24 Nov
2013
24 Nov
'13
5:43 p.m.
Evening,
What're the components most likely to fail in a BA23 PSU? I have one that
keeps pouring smoke out of it...yet I can never see a component that's
burnt, punctured, or other stuff like that. I also never hear an arc or
anything exploding. The PSU also squeals when powered down.
Yes. It seems to STILL work despite smoke coming out of it. Haven't
confirmed after today's magic smoke release though. I'll test again once
I get a warmer day.
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
24 Nov
24 Nov
6:38 p.m.
On Nov 24, 2013, at 17:43 , Cory Smelosky <b4 at gewt.net> wrote:
What're the components most likely to fail in a
BA23 PSU? I have one that
keeps pouring smoke out of it...yet I can never see a component that's
burnt, punctured, or other stuff like that. I also never hear an arc or
anything exploding. The PSU also squeals when powered down.
What does the smoke smell like? If it smells like burning paper, then look for blocky
translucent yellow EMI filtering caps made by Rifa. I don't have personal experience
with a BA23 yet, but I recently discovered the joy of exploding Rifa metallized paper
dielectric caps in a TRS-80 Model 12. I seem to recall that the same type of cap was one
of the things that blew out in a VCF thread about restoring a PDP -11/44. I now consider
those Rifa caps to be replace-on-sight items, and I'll be replacing them with
metallized polypropylene caps instead of *^!@$#&# metallized paper ones.
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
7:19 p.m.
On Mon, 25 Nov 2013, Mark J. Blair wrote:
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
On Nov 24, 2013, at 17:43 , Cory Smelosky <b4 at gewt.net> wrote:
Will do. It's certainly lot a metallic or plastic smell...I know those
smells quite well and it was nothing like those. ;)
What're the components most likely to fail in
a BA23 PSU? I have one that
keeps pouring smoke out of it...yet I can never see a component that's
burnt, punctured, or other stuff like that. I also never hear an arc or
anything exploding. The PSU also squeals when powered down.
What does the smoke smell like? If it smells like burning paper, then look for blocky
translucent yellow EMI filtering caps made by Rifa. I don't have personal experience
with a BA23 yet, but I recently discovered the joy of exploding Rifa metallized paper
dielectric caps in a TRS-80 Model 12. I seem to recall that the same type of cap was one
of the things that blew out in a VCF thread about restoring a PDP -11/44. I now consider
those Rifa caps to be replace-on-sight items, and I'll be replacing them with
metallized polypropylene caps instead of *^!@$#&# metallized paper ones.
25 Nov
25 Nov
4:08 p.m.
The wiring harness between the PSU and the backplane was known to overheat
and burn on highly loaded systems. Because the original harness had wires
of varying length, current was not evenly shared between wires. A
replacement harness had all the wires the same length.
On Sun, Nov 24, 2013 at 10:19 PM, Cory Smelosky <b4 at gewt.net> wrote:
On Mon, 25 Nov 2013, Mark J. Blair wrote:
look for
blocky translucent yellow EMI filtering caps made by Rifa. I don't
have personal experience with a BA23 yet, but I recently discovered the joy
of exploding Rifa metallized paper dielectric caps in a TRS-80 Model 12. I
seem to recall that the same type of cap was one of the things that blew
out in a VCF thread about restoring a PDP -11/44. I now consider those Rifa
caps to be replace-on-sight items, and I'll be replacing them with
metallized polypropylene caps instead of *^!@$#&# metallized paper ones.
Will do. It's certainly lot a metallic or plastic smell...I know those
smells quite well and it was nothing like those. ;)
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
On Nov 24, 2013, at 17:43 , Cory Smelosky <b4 at gewt.net> wrote:
> What're the components most likely to fail in a BA23 PSU? I have one
that
> keeps pouring smoke out of it...yet I can
never see a component that's
> burnt, punctured, or other stuff like that. I also never hear an arc
or
anything
exploding. The PSU also squeals when powered down.
What does the smoke smell like? If it smells like burning paper, then 
24 Nov
24 Nov
7:27 p.m.
Check the wiring harness between the PSU and the backplane. The original
rev was dodgy and prone to smoke.
7:55 p.m.
On Mon, 25 Nov 2013, Ethan Dicks wrote:
Okay. I grabbed some pictures and I think I've found the culprit(s).
http://uploads.gewt.net/me/psu-goo.jpg
I don't think that's any kind of potting compound and that doesn't look
like it should be there. I could be wrong, though.
http://uploads.gewt.net/me/deformed-component.jpg
I have NO idea what that...deformed...thing was once, but it does not look
quite right. (directly behind the...transformer?)
