laser printer dimensional accuracy?

[Chapman Flack]

Hi,

I picked up a used Phaser 750 laser. If I tell it something like:

true setstrokeadjust 0 setlinewidth
20 20 550 720 rectstroke
showpage

(that is, draw a rectangle 550 by 720 points with ll corner at 20,20)
and measure the rectangle with a ruler, it's 550 by 723 points.

On this printer, paper feeds in the y direction, so the dimensional
accuracy is spot on for transverse measurements, but about 0.4% off in
the feed direction.

I'm actually used to cheap inkjets being more accurate than that, and
this is kind of annoying because I sometimes print scales for
instruments and I want them dimensionally accurate. Of course by
measuring my particular printer I can just throw a scale factor into
the PostScript code, but that would make the file specific to my
printer ... PostScript was supposed to standardize dimensions so that
wouldn't be necessary ....

Is this typical accuracy for a laser? I suppose with an inkjet using a
paper feed stepper and a laser using a continuous feed, this kind of
thing could be seen on a laser if the feed speed is just a tad high.

Anybody know if there's a way to correct for it on a Phaser 750? The
front panel menu has adjustments for margin offsets but not for
stretch. One approach might be to add some code to the printer's
Sys/Start job (though I'm not sure what to put there); another might be
to look for a feed speed adjustment on the print engine. I have what
claims to be the service manual for this thing, but there's a great
deal it doesn't document; nothing about feed servo adjustments, and
nothing to describe any of the engineering test page procsets or the
RRCustomProcs procset visible in the postscript interpreter. At least
nothing I've found ... the organization of files on the CD is a bit
weird.

Thanks,
-Chap

 

[ato_zee]

On this printer, paper feeds in the y direction, so the dimensional
accuracy is spot on for transverse measurements, but about 0.4% off in
the feed direction.

In the feed direction accuracy would vary according
to feed roller diameter accuracy, since it's usually a stepper motor,
or shaft encoder, which assumes the specification for roller
diameters, this is of course subject to production tolerances.
For elastomer rollers shrinkage with curing, and pressure/aging
are pretty well uncontrolled. For new printers you may do better.

Paper dimensional change is also a factor, whether laser or
inkjet and may vary according to the lay of the paper, being
different on each page axis, similar to how wood changes with
moisture content, along or across the grain.

I investigated this in relation to printed circuit layout, you can't
produce a paper layout, stick it to a board, drill holes, and find
that the pins of the IC holders go through the holes. You have
to keep enlarging the holes and trying the IC holder/board in
a vice until it goes in. Well almost.

I found that you can solve it by using an HP or other professional
plotter, accuracy is the name of the game for these, that's
what they are made for. Residual errors for a plotter, with a
maintenance contract, are negligable. You can also have
colour, but be advised that for maximum accuracy, use mylar
drafting film for dimensional stability.

As an after thought amateurs often restring the drive cables
in plotters to save on a maintenance contract, or call out
charges. The invariably don't have the manual and
tension gauges, often restringing it wrongly so the drive
cables follow the wrong path, and calibration is all over
the place. Mfrs engineers tut, tut, oh my dear, who did
this. So eBay plotters should carry a health warning.

 

[DaveG]

I investigated this in relation to printed circuit layout, you can't
produce a paper layout, stick it to a board, drill holes, and find
that the pins of the IC holders go through the holes. You have
to keep enlarging the holes and trying the IC holder/board in
a vice until it goes in. Well almost.

I've done that before but used laser/copier compatible acetate sheets for
OHPs. Didn't seem to be any noticeable shrinkage on printing for boards
up to 8x8. Not enough to affect the hole pitch anyway.

Note for anyone trying this. Do NOT use normal OHP sheets. They melt and
ruin your expensive fuser. I had a call out to a site because this had
happened. £180 for a new fuser. Then the night shift guy did it again
the next night 'cos no one told him it was his fault in the first place.
D'oh!