[sdiy] Diode Matching

David G Dixon dixon at mail.ubc.ca
Sat Apr 11 02:23:24 CEST 2020

If matching is truly important, what I’d do is solder on sockets so I could
plug in different diodes.  They probably only need to be matched in pairs,
I’m guessing, so you could just plug in two and then try different other
pairs until you like the results.  I’m guessing you will like the results
after the first pair, because I’m betting there just isn’t that much
variation in 1N4148s.



From: Synth-diy [mailto:synth-diy-bounces at synth-diy.org] On Behalf Of Chris
Sent: Friday, April 10, 2020 2:49 PM
To: Schwarz Raphael
Cc: synth-diy at synth-diy.org
Subject: Re: [sdiy] Diode Matching


heresy perhaps, but have you tried -not- matching them? I'm curious how
poorly your ring mod performs with diodes just grabbed from the bag vs hand



On Apr 10, 2020, at 4:38 PM, Schwarz Raphael <raphschwarz at gmail.com> wrote:


Seems to me like you have a crappy multimeter, 

temperature only plays out when when you want to match diodes up to a few
mV. On a 1n4148 there is about 150mV variation between 0°C and 90°C . At
room temperature lets say 15°C to 30°C there should be less than 5mV . If
you need 2mV matching I would tape them on a sheet of paper and let them
rest a bit before measuring


Recently I started to test replacing hand matched diodes by matched
transistor pairs in a single SMD package to avoid having to match SMD
diodes. It works well in simulation but I need to test that

Le 10 avr. 2020 à 20:19, ColinMuirDorward <colindorward at gmail.com> a écrit :


Others will have more informed feedback, but did you rule out temperature
changes as the cause of your measurement drifts?




On Fri, Apr 10, 2020 at 4:48 PM Didier Leplae via Synth-diy
<synth-diy at synth-diy.org> wrote:

Does anyone have thoughts on the best way to do diode matching? 

I am building some ring modulators and need matched 4148 diodes, or similar.


So far I have tried three different methods of testing and find the results
to be confusing and unreliable.


    The first method I tried was using a DROK
01MS1FOYM>  transistor tester (which also tests diodes). I taped a strip of
new 4148 diodes to a sheet of paper in a notebook and one by one wrote down
the results. I noticed that most of them had pretty similar results, but
there were a few outliers that had wildly higher readings (forward voltage
drop, I think). So, out of curiosity I went and tested those again and found
that the readings the second time were much more similar to the others. 

    Then I started rechecking all of them and found that I was getting a
completely different set of readings. The differences between readings on
any one diode seemed to vary as much as differences between diodes. And as I
rechecked diodes I kept getting different results.

    So, I concluded that my transistor tester must not be very good.


    I then tried a the second method which is the diode testing setting on
my Etekcity multimeter <https://www.etekcity.com/product/100111> .  This
time I found that all the diodes read pretty much the same forward voltage
drop. (They seem to increase slightly the more diodes I tested but if I go
back to the first diode, for example, the reading will be at the same
increased value). I came back later and tested again, this time the results
were similar in that they all read the same value, except the value was
100mV or so lower. And again if I came back the next day similar results.
So, I concluded that my multimeter's diode testing mode might be unreliable.


    I then tried the third method of using the voltmeter, in mV setting, to
test forward voltage drop. Since my voltmeter's mV setting only goes to
2000mV, I used a 1.5V AA battery and a 10k resistor (to give about 0.15mA)
in series with my diode. The results were quite similar to the readings I
was getting using the diode testing mode on the multimeter. The values in
one set of tests would be virtually identical for all of the diodes, but if
I come back and test again they are quite different from the previous value,
but the same for all diodes.


I'm pretty confused by all of this. Do I just have a crappy multimeter? Or
are my diodes all miraculously matched?


Any thoughts would be helpful. 











Synth-diy mailing list
Synth-diy at synth-diy.org



- <https://www.instagram.com/colinmuirdorward/> 


Synth-diy mailing list
Synth-diy at synth-diy.org


Synth-diy mailing list
Synth-diy at synth-diy.org


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20200410/983cf983/attachment.htm>

More information about the Synth-diy mailing list