LP11 problems -- Help!

[Rob Warnock]

+---------------
| The first system is a VAX-11/750 running 4.2 BSD.  The printer
| is a Dataproducts B300-2 band printer with a Winchester interface.
| It is connected to a DEC LP11 controller (module M7258) via a
| 70-11212-25 cable.  We get consistent I/O errors...
| The kernel recognizes the lp on booting and gets the address
| and vector right.  The entire guts of the printer (it's brand-new)
| have been replaced to no avail.  Cables have been swapped, reversed,
| inverted, and turned inside-out to no avail.  Two different controllers
| have been tried.  Yes, the special file /dev/lp is there...
+---------------

One possible cause of your problem (posted, in case others need it):

Back at DCA several years ago, we had some problems with DEC LP8s like this.
The Dataproducts printers have a small resistor termination board, generally
mounted somewhere near the Winchester connector (but sometimes up in the
main electronics package), which contains the pull-up (or pull-up/down)
resistors which terminate the printer cable. The LP11 controller, if
it is like the LP8, does NOT have internal pullups on the open-collector
outputs or the ready/error/... inputs (some of the inputs are actually
pulled DOWN below zero, and take a stiff pullup externally to get them
up to logic levels). When a printer comes from DEC it is supposed to (!)
have the resistors configured for the DEC controller, but when you buy
the printer yourself, you have to take that responsibility.

The problems that result if the resistors are not correct can range from
none (everything's still o.k.) to flakey (SOME characters in SOME positions
are SOMETIMES corrupted) to dead (won't even come on-line). It depends also
on the length of the cable and the noise environment around the cable (e.g.,
how good is the frame/power ground between the CPU and the printer?).

Swapping cables and controllers doesn't fix it, of course, but in some
of the flakey cases it will "change" the problem, causing no small amount
of hair-pulling among the diagnosers. In some borderline cases, swapping
a controller will fix it for a while... only to fail later.

You will need the maintenance manuals for both the LP11 and the printer
to verify/reject/correct this problem (if it exists).

p.s.:	Q: "How many programmers does it take to change a lightbulb?"
	A: "Programmers don't change lightbulbs, lightbulbs are hardware!"

(p.p.s.	Santayana said, "Those who do not learn from history are doomed
	to repeat it." Netnews is a way of spreading history around, so
	we don't lose it.)

Rob Warnock

UUCP:	{ihnp4,ucbvax!amd}!fortune!redwood!rpw3
DDD:	(415)369-7437
Envoy:	rob.warnock/kingfisher
USPS:	Suite 203, 4012 Farm Hill Blvd, Redwood City, CA  94061