4P power supply C33/C32 replacement

[mlaferriere]

Hello Folks.

I picked up a 4P from eBay that worked fine for about half an hour when C32 blew on the power supply board. I would also like to replace C33 at the same time. This is one of the Tandy power supply boards as opposed to the Astecs.

I did come across http://www.vcfed.org/forum/archive/index.php/t-43594.html but was concerned about the physical size difference and those part numbers where for the Astec supply even though ECQ-UAAF224K should match up with my C32.

Does anyone have replacement part numbers for the Tandy power supply in a 4P?

Thank you.

 

[Chromedome45]

Well if you get the replacement caps and they are a little wider you can always bend the leads inward some till they match the holes and solder them in. Have done it myself. Might stand up a little taller but no big deal. Just my 2 cents worth!

 

[Chuck(G)]

If you use a modern polypropylene capacitor, it's likely to be smaller, rather than larger. Note if the original has an X/X1/X2/Y marked--if so, it's a "safety" cap and you want to replace it with one of a similar safety rating. (e.g., C33 has an "X" safety rating mark). Replacing a non-safety cap with a safety one is perfectly okay.

 

[NF6X]

I'm at work right now, so I don't have access to my records or my 4P. But I'll try to comment from memory.

Yes, definitely replace C33 at the same time as C32. There may be a third replace-immediately cap in there, too. I recall that there is one hiding in another corner on some of the Astec supplies in the Model 12 and Model 6000, but I don't remember if the 4P supply also has a third cap, and I don't think I've examined any non-Astec Tandy supplies yet. I recommend always replacing those yellow Rifa paper dielectric safety-rated caps on sight before applying power.

Most of the Astec supplies I've encountered in TRS-80 computers already had 3-4 holes for each of those yellow incendiary capacitors to support different lead spacings. But even if the caps you can find don't have exactly the same spacings as any of the hole pairs, close enough is good enough if you buy caps with long enough leads. Just avoid caps with very short leads that thus need to be mounted right against the PCB with holes at the precise spacing of the leads. I picked out Panasonic safety-rated film caps that had nice long leads, allowing them to be installed in spots where the holes weren't at the matching spacing.

Those Panasonic part numbers I listed in the thread that you linked to have been suitable replacements in all of my TRS-80 power supplies so far, including models II, 4, 4P, 12 and 6000. But I think that all of mine have had Astec supplies. There's a pretty good chance that they will also work OK in your Tandy supply, but I can't verify that yet.

 

[mlaferriere]

"Yes, definitely replace C33 at the same time as C32. There may be a third replace-immediately cap in there, too. I recall that there is one hiding in another corner on some of the Astec supplies in the Model 12 and Model 6000, but I don't remember if the 4P supply also has a third cap, and I don't think I've examined any non-Astec Tandy supplies yet. I recommend always replacing those yellow Rifa paper dielectric safety-rated caps on sight before applying power."

I did go over the Tandy power supply once or twice in good light. The larger electrolytics all appear clean with no swelling. I might ESR some of them Friday when the replacements for the RIFA's arrive just to make sure. I got distracted the other day googling about the RIFA capacitors. It seems the audio philes have had the same issues with RIFA's going solar sometimes even in a device that was not even turned on.



"Most of the Astec supplies I've encountered in TRS-80 computers already had 3-4 holes for each of those yellow incendiary capacitors to support different lead spacings. But even if the caps you can find don't have exactly the same spacings as any of the hole pairs, close enough is good enough if you buy caps with long enough leads. Just avoid caps with very short leads that thus need to be mounted right against the PCB with holes at the precise spacing of the leads. I picked out Panasonic safety-rated film caps that had nice long leads, allowing them to be installed in spots where the holes weren't at the matching spacing."

