14 Jun
2010
14 Jun
'10
8:39 a.m.
Tony,
I understand and respect your views and knowledge greatly. I have very
limited time these days, but at some point in the future I expect to have
time to enjoy the restoration of my collection. When that time comes, I
would like to have a spare set of processor boards strictly from the
standpoint that as the 11/05 has some LSI components, at some point it
seems likely that one of them will fail. As the boards are not becoming
any easier to come by, I would rather find a set now than wait until they
become unobtainium.
When I last looked at this machine, I did in fact verify that all of the
power supplies were correct and clean. It has been long enough ago that
I don't recall the exact error condition that the machine was exhibiting,
so I cannot describe it at the moment. The problem could well be with
some other portion of the system than in the processor proper, and this
is highly likely as you have stated, but it does not eliminate the fact
that I would still like to find a set of spare processor boards, and as
long as I am finding a spare set, I would like them to be known to be
working.
Cheers,
--tom
Tony Duell wrote:
Tony has a
good point here: it's usually the support stuff (power supplies,=
connectors) that bugger up a machine of this simplicity. -- Ian=20
Even if it isnt, I fail to see what good a spare CPU board set will do.
I know I am in aminority here, but I truely believe the only way to put
something right is to find the fault, and then to correct that fualt. Not
to randomly swap parts until the machine _appears_ to work again.
In any case, I don't know anyone who would be prepared to risk any PCB in
a machine where the health of the PSU isn't known.
The first thing to do is to discover just what it is and isn't doing. If
it were my machine I would start by checking the PSU with a meter and
'scope. PSU problems can make you think there's something wrong with just
about any part of the machine.
Then I would try reading and writing memory. If that fails. I'd stick in
a DL11 or something and try reading and writing that (if that works, it's
_likely_ there's a problem in the memroy section). And look at the Unibus
lines when doing meoroy accesses from the panel -- do they look right? If
not, there probably is a problem in the CPU -- then it's time to see
which signal(s) are wrong, what drives them, could there be a common
cause, check 'earlier' signals in the CPU logic, and so on.
-tony