schottky diode again..

From: Richard Erlacher <edick_at_idcomm.com>
Date: Wed Nov 3 18:42:59 1999

SInce I sometimes take testing fairly seriously, I have several meters with
several slightly different diode testing modes. I've found none which
indicate a bad diode is good, nor do any of them indicate a good diode is
bad.

One of my testers is something I built for matching diodes, and and
transistors, and it runs a pretty constant 1.26 or 12.6 ma (from an LED
biasing a 2N3904) through a diode, and a panel meter displays the forward
voltage.

Not one of my meters exhibits behavior as described by "Wizard" however. I
can't imagine what the problem is, but I agree that digital meters can
produce unpredictable results when used in a way not intended by the
designers.

Generally speaking, the DMM is a Digital Panel Meter with a fixed reference
voltage built into it, looking a the output of an analog or even mechanical
switch, the purpose of which is to run the appropriate current through/into
the component under test to generate a voltage scaled to display a value
reflecting the value of the component under test. In the case of diode
test, that's really easy, since it's a current within a fairly forgiving
range, and which will, with most any diode produce a voltage reflecting what
that particular diode drops at that current. Almost all measurements can
easily be done in this way. Voltage is easiest, of course.

Dick
-----Original Message-----
From: sms_at_antinode.org <sms_at_antinode.org>
To: Discussion re-collecting of classic computers
<classiccmp_at_u.washington.edu>
Date: Wednesday, November 03, 1999 4:40 PM
Subject: Re: schottky diode again..


>> From: "Richard Erlacher" <edick_at_idcomm.com>
>>
>> If you have a diode of any kind which behaves more as a resistor and less
as
>> a diode, i.e. it measures more or less the same in both directions,
though
>> differently on different settings on your DMM, I'd say it's broken.
>
> I'd say you have a modern meter, and neither of you knows how to use
>it. The "diode" range on the meter is there because the other
>resistance ranges are not useful for testing diodes. Read the manual
>for the meter, or buy a curve tracer, or build a continuity tester from
>an old flashlight (two-cell preferred), or ...
>
> I'm curious. Have you tried the same family of (mostly misleading)
>tests on a known-good, plain-old silicon diode?
>
>> > From: jpero_at_cgocable.net <jpero_at_cgocable.net>
>> > ...
>> > says "good" but I know it's very low current also low voltage so that
>> > would usually not screen out sick diodes.
>
> The current and voltage for the resistance ranges are probably much
>lower, which is why the readings are so useless. (Repeat previous rant
>here.)
>
>> That why I used resistance ranges just to be sure.
>
> This must be some new meaning of "sure".
>
> There are some subtle defects possible in a diode which will not be
>revealed by the "diode" test on a multimeter, but that's almost
>certainly the best test offered by the meter.
>
>> Wizard
>
> Apparently a misnomer.
>
>------------------------------------------------------------------------
>
> Steven M. Schweda (+1) 651-645-9249 (voice, home)
> 1630 Marshall Avenue #8 (+1) 612-754-2636 (voice, work)
> Saint Paul MN 55104-6225 (+1) 612-754-6302 (facsimile, work)
> sms_at_antinode.org sms_at_provis.com (work)
Received on Wed Nov 03 1999 - 18:42:59 GMT