Ugh, I should have waited for the email. One of my concerns was ordering something electrically similar but physically a different size. Even with the long leads I would like to keep it as close in to the board as possible and this was one of my concerns. I haven't removed the old caps and will do that when the new ones come in on Friday. The ones I wound up ordering I think are similar (or identical to your Panasonic's) from Digikey

C32 .22μf http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail&itemSeq=200030183&uq=636020474327819893
C33 10000PF http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail&itemSeq=200030190&uq=636020474327819893

I think I found that third capacitor. I didn't give it a second look earlier as it didn't have the traditional cracked plastic RIFA case. It is C24 in my Tandy supply .1K 250. Do they use Nano farads on these...? .1K nF would match the third capacitor from your comment. I think I am going to have to just order it and let it sit in the in box until I need other stuff as well since I need to pay $7 or so for shipping each time from Digi-Key or Mouser.

BTW, are the newer KEMET equivalents of the RIFA's any good or is it just a good best practice to avoid the metallized paper and replace with metallized polypropylene?



Thanks!

 

[NF6X]

It seems the audio philes have had the same issues with RIFA's going solar sometimes even in a device that was not even turned on.

That makes sense to me. That particular type of capacitor is used for EMI (electromagnetic interference) filtering on the AC line, and EMI filters are sometimes placed ahead of the main power switch such that they always see line voltage even when the equipment is turned off.

The ones I wound up ordering I think are similar (or identical to your Panasonic's) from Digikey

C32 .22μf http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail&itemSeq=200030183&uq=636020474327819893
C33 10000PF http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail&itemSeq=200030190&uq=636020474327819893

Those are volatile links to your search session, so I wasn't able to see what specific caps you ordered.

I think I found that third capacitor. I didn't give it a second look earlier as it didn't have the traditional cracked plastic RIFA case. It is C24 in my Tandy supply .1K 250.

C24 doesn't look like any sort of capacitor that I've learned to distrust, so I'd leave it alone unless it acts up.

Do they use Nano farads on these...? .1K nF would match the third capacitor from your comment.

I've never seen "kilonanofarads" used. I suspect that it's a .1 microfarad cap, that the "K" means something else, and it's just a coincidence that 0.1knF has the same numerical value as 0.1uF.

I think I am going to have to just order it and let it sit in the in box until I need other stuff as well since I need to pay $7 or so for shipping each time from Digi-Key or Mouser.

It doesn't hurt to have some spares sitting around, because you're likely to keep encountering very similar Rifa caps in lots of different 80s era computers. I keep the values I've encountered so far in my Radio Shack computers on hand, and I've generally been able to de-Rifa new gear as soon as it arrives without needed to wait for another order.

BTW, are the newer KEMET equivalents of the RIFA's any good or is it just a good best practice to avoid the metallized paper and replace with metallized polypropylene?

I have no specific information about new-production Kemet metallized paper caps. But at the risk of talking out my back end, I'll say that it's my expectation that they're the same design with the new parent company's name printed on the side, so I would expect them to fail in the same way after 20-30 years unless they have improved the composition of the outer plastic case. As I understand it, metallized paper caps entirely depend on the integrity of their outer case to keep out moisture. I think that the visible cracking of the cases in most of the 80s era Rifa caps I've seen lets atmospheric moisture seep inside, where it causes the paper dielectric to break down under line voltage. I've read that polypropylene film dielectric is much, much less susceptible to absorbing moisture, so I expect that poly film caps are much less likely to fail in the same manner even if their outer case sealing fails. Now, I don't know if poly film caps have their own failure mechanisms that could bite us all in the future, but the paper dielectric caps are the only ones that I have personally learned to completely distrust so far.

BTW, I forgot to congratulate you on your new 4P. They are neat machines, and I hope that you have lots of fun with yours.

 

[pski]

I have been replacing the filter caps in my TRS-80 power supplies, including MI/II/III/4/12/16/16B/6000, with the identical modern "RIFA" caps from Kemet.

Like these: https://www.digikey.com/product-detail/en/kemet/PME271YB5100MR19T0/399-7501-1-ND/3459487

So far, they have worked flawlessly. I don't know if they have improved the build quality of these caps since circa 1980ish, but I'm assuming they have learned a bit from the past. However, if they start failing in 30 years, well, I'll be happy to be alive in my 80s with working TRS-80s and a project list to replace them!