Need advice on capacitors for power supply bypass (Hafler P-500)

[Aaron D]

I think this is my first post on this forum. I mainly hang out on the Parts Express board. Most of the topics here are a little over my head... I tried a search on this subject a couple of weeks ago but I did not find anything.

I am interseted in modding my Hafler P-500. It seems that a good place to start is with a bypass cap(s) on each of the power supply caps (20k uF each, rated at 100V).

I have been told to try a newer low ESR 470 mF cap in parallel with each of the PS filter caps. Then, also use a 5 uF polypropylene film cap to bypass each of the filter caps. I am planning on using a Dayton Poly cap for the 5 uF but am having trouble finding a "newer low ESR 470 mF cap". I was told to check Digi-Key or Mouser or Michael Percy for possible sources for this cap. I have searched each of those site but w/o knowing exactly what to look for I think I am limited by the 100V spec.

The only thing I did find so far was this a BlackGate from Musical Concepts. It is rated to 160V and is $16/each. Also it is an axial mount and I would prefer a radial mount. I am not sure how much more I am paying for the 100V rating.

(first cap listed)

http://www.musicalconcepts.com/electrolytics.htm

My question(s) in a nutshell... What would you recommend I use for the 47uF 100V cap, where do I get it, and how much? I would prefer a radial mount.

As a side note, I am planning on putting a 6.8k bleeder resistor in parallel w/ each cap.

Thanks for any input.

Aaron

[Aaron D]

Not sure what to think. Am I asking a dumb question or was it just poorly timed with the holidays? Does anyone care to know the answer if I find it elsewhere? Anyone have any good resources I cold use to help me figure this out?

Thanks,
Aaron

[Gigapod]

Sorry, but can you post a picture with a detail of the original schematic (power supply), so I can have an idea of the voltages and currents involved?
What is the purpose of your mod? Hum reduction, "audiophile" improvement, replacing a damaged part, increase in power consumption so you have to reinforce the bypass caps?
Using capacitors much above the rated voltage is counter-productive: higher voltage capacitor have a higher ESR.
I am not sure what the purpose of the bleeder resistor may be, but in any case don't use a 1/4W part: it'll burn!
Ceramic caps have very low ESR, just use a 100nF, 220nF or 470nF 200V ceramic cap across the big eletrolytic caps and that should be OK.
Some people also add small electrolytic caps in parallel with the big ones, again to get a lower ESR. Imho anything in the 10uF ~ 100uF range is OK, just make sure to use the correct voltage rating. More is just a waste of money: the final result you want to achieve is to reduce PS impedance.
As mentioned by Rod Elliot on his website, using smaller capacitors in parallel is better than one big one, because of the smaller ESR of the smaller caps.

[boholm]

The "470 mF" part of it all had me confused a bit - especially since you also write "20k uF".

So my guess is that "470mF" really should be "470 µF". And Digi-Key has something: http://dkc3.digikey.com/PDF/T071/1335-1345.pdf Page 1341 (or 7 of 11)

On the buttom of the page you are linking to there is a set of capacitors for the DH-500. Maybe that could be something to think about too?

Otherwise it would be a good idea to have some 470µF capacitors as decoupling as close as one can get to the output stage - preferably on the board itself, maybe on the backside. I do not know the DH-500 so you (or someone else) will have to enlighten me regarding the circuitry of and around the output stage, if more help/info is wanted.

[BobEllis]

The schematic is available from Harmon, perhaps the original poster can find us the link. I downloaded it ages ago.

The DH-500 runs at ~90V rails, so you must have 100V caps there. Digikey P10874 fits the bill there (radial cap). Panasonic FCs are nice sounding caps. You could also use them to replace C12 and C14. They are higher in value than the originals but you should have no problem with that. If you want to stick with 100 uf, then use P10775 or P10776.

The feedback cap C8 is a good spot to put a quality non polar. Be sure to use at least 470 uf. Here you can get away with lower voltage here - 16V is plenty, Hafler used a 6.3V cap.

You might want to replace C7 with a (larger value) polypropylene or polystyrene film cap. Same with C6. Either cap could be bypassed with one of Mouser's $.40 10nf polystyrene caps instead of replacement.

