Old caps...

[eccerr0r]

I'm on the verge of acquiring old equipment that may have ancient 1950s-1960s electrolytic capacitors in them. It definitely would be best to simply replace the capacitors but I wonder how much luck people have had reforming them if they haven't yet shorted out?

I do have a Sencore LC meter with reform function (it can get to several hundred volts DC, measures leakage and value - too bad no ESR) and wonder what the success rate is - or should I still look into finding new high voltage capacitors to replace?

[goontron]

Re: Old caps...

Ive tried reforming before (im a valve radio geek mind you) some go bang during the process, some work, but then fail very soon anyway... if i wasn't living in central city i would be willing to come down there and see just if they would last after reform.

[electricboyo]

Re: Old caps...

Even more important than the electrolytics, electronic gear from the 1950's and 1960's contains many "paper" capacitors. These are sometimes referred to as "wax paper," "bumblebee," or "black beauty" capacitors. In most cases these will ALL have compromised performance after 50-60 years. DC leakage, even direct shorts, are common. Also, the uF capacitance value often changes with time. For some reason it often increases: A capacitor marked 0.1uF may measure as 0.15uF.

Check "Antique radio forums" as well as "audio karma" forums for lots of info about replacing capacitors in older gear. In some cases, collectors will remove the insides from the old capacitors and install new capacitors inside the old paper tubes. This is called "restuffing."

Regards, EB

[stj]

Re: Old caps...

old electrolytics unless they are soviet are a waste of time because the rubber bung will have hardened and perished.

the old soviet stuff from the 60's uses a plastic bung with a rubber o-ring under the can crimp - they seem a lot better.
but probably still internally half-dead.

[eccerr0r]

Re: Old caps...

This is actually a signal generator.

It looks like the B+ voltage will only be around 110V or so, and 200V modern electrolytics should work. May just steal some from dead PSUs from my junk pile.

And yes it looks like it will have paper caps in them. Two are Y-caps that probably can be replaced with modern Y-caps that I can swipe from other devices. However there are some other coupling caps that I will have to think about how to replace (and/or find a substitute)... I do wonder since B+ is 110VDC or so, whether I need to look for 400V devices or not despite the originals may be rated that high...

(Also I wonder if I changed the selenium halfwave rectifier to a silicon bridge, I should be able to reduce the capacitance needed on the filter cap... hmm.)

[goontron]

Re: Old caps...

(Also I wonder if I changed the selenium halfwave rectifier to a silicon bridge, I should be able to reduce the capacitance needed on the filter cap... hmm.)

no. do replace it though. now selenium rectifiers have a resistance to them, so depending on the design of the circuit you will need to add a resistor.

[stj]

Re: Old caps...

yea, replace the selenium rectifier, they are crap and when they fail they fucking stink!!!!!!

[eccerr0r]

Re: Old caps...

There's no question I should get rid of the Se rectifier if I can, but the question is using a full bridge instead of a half bridge, there are now two diode drops (or maybe I can use 6 diodes and have 4 drops). With double the number of positive going pulses I should be able to halve or at least somewhat reduce the capacitance needed to filter the output... or so goes the theory.

I have plenty of diodes. Not many high voltage caps

[eccerr0r]

Re: Old caps...

Well, closure, I got the sig generator, and decided to use a variac to slowly turn on power. Looks like it's OK. When I opened it to check out the guts, it llooks like it's one of the "later revs" and actually had a silicon diode like a 1N4004 installed already. The 2-section cap is old. The other caps look like they're all fairly new caps and not those large paper caps...

The signal waveforms that get generated are pretty messy, they don't look very sinusoidal, but my scope can and can't pick them up (when it's beyond the horizontal sweep frequency)...

[PeteS in CA]

Re: Old caps...

ecc, since you're going for function (pun not intended) rather than "authentic" restoration, I'd recommend checking whether the rectifier is silicon or germanium. I don't remember whether higher voltage parts were made in germanium, but if it is, I'd recommend replacing it with a 1N400X or 1N540X part of suitable PIV. Also, I'd replace any lytic that is more than 10 years old. Not launching isn't a standard of excellence, and after 15 or 20 years even good 1980s or 1990s lytics are getting long in the tooth.