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
8:17 p.m.
On Nov 24, 2013, at 10:55 PM, Cory Smelosky wrote:
On Mon, 25 Nov 2013, Ethan Dicks wrote:
Okay. I grabbed some pictures and I think I've found the culprit(s).
http://uploads.gewt.net/me/psu-goo.jpg
I don't think that's any kind of potting compound and that doesn't look
like it should be there. I could be wrong, though.
That COULD be hot glue. Not sure, but it doesn't look like something that would ooze
from a cap, and hot glue is sometimes used for vibration tolerance.
http://uploads.gewt.net/me/deformed-component.jpg
I have NO idea what that...deformed...thing was once, but it does not look
quite right. (directly behind the...transformer?)
I would call that "detonated". It looks like a cap that's blown up. Looks
like it may have lost its cover, even. If you look up the schematic, you can probably
figure out what C102 was supposed to be.
- Dave
8:22 p.m.
On Mon, 25 Nov 2013, David Riley wrote:
I can't find BA23 PSU schematics, unfortunately. Someone on the list
probably has the needed cab memorised.
On Nov 24, 2013, at 10:55 PM, Cory Smelosky wrote:
Possible. I can investigate further once everything discharges and it's
safer to touch the system.
On Mon, 25 Nov 2013, Ethan Dicks wrote:
Okay. I grabbed some pictures and I think I've found the culprit(s).
http://uploads.gewt.net/me/psu-goo.jpg
I don't think that's any kind of potting compound and that doesn't look
like it should be there. I could be wrong, though.
That COULD be hot glue. Not sure, but it doesn't look like something that would ooze
from a cap, and hot glue is sometimes used for vibration tolerance.
http://uploads.gewt.net/me/deformed-component.jpg
I have NO idea what that...deformed...thing was once, but it does not look
quite right. (directly behind the...transformer?)
I would call that "detonated". It looks like a cap that's blown up. Looks
like it may have lost its cover, even. If you look up the schematic, you can probably
figure out what C102 was supposed to be.
- Dave
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
8:37 p.m.
On Nov 24, 2013, at 11:22 PM, Cory Smelosky wrote:
I can't find BA23 PSU schematics, unfortunately. Someone on the list
probably has the needed cab memorised.
Alternatively, its faceplate may be somewhere in the box still
(check the lid; I've definitely had things in PSUs explode and
found bits of the components glued to the roof of the enclosure).
- Dave
http://uploads.gewt.net/me/deformed-component.jpg
I have NO idea what that...deformed...thing was once, but it does not look
quite right. (directly behind the...transformer?)
I would call that "detonated". It looks like a cap that's blown up. Looks
like it may have lost its cover, even. If you look up the schematic, you can probably
figure out what C102 was supposed to be.
8:42 p.m.
On Mon, 25 Nov 2013, David Riley wrote:
On Nov 24, 2013, at 11:22 PM, Cory Smelosky wrote:
I can't find BA23 PSU schematics, unfortunately. Someone on the list
probably has the needed cab memorised.
Alternatively, its faceplate may be somewhere in the box still
(check the lid; I've definitely had things in PSUs explode and
found bits of the components glued to the roof of the enclosure).
The PSU has been moved around and flipped in all manner of ways...if it
didn't lodge somewhere it's gone for sure.
http://uploads.gewt.net/me/deformed-component.jpg
I have NO idea what that...deformed...thing was once, but it does not look
quite right. (directly behind the...transformer?)
I would call that "detonated". It looks like a cap that's blown up. Looks
like it may have lost its cover, even. If you look up the schematic, you can probably
figure out what C102 was supposed to be.
- Dave
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
9:08 p.m.
On Sun, Nov 24, 2013 at 11:42 PM, Cory Smelosky <b4 at gewt.net> wrote:
I think you might find
http://www.vintage-computer.com/vcforum/showthread.php?24876-H7864-power-su…
useful.
Jim
Alternatively,
its faceplate may be somewhere in the box still
(check the lid; I've definitely had things in PSUs explode and
found bits of the components glued to the roof of the enclosure).
The PSU has been moved around and flipped in all manner of ways...if it
didn't lodge somewhere it's gone for sure.
8:58 p.m.
On Nov 24, 2013, at 19:55 , Cory Smelosky <b4 at gewt.net> wrote:
On Mon, 25 Nov 2013, Ethan Dicks wrote:
Okay. I grabbed some pictures and I think I've found the culprit(s).
http://uploads.gewt.net/me/psu-goo.jpg
The white glorp might be hot melt glue, or it might be some leaked innards of something.
