Replacing "micamold" caps in oscillator section
 

In a very slow repair of an AR88 I've pulled out various brown shaped bakelite things all marked micamold and with colour coded values. The dot pattern seems to suggest that they are mica (see attached chart) but the leakiness and general information around here suggests they are actually wax-paper in a bakelite case.

I have replaced most of them with 630V film caps, which is clearly fine for decoupling AVC, screen grid etc. but now have come to the ones in the oscillator section which have quite large values (for mica) of 525pF, 1550pF, 2700pF, 3000pF, 3900pF. They are in the resonant circuit.

Now I imagine the temp coefficient of these will matter as usually it is chosen to be negative to compensate for the +ve temp-co of the coil. I could replace by mica rated at 500V (it seems very hard to find 1kV at a non-astronomical price) which has a temp co of around +50ppm / deg. But since the originals may not actually be mica this might not have the desired effect.

Maybe my expectations of temperature stability are too much for the era, but there's a nice stable HP synthesizer ready to check once the set works.

I'm wondering if anyone who has trod this path before me has any advice? I was always given the advice to not touch or change things in a VFO, but with the age of this set clearly this is not advice I can heed.

Re: Replacing "micamold" caps in oscillator section
 

You do not state if your receiver is an AR88, AR88D or AR88LF.

In the AR88D, capacitors C13 and C14 isolate the oscillator grid and anode from the tuned circuits.

If C13 and C14 are not leaky, it may not matter very much if the other capacitors are leaky since they have no DC voltage across them.

Re: Replacing "micamold" caps in oscillator section
 

I've opened some and found paper. 'Mr Carlson' has recently put videos of restoring a CR88 on youtube and swears they're really mica. But the CR88 is a post-war variant. I suspect paper got used in the war years and maybe reverted to mica by the time of the CR88.

I just don't trust the little beggers and swap on sight.

Oh, the mechanism of leaks can be effective creators of significant noise.

David

Re: Replacing "micamold" caps in oscillator section
 

But only if the exact capacity is not important. The dielectric constant of water is sufficiently different that leaky capacitors will often be significantly out of tolerance.

Re: Replacing "micamold" caps in oscillator section
 

If it says 'Micamold' on them, they're moulded paper. They were used extensively in the war years in military equipment, and that fact was used as an endorsement of reliability. The 1944 Micamold Radio Corporation advert in the August 1944 advert in Electronics Industry stated:

Quote:

'Molded versus Mica: Wartime equipment has definitely proven the reliability of Micamold Molded Paper Capacitors to function satisfactorily in by-pass, coupling and filter applications above .001 mfd. As strategic mica is scarce, Micamold Molded Paper Capacitors not only fill the breach but materially assist in maintaining the flow of equipment to the Armed Services'.

End quote.

That scarcity of mica would almost certainly have continued into the late 1940s.

In my view, any capacitor which looks like a Micamold paper cap is one, unless it has 'Mica' stamped on it. That doesn't in itself mean that it's duff, but after seven decades, it's dubious. A leakage and capacitance test would confirm that, one way or another.

As to the 1,000V rating, are the capacitors in locations in the oscillator section where they have DC Voltage on them which is anywhere near that high, or even a tenth of that voltage? If not, 500V ones would be fine.

Hi-Fi collective stock silver mica up to 10,000pF, and price wise, your highest value of 3,900pF would be the most costly, at £4.30 for a +/- 5% one:

https://www.hificollective.co.uk/cat...55.html?page=1

Or if you wanted 1% 500V ones, which are 'audio grade' (and hence, sprinkled with fairy dust), a 3,900pF one would be £13.00:

https://www.hificollective.co.uk/cat...rs.html?page=1

I wouldn't have any faith in Mr Clarkson 'swearing that they're mica'. The definition of faith is 'a firm unshakable belief in something without proof or evidence'. If he cracked open a few and found that they were indeed mica, then yes, he would have underpinning evidence to support his, as yet, unsubstantiated opinion.

This earlier thread might be of interest:

https://www.vintage-radio.net/forum/...10#post1189410

Every success in your ambition AR88 restoration Alan.

Re: Replacing "micamold" caps in oscillator section
 

So far as I know Micamold made paper and also mica capacitors.

Lawrence.

Re: Replacing "micamold" caps in oscillator section
 

Take one out and test it for value and leakage at working voltage - if it tests good it doesn't matter what it is.

Some of the micamolds in a BC348 were a near dead short with boiling wax spitting out the ends.

Re: Replacing "micamold" caps in oscillator section
 

Yes, "Mr. Carlson" is pretty sound most of the time (in the limited number of his videos that I have seen), but being from a later generation, has picked up a few "latter day myths", & "gets the bull by the backside" sometimes!

I remember that "micamold" caps were a bit "suss" back in the day & were often amongst the "replace on general principle" components.
I always thought they were just very crappy mica caps.

In Oz, there was always a distinction made between just plain "mica" capacitors & "silver mica" ones, with the former regarded as less reliable &/or accurate.

Re: Replacing "micamold" caps in oscillator section
 

So firstly on the model; there is no LW band so not an AR88LF. Unfortunately the serial number plate appears to have been removed at some point in the past (to fit a large round antenna connector). Dials are all yellow (no black bands).

As for AR88 vs AR88D, the plate in place of the S meter says AR88 and I don't find any other identifying marks on the set. The original chokes for HT are round rather than "rounded corner square".

