Roman Black wrote:
>Maybe as another suggestion for a competition
>idea, what about a universal semiconductor tester?
>Like three leads, A,B,C, and a 2-line LCD display.
>You can clip it on any transistor or diode and
>it tells which leg is which, NPN, PNP, fet,
>darlington, etc etc, even measures beta and
>saturation drop. All these things would be easy
>to measure if you could adjust voltages and measure
>voltages, say with a 16F876 or even a F628.
>Just an idea. Hey, i'd like one!!


Hey, Roman, I'll tell you what my idea is, and maybe you can tell
me what is wrong with it :).

I have been mulling this over for several months - even before
that UK company started selling exactly what you mention above.
My device would work with the little virtual instruments I have
been producing, so it relies on a PC for control/display, rather
than having its own LCD. This does make it easier to extend to
different devices, since all you have to do is change the PC
s.w. rather than the PIC firmware..

The plan is to have it work symmetrically on 3 pins, and
automatically identify transistor C, B, E, etc, as the first step,
then go on to test and identify the type of device. Should work
with BJTs, FETs, SCRs, diodes, etc.

The key is to use 6 pins in groups of 2 for different stim and
record regimens. This is purely digital, but with more complex
circuitry could be made more sophisticated. The current design
was made to be easily added to an existing device. It would not
work with zeners > 5v.

  PIC
-------+
       |                                   D.U.T.
   RB7 +-----3.3K----+-------> A ------------+
       |             |                       |
   RB6 +-----330-----+                       |
       |                                  | /  e
   RB5 +-----3.3K----+-------> B ---------|
       |             |                 b  | \  c
   RB4 +-----330-----+                       |
       |                                     |
   RB3 +-----3.3K----+-------> C ------------+
       |             |
   RB2 +-----330-----+
       |
-------+

The D.U.T. connects at A, B, C. The 330 R's are in my existing device,
but could be eliminated in a dedicated device. Basically, you use
pins RB2-RB7 as either outputs hi/lo, or as inputs for sensing.
Of each pair, normally one will sense while the other stimulates.
Flipping the pins in a pair changes the level of stimulation.

To test diode action between A and B, set RB7=out_hi, RB6=in_sense,
RB4=out_lo, RB5=in_sense. For diode conduction A->B, both RB6/RB5
will read low. By doing the inverse to that just described, you can
show no conduction in the reverse direction.

You go thru this regime for all pin combinations, discover you
have back-to-back diodes, and it's a PNP with the base at B.
Discovering whether A or C is the emitter/collector takes a
dynamic measurement.

By connecting 3 A/D channels to points A/B/C, and also possibly
using 3 more digital pins with 33K R's, you can measure more
complicated parameters, like beta, etc.

There are probably many improvements that can be made to this basic
idea. Of course, you could go whole hog and do curve tracing/etc.

best regards,
- dan michaels
www.oricomtech.com
=======================

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu