🚀 go-pugleaf

In article <1992Oct2.121943.8893@bmw.mayo.edu> brunkhorst@mayo.edu (Geoff  
Brunkhorst) writes:
  > In article <5237@rosie.NeXT.COM> John_Graves@next.com (John Graves)  
writes:
  > > In article <1992Sep28.183056.20440@crash> coconut@crash.cts.com  
(Brian Dear)  
  > > writes:
  > > > NeXT tells me any lengthening of the cable would result in
  > > > loss of signal or degraded signal.  I say a properly shielded  
cable ...etc...

I am not recommending this, but:

From: william hunt <willie@iuvax.cs.indiana.edu>
Subject: Re:  lengthening a NeXT printer cable

I made a cable using seperate coaxs for each signal.  It
works fine, except that when the printer runs out of paper, putting in
more paper does not get the printer printing again.  Wagle found that
he had to turn the printer off using the soft window, and then it
starting printing again.

As far as signal skew goes, there should be no problem with data going
to the printer, since the CPU provides data and data clock.  However, the
return data may be suffering skew problems.  One solution, is to
adjust the length of the cable so the return data is skewed by exact
one clock cycle.  I will try this sometime and let you know how it
works.

Willie

From: william hunt <willie@iuvax.cs.indiana.edu>
Subject: Re:  lengthening a NeXT printer cable

I measured the delay with the oscilloscope and found that that I made
the cable a poor length for skew.  I added enough cable to the return
data line so that the cable is now equal to 80 feet in length.  This
adjusted the skew for the return data one full clock cycle.  Wagle and
I tested to see if the running the printer out of paper still caused a
problem.  We found that the problem is fixed.

So, the moral to this story is to make long cables multiples of 80
feet (for 81% c coax).  The cable can be slightly longer that 80 feet but
not by more than 15 feet.  The cable can not be shorter that 80 feet.
These cable length should work:
    0 to  15 feet  
   80 to  95
  160 to 175
  240 to 255
Of course the lengths will have to be adjusted for the velocity factor
if it is different than 81 percent.  Mabey, Next can make a special cable
and sell it!

Willie




--
-------------------------------------------------------
James L. Mynatt	Regional Systems Engineer	NeXT Computer, Inc.
USPS:		Echelon IV, Suite 400 
		9430 Research Blvd.
		Austin, Tx. 78759
Internet: Jim_Mynatt@NeXT.COM