[rescue] Help me rescue my Sun Ultra 10 (2x)

[HP van Braam]

Ideally I'd like to replace as few caps as possible, mostly because it seems to be just wasteful (and potentially dangerous) to replace more caps than are necessary.



I did do the old 'finger test' on all caps and none of them get appreciably warm. And visual inspection doesn't show any obvious failures either. They are also all brandname caps, so maybe they are just... fine? The 8 big caps next to the CPU are all dead on at 1.90 volt with no fluctuation, but there are the 5 other caps that I could not measure, and only did the finger test/visual inspection on.

I don't really have a good way of doing this with the SMD caps, and replacing those should also be a bit gentler on the board than the thru-hole ones, so I'm a bit less worried about yanking and replacing those.



But your point is well taken. I should probably be prepared that it might not work, and that I might have to replace the 13 big cans too, maybe also the the 2 "audio grade" caps on the CPU carrier board... Finding replacements for those might be a problem...



- HP



-----Original message-----
From: John Hudak via rescue <rescue at sunhelp.org>
Sent: Monday December 15th, 2025 16:03
To: The Rescue List <rescue at sunhelp.org>
Cc: John Hudak <jjhudak at gmail.com>
Subject: Re: [rescue] Help me rescue my Sun Ultra 10 (2x)

You may want to consider checking the PS caps with an ESR (Equivalent Series Resistance) meter. High ESR caps can cause voltage drops, internal heating that changes the capacitance and filtering effectiveness at high frequencies.  In many cases one can check the ESR with the capacitor in place but it is best to have it disconnected from the circuit. The caps on the MB can also be checked.   OTOH, given the age of the boards and PS, you might want to just replace them all.
It sort of depends on your vision of the future for the system.
Good luck

On Mon, Dec 15, 2025 at 9:45 AM HP van Braam via rescue <rescue at sunhelp.org <mailto:rescue at sunhelp.org> > wrote:
Okay, I have measured voltages on all of the big caps, and they all have cromulent and stable voltages. So that just leaves the forest of SMD caps in the bottom right of the board.

I'll be replacing those on Wednesday on one of the boards... Wish me luck... :)



- HP



-----Original message-----
From: John Hudak via rescue <rescue at sunhelp.org <mailto:rescue at sunhelp.org> >
Sent: Sunday December 14th, 2025 1:27
To: The Rescue List <rescue at sunhelp.org <mailto:rescue at sunhelp.org> >
Cc: John Hudak <jjhudak at gmail.com <mailto:jjhudak at gmail.com> >
Subject: Re: [rescue] Help me rescue my Sun Ultra 10 (2x)

I think that is a valid problem.   

On Sat, Dec 13, 2025 at 6:43 PM Dan Moisa via rescue <rescue at sunhelp.org <mailto:rescue at sunhelp.org> > wrote:
I have a 3/80 that is flaky in a very similar way, I'm suspecting on-board caps at this point.

Dan.

On Sat, Dec 13, 2025 at 3:40 PM HP van Braam via rescue <rescue at sunhelp.org <mailto:rescue at sunhelp.org> > wrote:
Hello all,

 Well I guess I celebrated a bit too early... The Ultra 10 worked for a
 day or two but today it was running VTS while I was having dinner, I
 came back to it on the ok prompt, presumably having crashed.

 When I tried rebooting the machine now I'm back where I started. With
 the power led staying off when booting the machine. I've tried moving
 my now one new NVRAM chip to the other board, but this also does not
 work.

 The same thing happened as the first time too. Earlier today the
 machine booted but complained about NVRAM contents. Then I re-did the
 whole nvram mkp stuff, and `banner` was happy again. But later the
 machine did fail again.

 Now, I did use a 48T58Y and not a 48T58 (without the Y), is there any
 chance that that just didn't work right?

 Thanks!

 - HP

 On Tue, 2025-12-09 at 20:18 +0000, HP van Braam wrote:
 > Hello!
 >
 > I have two Sun Ultra 10s with the same fault, and they came to get
 > the same fault in the same way.
 >
 > They both did the following:
 >
 > * Started to get IDProm errors
 > * Used the "normal" instructions for fixing this
 > * Some time later after a reboot things started to go wrong
 >
 > This sort of worked for a moment, but now when I apply power to
 > either machine the power led stays off (most of the time) in the very

 > rares time when the power led does turn on I can't succesfully boot
 > the machines. Sometimes they end up booting into Solaris but they are

 > not stable.
 >
 > For instance even dtlogin will just start and cycle on and off until
 > ultimately the system hangs entirely. At this point pressing stop+a
 > and resetting usually doesn't boot correctly.
 >
 > Swapping around ram/CPUs/power supplies between the two systems has
 > no effect. Removing the AFB doesn't really seem to change anything
 > either.
 >
 > It seems kind of unlikely that both computers have the same obscure
 > hardware fault in the same component, I hope this is all just the
 > IDProm fucking up.
 >
 > I tried both of the IDProm chips in my minipro reader and they seem
 > to not really read that same each time. And writing data to the chips

 > doesn't seem to "stick".
 >
 > Help would be very much appreciated! Thank you! 
 >
 > - HP

 _______________________________________________
 rescue list - http://sunhelp.org/mailman/listinfo/rescue_sunhelp.org <http://sunhelp.org/mailman/listinfo/rescue_sunhelp.org>
_______________________________________________
 rescue list - http://sunhelp.org/mailman/listinfo/rescue_sunhelp.org <http://sunhelp.org/mailman/listinfo/rescue_sunhelp.org>
_______________________________________________
 rescue list - http://sunhelp.org/mailman/listinfo/rescue_sunhelp.org <http://sunhelp.org/mailman/listinfo/rescue_sunhelp.org>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sunhelp.org/pipermail/rescue_sunhelp.org/attachments/20251215/a1038e27/attachment.html>