9825B

[Danny Van Braband]

Hello,

I'm Danny Van Braband an old-computer-enthousiast from Belgium. Yesterday i found two 9825B's on a junkmarket. One works like a charm. The second one however has a problem. It starts with a lot of ventilator noise, so i guess the power supply is ok! The display stays black though. After a few seconds the printer begins to eject paper and printing some lines. And then the speaker begins to produce a continuous beep.
Ofcourse i'd like to solve the problem. Has anyone have an idea to what could be wrong? I can use all advise i can get. I'm also looking for the schematics. Can someone help me?

With friendly greetings,

Danny Van Braband
vbd3@yahoo.com

[Tony Duell]

Unfortunately, the fact that the fan is running does _not_ mean that the power supply is working. The fan runs off 120V
(obtained from the primary of the mains transformer) in these machines.
Start by checkling the PSU output voltages. To get inside, first unplug the mains plug. Then undo the 2 screws on the back and
lift off the cover. Then loosen the 2 similar-sized screws just behind the keyboard/display assembly inside.
Slide the keyboard forward to free the front edge and lift it
up carefully. There are 3 cables linking the keyboard assembly to the rest of the machine -- a 20 wire ribbon cable to the tape drive, a 40 wire ribbon cable (I/O bus) and a power connector (many coloured wires).
You can run the machine without the keyboard assembly connected for testing the PSU. The voltages can all be found on the power connector that you unplugged from the keyboard
The wire colours there (at least in my machine) are : Blue +20V (unregulated) ; Yellow +12V ; Orange +7.5V ; Red +5V ; Black Ground ; Brown Ground ; White/Orange -5V ; White -20V unregulated.
All voltages are with respect to the ground wire, the metalwork over the power supply board and round the I/O edge connectors at the back is also connected to ground.
There is also a -12V supply, White/Yellow wire, on the CPU board and tape controller board power connectors. Check all the supply voltages with a
voltmeter, and if possible use a 'scope to look for excessive ripple.
If it's not a PSU problem, unplug all the ribbon cables in the unit, clean the connectors (use propan-2-ol) and reconnect them. There are 3 such cables :
From the rear edge connector on the CPU board to the lower RAM board (extra timing signals for the MMU, etc) ; from the ROM backplane at the front to the front CPU board edge connector and then to both memory boards -- Memory bus.
And from the tape controller to the I/O backplane to the header plug at the back of the CPU board, and then to the keyboard/display -- I/O bus.
AFAIK, HP never provided official schematics for this machine, but I have repaired mine a few times so I might be able to help.

[Danny Van Braband]

Hallo Tony,

As you suggested i checked the tensionlevels on te power supply first. They seem OK though. Then i cleaned the edge connectors and even removed some cobwebs. I disassembled the keyboard to check for "sticking keys".
I mended the spacebar (didn't give contact).
The machine still reacts the same though.

I've been thinking, as the machine starts out with printing, maybe it's trying to tell me something. I also saw that the previous owner used a roll of plain paper instead of thermal. So now i'm trying to find some thermal paper.

Thanks for wanting to help me!

Greetings,

Danny Van Braband

[Tony Duell]

What test equipment do you have? If at all possible, I would recomend looking at the
power supply outputs with an oscilloscope -- You might have ripple on some of them (due to dried-up capacitors on the PSU board, for example).
I think a 'scope or logic analyser is going to be very useful for finding this fault, anyway...
One 'nasty' thing about the 9825 is that if the I/O chip and/or the processor are not running, it's possible for a couple of pins on the I/O chip (the 40 pin ceramic chip on the PCB under the keyboard) to
be stuck in states where the beeper sounds continuously and the printer feeds paper (unlike the 9815, there is some circuity to repeatedly pulse the feed solenoid I think on the 9825).
So it could be a problem with the processor, ROM, RAM (so the processor is not running the right code), the I/O chip. the PSU, etc...
Maybe we should take this to private e-mail, it's easier to send long messages that way.