• Please review our updated Terms and Rules here

Parallel printer suddenly not working

Overmann

Experienced Member
Joined
Apr 21, 2017
Messages
102
Location
Norway
I have a very nice Star LC24-200 color that I've been using with my old 8088 and 386. Now it's been several months since last time and now it suddenly isn't working. I've tried it with my 8088, my 386, my Win98 pentium and my Amiga.
On the DOS machines the printer prints but only garbled nonsense. Some of the original text might be there but it only passingly resembles actual words. The letters are far apart and printed on top of each other in places. It's very strange. The Win98-machine and the Amiga won't print at all, despite proper drivers being installed.

The printers self-test (both short and long) come out looking perfect. It is set to Standard emulation mode (Epson) and not IBM which I think it proper.

Have anyone seen this? Can this occur if the parallel cable is pulled out while the printer is turned on? I have spent the day troubleshooting, but have now arrived at a dead end.
 
I've seen a couple of cases where the drivers/receivers on a parallel printer interface have been zapped by ESD. I don't know if this is the case for your unit, however.
 
Do you know of a way to test it? I don't know if I ever unplugged the cable while it was on, but I just can't see how this just happened by itself.
 
It happens. This is one of those rare times that I would just shotgun replace the drivers/receivers.
 
@KC9UDX
What do you mean by shotgun replace the drivers/receivers?

@Chuck(G)
I do take pretty good care of my things, and I can't see tit being subjected to static electricity. Wow, this really annoys me. I was looking forward to using it on my Amiga (that I finally got set up properly) and it's all dead-ish.
 
"Shotgun replacing" is a poor troubleshooting technique where components are replaced without testing. Most of the time this can introduce new problems, making troubleshooting very difficult. But occasionally there's good reason to do it. This is one of those times, in my opinion.
 
I've opened it up but don't know what to look for (not surprising, sadly), but some of the chips are socketed. I tried reseating them but it had no effect.
I attached a couple of pictures of the area around the port. I just now realised that the traces are not visible, but I thought it might be interesting either way.


IMG_20190316_204337.jpg
IMG_20190316_204413.jpg
 
Well, you may be out of luck, but I'd start with taking one of the data lines from the port connector (pin 2 through 9) and your continuity tester and see what pin on what chip shows continuity. That at least might give you a rough idea.

Another approach would be to hook the thing to a system that allows direct access of all pins (e.g. a PC running DOS) and see what bits might be stuck by manipulating the I/O port pins directly.
 
Well, you may be out of luck, but I'd start with taking one of the data lines from the port connector (pin 2 through 9) and your continuity tester and see what pin on what chip shows continuity. That at least might give you a rough idea.

Another approach would be to hook the thing to a system that allows direct access of all pins (e.g. a PC running DOS) and see what bits might be stuck by manipulating the I/O port pins directly.

Interesting! I'll have to read up on this. Perhaps I can finally get some use out of my fancy digital oscilloscope :p
The manual for the printer lists a rather useful basic program for use with the printers HEX DUMP function.

10 LPRINT WIDTH “LPT1 : “ ,255
20 FOR 1=0 TO 255
30 LPRINT CHR$(I) ;
40 NEXT I
50 LPRINT
60 END

I've tried typing it in qbasic with my 386 but I get errors when pressing RUN i get an error already at WIDTH:
"Expected: TAB or SPC or , or ; or expression or USING or end-of-statemen"

I've tried typing the code into wordstar and printing and it gives me a line of HEX but Wordstar isn't running code and I guess the printer isn't either.
 
Last edited:
Interesting! I'll have to read up on this. Perhaps I can finally get some use out of my fancy digital oscilloscope :p
The manual for the printer lists a rather useful basic program for use with the printers HEX DUMP function.

10 LPRINT WIDTH “LPT1 : “ ,255
20 FOR 1=0 TO 255
30 LPRINT CHR$(I) ;
40 NEXT I
50 LPRINT
60 END

I've tried typing it in qbasic with my 386 but I get errors when pressing RUN i get an error already at WIDTH:
"Expected: TAB or SPC or , or ; or expression or USING or end-of-statemen"

I've tried typing the code into wordstar and printing and it gives me a line of HEX but Wordstar isn't running code and I guess the printer isn't either.

try changing line 10 to:
10 WIDTH "LPT1:", 255

I suspect though this is intended for gwbasic.

Frank IZ8DWF
 
try changing line 10 to:
10 WIDTH "LPT1:", 255

I suspect though this is intended for gwbasic.

Frank IZ8DWF

I tried that now and get a "device fault at 30"
I downloaded GWbasic and typed it in there and got a "syntax error at 10". I then changed the line as you suggested there as well and got a "device fault at 50".
 
I tried that now and get a "device fault at 30"
I downloaded GWbasic and typed it in there and got a "syntax error at 10". I then changed the line as you suggested there as well and got a "device fault at 50".

Very interesting. It means that probably your fault lies in some handshake signals of the centronics port (and maybe some data lines). I would check the various signals (strobe, select, busy, ack, ...) for any stuck level.
It would be interesting to know on what occasion Basic/DOS puts out a device fault signal when dealing with an LPT interface.
Frank IZ8DWF
 
If the self-test works perfectly, then I'd suggest that the bulk of the innards of the printer is fine.

I'd suspect the printer cable. If you have another one, try that. Does it do the same? Can you borrow another from someone else, just to test.

If the problem persists, then the problem is with the interface within the printer, which may be impossible to fix.

The DOS machines are sending specific codes for specific characters, sounds like some are getting thru, and others are not, and others are getting garbled and being interpreted as other things (like TAB, or backspace, or some other char entirely). This is quite normal where certain data lines in the cable are broken. The WinDoze drives are creating a graphics image in the computer and trying to send that, and the printer is getting nothing viable at all, and the computer is recognising that and reporting that device is not there.

Check the cable you have for any hint of damage, incl crushing. Check connectors at both ends (they are different) again for damage. And dirt or corrosion of any pins or connectors (the printer end does not have 'pins').

Geoff
 
I've tried 3 different cables now and none of them work. But I noticed something strange. The printer prints something different each time. Even using the same cable on the same document. That can't be normal? :p
I know I have more cables around so I'll have a look. You never know, these three could be bad. Some of them are Amiga cables, but I assume that a parallel is a parallel..?
I'm reading up on checking signals (strobe, select, busy, ack, ...) for any stuck level. Brand new stuff to me.
I did another long diagnostics test now just to be sure and it comes out prestine. I have set the printer back to standard startup settings. Emulation Standard (Epson) and input buffer (not download buffer).

It looks as though I'll be picking up another, brand new, parallel printer during the week, so atleast I'll know if the cable is good.
 
Back
Top