[eccerr0r]

Re: Old caps...

Yes, mostly want it for function but not going to completely trash it for authenticity - but I doubt I'd go buy a new 2-section cap to replace the aging unit. But I was kind of surprised as it had a black DO-41 case diode installed, but didn't look closely at the numbers on it. I'll need to open it again as for some reason the modulation oscillator doesn't work, but the main oscillator does (as verified by two radios). I only hear 60 cycle hum, indicating perhaps one of the tubes has a heater to cathode short or something. Don't know yet.

This is actually an Eico so theoretically it could have been a kit, but it looks like this wasn't one of the kits, rather it was factory assembled. Everything is riveted in except the main case, including the mount that the selenium rectifier should have been.

As for authenticity - I already removed the silly custom output connector and stuck in a BNC. Of course I kept the old connector in case I need to revert back...

[eccerr0r]

Re: Old caps...

Speaking of the bridge, the main reason for the bridge? I was curious of seeing if the power factor could be helped slightly with a bridge instead of having a halfwave rectifier. However I plugged it into my Kill-a-Watt and saw that this vacuum tube device had a PF of 0.96 - probably due to the near-unity filament heaters being a large portion of power consumption.

The vacuum tube signal generator eats about 10 watts. Not too horrible.

[eccerr0r]

Re: Old caps...

Wonderful. I found that the modulation circuitry doesn't actually work.

The culprit?

Yup, the cap that I thought was OK wasn't. The main filter electrolytic, a dual 20 microfarad 150V unit was indeed dried up.

Here's the current head scratcher: The original diode is indeed silicon, though I couldn't cross reference it, the drop across it was around 0.522V with my multimeter.

The isolation transformer output around 130VAC...

See a bit of irony here? Well, rectifying 130VAC (RMS) will be around 170VDC, so the pulses go well over the working voltage of the capacitor!

I stuck in another cap in parallel with the suspect capacitor and the modulation started working again. I was wondering why the B+ voltage only got to 60V which is just barely enough to get a tube to work, and there were series cathode followers in the design, and 60V just isn't going to cut it.

[goontron]

Re: Old caps...

something is holding B+ down. and it was a dry cap. find a service manual for the thing and test everything!

[eccerr0r]

Re: Old caps...

I just hate testing things that can't be tested in-circuit.

Fortunately, the design is very simple. I hope there are no other problems. I'm not sure what the remainder of the caps are - a bunch of small gumdrop shaped capacitors that have fairly high capacitances versus size. Not sure what material they are made of. The only known 'ugly' components are that electrolytic capacitor and possibly the vacuum tubes themselves.

Now it's just time to tweak it... I wonder if I should get some tubes for it as spare...

[goontron]

Re: Old caps...

I just hate testing things that can't be tested in-circuit.

Fortunately, the design is very simple. I hope there are no other problems. I'm not sure what the remainder of the caps are - a bunch of small gumdrop shaped capacitors that have fairly high capacitances versus size. Not sure what material they are made of. The only known 'ugly' components are that electrolytic capacitor and possibly the vacuum tubes themselves.

Now it's just time to tweak it... I wonder if I should get some tubes for it as spare...

post some pics and lets see if valvedubstep from the antiqueradio forms can help.

[eccerr0r]

Re: Old caps...


Looks ceramic-ish but it couldn't be...

...could it?

[goontron]

Re: Old caps...

paper. seen them once before in a GE radio.

[Sparkey55]

Re: Old caps...

old electrolytics unless they are soviet are a waste of time because the rubber bung will have hardened and perished.

the old soviet stuff from the 60's uses a plastic bung with a rubber o-ring under the can crimp - they seem a lot better.
but probably still internally half-dead.

I have seen just as many USSR electrolytic caps in the plastic bung fail as American made ones too. The problem is that the quality of the paper was not as good as needed to be to stand up to the decades of service. The paper was not acid free and would leech into the electrolyte and rot the rubber o-rings with time regardless of the sealing methods at the time.

[eccerr0r]

Re: Old caps...

Ah. someday I should get one to tear apart to see what's inside, but at the moment things are working about as expected... Though I haven't decided yet whether I should gut the old electrolytic and put two modern caps inside or not.