You probably have already been told, but consider replacing the main PSU caps. Higher value (30-40K uf) would be a good idea, but 100V or higher is required. My DH-500's caps are still in good shape (very low ripple) after 23 years of everyday use, though.

Edit: be sure to use a 2W or better resistor for the bleeder. I'd go at least 10K or up to 20K. You won't get a partially discharged cap thump with the speaker delay circuit. It will take quite a while to open up and get at the caps, so a long bleed down time is acceptable.

[Aaron D]

Thanks for the replies. I will try to clear up any loose ends as much as possible.

1) motivation for this is ““audiophile" improvement”

2) the power supply runs at 90V the main PS caps are rated to 100V

3) yes, 470mF should have been 470uF, that is a cut and paste error that caused me some confusion as well. I also do not know how to type the funky “micro” symbol, what is that “mu”? I think I recall it from my physics class as the coefficient of friction as well. I’m better with mechanics than electronics (I swear).

4) Schematic is available at http://www.hafler.com/techsupport/pd...00_amp_man.pdf the fun stuff starts on page 16


Finally, just to clarify. I assuming that all the caps listed are polarized unless other wise stated, correct?

Bob, you stated

You might want to replace C7 with a (larger value) polypropylene or polystyrene film cap. Same with C6. Either cap could be bypassed with one of Mouser's $.40 10nf polystyrene caps instead of replacement

How much larger is good/practical?

Also

You probably have already been told, but consider replacing the main PSU caps. Higher value (30-40K uf) would be a good idea, but 100V or higher is required. My DH-500's caps are still in good shape (very low ripple) after 23 years of everyday use, though

I have considered it but w/o actually testing them I think they are in decent shape. I am wanting to get them checked someday (I want to learn how to do that too but one thing at a time…). If I do find out mine need replaced I will probably go for around 80k uF total, good suggestion.

Anything else you guys have to add is much appreciated. This is a hobby for me and I like learning about new things. If you guys take the time to type it, I will consider it worth listening to.

Thanks again,
Aaron “D

[BobEllis]

Originally posted by Aaron D
Schematic is available at http://www.hafler.com/techsupport/pd...00_amp_man.pdf the fun stuff starts on page 16


Finally, just to clarify. I assuming that all the caps listed are polarized unless other wise stated, correct?

The caps are non polarized unless listed as electrolytic. Note that C8 is a non polarized electrolytic.

How much larger is good/practical?

Basically whatever you can fit. C7 is in the feedback loop, bypassing whatever potential nasties are generated by C8. C8 is one of those Black Gate might make sense spots. A couple uf of polypropylene and a polystyrene bypass for C7 should be suitable.

C6 is another spot where a good cap is important. A while back one of the real experts (IIRC Hugh Dean of AKSA) said that the Vbe multiplier bypass cap has a big impact on sound quality. I've seen values from 100nf up used. My Leach amps sound great with 100nf polypropylene stacked film caps. Polypropylene ought to sound better than the unspecified film cap, or just bypass it with a 10 nf Wima FKP2 (available at Mouser)

I have considered it but w/o actually testing them I think they are in decent shape. I am wanting to get them checked someday (I want to learn how to do that too but one thing at a time…). If I do find out mine need replaced I will probably go for around 80k uF total, good suggestion.

Basically, the electrolyte dries up so capacitance decreases. Leakage current can also increase. I look for low ripple at idle - measured in volts means time to replace. You'll get more ripple under load.

Anything else you guys have to add is much appreciated. This is a hobby for me and I like learning about new things. If you guys take the time to type it, I will consider it worth listening to.

Thanks again,
Aaron “D

Take my mod advice with a grain or two of salt. I haven't actually done anything I have suggested - my DH500 does subwoofer duty so modding it isn't high on my priority list. Take a look at the DH220 mod thread for ideas - the driver boards are similar. Replacing old electrolytics seems to be generally accepted as the first step in refreshing an old amp.

[William Meyer]

Could someone please show a schematic of the values of the mods they did on the Hafler DH-500/P-500? I would like to make some changes to my D-500 too but if I see it on the schematic it wiould be easier to follow. Is it better to use smaller cap say 2x 10,000uF=20,000uF or even 3x10,000uF=30,000uF or just stick with a single 20,000 uF??? Do I need to put a smaller cap and bleeder resistor on each larger cap too. See why I need a schematic. Thanks