Percom Drives

Date: 12/03/94-03:27:51 AM Z

Subject: Percom Drives
Date: Sat Dec  3 03:27:51 1994

Continuing my vain effort to catch up my email, I've followed with
interest the thread on Percom drives.
I just completed rebuilding 6 of them & am in the process of preparing
a manuscript on the subject for AC (yes, The Alchemist is yet alive!).
The controller on the RFD40-S1 does support 80 tracks but does NOT
support double-sided disks.  I've not had a double-sided model in my
hands but an ad from the place in Chantilly Virginia that used to
service Percoms indicates the double-sided/double-density model was
designated RFD44-S1.
What I discovered about the RFD40 (using SpartaDOS 3.2d) was that it WILL
lay down a true 720K format on a 3.5" drive.  However, the controller
will only access the first 360K of it.  So, in effect, the 3.5" drive
becomes a 360K drive with a single-sided controller.  The RFD series drivs
can also be programmed using the switches to properly recognize a double-
density slave drive upon bootup.
The same cannot be said of the AT88-S1PD drives.  With a double-density
mech installed, the AT88-S1PD WILL format single-sided/double -density
on either a 5.25" mech or a 3.5" mech.  However, a double-density
mech slaved to an AT88 will onlt be recognized as a single density mech
upon bootup.  But-- if you format a double-density disk on the slave,
it will thereafter recognize the slave as double-density (until, of
course, you turn it off).  Sheesh!
Percom could never make up its mind which type of connector to use for
the slaves.  Some models used 34-pin mass termination pins, others used
card-edge connectors.  And just to complete the confusion, the AT88-SPD
used BOTH.  On those models, the mass termination pin connector is for
the parallel printer port.  The printer port on these models works fine
when directly connected to an Epson printer, but when you place a print
buffer in series with the printer there is weirdness.  (The print buffer
problem will be the subject of a whole nuther article).
I've seen many horrors involving the power supplies in Percom drives.
From the very beginning to the very end, NO ONE at Percom EVER under-
stood how to properly mount voltage regulators.  I've replaced about
a million failed regulators in Percoms, all because of the way Percom
mounted them.  The power supply failures gave the Percom drives a bad
reputation which I feel they did not deserve.  The sad part of it is,
the fix is ridiculously simple.
Percom tech people also didn't know how to diagnose hardware problems
over the phone.  When my friend's At88 died & he called Percom, they
charged him $25 to replace the SMC-made 1795 FDC chip on the controller
card (they mailed him the chip).  He plugged it in, the drive was still
dead.  (It was a failed regulator- as usual- that was the culprit!)
Percom claimed the SMC-made chips tended to overheat.  Baloney!  Several
of my Percoms contain SMC chips & run just fine.
I disagree with recently proposed nomenclature I saw posted here on
RFD-series drives.  It read: RFDxx-Sy where is number of tracks on the
where xx is the number of tracks on the originally installed mech &
y is the number of sides on the mech.  I don't think so.  Such literature
as I've seen indicates the model nomenclature should be as follows:
                   x=0 --> a single-sided drive
                   x=4 --> a double-sided drive
                   y=1 --> a single-mech drive unit
                   y=2 --> a double-mech drive unit
So, is everbody suitably confused now?    :)
           - Ben Poehland
             The 8-Bit Alchemist

Return to message index