The yellowish block to the right of the glorp appears to be one of those ^#%!$$#%
metallized-paper dielectric EMI filtering caps that I mentioned. The TRS-80 Model 12 power
supply of mine which blew out one of those caps was also made by Astec. I would replace it
with a class "X" safety-rated EMI filtering cap with the same voltage and
capacitance ratings, but NOT a metallized paper type. Identical Rifa caps (actually
Kemet-Evox-Rifa due to acquisitions) are still available from Digi-Key, but I would NOT
use those blasted things.
I theorize that the failure mechanism is that after a few decades, the plastic cap shell
cracks, allowing moisture to get it. That gets absorbed by the paper dielectric, and
eventually the dielectric breaks down under power. Polypropylene film dielectric is
supposed to have very low moisture absorption, so I figure they should last longer, even
after their seals degrade.
Notice the cracking of the plastic on that cap in your picture. If it hasn't already
blown, it will.
http://uploads.gewt.net/me/deformed-component.jpg
I have NO idea what that...deformed...thing was once, but it does not look
quite right. (directly behind the...transformer?)
The one right above where it says "C102"? That looks like another one of those
&#%!*&$* metallized-paper dielectric EMI caps which has blown its top. If you
can't find a schematic diagram to get its ratings, then maybe somebody else here with
a BA23 can try to read the markings off that cap in their supply.
I found that safety-rated EMI caps other than the (#&%*^$ Rifa ones have different
lead spacings. At least in my Model 12 supply, there were extra holes to allow caps with
different lead spacings to be substituted, and the Panasonic caps I bought dropped right
in. Their lead spacings weren't quite right for the Rifa caps in my Model 4 supply
(also made by Astec). Pick out caps with long wire leads instead of short pins in case the
lead spacing isn't the same as any of the holes on your PCB.
If you can get the numbers off the top and sides of those caps, then I can help you pick
out some replacement parts if you don't know how to. I did this just last week for a
couple of my TRS-80 systems, and I expect I'll be doing the same thing again once I
get started on my PDP-11/44 project soon.
I recommend replacing every one of those metallized-paper caps that you encounter, whether
it's already blown or not.
--
Mark J. Blair, NF6X <nf6x at nf6x.net>
http://www.nf6x.net/
9:09 p.m.
On Mon, 25 Nov 2013, Mark J. Blair wrote:
Yeah. I'm not quite sure if it's worth the effort...the PSU needs a VERY
good cleaning...which will prove difficult. Transformers don't look too
good...but they don't look riskily bad, either.
I might also do it merely for the sake of repair and restoration practise.
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
On Nov 24, 2013, at 19:55 , Cory Smelosky <b4 at gewt.net> wrote:
Yup.
On Mon, 25 Nov 2013, Ethan Dicks wrote:
Okay. I grabbed some pictures and I think I've found the culprit(s).
http://uploads.gewt.net/me/psu-goo.jpg
The white glorp might be hot melt glue, or it might be some leaked innards of something.
The yellowish block to the right of the glorp appears to be one of those ^#%!$$#%
metallized-paper dielectric EMI filtering caps that I mentioned. The TRS-80 Model 12 power
supply of mine which blew out one of those caps was also made by Astec. I would replace it
with a class "X" safety-rated EMI filtering cap with the same voltage and
capacitance ratings, but NOT a metallized paper type. Identical Rifa caps (actually
Kemet-Evox-Rifa due to acquisitions) are still available from Digi-Key, but I would NOT
use those blasted things.
I theorize that the failure mechanism is that after a
few decades, the plastic cap shell cracks, allowing moisture to get it. That gets absorbed
by the paper dielectric, and eventually the dielectric breaks down under power.
Polypropylene film dielectric is supposed to have very low moisture absorption, so I
figure they should last longer, even after their seals degrade.
Notice the cracking of the plastic on that cap in your picture. If it hasn't already
blown, it will.
Yeah.
http://uploads.gewt.net/me/deformed-component.jpg
I have NO idea what that...deformed...thing was once, but it does not look
quite right. (directly behind the...transformer?)
The one right above where it says "C102"? That looks like another one of those
&#%!*&$* metallized-paper dielectric EMI caps which has blown its top. If you
can't find a schematic diagram to get its ratings, then maybe somebody else here with
a BA23 can try to read the markings off that cap in their supply.
I found that safety-rated EMI caps other than the (#&%*^$ Rifa ones have different
lead spacings. At least in my Model 12 supply, there were extra holes to allow caps with
different lead spacings to be substituted, and the Panasonic caps I bought dropped right
in. Their lead spacings weren't quite right for the Rifa caps in my Model 4 supply
(also made by Astec). Pick out caps with long wire leads instead of short pins in case the
lead spacing isn't the same as any of the holes on your PCB.
Yeah. I'll always go with long-lead ones incase I make a mistake anyway.
;)
If you can get the numbers off the top and sides of those caps, then I can help you pick
out some replacement parts if you don't know how to. I did this just last week for a
couple of my TRS-80 systems, and I expect I'll be doing the same thing again once I
get started on my PDP-11/44 project soon.
I have an H7140 with a present 300VDC rail that does nothing else. (It
measures 320VDC but I can't see anything obviously blown). I need to get
the replacement for it, though. I just need to pick it up...not sure if
it'll work.
No DC OK light comes on.
I recommend replacing every one of those metallized-paper caps that you encounter,
whether it's already blown or not.
I'd agree with that. I need to do the same on some wax-paper cap gear
that still works fine.
25 Nov
25 Nov
12:52 p.m.
-----Original Message-----
From: cctalk-bounces at classiccmp.org [mailto:cctalk-
bounces at classiccmp.org] On Behalf Of Cory Smelosky
Sent: 25 November 2013 05:09
To: General Discussion: On-Topic and Off-Topic Posts
Subject: Re: BA23 common component failures?
On Mon, 25 Nov 2013, Mark J. Blair wrote:
>
> On Nov 24, 2013, at 19:55 , Cory Smelosky <b4 at gewt.net> wrote:
>> On Mon, 25 Nov 2013, Ethan Dicks wrote:
>>
>> Okay. I grabbed some pictures and I think I've found the culprit(s).
>>
>> http://uploads.gewt.net/me/psu-goo.jpg
>
> The white glorp might be hot melt glue, or it might be some leaked
innards
of something. The yellowish block to the right of the
glorp appears to be
one
of those ^#%!$$#% metallized-paper dielectric EMI
filtering caps that I
mentioned. The TRS-80 Model 12 power supply of mine which blew out one
of those caps was also made by Astec. I would replace it with a class "X"
safety-rated EMI filtering cap with the same voltage and capacitance
ratings,
but NOT a metallized paper type. Identical Rifa caps
(actually Kemet-Evox-
Rifa due to acquisitions) are still available from Digi-Key, but I would
NOT use
those blasted things.
>
I think the filter caps on every single one of my BA23 and BA123 power
supplies have blown up. I have replaced them all, and yes you need slightly
longish leads as the replacements I have found don't quite fit (but almost).
Yup.
> I theorize that the failure mechanism is that after a few decades, the
plastic
cap shell cracks, allowing moisture to get it. That
gets absorbed by the
paper
dielectric, and eventually the dielectric breaks down
under power.
Polypropylene film dielectric is supposed to have very low moisture
absorption, so I figure they should last longer, even after their seals
degrade.
>
> Notice the cracking of the plastic on that cap in your picture. If it
hasn't
already blown, it will.
Yeah.
>>
http://uploads.gewt.net/me/deformed-component.jpg
>>
>> I have NO idea what that...deformed...thing was once, but it does not
>> look quite right. (directly behind the...transformer?)
> The one right above where it says
"C102"? That looks like another one of
those &#%!*&$* metallized-paper dielectric EMI caps which has blown its
top. If you can't find a schematic diagram to get its ratings, then maybe
somebody else here with a BA23 can try to read the markings off that cap
in
their supply.
Yeah. I'm not quite sure if it's worth the effort...the PSU needs a VERY
good
cleaning...which will prove difficult. Transformers
don't look too
good...but
they don't look riskily bad, either.
I might also do it merely for the sake of repair and restoration practise.
the
I found that safety-rated EMI caps other than the (#&%*^$ Rifa ones have
different lead spacings. At least in my Model 12 supply, there were extra
holes to allow caps with different lead spacings to be substituted, and Panasonic caps I bought dropped right in. Their lead
spacings weren't
quite
right for the Rifa caps in my Model 4 supply (also
made by Astec). Pick
out
caps with long wire leads instead of short pins in
case the lead spacing
isn't
the same as any of the holes on your PCB.
Yeah. I'll always go with long-lead ones incase I make a mistake anyway.
;)
>
> If you can get the numbers off the top and sides of those caps, then I
can
help you pick out some replacement parts if you
don't know how to. I did
this
just last week for a couple of my TRS-80 systems, and
I expect I'll be
doing
the same thing again once I get started on my
PDP-11/44 project soon.
I have an H7140 with a present 300VDC rail that does nothing else. (It
measures 320VDC but I can't see anything obviously blown). I need to get
the replacement for it, though. I just need to pick it up...not sure if
it'll
work.
