took them out - I can respect that position as much as my own. Its such a simple and quick thing to try that I cannot understand why anyone would lose any sleep over just sucking and seeing.

Shoog
 
Last edited:
I think you just need to accept that our ears are better measurement instruments than any man made device, for certain things. We have millions of years of evolution behind the design of our sensors. It's a matter of life and death, at times.

the ear as a measurement instrument is a lousy tool.......

but as a means to listening pleasure, yes it is very reliable, and yes it matters.....
 
Shoog -perhaps you use large main filter caps with poor ESL, or place your bypass cap in a much better position? Large caps typically start rising in Z above 100kHz, and a 1uF film bypass is only going to get down to low Z above 1 Mhz - do you use a smaller electrolytic for the bypass? A bypass on the main filter cap can serve two masters - do you place yours for smaller loop area of the power supply, or the output signal stage?

Ciao, Tim
 
I have generally used large computer grade caps (220-1000uf) and then bypassed them with something small such as a film cap of 1uf. Every situation is necessarily different. I add the film cap on the last electro in the power supply.
I also tend to bypass my cathode bypass electros (generally effective values of 500uf) with 10uf film caps. on the bypass cap, but again this depends a great5 deal on what is at hand.

Shoog
 
me too use those kind of caps in my tube amps, a capacitor bank per channel, capacitors shown are 470ufd/450volt ecaps.......this is a 4D32 pp amp i am building...a work in progress.....

IMG_1344.jpg
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.