H7874 power supply

-----Original Message----- From: cctech <cctech-bounces at classiccmp.org> On Behalf Of Robert Armstrong via cctech Sent: 29 March 2020 23:15 To: rob at jarratt.me.uk; 'General Discussion: On-Topic Posts' <cctech at classiccmp.org>; 'Maciej W. Rozycki' <macro at linux-mips.org> Subject: RE: H7874 power supply Remembering that I have the R400x enclosure and I'm only interested in powering disk drives - I looked up the specs for the RF7x drives, and they all use about 1.5A at 5V and 2.5-3A operating at 12V. There's a 5-6A spinup surge on the 12V supply, but as long as you spin them up in sequence you can probably gloss over that. Anyway, I have 4xRF7x drives and two SCSI drives, so figure maybe 20A at 12V and 10A at 5V. That's entirely doable with an ATX supply. The fans, annoyingly, are 24VDC so those would either have to be replaced with 12V fans or I'd have to kludge up an extra 24V supply. My next step was to figure out the pinout of the connectors on the back of the H7874. The big, high current, pins aren't marked anywhere inside the H7874 that I could find but, conveniently, they ARE marked on the R400x backplane. There are also a couple of high density connectors sandwiched in between the power pins; the lower one doesn't appear to be used but the upper one has a few connections. From looking at the backplane it appears that only four pins in the upper connector are actually connected. That's a guess though, because I can't see that back side of the backplane. I needed to see more of the backplane to figure out where these four connections went, so I removed all the drives from the R400x. Then, on a whim, I stuck the H7874 back in there and tried to power it up. It works! It appears that I don't have a bad power supply after all; I have a bad drive somewhere. Don't I feel silly :)

So I don't need to fool with the ATX supply after all, or at least not until my PS really does die. Thanks for the help, and sorry for the false alarm. Bob -----Original Message----- From: Rob Jarratt [mailto:robert.jarratt at ntlworld.com] Sent: Sunday, March 29, 2020 2:05 PM To: bob at jfcl.com; 'General Discussion: On-Topic Posts'; 'Maciej W. Rozycki' Cc: rob at jarratt.me.uk Subject: RE: H7874 power supply I have some notes I made about trying a PC PSU, I don't think it really worked, there are signals to indicate power OK and perhaps others. But perhaps a more concerted effort might yield something that could work. Regards Rob > -----Original Message----- > From: cctech <cctech-bounces at classiccmp.org> On Behalf Of Robert > Armstrong via cctech > Sent: 29 March 2020 19:15 > To: 'Maciej W. Rozycki' <macro at linux-mips.org>; 'General Discussion: > On- Topic Posts' <cctech at classiccmp.org> > Cc: rob at jarratt.me.uk > Subject: RE: H7874 power supply > > >Maciej W. Rozycki <macro at linux-mips.org> wrote: > >numerous Chemi-Con SXF parts scattered across all boards, some > >leaking > > Yeah, mine is full of "Nippon Chemi-Con" electrolytic too. > > What about just gutting the PS chassis and sticking in an ATX power > supply? I don't know how the maximum current per output breaks down, > but as I understand it the H7874 is only a 600W supply. That's not a > problem for an ATX supply, and it would have all the required output > voltages. You'd have to run the cabinet fans at a fixed speed and > you'd lose the power control bus feature, but that's not a show stopper. > > Actually my unit is in an R400x chassis full of RF drives, so I > believe I only need > +12 and +5. I don't think the 3.3v and -12 are used in that case. I > +wonder how > much current I really need for the drives? I should look that up and > I might be able to find a surplus dual output supply that would do the job. > > Thanks again, > Bob