H7821 power supply in MicroVAX 3100, SCSI disk enclosures and others
ucespamdump at gmail.com
Fri Jan 31 12:29:52 CST 2020
General rule for Elecrolytic caps is age and heat is not their friend.
So the assumption is if it powers up they may be leaky(electrically) or
and not completely bad so inspection is well advised.
My experience is if powered regularly (at least once a year for a hour or
they seem to behave well. Same for other electronics (non computer).
On Fri, Jan 31, 2020 at 8:24 AM Peter Coghlan via cctalk <
cctalk at classiccmp.org> wrote:
> The recent discussion on BSC protocol prompted me to dig out my Microvax
> with DSH32 synchronous serial interface. It had been idle in storage for
> several years and it wouldn't power up, only giving a brief flash on the
> diagnostic LEDs and a quick twitch of the fans. There was a slight smell,
> the stale air that comes out of a deflating tyre.
> I took out the H7821 power supply and found that five identical brown
> 1800uF 25V
> electrolytic capacitors on the output side had leaked.
> The SCSI disk enclosure where the machine's system disk lives required
> power cycles to get it to run at all and it died as soon as the disk tried
> spin up. It turned out to also contain a H7821 power supply which had a
> similar issue with the same five brown capacitors, although not as
> as in the main unit.
> I found a second disk enclosure which had seen little use and grabbed the
> supply out of that to put in the MicroVAX. It worked well enough to test
> but there was a ring of goo around the bottom of one of the brown
> which was worst affected in the other units. Time to order a batch of
> replacement capacitors and figure out what else has been damaged. While
> it is
> not the worst I have seen, access to these power supplies for repairs is
> difficult and it is really difficult to debug them safely while they are
> running with the cover off :-(
> If anyone has anything with H7821 power supplies in them, I suggest
> on these capacitors. If anything with these power supplies is in storage,
> suggest ensuring it is stored the normal way up as this should limit the
> ability of the goo to escape and spread around the power supply.
> And there I was thought I was being safe enough by removing the nicad
> packs some years ago...
> Peter Coghlan.
More information about the cctalk