No DC OK light comes on.
Lucky you , I have been trying for the last 2.5 years (not continuously
mind) to fix my H7140. It still doesn't work :-(
Regards
Rob
I recommend replacing every one of those metallized-paper caps that you
encounter,
whether it's already blown or not.
still works fine.
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
1:06 p.m.
On Mon, 25 Nov 2013, Robert Jarratt wrote:
I think the filter caps on every single one of my BA23 and BA123 power
supplies have blown up. I have replaced them all, and yes you need slightly
longish leads as the replacements I have found don't quite fit (but almost).
Have the value on hand?
Lucky you , I have been trying for the last 2.5 years (not continuously
mind) to fix my H7140. It still doesn't work :-(
Aww. This one might not work either. :(
Fingers crossed!
Regards
Rob
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
1:49 p.m.
-----Original Message-----
From: cctalk-bounces at classiccmp.org [mailto:cctalk-
bounces at classiccmp.org] On Behalf Of Cory Smelosky
Sent: 25 November 2013 21:07
To: General Discussion: On-Topic and Off-Topic Posts
Subject: RE: BA23 common component failures?
On Mon, 25 Nov 2013, Robert Jarratt wrote:
>
>
> I think the filter caps on every single one of my BA23 and BA123 power
> supplies have blown up. I have replaced them all, and yes you need
> slightly longish leads as the replacements I have found don't quite fit
(but
almost).
Have the value on hand?
I was told it didn't actually matter that much, for the record there are two
of different sizes, namely 0.22uF (X2) and 0.47Uf (X).
Regards
Rob
Lucky you , I have been trying for the last 2.5 years (not
continuously
mind) to fix my H7140. It still doesn't work :-(
Aww. This one might not work either. :(
Fingers crossed! 
26 Nov
26 Nov
4:47 p.m.
Robert Jarratt wrote:
I think the filter caps on every single one of my BA23
and BA123 power
supplies have blown up. I have replaced them all, and yes you need slightly
longish leads as the replacements I have found don't quite fit (but almost).
I have a number of BA23 and BA123 boxes with the
normal power supplies and backplanes.
Some have been used within the past couple of years.
Many have not been powered on for at least 10 years.
Should I be concerned about all of then equally at this
point or just the BA23 and BA123 which have not
been used for at least 5 years?
Jerome Fine
27 Nov
27 Nov
12:02 p.m.
Robert Jarratt wrote:
I think the filter caps on every single one of my
BA23 and BA123 power
supplies have blown up. I have replaced them all, and yes you need slightly
longish leads as the replacements I have found don't quite fit (but almost).
I have a number of BA23 and BA123 boxes with the
normal power supplies and backplanes.
Some have been used within the past couple of years.
Many have not been powered on for at least 10 years.
Should I be concerned about all of then equally at this
point or just the BA23 and BA123 which have not
been used for at least 5 years?
25 Nov
25 Nov
1:09 p.m.
On Mon, 25 Nov 2013, Ethan Dicks wrote:
Okay. I grabbed some pictures and I think I've found the culprit(s).
http://uploads.gewt.net/me/psu-goo.jpg
I don't think that's any kind of potting compound and that doesn't look
like it should be there. I could be wrong, though.
That looks to me like hot melt glue, which is commonly used on SMPSU
boards to prevent components from moving. I think it's somesafety
requirmeent to prevnent mains-side and output-side components getting too
close to each other.
http://uploads.gewt.net/me/deformed-component.jpg
I have NO idea what that...deformed...thing was once, but it does not look
quite right. (directly behind the...transformer?)
The thing taht looks lik a transdroemr may well eb a bifilar-wound
indcutor, used as part of the mains supressor.
The melted thing is a capcitor (note the Cxxx marking on the PCB) and is
one of the typically antisocial ones that fails in interesting ways. It's
part of the mains filter. For testing, just remove it, the supply will
still work. Rmeplace it with a correctly-rated (designed for this
application) capacitor of the same value. They are not hard to get.
I don't normally approve of 'shotgun debugging', but when I find one
capcitor i na mains filter circuit has failed, I normally replace the
lot. Saves taking the case off again a couple of days later :-)
-tony
4048
days inactive
4050
days old
18 comments
9 participants
participants (9)
-
ard＠p850ug1.demon.co.uk -
b4＠gewt.net -
chd＠chdickman.com -
ethan.dicks＠gmail.com -
fraveydank＠gmail.com -
jhfinedp3k＠compsys.to -
jim＠deitygraveyard.com -
nf6x＠nf6x.net -
robert.jarratt＠ntlworld.com