|
Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets. |
|
Thread Tools |
5th Jan 2022, 7:54 pm | #1 |
Hexode
Join Date: Apr 2021
Location: Crawley, West Sussex, UK.
Posts: 442
|
Old caps and apparent measurements?
Having got into restoring older equipment (having worked on more modern valve stuff for quite a while), I have observed up the generally-accepted view that many 70-year old caps of the electrolytic or wax & paper construction are probably best assumed 'failed' or likely to.
I've just pulled two caps from a radio I'm restoring: one a 2µ2 200WVDC elec. and the other a 0µ05 350WDC Dubilier 'Type 460' wax/paper. Out of curiosity I wound up to 160V DC (as I happen to have a current-limited DC PSU) across each and measured the steady-state leakage. The elect settled at ca. 150µA and the wax/paper at 30µA, compared with a new Kemet 4µ7 450 WVDC elec of ca. 15µA, and a CDE 0µ047 400WVDC of < 1µA. So, I know neither old ones are 'good' but they don't seem to me to be all that 'bad'. I only gave each one 10 mins soak however. FWIW, I tested the parallel resistance of the Dubilier waxy and it was 24kΩ at 10kHz (cf. 160kΩ on the CDE new one) . The ESR of the elec was 6Ω (vs. 2Ω for the new Kemet). I personally have no inclination to trust a 1955 electrolytic in HT duty, almost irrespective of it's apparent measurements, but without getting too deep into this, I'd be keen to understand what expertise on here thinks about my findings on this and the waxy? |
5th Jan 2022, 8:07 pm | #2 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,998
|
Re: Old caps and apparent measurements?
Electrolytics are intrinsically a bit leaky - they need to use a smidgin of energy to keep the surface-etching of the foils active. A few tens of Microamps leakage in a smoothing circuit is not going to be an issue but a few tens of Milliamps can lead to heating and thermal runaway. Electrolytics make one hell of a mess when they explosively-disassemble.
Waxies etc - IMHO any detectable leakage is justification for replacement: a few Microamps of leakage in a coupling or AGC-line application [where those microamps may end up across a Megohm or two of resistance] and you can have serious problems. Personally, I don't generally bother testing any parts I suspect: in the time it takes to lift-one-end-and-test a component I can lift both ends and replace the suspect part with something that is better than the original and will be good for the _next_ 70 years.
__________________
I'm the Operator of my Pocket Calculator. -Kraftwerk. |
5th Jan 2022, 8:30 pm | #3 |
Dekatron
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,934
|
Re: Old caps and apparent measurements?
I recall starting to re-build my 1940 HRO. The big electrolytics had completely dried out and were light as a feather. Some of the paper caps just uncoiled like some sort of tinfoil Christmas decoration when the two leads were no longer holding them together and so too, some of the resistors fell apart when the leads were no longer offering structural strength.
B
__________________
Saturn V had 6 million pounds of fuel. It would take thirty thousand strong men to lift it an inch. |
7th Jan 2022, 9:48 pm | #4 |
Rest in Peace
Join Date: Apr 2008
Location: London, UK.
Posts: 2,508
|
Re: Old caps and apparent measurements?
As per @G6Tanuki, the electrolytic will probably still do its job, because any sane circuit design will allow for a bit of leakage from any normal electrolytic, and most applications will tolerate as much leakage as the cap itself can endure without self-destructing. Your 150μA might well drop to a genuinely acceptable figure after a few hours. I would not judge a cap leaking that amount after only 10 minutes, I would leave it on.
OTOH a paper capacitor that leaks 30μA is completely unacceptable in many circuit positions and could cause complete malfunction. That equates to an insulation resistance of only 5 MΩ which is in the same order as a lot of other circuit resistances, whereas hundreds of megohms or better would be expected of a new part. Of course there are equally many circuit positions such as RF stage cathode bypass where a few kilohms is good enough and megohms completely trivial. That is why the Bush DAC10 in my dining room still works perfectly with all but a couple of its dozen original waxes. The ones that can't be allowed to leak have been changed. |
9th Jan 2022, 5:31 pm | #5 |
Dekatron
Join Date: Feb 2003
Location: Cottingham, East Yorkshire, UK.
