value of classic DEC machines?

From: Doug Carman <pdp11_at_bellsouth.net>
Date: Sun Oct 28 23:27:15 2001

Tony Duell wrote:
>
> > Anybody know if an 11/70 can be fit into a single rack?
> >
>
> The CPU is 1 BA11-F box IIRC, which along with PSUs and cooling space is
> going to take up half the rack. You could probably fit one or two of the
> memory boxes above it without impeding the airflow through the top CPU
> fans too much.
>
> So you could get the CPU and (some) memory into a single rack. If you can
> make do with one memory box then you can get a disk drive in there as well.
>
> But I would think 2 racks (one for CPU + memory, the other for
> peripherals) are more practical. 2 racks would give you enough space for
> a reasonable system.

Trying to put large things above the CPU will reduce the airflow to the
CPU. After having someone kindly give me two H960 racks, I was able to
put my own 11/70 system into a three rack configuration. This
eliminates everything from the CPU cabinet except the CPU and two large
fans in the top of the same cabinet. My MK11 memory, RX02, and SA72
storage are in a second cabinet. My BA11 expansion shelf, with all the
I/O panels that go with it, are in a third cabinet. This is closer to
what DEC would have configured, is much nicer to work on, and has much
better air circulation through the cabinets. Even at that, I don't have
the back doors on the cabinets, mostly because of all the cables.

The CPU needs lots of unrestricted airflow, especially if you are not
running it in a very well air conditioned equipment room. I believe
that DEC would only allow certain types of equipment to be placed above
the CPU. Basically, things that are not very deep, and would not
restrict airflow.

The only way I could see reducing it to a single cabinet would be to use
a third party memory solution (one that fits in the CPU), and not extend
the UNIBUS outside the CPU cabinet. This reduces you to using only the
4 SPC slots in the CPU.
Received on Sun Oct 28 2001 - 23:27:15 GMT

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:34:21 BST