SBC6120 (a build-it-yourself PDP-8 clone) grammer checked
ard at p850ug1.demon.co.uk
Wed Jun 14 19:58:19 CDT 2006
> On Saturday 10 June 2006 05:40 pm, Tony Duell wrote:
> > For electronic test instruments (not including hand tools, etc), my order
> > would be something like
> > 1) A good VOM (analogue) or DMM, whichever type you prefer.
> You need both! Though I do tend to reach for the digital stuff more often
I would agree, but you can get away with only one when you're starting out.
For many years I used an analogue VOM only. To be honest, it's rare that
you need to make an accurate meashuremnt when repairing a classic
computer (you want to know if the 5V line is pressent and correct, but if
it's acutally 4.95V, things will still worl fine). Then I picked up a
cheap Fluke 85. I use that almost exclusively now, but the analogue meter
stays on my bench, it's better for osme things.
> than not, but that's more a matter of physical convenience than anything
> else (little teeny DMM sits on my desk AAMOF).
> > 2) A logic probe
> I've not found those as useful as I'd have thought at one time. Got one here
Tastes differ, I use one all the time. Actually, I've not used a normal
logic probe for the last 8 years or so, that's only because I bought a
LogicDart. But I debugged an awful lot of old computer stuff with a cheap
Radio Shack probe, and could do it again.
> > 3) A bench power supply (at least 5A output)
> 5A at what voltage?
Adjustable, at least 0-20V. I have a nice 0-30V, 0-10A PSU on the bench
(acutally, the Velleman kit). I use it a lot.
> > 4) The other of the VOM/DMM pair
> > 5) A 'scope or logic analyser, depending on what sort of work you're doing
> > 6) The other one of those instruments
> Scopes I've been using for ages, but never had occasion to use a logic
> analyzer. Or could afford one.
There are plenty of second-hand analyesers about that are fine for
classic computer repairs. I have an old Gould 100MHz unit.
As regards using it, it depends one what you work on, but I've found it
very useful for tracing microcode in some older machines. See what the
processor is really doing (and then work out if it should be doing it).
> > I find it helps when prototyping to be able to build one bit of the
> > circuit, check it works properly, then add a bit more, and so on. This is
> > easy with wire wrap, harder on a PCB.
> That was the way I learned how to build things as well, only we started with
> a hunk of blank sheet metal and ended up with a 5-tube radio. :-)
Somewhere I have a pile of 1950's Practical Wireless magazines which
include a set of articles like that. Start with a crystal set, add valve
amplifiers, etc, finally end up with a superhet. The only thing is, it
was lethal. The valve heaters were supplied by a 6.3V transformer, but
the anode supply (B+ line) was got by half-wave rectifying the mains. I
dislike a live chassis at the best of times, on an expermintal set-up
with headphones, it would seem to be suicidal!
More information about the cctalk