• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Bypassing Capacitors

Grainger49

Grainger49

Old Fart
Just looking for what you expect from bypassing capacitors.

Really two questions:

1) what do you expect from bypassing a power supply capacitor?

2) what do you expect from bypassing an audio path capacitor?

I'm not asking for dissenting opinions, just your expectations.

My answers:

1) better effective ESR, quicker energy release (the low ESR bypass has energy that will dump much faster than the electrolytic). It imparts some, but not all, of the effects of having all better capacitors.

2) I know it is heresy, but I have used some Teflon KK 0.1uF to bypass audio path caps and it has smoothed the high. It hasn't, to my aged ears, smeared the sound. I can't afford 2 2uF Teflon caps but the KKs are a great deal.

So how about you guys?
 
Last edited:
Register to hide this ad
I don't know about bypassing power supply capacitors equating to higher ESR/Energy release.... If you're bypassing an large filter electrolytic with a cap... film or otherwise, you would be lowering the ESR. ESR is a function of frequency. Electrolytics have poor ESR that gets worse with higher frequencies. The bypass caps lower the power supplies impedance or ESR at higher frequencies.

Teflons???? smoke'm if you got'm. Probably the best cap dialectric material. Hard to find and expen$ive.
 
I edited the OP. I didn't express my low ESR thing very well. I'll edit that too.

What I meant was a low ESR cap added to a supply gives some energy storage that is faster than the bulk electrolytics.

The Teflons I use are Kommie Kaps. A nick name from Bottlehead. I look at it as the most stringent quality control ever. You don't meet the USSR Military Specs, you disappear!
 
Grainger49, your understanding of the subject agrees with mine...but I'm not sure that makes us correct!
 
I was just taking a poll of what you guys understood was the result of bypassing.

I think there are a lot of threads about how to and what to do, what not to do.

I'm just curious what everyone expects from bypassing.
 
There are advantages to doing this in the power supply, but as for coupling caps in signal-processing circuitry there can be a tremendous penalty. Here's an example.

A preamp board comes in for repair because the owner is complaining of a persistent hum in the audio. Inspection reveals that he has bypassed several dozen small electrolytic coupling caps on the board with boutique-quality films. You know, the big juicy ones that make good grid capacitors for high-end tube power amps.

The problem is that the large physical size of the bypass caps and the necessity of mounting them slightly away from the board due to crowding constraints makes them superb pickup devices for stray field energy, most notably hum from the AC line. But of course the owner REFUSES to believe that the bypass caps could be the source of the hum, largely because he would rather believe audiobabble rather than the laws of physics.

The lesson to be learned here is that the bypass capacitors should be SMALL and should be mounted tight to the board.

Fred
 
Last edited:
Great that this was brought up so I could add my 2 cents. I just got done replacing the original power supply caps in my SX-1250 and also added 2.2uF 100v film bypass caps across each power cap. First I installed the new main caps and gave it a listen for several days. Then I added the bypass caps hoping that the high freqs. would improve more. I was not disappointed, they became more defined and crisper alsmost to the point where listening to a favorite LP or CD I would get startled by a flick of a guitar string or cymbal that I never really heard before. End result was I got more than what I expected without being overwhelming to the ear at minimal cost. :thmbsp:
 
So, what does it make sense to bypass: big power supply caps? DC-blocking caps? What else?

(Thanks for this thread, I've been wondering about this myself!)
 
I prefer to relate my experience only. There is science and black magic on this subject and we could get buried in it.

I have bypassed DC blocking, audio path (coupling) capacitors with good results. BAT used some huge Teflon coupling caps. Somewhere in the 4uF range. They are physically enormous, not to mention prohibitively expensive. Teflon is "reported" to be the finest sounding coupling cap.

I put some 0.1uF KK Teflon caps in parallel with my larger (2.0uF) coupling caps and I liked the extension in high frequencies that I got. I also got smoother highs. But others might warn of the "smearing" that happens when the audio signal passes through two parallel caps. And I know what it is. I just didn't hear it.

I have also bypassed several power supply capacitors. The sound was, again, smoother with the bypass caps in place.

Just one man's experience.
 
Fred Longworth said:
There are advantages to doing this in the power supply, but as for coupling caps in signal-processing circuitry there can be a tremendous penalty. Here's an example.

A preamp board comes in for repair because the owner is complaining of a persistent hum in the audio. Inspection reveals that he has bypassed several dozen small electrolytic coupling caps on the board with boutique-quality films. You know, the big juicy ones that make good grid capacitors for high-end tube power amps.

The problem is that the large physical size of the bypass caps and the necessity of mounting them slightly away from the board due to crowding constraints makes them superb pickup devices for stray field energy, most notably hum from the AC line. But of course the owner REFUSES to believe that the bypass caps could be the source of the hum, largely because he would rather believe audiobabble rather than the laws of physics.

The lesson to be learned here is that the bypass capacitors should be SMALL and should be mounted tight to the board.

Fred
Click to expand...

You could demonstrate this for him by tapping on the outside of the caps with a chopstick (or whatever you use). You could also tell him that he can refuse to believe you after he has done as much service on this stuff as you have.
 
Bypass caps

I by passed the power supply caps in my Ampex 6973's and it removed all the hum at idle .I don't possess golden ear ability but these little 15 watters push the Klipsch La Scala's at crazy levels .I'm a firm believer in bypassing the power supplies on amps and will do the same to my W5M's I'm doing right now .They sing well enough that the kick drum almost sounds like you can hear the drummers foot hitting the pedal,very clean and detailed .
 
I bypass the hell out of power supply caps, especially if I'm forced to use a lytic. I bypass crossover caps out 5 times usually, and to date have not experimented much with bypassing coupling caps. I kind of look at it like getting a little of the best of all worlds, while lowering ESR. It's very cheap, and very easy (ha, reminded me of my prom date), so I figure why not.
 
What I expect? Nothing. I've done blind testing and nothing is to be gained. And the facts speak for themselves - it's BS.
 
I know this works with tweeter caps but don't expect fireworks and a parade of unicorns. It is a rather subtle difference but it is there.

If your ears are good you can even hear the difference between different bypass caps like the Audiocaps and the Daytons.
 
More important in tuners than in an amp, unless you have a very weak power supply in the amp. Maybe more important in bias circuits with their higher impedance and low signal levels.
 
Last edited:
What I expect by-passing a power supply capacitor is to have a better shunt to ground for the very high freqs riding on the DC.

I don't know if this is correct, but this is what I'd expect in that application.
 
elnaldo said:
What I expect by-passing a power supply capacitor is to have a better shunt to ground for the very high freqs riding on the DC.

I don't know if this is correct, but this is what I'd expect in that application.
Click to expand...

Sounds right to me.
 
You must log in or register to reply here.
Back
Top Bottom