Salvaged caps - what to keep?

[Phil_S]

These came from the Dukane 1F460 that I've been stripping slowly so I can repurpose it. I found one e-cap dated 6652 and some older things, so I guess that makes the amp 1967 issue. I know all the e-caps are too old and have to go, but what about these?

I'm thinking the 160P's, all .047mf 400V have eBay value of $7 each and maybe more on a good day. That's tempting. The 160P's are all in the range of .051-.058.

Those Aerovox's are big. They meter reasonably close to indicated value (.5u and .25u).

What about the domino caps -- El Menco brand? The 5 domino caps that are coded alike meter 39p (should be 50p per schematic) and the other .001mf.

The bumble bees?

If these were yours, would you keep any? Would you use them in a personal (not for resale) build?

Thanks.

[overtone]

I would check them for leakage.

• You will probably find the El Mencos are fine. I always keep them and use them when ever I can instead of the modern silver-micas.
  Bumble bees I retire to guitar tone controls or the Bay.
  No experience with the Aerovox caps.
  The other black ones, I think are Sprague. I found a load in a GA-40, one was leaking too much. I have seen both good and bad opinions on those. I would try them.

tony

[Gibsonman63]

Maybe something fenderish with the dominoes. I have a brownface Pinceton with those on the tone controls. Not sure the values, though.

[Phil_S]

Amateur status showing here...how to I check for leakage out of circuit? I don't have any really fancy tools, just a few basic DMMs and a scope.

Thanks for the input.

[cbass]

Those Sprague's are most likely good they are mylar paper construction and are uasuly not leaky

[overtone]

Amateur status showing here...how to I check for leakage out of circuit? I don't have any really fancy tools, just a few basic DMMs and a scope.

Thanks for the input.

The coupling caps have to block the DC but let the AC through.
I measure as taught by R.G. Keen over a 1Meg resistor to ground:

• DC voltage on one side of the cap then clip the other side to a 1 Meg resistor which is grounded.
  Measure how much DC is coming through at the two ends of the resistor.
  Should be around 1mV to zero for a new one or old Philips mustard.

The last ones I measured were '58 yellow Astrons which were leaking up to 100mA. I carried on using them.

Usually, I test before fitting on a home brew 5F1 mule amp which I listen to music through.
I made a kind of coupling-cap dog house on the top and use it to test caps:

• signal in with 110V DC on it > CAP > (test point) 1M resistor > ground (test point)

Anyone do this differently?

[Phil_S]

Thanks for the primer on cap testing.

I'll think about an easy way to find 110VDC +/- a bit. I've got a filament PT that has a 25V winding. Under no load (30VAC) with a doubler it should give me ~80VDC. Maybe that's the one?

[overtone]

I guess coupling caps in our applications are going to see somewhere between 100 to 200V DC.
Probably better to test higher towards the cap rating to be sure, but 80V DC will also give you an indication of the state of them.

[Phil_S]

I see...I have a small collection of PT's that I've acquired from used equipment and bargain hunting on eBay (minor addiction). Rolling through, I have one that is 114-114 no load (no CT) so I suppose that one will put out around 150VDC with a simple cap input full wave bridge rectifier. That will be the one to use and it is easier to build out than the doubler. I need to find a chassis for it. I must have something here...Thanks again.

[rp]

Those 160Ps are excellent, as good as Astrons but don't tell anyone. I found if they are on spec with just a Fluke for capacitance you're probaly lucky for leakage. Bumblebees always leak however. .05 is perfect for early 50s tweeds, 5C8 is .05 caps straight across.

Build something cool using new cheap caps. Get it 100%. Then swap in the 160Ps. If it sounds perky, with correct voltages an no scratchy pots you're likely golden. Bet you dig the sound a lot more with the old caps. Otherwise sell them to me for my 5C8.

[Phil_S]

I tested them today at 300VDC. DC leakage is between 11mV and 41mV. That is what the Fluke reports across the 1M resistor. I just read this from the meter, is that right? No math to be done? BTW, I was able to find a resistor that metered .999M!

I think, not bad, but curious what you have to say.

They meter on my Amprobe DMM at between about .051mf up to about .058mf. They are supposed to be .047mf. I think this is not too shabby either considering spec was probably +20%, but I don't know.

[overtone]

Sounds good, I would use them like RP suggested.
Try a modern cap too, just to give you a reference.

[Phil_S]

This is good news, bad news. I had thought I would sell them to finance the parts for the new build. Now it seems I need to use them in the build!

Thanks for the help on the cap tester. I needed a quick project to keep the monkey off my back. It did the job.

[overtone]

Glad to help. I think the suggestion from RP to build with new perfect caps, dial it all in, then fit the old ones is really very good.

Way better to use them in an old circuit than sell them.
You don't get enough for these on the open market.

A while back Ian, Gingertube, suggested up to 300mA Dc leakage is tolerable in conjunction with cathode biased but only 100mV DC leakage with fixed bias. I kept my up to 100mV leaking Astrons in the '58 5E3 and it sounds incredible.

tony


Edited: mA to mV. Thanks Structo

[Phil_S]

Strategically speaking, where would you expect the sweet spots to be for the 160P's? I asking about the octal thing and the 6G4A. You can find the 6G4A and I'm posting the all octal matchless thingy.

Dang, what am I going to do with that pile of .047 Mallory 150's that just arrived from Mouser?