5 Mar
2008
5 Mar
'08
11:14 a.m.
From: ard at p850ug1.demon.co.uk (Tony Duell)
Subject: Logic testing
I fail to see what silicon .vs. germanium has to d owith the
polarity of
the logic cignals. In general PNP transistors, and for that matter
PMOS
fets, imply -ve logic levels, and plenty of machines were built using
those components. Also ECL chips have -ve logic levels (around -2V)
wrt
ground.
I bow to your greater knowledge. I was employed as a programmer and
picked up hardware later.
The HP Logicdart (a handheld 3-input logic analyser) can certainly
handle
the PMOS logic signals in older HP handheld calculaotrs and the -15V
logic levels in the discrete transistor circuitry of the HP9100. I
don't
see why it'd not work for you. And most _decent_ logic analysers (as
opposed to the TTL-only toys...) can handle ECL levels.
In cany case, I cna't believe the signals in your machine are all that
fast. Is it not possible to make up level shifter stages to turn them
into TTL-level sgiansl for testing?
Indeed, the clock is 1MHz, 750ns high and 250ns low, with lots of
wobbling about as the signal overshoots.
Thanks Tony, if I see a logic analyser going cheap (though that is
unlikely) I will buy it.
6 Mar
6 Mar
3:13 a.m.
Roger Holmes wrote:
Tony Duell wrote:
Roger Holmes wrote:
Not so useful
with peripherals though, I find a storage scope is
invaluable for them, the more traces the better, so I suppose a logic
analyser would be useful, but does one exists for MINUS 6.3 volt
logic? I always presumed they were only invented after silicon
replaced germanium and so only work with positive logic voltages.> >
I fail to see
what silicon .vs. germanium has to d owith the
polarity of
the logic cignals. In general PNP transistors, and for that matter
PMOS
fets, imply -ve logic levels, and plenty of machines were built using
those components. Also ECL chips have -ve logic levels (around -2V)
wrt
ground.
I bow to your greater knowledge. I was employed as a
programmer and
picked up hardware later.
Most Ge transistor equipment (pre-mid-60's) was PNP/-V, most post-mid-60's
was Si/(NPN if bipolar)/+V, as per the supposition of your (Roger's)
original question.
Historically, or in terms of the prevalence of according systems, there
is a correlation between the material and the polarity, although it's
not a technically imperative correlation (there are exceptions).
I'm not sure of the cause of the correlation, a guess would be that
manufacturing issues lent a preference to Ge-PNP and Si-NPN.
6087
days inactive
6088
days old
1 comments
2 participants
participants (2)
-
hilpert＠cs.ubc.ca -
roger.holmes＠microspot.co.uk