Replacing Power Supply Capacotors

[artto]

If electrolytic capacitors are being used in a amplifier or preamplifier power supply, should these be replaced with electrolytic of the same values or can other "better" capacitors be substituted? The tube preamp in question is 34 years old. ARC SP6B

 

For example, I read a post on diyAudio that mentioned not using metalized polyester caps across a mains supply where it might encounter pulse stress and cause failure.

 

It looks like everything in the audio signal path is polypropylene or polyester. It's my understanding that those don't need replacement.

Edited September 16, 2014 by artto

[tube fanatic]

It depends on what you mean by "better" electrolytics.  You don't want to go too high in value as that will increase the turn-on current surge.  And even doubling the value will not have all that much effect on the ripple (which was always very low in ARC products anyway).  About the only thing you could do if replacing them is to go a bit higher in the working voltage; however, be very sure to measure the original caps and choose replacements which are the same size.  I would leave the other caps alone,  and even 34 year old electrolytics may still be in perfect shape.  As far as putting a cap across the mains, you need to use either a class  X1 or X2 (only if the circuit is fused), or class Y1 or Y2 (which is supposed to fail by opening, not shorting, so a fuse isn't required) suppression cap.  For bypassing the power cord to the chassis, you must use a class Y1 or Y2 cap.  Usually .1uf is a good all around value.  Find someone really good if you decide to proceed- reworking pc boards like those isn't fun!

 

Maynard

Edited September 17, 2014 by tube fanatic

[artto]

Maynard, thanks for your insight.

 

As I dig into this further I'm finding it isn't as simple as just finding a cap of the same capacitance value, voltage rating & type, which in itself is proving somewhat difficult (ie: a 2x1500uF 50VDC electrolytic can/cylinder of the same mounting type).

 

ARC says they stock the parts, but I'm sure they will charge rip-off prices which is one of my motivations for doing this myself.

 

"Find someone really good if you decide to proceed- reworking pc boards like those isn't fun!"  Yes, I'm inclined to agree at this point. ARC is well known for putting together some pretty tricky stuff and this is increasingly appearing to be out of my league. I already talked to two guys about it and neither would to touch it.

 

There isn't anything wrong with the preamp (as far as I know). It doesn't get used very often anymore but it has a glorious (albeit inaccurate) coloration that makes LP sound wonderful which I guess is why a used current model ARC PH6 sells for about the same price as a 34 year old SP6B. That in itself says something which is another reason why I don't want to "trade up".

 

I'm beginning to think it might be best to just leave things alone if you (and some others) think that the electolytics could still be good.  ARC used military grade parts in this thing. The preamp hasn't been exposed to any extreme conditions, spending all of its time in the same room/temperature/humidity conditions. Since it hasn't been turned on for a couple years I'll bring the power up very slowly with a variac & see what happens. If worst comes to worst, I guess I'll just have to send it in to ARC for repair.

 

EDIT: What I meant by "better" capacitors is substituting something like polypropylene or polyester for the electrolytics.

Edited September 18, 2014 by artto

[tube fanatic]

Yes, finding a dual 1500 uf cap of the same size and pinout as the original (or any dual 1500 uf cap for that matter) may be very tricky.  It wasn't unusual for companies to use proprietary components to force those needing service to let them do it, or at least sell them the parts.   As far as aging of electrolytics goes, there are many factors which have to be considered such as how close to the rated voltage it is used, how much ripple current it handles, and internal core temperature (related to ripple current).  If used well below its maximum ratings, life can be absolutely huge (I'm talking even 50000 hours or more).  Preamps obviously don't have the same temperature issues as power amps, so ambient temperature heating of the capacitor (which can further raise its core temperature) is not an issue.  I have no idea what ARC would charge for a total electrolytic replacement, so it might be worth giving them a call to find out.  If they offer a substantial warranty, perhaps it would be worthwhile.  Regarding going to caps with different dielectric materials, I'll tell you that I'm in the camp which doesn't believe that it makes any difference at all.  Others will disagree.  It simply becomes a cost issue, but keep in mind that there's nothing wrong with ordinary electrolytics as you have determined by finding that it sounds glorious to your ears.  And, lastly, after a couple of years of disuse it can't hurt to use a variac and bring it up slowly.  You don't have to go nuts with that though- start out with around 25V, and every couple of minutes increase it by the same amount until you're at line voltage.  One last point, since you have a variac, is to consider using the preamp all the time at the minimum rated line voltage (usually about 105).  That will further increase the potential for long electrolytic life.  Hope all this helps!

Maynard

[hydro_pyro]

Good luck finding a 1500+ uF cap that ISN'T an electrolytic. A polyester cap of that value would need to be the size of a small file cabinet.

If it ain't broke, don't fix it.

[tube fanatic]

The metallized polypropylene caps can be used in place of electrolytics.  Depending on value, as you stated, they can be a bit larger (but the size of a small file cabinet, no).  However, as to life span, they are hugely better than aluminum electrolytics.  I know some amp guys who have successfully installed them in some equipment.  Sonically, it seemed to be a waste of time and money imo.

 

Maynard

[hydro_pyro]

Have you ever priced out a poly cap of that size/value? It's nowhere near practical. That's why they're rarely used in values above 1 uF.

[artto]

I haven't hear back from Audio Research yet. I asked for a quote for repair and a quote for the parts.

 

A friend is going loan me his ESR meter. I guess I'll just do the variac start up, take some measurements & see what I get.

 

I hooked up the Linn/Morch UP4/Decca Jubilee to a solid state phono preamp module in my NAD C390DD & didn't quite like what I heard. I suspect the Decca isn't quite dialed in (reinstalled the Decca on a different arm wand - finicky thing that Morch because of the way the cartridge is mounted - and the Decca too). Should be interesting to hear it again through the SP6B.

[grindstone]

Some ESR meters require the caps to be drained to test in-circuit so maybe heads-up to check that and prevent hurting the meter. Maybe the ARC has means to do that built-in already--no clue.

[djk]

"That's why they're rarely used in values above 1 uF."

 

I've used the 200µF Solen PP for supply bypass before, they weren't all that large, and they weren't all that expensive

[hydro_pyro]

Sure, but power supplies use values many times higher. 1500 uF like the OP mentioned is somewhat on the small side, and only because it's a low-current pre-amp with a small supply.

WHY NOT electrolytic? Good quality components last for decades or more.

[artto]

Sure, but power supplies use values many times higher. 1500 uF like the OP mentioned is somewhat on the small side, and only because it's a low-current pre-amp with a small supply.

WHY NOT electrolytic? Good quality components last for decades or more.

Actually, I believe (not absolutely sure) from the schematic the 1500uF cap looks like its part of preamp's "turn-on delay" circuit.

 

I might be misunderstanding what you're saying regarding "low-current pre-amp with a small supply". The SP6B is a hybrid with a solid state power supply that used something called the "Richardson Effect" to multiply effective capacitance in the power supply. From what I know, the SP6B is pretty formidable in gain and output being able to supply 60 volts rms at the main outputs. Current wise, this may still be low, I don't know. (I can read but not write so to speak). 

[artto]

Some ESR meters require the caps to be drained to test in-circuit so maybe heads-up to check that and prevent hurting the meter. Maybe the ARC has means to do that built-in already--no clue.

Excellent. Thanks for the heads-up 

[erik2A3]

...hurting the meter isn't good; hurting yourself much worse. PSU capacitors can kick hard if they use you as a convenient path to ground. Measure stored charges before you touch anything. If ARC uses a bleeder circuit in that amp, you're in much better shape.