#Parallel Ports Quit
11 messages in this thread
I need some help from some of the Guru's on here! My parallel ports quit
today, its weird, neither the gvp I/O or my amiga parallel ports work no
more. I can print to a fax and I can print to the printer thru the same
cables from a laptop but not from the amiga, it just sends the wrong
characters and it seems the linefeeds and carriage returns are missing,
from pagestream it looks normal it sends and sends and completes but the
printer never ejects te page so U take it off line and form feed it and
all i get is either a blank page or one line of crap. I swapped the 8520
chips and it still does same thing (swapped the 2 on the motherboard to
see if problem changed). I have also done a restore from tape from a time
wwhen I knew it worked and booted from a floppy. It also seems I cant
mpount rad disk anymore either.
The Machine is a 2000 with a GVP 030/33 board with 16 megs of ram, a
picassoII board, 3 harddrives a DAT and a scsi scanner. I have made no
changes to the OS except to add CDROM drivers last month and it worked
since then.
All ideas welcomed!!! Thanks
Keith Shook
on AutoPilot!
Keith:
Someone else with your hardware configuration might be able to respond more
specifically (and I hope they do so), but here are a few ideas:
1. Hardware. From the symptoms you describe, the hardware seems to be working
at least partially ("sends the wrong characters"). About the only possibility
I could see here is one of the parallel lines being "stuck at zero" or "stuck
at 1". These wrong characters: do they relate in any way to what you're
sending? Are they reproducible; same output every time? Many printers have a
diagnostic mode, where they report every character they received as a hex
number. Worth a look.
2. Preferences. Double check; make sure you haven't somehow bumped to the
wrong printer driver.
3. Printer driver: could this program have somehow been corrupted? Try
comparing it to a backup copy, or replacing it.
4. If specific tests turn up nothing, it might be time to look for system
troubles that are manifesting themselves first in this area. Sure your power
supply is coping happily now you have the new CDROM interface installed?
Perhaps you might try memory diagnostics to see if there's any sign of RAM
failure. Not sure if 68030 caching might have anything to do with it, but you
could try playing with it.
"Linefeeds and Carriage Returns are missing" does make me first suspect the
hardware area. Might be a bad connector or a bad connection; since you've
tried swapping the 8520's, the problem does not seem to be there.
Hope you can spot it quickly; must be REALLY frustrating.
–Jim
Jim,
Thanks for the promt reply! The characters seem to be the same with the
same text sent ie i send test and I get paqp, I send test test I get paqp
paqp I send:
test
test
I get paqp paqp
Im sure a bit is visiting elvis somewhere but where?? I have a laserjet
II nad I cant seem to find anything about hex dump. I have restored
system files from DAT back 2 months and even reinstalled workbench 40.42
twice. Im sure the cables are good I had no problems printing from a 486
laptop thru the same cable. The funny thing is the GVP I/O extender
Parallel port is dead as well?? Not dead actually does the same things.
Ive tried printer drivers of all kinds from all kinds of disks.
Ive run virus checkers from all over, Ive stripped the system to a 68000
with 2 megs and it still does it. The CDRom is external on the bus with
the dat, scanner and NEC 4X CDROM Power supply voltages under full load
righ on the money. Everything was running smooth with no system changes
(except the CDRom a week or so before) and one minute it printed and the
next it dosent and hasent. Yes this is frustrating I have to use a 486
:<
Thanks for the help!! Got anymore anyone??
Keith Shook
on AutoPilot!
Keith,
A complete alphabet might be helpful. right now you're sending (in
hex) 74, 65, 73, 74 and getting 70, 61, 71, 70… it looks like something
has happened to both bits 1 and 2 (held hard off). That would also
account for the missing linefeed.
Any idea what chip may be causing this?? Ill run the alphabet today and
post it. Thanks!
Keith Shook
on AutoPilot!
Keith,
I'd guess the 8250 but you say you've swapped it.
Aha! so "test" becomes "paqp", does it? Then you're dropping bits 1 and 2!
(if you do hex, "test" is 74 65 73 74 …
"paqp" is 70 61 71 70 …
.. see how the value is dropping by a value of 2 [bit 1] or 4 [bit 2]?)
Looks certain you're into some hardware-chasing. You say the cables and such
work OK on another computer, in which case they must be OK. You've tried
swapping I/O chips and see no change in the problem; so unless you've blown
both of them (which would be a NEW problem you'd better solve before buying new
chips), they are OK. Now it's a matter of chasing wires from the chip to the
serial port socket. Either they are "open" (break in wiring, bad connection)
or they are "grounded" (something laying across the wires somewhere).
If you don't do hardware stuff, see if you can find someone who knows their
way around to do the looking for you.
–Jim (Any other suggestions, anyone?)
I have 2 new CIA's on the way and im going to field strip it again today
and take a good look around, Does the GVP I/O card pull its signals from
the same place?? Since it does the same thing?? It sure would be nice to
have some schematics!!
Thanks Again!!
Keith Shook
on AutoPilot!
Looks like the reply from Ron Legro should tell you about the GVP I/O card (I
don't have one so can't comment). The new CIAs are certainly the thing to try
now, although I would suggest you keep a watchful eye on the whole setup. I
find it odd that BOTH your existing chips should lose i/o lines 1 and 2, and
wonder: is there connected circuitry that's running those lines too hard?
–Jim
The GVP I/O board redirects parallel output, so it's not even letting that
output get to the normal parallel port. Therefore it's something before
that point that apparently is the cause. I don't know if bad CIA chips
could affect a file on its way to the port but probably yes. Maybe you
need to try TWO new CIA chips. They're cheap and having extras is good
insurance, anyway. Try printing to a file or using the system CMD utility
if that ability isn't in your application. Then inspect the output with a
text viewer and see if the same garble exists. That might help you narrow
the problem. I suspect that a print to file or CMD won't make use of the
CIA chips or the ports themselves, so continued garble might tend to point
at a software problem.
— RON LEGRO / Milwaukee via Autopilot 1.73
"Accept nothing less than a high standard of standardness."
Last time I had CIA problems (man, that sounds like I'm in a HEAP of
trouble!) moving them moved the problem from the printer to the floppy
drives.
DJ