Posts: 5,761
|
Re: Old caps and apparent measurements?
I don't 'do up, mend, get going, or mess about' with radios. I only fully restore them, mostly 40s and 50s valve radios. If they don't have enough appeal, I don't bother with them. Others may have a different outlook and want to do as little as possible, maybe focusing more on cosmetic aspects, which is fine by me - it's their radio and wouldn't presume to dictate what they do with it.
Where waxy paper capacitors are concerned, sure, if the audio coupling capacitor is replaced and if fitted, the RF capacitor across the mains input, the radio might work after a fashion, and just because a capacitor says '350V working', doesn't mean that it has anything like that voltage across it, or that it's in a critical position. But I routinely replace all waxy capacitors with 630V polypropylene ones. Why? Because to test them for capacitance and leakage at their rated voltage means lifting one end, and the test will almost always show that the capacitors is past it's best, if not completely jiggered. By that time, it could be out of the radio and a replacement one soldered in. I don't subscribed to the view that to routinely replace paper capacitors on radios which are 70 - 80 years old amounts to a 'shotgun' approach, or 'compulsive capacitor replacement disorder' as I've heard it described. Experience over the years tells me otherwise. I should also add that I change any Hunts 'Mouldseal' capacitors on sight and the pitch covered Philips ones, as they invariably exhibit high leakage and out of spec capacitance. I'm presently restoring a 1948 Murphy A124 in which - among other things - I've replaced all the waxy paper capacitors, and a couple of metal clad ones. Purely out of curiosity, I've tested each one for capacitance and leakage. The results are attached. Some were in roles such as HT decoupling, others in non-critical parts of the circuit. Maybe if I'd left them alone the radio would have worked after a fashion, but why would I want to do that? The radio wasn't designed to have a parallel resistor across each capacitor. I have a homebrew electrolytic capacitor reformer, (designed by forum member Stef, 'Saddlestone Man' and published in Radio Bygones and the BVWS Bulletin), which serves me well. Sometimes smoothing & reservoir capacitors often do reform, but not always. In this radio I ended up re-stuffing the cans. Invariably the audio output valve cathode bypass capacitor - often 25uF/25V - has had it, and needs replacement. In this particular radio, six of the 31 resistors were also well out of spec so I replaced them. Hope that's of interest.
__________________
David. BVWS Member. G-QRP Club member 1339. |
9th Jan 2022, 6:31 pm | #6 | |
Dekatron
Join Date: Sep 2011
Location: Rugeley, Staffordshire, UK.
Posts: 8,831
|
Re: Old caps and apparent measurements?
Quote:
__________________
A digital radio is the latest thing, but a vintage wireless is forever.. |
|
9th Jan 2022, 6:44 pm | #7 |
Moderator
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,944
|
Re: Old caps and apparent measurements?
I agree with David in #5. It's very difficult to test old wax caps in any meaningful way, so it's more trouble than it's worth to try to do it. If you don't change them, they can produce all sorts of odd symptoms which make fault finding difficult.
Old electrolytics can indeed often be restored by reforming, but I tend to think they're also more trouble than they're worth. It does depend on the purpose of the restoration though, and the attitude of the restorer. |
9th Jan 2022, 7:13 pm | #8 |
Dekatron
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
|
Re: Old caps and apparent measurements?
Another often ignored variable affecting leakage is the operating temperature which can be much higher than ambient under the chassis. I have yet to test a vintage paper capacitor that would meet the original very low leakage specification.
|
9th Jan 2022, 7:54 pm | #9 |
Pentode
Join Date: Dec 2009
Location: Chatteris, Cambridgeshire, UK.
Posts: 176
|
Re: Old caps and apparent measurements?
A striking example of this is in a Quad II amplifier. The coupling capacitors are clamped directly to the KT66 bases and after a short time, will become quite hot. This is exacerbated with the use of more modern valves with the anode structure sitting much lower (and nearer to the base) than in an original KT66.
__________________
Andrew |