Matching diodes

[LektroiD]

What's the best (or easiest) way to match the diodes for the EMS filter?

Are diodes heat sensitive like transistors? Presuming it is a similar technique to transistor VBE matching...

I would also like to know how to set up the test gear and what to look for on the multimeter/scope?

If anyone has a schematic, or breadboard/stripboard layout to make a simple diode matcher that would be most helpful.

(OR)... can I just throw any diodes in there and expect a reasonable sounding filter - does it really affect the sound that much?

[analog_backlash]

Hi LektroiD.

Most multimeters have a diode test setting (as part of the resistance range(s)), you just have to match the voltage drop across them. Alternatively, you can put a resistor in series with the diode, apply a voltage across the resistor/diode and measure the voltage across the diode, as in this previous posting:

http://electro-music.com/forum/post-188963.html

A "schematic" (for an entirely different end use) is shown at the bottom of this page:

http://www.qrp.pops.net/xmfr.asp

Yes, they are temperature sensitive (you can use them for making electronic thermometers) so you need to let the temperature stabilise before taking a reading.

I hope that this helps,

Gary

P.S. I think that I may have been a little ratty with you once before. Sorry, I was in one of my moods...

[LektroiD]

Thanks Gary, the only thing I'm not really sure about is 'voltage drop'. What's the difference between testing that and straight voltage? I have a diode setting on my tester, but not sure what I'm looking for exactly (it's all new to me).

*EDIT* I found a video on Youtube demonstrating what I need to know.

Don't worry about being ratty; I'm not one to hold grudges, which is probably why I don't remember the incident.

[analog_backlash]

I'm glad that you've worked it out now LektroiD.

Any forward biased p-n junction (i.e. where the p is held more positive than the n), whether in a silicon signal/rectifier diode or a silicon transistor will show a voltage drop of (very approximately) 0.6V across it. The value is lower for germanium diodes/transistors and also for Schottky diodes. It is quite variable from one diode to another, so you need to match them as well as can (depending on the accuracy of your multimeter).

Thanks for not remembering my last post to you - it wasn't offensive, just a little bit condescending .

Gary

[Parkchemiker]

Hello,

i built a diode matcher using the well known Wheatstone bridge. The circuit needs following parts:

9V battery, 1X 1K, 2 X 1K precision, 1X 100K, 1X Poti 1K lin, a pair of silicon diodes to be matched and a digital multimeter.

I have tested the circuit with LTspice and on a breadboard.

Matching here means turning the knob on the 1K-Pot and looking at the multimeter. If the diodes are equal, the multimeter always shows 0mV, otherwise not.

---

DiodenMatcher.jpg
Description:
Filesize:	120.97 KB
Viewed:	325 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

---

DiodePairing.jpg
Description:
Filesize:	521.53 KB
Viewed:	303 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[Freq Band]

I don't know if testing for reverse current is necessary, but I thought I'd add this if it's ever needed.
Also remember that some clear glass diodes are sensitive to light, so cover them from light when testing..... or get the black epoxy type.

[/img]

---

my_diode_rev_current_test.jpg
Description:	reverse current test for diodes.
Filesize:	20.23 KB
Viewed:	270 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[LFLab]

I just matched them for forward voltage drop using my DMM, for my yusynth EMS-esque VCF build.
Works just fine (also matched the caps in the ladder and a bunch of transistors in the discrete OTA).

Have a bunch of PCB's left if anyone is interested (parallel mount, no wiring, also has a VCA)



The VCA is a slightly adapted version of Yves' VCA.