Replacing old paper in oil capacitors

[Michael Maurice]

I'm restoring a rather old Dynatron radiogram and have to replace the smoothing capacitors which are as the pictures below.


I'm not sure what they're made of but I presume paper in oil. I'm sure they sahould be replaced, but what with? They are bolted to the chassis.


Should one just open it up using a hacksaw, remove the innards and fit new capacitors?



What should I use I was thinking of these in the appropriate value


https://uk.rs-online.com/web/p/polyp...itors/1719163/


or


https://uk.rs-online.com/web/p/polyp...itors/8829310/


Can someone guide me on this. I was thinking of doubling and quadrupling these to make either 2uF or 4uF.

[Argus25]

I'm surprised those original caps have failed, usually that sort of PIO cap is super reliable.

You can replace them with practically any kind of poly cap. Probably because of the space inside those grey canisters you could even fit 600v rated ones of the same value, or increase the capacity and keep them 400V rated. If they are just for for an HT smoothing application you could also use electrolytic caps. I tend not to if a poly will fit.

If you were going to re-stuff them, probably they are better opened at the base under the clamp area, so they will still look cosmetically good afterwards.

If you want PIO replacements, Surplus Sales Nebraska has a good selection, or perhaps some of the Russian ones on Ebay, I find them very good quality as most are mil spec parts.

[The Philpott]

Are they actually, u/s Michael? I think of those ones as hard as nails.

Dave

[turretslug]

It's possible to get capacitors of this value and working voltage as polypropylene motor-run types. Farnell certainly stock Ducati made types rated at 450VAC, 630VDC, RS almost certainly do too, possibly RS-branded. They are stud-mounting and surprisingly compact for non-polarised high voltage capacitors, yet rated for arduous motor current applications.

I really wouldn't be inclined to open the old ones up- it's difficult to know if the unpleasantly hazardous PCB oils are present inside.

[Argus25]

When you open them up, make a point to wear gloves and don't drink the oil.

[Craig Sawyers]

Originally Posted by The Philpott

Are they actually, u/s Michael? I think of those ones as hard as nails.

Dave

Seconded. The hermetically sealed ones like that rarely fail. Which means that I have not seen or heard of one fail.

As an example, I'm just restoring an Advance high voltage supply (a valved one) made in 1964. It has two constant voltage transformers with monster resonating capacitors made by TCC. 20uF and 4uF. the 20uF one measures 21uF with an ESR of 0.4 ohms and no measurable leakage, and the 4uF one measures 4.05uF with an ESR of 0.6 ohms and no measurable leakage.

So absolutely no pressure to replace them.

Craig

[Michael Maurice]

I haven't actually tested them, I will a bit later, I thought those capacitors should be changed because of their age.

Maybe I'm wrong.

[Craig Sawyers]

Tubular paper capacitors with axial leads - yes, replace on sight. Particularly, but not uniquely, Hunts. But the metal cased, hermetically sealed paper/oil bricks rarely have a problem.

Craig

[Al (astral highway)]

These metal-cased paper in oil types always seem to me to be ageless. Def worth testing before replacing !

[Argus25]

Originally Posted by Michael Maurice

I haven't actually tested them, I will a bit later, I thought those capacitors should be changed because of their age.

Maybe I'm wrong.

These capacitors even today, will be better than any modern plastic part you would attempt to replace them with. They are made of very long lasting and stable materials. The only reason these are not made today in great numbers are the size of them and the cost. Plastic film capacitors replaced them only for those reasons. Plus, unlike modern capacitors, these ones have a fire risk that is so close to zero, you can regard it as zero. So it will be a really big surprise if they have failed.

[GrimJosef]

And ... here's the exception that proves the rule. These canned PIOs are usually very good. But here's a Beam Echo DL7-35 amp that I currently have in for service. I've taken the transformer cover off. The sharp-eyed amongst you might be able to read that the reservoir cap, behind the right-hand EL34, is a Hunts PIO 4uF. You'll have to take my word that the capacitance meter leads are slipped through a hole in the chassis and wired across the cap, one terminal of which has been disconnected and is therefore floating. The meter reads 1.62uF. Having opened a couple of PIOs in the past I've found they're often divided into paralleled sections inside. I imagine that this one has two sections and one has failed open-circuit. So while they're often good, they sometimes aren't. If they test OK though then I'd certainly retain them.

Cheers,

GJ

[CambridgeWorks]

I've seen a TCC 8uf 600 or 800v block with that very problem, read about 4uf. I guessed maybe 2 in parallel and a wire had come adrift.
Rob

[Argus25]

Originally Posted by GrimJosef

And ... here's the exception that proves the rule.
Cheers,

GJ

That is pretty rare. Also it sounds more like a manufacturing fault with poor welds to the foils or interconnects, rather than a problem intrinsic to the life cycle & design of the capacitor. Maybe it was made on a Friday afternoon. I don't know what their life is, but I wouldn't be surprised if many make it well past 100 years. Of course a part like that is incompatible with modern life-cycling and re-cycling philosophy.

Apparently Henry Ford had researchers go around to car scrap yards to examine the components in the end of life Fords and find the components that were still working. Then he knew he was making that part too well.

[FrankB]

I used to get the PIO caps by the hundredweight at the Hamfairs here. Nobody wanted them..until the audiofools discovered them.
I have seen some fail, often spectacularly, and take out the power tx & choke in a linear amp. PS I got my hands on. I think they protected the 29 cent fuse quite well.