Oscillator is shunt fed from the triode (so as suggested no HT on tank caps). Seems to have a mix of transformer and capacitor feedback.

But every single decoupling cap labelled "micamold" has been either faulty (sometimes visibly bulging, weeping) or removed by a previous owner. This rather suggests replace on sight might not be a bad idea.

Thanks for Hificollective link, they do indeed have them all!

Re: Replacing "micamold" caps in oscillator section
 

For decoupling, I am happy to use the yellow plastic Vishay capacitors that are widely available; I'd only go with Mica ones [I'm fond of the maroon resin-dipped ones made by CDE/Cornell-Dubilier] in applications related specifically to tuned-circuits, where thermal stability may be an issue - but honestly, a snail with a couple of drawing-pins stuck into it would probably be more thermally-stable than a 70-year-old waxed-paper capacitor!

Re: Replacing "micamold" caps in oscillator section
 

Just to note again....Micamold made capacitors with a mica dielectric as well as paper dielectric, the lower value ones are most likely to be mica dielectric.

Ie: The Micamold type PO were silvered mica, those types were used in lot's of tuned circuits in various lumps of WW2 kit, eg: the BC348 comms. receiver I had used them in tuned circuits and for oscillator coupling etc.

Also note that Micamold made resistors that used the same type of encapsulation as their molded capacitors, those molded resistors also used coloured dots to identify their nominal value.

Lawrence.

Re: Replacing "micamold" caps in oscillator section
 

My understanding is that the micamold range was specifically created to deal with a probem during WW2 that the USA was very short of mica so an alternative was needed.

In my experience, every micamold capacitor I have encountered was badly leaky.

Re: Replacing "micamold" caps in oscillator section
 

The quote suggests that every micamold value above 1nF is paper, so the values below 1nF are somewhat likely to be mica types.

Re: Replacing "micamold" caps in oscillator section
 

I have done several AR/CR/SC-88's and this is exactly my approach and it seems to work well. Have not tried the snail approach;D

Double check all of the resistors while you are in there. You may find many that are near or out of tolerance.

Tom
W3TA

Re: Replacing "micamold" caps in oscillator section
 

Which would be just one of those listed by Alan - namely, 525pF:

"... have come to the ones in the oscillator section which have quite large values (for mica) of 525pF, 1550pF, 2700pF, 3000pF, 3900pF. They are in the resonant circuit".

Would they have used four Micamold paper caps and just one Mica? I think not, and on the balance of probabilities, they'll all be Micamold. To replace all except the 525pF one would save £2.00 off the shopping list. As I said in my earlier post, there is always the option to test them for leakage and capacitance, and if OK, leave them in place, especially if in awkward positions to replace.

But it's not my radio - it's Alan's. He's already replaced the capacitors in less critical roles in the radio, and it's for him to decide whether to leave those in the resonant circuit of the oscillator, which he's clearly concerned about or he wouldn't have started the thread.

Re: Replacing "micamold" caps in oscillator section
 

As the capacitors in question are oscillator padders, necessarily of reasonably precise (and often non-standard!) values to ensure good tracking- particularly with a relatively highly-resolving tuning scale that was also expected to stay accurate with temperature and time (how long does a war go on for....), I'd expect these particular capacitors to be of a dependable and precise-tolerance type- i.e. a likely exception to any broad "paper dielectric above 1nF" rule that might otherwise be applicable to mere decouplers.

Re: Replacing "micamold" caps in oscillator section
 

As another example, Micamold types employed in WW2 IFF kit RC-150, RC151 etc.

Source: Parts list in training manual TM11-1517.

Some extracted info (converted to pF and uF):

40pF, 75pF, 100pF, 180pF, 200pF, 250pF...All listed as Mica, Micamold, Type O.

50pF, 100pF...Both listed as Silver Mica, Micamold, Type PO.

0.001uF, 0.002uF, 0.005uF...All listed as Mica, Micamold, Type W.

0.01uf, 0.05uF, 0.1uF...All listed as Molded Paper, Micamold, Types 340-21, 342-33 and 345-21 respectively.

https://www.radionerds.com/images/2/2b/TM_11-1517.pdf

Types O, PO, and W are of the molded type.

Make of it what you will.

Any errors then guilty as charged.

Lawrence.

Re: Replacing "micamold" caps in oscillator section
 

In a provisional EMER document for for the AR88 I'm looking at, it doesn't give any indication as to the construction of the oscillators padder capacitors (525pF through to 3,900pF) or the two coupling capacitors to the audio output valve control grid (4,700pF each) But in a later EMER document for the AR88D they are all listed as Mica.

Lawrence.

Re: Replacing "micamold" caps in oscillator section
 

Paper seems like a sub-optimal choice for padders so they might all be mica.

Maybe to make sure, measure the capacity and the leakage and leave them alone if in spec.

Re: Replacing "micamold" caps in oscillator section
 

Thanks for all the helpful advice. The micamolds have all gone thanks to genuine mica replacements from HiFi Collective following the link above.

Final rationale was that I had removed some parts to get at the decouplers in the oscillator enclosure so it made sense to go in once and then realign rather than have to repeat the alignment when I find the inevitable dodgy capacitor (I took the opportunity to replace a few Rs as well which were well outside their tolerance).

I have to admit that they are quite a pain to get at being buried under the bandchange switch, but it would be unfair to complain that RCA didn't make things easy for the 80ish years service on this set.

Adding some mains power to the set doesn't seem quite so distant day any more...