cluster size

rca@rle-vlsi.mit.edu
Wed, 3 Jul 91 13:29:26 edt

I have comments on two topics:

Re cluster size hacking: I had no floppy disk for my machine, so I just
backed up via LapLink to another machine, tweaked the magic locations and
booted from the internal ROM disk D: by pressing control-alt-Fn-break.
Then I ran the format command with the command line given in the article
and used LapLink to restore the contents of C:. It does save quite a bit of
space!

Re power connector: I have a friend who has opened up his power supply and
drawn out a schematic of the contents. It is fairly straightforward, but
there are a few odd things in it. Including a transistor and diode which
do nothing except connect to a couple of pins on the power supply end of
the cable. One probably doesn't want to know what the heck that is all about.
It may just be a special circuit for notifying the computer that this power
supply is present. I will post the pinout of the power cable as soon as
I get the final copy from my friend (he has no email connection.)

Regarding the power extension cable idea, here are some reasons why a MAC
serial cable might not be a good candidate:

1) The individual wires may not be large enough for the current load.
This is an issue because the MAC cable is intended for serial use,
and would use the smallest practical wires. My friend reports that
the disk drive may draw large currents from the power supply, thus
the size of the wires is a bigger issue when the disk is in use.

2) The presence of the longer cable may upset the voltage/current
regulation circuitry in the supply module. I would hope that
appropriate precautions were taken in the design of the power supply
system (inside and outside the computer), but you can never tell
what corners may have been cut. In any case, a longer cable will
introduce more voltage drop in the current-carrying wires and
under high load (disk drive) this may be too much for proper
operation.

3) There may be an RFI problem. The basic problem can usually be
handled inside the box, but I am suspicious of the shortness of
the cable and the presence of not one, but two RFI warts. This may
mean that they took shortcuts inside the box to save space and
stuck the warts on outside to compensate.

My friend and I will probably try the MAC cable or something like it, just
to see what happens. I will report back any problems that we see. My
gut feeling is that a short extension will be OK for non-disk use, but that
operation with the disk will want the shortest possible cable.

Bob

----------------------------------------------------------------
Robert C. Armstrong rca@rle-vlsi.mit.edu