HP9845 PSU (was HP Integral)
ard at p850ug1.demon.co.uk
Tue Mar 15 17:15:01 CST 2005
> >> a couple of them. I got lucky with my first one. The wires on the large
> >> heavy inductor had sheared off at the PCB board so it wasn't hard to fix.
> >Do you mean the black plastic block that's bolted to the metalwork that
> >also holds the transformer for startup? That's part of the mains input
> >filter, of course. Or one of the 2 main chopper transformers? Or one of
> >the inductors on the output PCBs?
> I don't remeber which it was but it was an inductor and not a
> transformer. The thing was working before it was shipped and broken when it
There are a lot of small-ish inductors on the output PCBs, of course, but
those would have been rattling around in the case if they'd come off the PCB.
For those who've not seen this PSU, it's on 5 PCBs. At the top is a large
board containing the mains input rectifiers and smoothing caps, the
rectifiers for the (linear) startup supply, the 2 main chopper
transformers, the current sense transformers, and not a lot else. Mains
input comes into this board via a 4 way AMP connector, which mates with
one strapped to the side of the PSU casing.
Plugged into that are 2 output PCBs, one for the logic supplies, one for
the mechanical supplies. These contain the smoothing caps, inductors, the
chopper driver trnasistors, the crowbar (which operates on the printhead
supply only, for some odd reason). These also plug into the main
backplane in the base of the machine, and are the connection between the
PSU and the reat of the machine.
There's a chopper PCB that pluges into the top board and one of the
output boards. It contains the 4 main chopper transistors (2 push-pull
choppers), the fast-turn-off circuits, the pulse transformers to couple
the drive to the choppers (all the control circuitry is on the output
side of the isolation barrier).
And there's the control PCB. 5 chopper control ICs (for the main choppers
and the 3 switching regulators), the chopper ICs (PIC6xx series) for the
output switching regulators, 3 quad comparators (protection circuit), 2
quad op-amps (assorted current sense and reference buffering functions),
2 TTL monostables (power-on Init/ signal), 2 3-terminal linear regulators
(startup supplies), etc.
As I said, complicated.
I still want to know what the 'discharge tool' mentioned on the PSU case
consists of. The instructions are to disconnect the mains, wait 5
minutes, insert the discharge tool into a pair of holes and then into a
second pair of holes (these holes line up with pads on the top PCB
connected to the termianls of the mains smoothing caps), then check
they've discharged to < 25V.
All I can add is that there's a fad 22k bleeder resistor across each of
said capacitors, so I can't see any need for anything else...
[Back to the Integral]
> I susepct thrre was a service ROM cartridge that fits in
> >place of the HPUX ROM too.
> There is. I have one. :-)
Right. Does it do anything useful (in other words, can you find faults
with it that you can't easilty find witnout it, but with a logic analyser?)
> >No problem. My view is that if I post something to classiccmp, it's been
> >made public (after all classiccmp is archived publically, I think). I
> >therefore have no problem with my postings being made public in other
> >ways. However, I do expect an acknowledgement somewhere on the web page
> >or file...
> Not a problem. I'd already planned on that. I don't know if you want
I didn't expect it to be a problem (and BTW, this applies to anyone on
classiccmp who wants to quote something I've posted)
> you e-mail address included or not or you want it included but made
> unreadable to the SPAMbots. Let me know what you prefer.
Well, I've posted my e-mail address un-mangled to a lot of places, so I
don't think one more will make much difference... I don't really care.
More information about the cctech