PSU capacitors

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

Yes it's common practice these days and many people just blindly do bypassing without thinking a bit depper what it does. Back in the days of Sound Practices magazine it was already mentioned in write ups of advanced amp developers that they are not in favor of it.

There is the potential risk of resonances between the small bypass cap and the lead inductances to the big one.

I have heard amplifiers which had their sound quality completely destroyed by bypassing. After removal of all bypass caps they sounded really good.

I think many people just blindly apply a bypass cap. If they hear a difference they think it must be better while actually it might be worse.

I know this is in contradiction to most peoples believs and practices. So I want to make clear all this is IMHO and IME.

I'd say: Get a cap which sounds good to you without bypass. Especially electrolytics improved dramatically in the recent decade (I still don't like them though) so what might have been a good practice on elcaps 20 years ago is not needed any more these days.

Best regards

Thomas
 
Hi!

In the case of an OTL headphone amp using a lytic 220uF as the output cap maybe a good quality bypass cap (1uF film or PIO) can be beneficial?
Well as I said: In my opinion NO!

My suggestion would be: Determine the minimum capacitance required. Do you really need that much? Maybe less is sufficcient, like 100 or 50uF. You can get ASC MP in oil caps in 50uF. Give them a try.

My experience: Smaller caps (smaller capacitance values) sound better than larger ones, so I always use the minimum capacitance which is needed. Don't bilndly oversize cap values, this is bad practice. Second: Caps with higher voltage rating sound better than those with lower voltage rating.

If you must use an electrolytic, get a modern one (recent manufacture). Try it without bypassing first.

Of course you may bypass it and if you prefer the sound of the bypassed cap leave it like that.

Best regards

Thomas
 
Any modern electrolytic designed for Inverter or SMPS duty will do, Audio duty is a walk in the park compared to off-line switching duty with it us rise times and high currents, I rarely see any additional bypass capacitors used with these capacitors. If you can hear the difference between capacitors then they are too small, Elna, rifa, BHC aerovox, sprague (when they were in business) hitachi, nichicon just to name a few all make quality capacitors which last decades in industrial environments
 
Member
Joined 2009
Paid Member
Hi!

Go for ASC X386 series caps. MP in oil. Better than any electrolytic, no matter how well you bypass it IMHO. I'm not a friend of bypassing I think it can create more problems and colorations than it solves. Use caps which work well without any bypassing.

best regards

Thomas

Thanks for the input, Michael Percy has 20uF/440VDC & 40uF/440VDC I want to replace lytics 220uF PSU & 22uF near the tube, wich values I have to choose to replace the lytics?

I dislike also bypassing good lytics with films;)

You are right your kit it's not expensive if looking for the best:)

Cheers

Felipe
 
Member
Joined 2009
Paid Member
Hi!

Yes it's common practice these days and many people just blindly do bypassing without thinking a bit depper what it does. Back in the days of Sound Practices magazine it was already mentioned in write ups of advanced amp developers that they are not in favor of it.

There is the potential risk of resonances between the small bypass cap and the lead inductances to the big one.

I have heard amplifiers which had their sound quality completely destroyed by bypassing. After removal of all bypass caps they sounded really good.

I think many people just blindly apply a bypass cap. If they hear a difference they think it must be better while actually it might be worse.

I know this is in contradiction to most peoples believs and practices. So I want to make clear all this is IMHO and IME.

I'd say: Get a cap which sounds good to you without bypass. Especially electrolytics improved dramatically in the recent decade (I still don't like them though) so what might have been a good practice on elcaps 20 years ago is not needed any more these days.

Best regards

Thomas


+1:)
 
Hi Felipe,

Thanks for the input, Michael Percy has 20uF/440VDC & 40uF/440VDC I want to replace lytics 220uF PSU & 22uF near the tube, wich values I have to choose to replace the lytics?

You can also get them at alliedelec.com They have better prices and also more choice in cap values. Or get them from me, if you don't want to go through the hassle of EU import duties and related paperwork ;)
Be aware that sometimes the ASCs come with little dents. They are uncritical just don't look too nice. Sometimes they come spotless. Since these are manufactured for industrial use, things like cosmetics are not taken care off. I select the best ones and get them varnished with various colors, like metallic white, black, metallic silver and graphite grey. But obviously that comes at a price ;)

Find out what the lowest capacitance value is which your circuit needs and then replace it.

Best regards

Thomas
 
Yes it's common practice these days and many people just blindly do bypassing without thinking a bit depper what it does.

I think that's a fairly reckless statement to make. Bypassing lowers the impedance of the cap at high frequencies. In other words, bypassing gets you closer to an ideal cap.

Also bypass caps on ICs (op-amps, chipamps) are not optional. They are needed for stability.

Back in the days of Sound Practices magazine it was already mentioned in write ups of advanced amp developers that they are not in favor of it.

Did they back it up with data or were they just stating an opinion?

There is the potential risk of resonances between the small bypass cap and the lead inductances to the big one.

Bypassing lowers the effective ESL of the total capacitance formed by the capacitors in parallel. Recall, that the total ESL of the combined cap is L1 || L2 || L3, the ESR is similarly R1 || R2 || R3, and the capacitance is naturally C1 + C2 + C3. Hence, even if parasitic oscillations were a problem the oscillations would occur beyond the self resonance of the smallest cap (in the GHz range).

I'd say: Get a cap which sounds good to you without bypass.

Obviously, that's the ideal world. Sometimes, you have to compromise though.

~Tom
 
I think many people just blindly apply a bypass cap. If they hear a difference they think it must be better while actually it might be worse.

I pretty much did this exact thing. Most people said that it sounded better with the film caps so I just did it. Someone told me to remove them on my Int. Amp and have a listen. It actually sounded better to me. It cleared things up and I have never put them back on. This is only one case but I think Thomas is right that it should be something you don't just automatically do.
 
Hi, to be honest I have only been messing with audio for a year and a half....but I am begining to ask the question does the maths reflect the listening experience??this arguement of bypassing is surely an example...I have removed ceramic bypass caps in circuit boards tried various caps with lower esl/leakage/etc/etc and gone back to ceramic....or taken them all out....the permutations of it all are endless....a plate of spaghetti...... I am not saying that these caps do not make a difference...but what suits your ear is surely that is the answer....
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.