9 Sep
2004
9 Sep
'04
10:50 a.m.
The bad news was that it would only print to the rear
output slot, and
any attempts to print to the main upper output bin resulted in front
panel error messages about jams. I've called PrinterWorks (great laser
printer gurus, www.printerworks.com) and they advised me to replace the
I'll second the recomendation for The PrinterWorks. I've bought spares
for my CX and SX printers from them, and they've shipped the right stuff
promptly.
I've found they sell individual parts for most mechnanical sections,
although they recomend swapping out the whole assembly unless you know
what you're doing. I don't know what I am doing, but I still change the
little bits...
So I went looking for a service manual. Found found
on eBay in PDF
on CD, bought it, had to wait for the CD to arrive, then go to Kinko's
and have the PDFs printed. It was a pretty good deal: in addition to
the combined IIISi/4Si service manual from HP (which is what made me
realise that they are the exact same printer), there was also a generic
laser printer repair book that taught a lot of general principles not
specific to any given make or model, and approched things from a rather
Tony Duell-like perspective.
OK, my story of laser printer repair (which I am sure I've posted before,
but...)
I was given a Canon LBP8-A1 (CX engine) that didn't work. I'd never been
inside a laser printer before, but 'they can't be that complicated',
right ;-)...
Anyway, I got the casing off and did some tests on the formatter board
(it's in the lid on the CX).It was pretty clear that the ROMs were never
being selected (so the 6800 was going crazy), it looked to be a fault in
one of the PAL/HAL chips on the board. Contacted Canon, they were no help
at all. No spare chips, heck they wouldn't even sell me a board or a
service manual (I didn't know about the PrinterWorks back then).
Figuring I had nothing to lose, I took the machine totally apart. Even if
I couldn't get it working, I might at least learn about the insides. I
made lots of notes, diagrams of connections, etc.
And then I got a PERQ. Bob Davis told me that a related printer the
CX-VDO was used on the PERQ with a sepcial interface card in the PERQ. I
managed to locate most of such a card, and the schematics for it. Bob
buzzed out his printer's cable, I got the card together and made the
cable. But of course I still didn't have the printer.
But then I noticed, comparing my notes on the CX and the PERQ cable that
the CX-VDO didn't use the formatter board (where the dead PAL was), but
that a straight-trhough ribbon with a DC37 socket on one end and a 34 pin
header socket on the other, the former to connect to the PERQ cable, the
latter to the DC contorller, should do the trick. I made one, put the
printer back togther, got it aligned (suprisingly easy -- yes I'd
stripped down the optical block as well), and tried it out.
No, it didn't work, but it almost did. In particular it did feed a sheet
of paper. It did print _something_ -- just a column 16 dots wide. The
fault, fortunately was on the PERQ end -- a FIFO wasn't being reloaded.
I'd missed out one of the kludgewires on the board. As I had schematics,
this wasn't too hard to find.
That wasn't quite the end of it. There were a couple of minor mechanical
rapairs (had I known about the Printerworks I'd have bought the bits, as
it was I had to make brackets, etc) and I ahd to make a new control
panel. The LBP8-A1 has a panel with a few switches, a 7-segment display
some lEDs, etc that connects to the formatter. The CX-VDO has 5 LEDs
connected to the DC controller. Well, I found the header and identified
the signals (Bob told me what the LEDs were, they were things like
paper-out, so I could see which signal changed state when I frobbed the
paper sensor, etc). I took the orignal panel apart, stuck a new label on
the front, used 5 of the existing LEDs (which I labelled appropriately),
desolded the switches, display, and formatter cable, and fitted a cable
to connect to the DC controller.
It worked. It still works. Pity I can no longer get toner for it (I am
going to have to work out how to refill those cartridges, and what to
use).
After that, fixing SX printers was almost trivial!
With the help of the service manual (OK, part swapper guide) I figured
out how to take the sucker apart. I got to the job offset assembly and
took it out for examination. It was dirty with grease, but not in the
paper path, and there wasn't anything obviously wrong with it. I was
hesitant at first about just buying a new job offset assy and trying to
swap it in and see what happens, having heard Prof. Duell's admonitions
against randomly swapping parts. But then I broke down and did it, as
Thing about mechanical problems is that you can often observe the unit in
opeeration and see what doesn't move when it should (or does move when it
shouldn't).
Yes, like just about every laser printer in the world,
it's a Canon engine,
FWIW, the DEC LN03 (I have one, I think mine is the Postscript model) is not
a Canon engine...
and this one is NX. The word engine is a bit
misleading: I originally
thought it meant something internal and did not include the outside
appearance of the printer and input/output options like paper trays.
As it turns out, NX engine means not just some inside guts like the
laser or the motor, but the entire printer from plastic covers outside
to the last screw inside is made by Canon. The ONLY HP part in there
Not necessarily. The Apple LW and LW2 are CX and SX engines, but the
outer cases are different to the normal Canon ones (they may still be
made by Canon, but they're not the same).
The NX engine has 300 DPI and 600 DPI versions, and
the difference is
only in the DC controller board. Every other part is the same, as are
Are the scanners the same? Normally the higher resolution scanner needs
to rotate faster (so as to put more lines down in the same time). I know
that on the CX engine there was a 415 (!) dpi version and a crystal on
the sannner motor PCB [1] needs to be changed.
[1] The chips on this board are standard consumer-electronics devices
(used in stereos, VCRs, etc). The motor control PLL even has a 33/45
select pin, obviously it's from a record turntable. This helped a lot
when sorting out that machine.
-tony