peterbudduk
Newbie
- Credits
- 18CR
[ukvac] TECH: IC Tester
Guys,
I am about to build an in-circuit IC tester based on something I saw on
the 'net a few months ago (but I cannot find it now).
It works as follows, it compares the outputs of the IC on the PCB
("device under test" or "DUT") with a known good device inserted into a
ZIF socket of the tester. The way I was going to build it was to have
an array of switches one for each pin of the DUT. If the pin is an
input then the switch would directly connect the DUT with the known
good device. If it is an output then the switch would connect the
appropriate pins to an exclusive or gate to indicate via a transistor
and a LED (and possibly a pulse stretcher) whether they are behaving
the same way.
It strikes me, however, that there could be a simpler way that avoids
the use of switches and exclusive or gates. If every pin (ignore
power) of the DUT was connected to the known good device via a
resistor, then the transistor and the LED could simply detect whether
there is a voltage drop across the resistor. Unfortunately, I do not
know enough about the input impedances of TTL ICs to know whether this
is feasible. Bear in mind that I will probably be testing a mixture of
TTL and LSTTL ICs. Any comments would be very welcome.
Note that I am aware that an IC tester of this sort will not be a
panacea - it is primarily intended as a quick test to spot likely
problems.
Peter
------------------------------------------------------------------------
ebates.com. Earn up to 25% cash back for shopping online at 75 stores
like Borders, CDNow and Beyond.com. Refer a friend and earn even more!
http://clickhere.egroups.com/click/690
eGroups.com home: http://www.egroups.com/group/ukvac
http://www.egroups.com - Simplifying group communications
Guys,
I am about to build an in-circuit IC tester based on something I saw on
the 'net a few months ago (but I cannot find it now).
It works as follows, it compares the outputs of the IC on the PCB
("device under test" or "DUT") with a known good device inserted into a
ZIF socket of the tester. The way I was going to build it was to have
an array of switches one for each pin of the DUT. If the pin is an
input then the switch would directly connect the DUT with the known
good device. If it is an output then the switch would connect the
appropriate pins to an exclusive or gate to indicate via a transistor
and a LED (and possibly a pulse stretcher) whether they are behaving
the same way.
It strikes me, however, that there could be a simpler way that avoids
the use of switches and exclusive or gates. If every pin (ignore
power) of the DUT was connected to the known good device via a
resistor, then the transistor and the LED could simply detect whether
there is a voltage drop across the resistor. Unfortunately, I do not
know enough about the input impedances of TTL ICs to know whether this
is feasible. Bear in mind that I will probably be testing a mixture of
TTL and LSTTL ICs. Any comments would be very welcome.
Note that I am aware that an IC tester of this sort will not be a
panacea - it is primarily intended as a quick test to spot likely
problems.
Peter
------------------------------------------------------------------------
ebates.com. Earn up to 25% cash back for shopping online at 75 stores
like Borders, CDNow and Beyond.com. Refer a friend and earn even more!
http://clickhere.egroups.com/click/690
eGroups.com home: http://www.egroups.com/group/ukvac
http://www.egroups.com - Simplifying group communications
