'Matching' transistors

[The Captain]

Hi- can anyone explain how I measure transistors for matching? Ive done the obvious with a built board and just attatched a battery up to it, but I cant get any ~1.6 mv readings which Im expecting, only ~2.5v between pins * and %.
Appreciate any thoughts, Captain

[Mark Hammer]

It's going to vary by what it is you are matching them for.

One easy way around this is to buy matched transistor arrays on a DIP. Just keep in mind that sometimes not all the pins you need to be disconnected from each other are.

[The Captain]

Ah yes sorry- its for a Phase 90 (mxr phaser from tonepad.com). There are 4 needed to match, so i got a good few but alas Ive no idea how to test them.. I noticed a npn/ pnp thingy on my MM & also a hfe bit, but neither these nor the 200mV setting show me ~1.6$ mV probing the transistors every obvious way. Can anyone explain what needs doing? thanks

[Steve Conner]

The hfe measuring function on a multimeter works by sticking the transistor legs into the special socket provided on the front panel. You don't use the test leads. The readout shows the current gain of the transistor which is typically between 50 and 500. Choosing transistors that give a similar hfe reading would be as good a way of matching them as any.

I don't see what makes you expect 1.6mV anywhere. Probing a transistor with the diode function on a DMM usually gives around 0.7V.

BTW, what are pins * and % on a transistor?

[Mark Hammer]

The matching is for Vgs on the FETs. Go to www.geofex.com and you'll find several documents about how to match and select the FETs. The "ideal" ones can be gotten from small bear electronics at a reasonable price.

[The Captain]

Ah Im with you now.. thanks for that. Reading the build reports of said Phase 90 on tonepad is where I see ~1.6mV as being folks' readings; matching along lines of about 1.61 or 1.62 seems consitently noted- although Im using four 2N5457s for 2N5952s.

BTW.. * and % just my lazy Ids for a leg of transistor. ie x and y/ anyfink and anyfink.

Thanks Steve will try that now.

[The Captain]

Thanks but I cant understand a word of that its so complicated, or how to actually measure them bar this way: with the hfe thing on the DMM- some of mine read around 515 (515 'hfe'?) although the reading steadily creeps up from say 510 to 520 over 5 mins which is rather unhelpful. Whether these are a 'match' or hopelessly unmatched, Ive no idea though. I guess I'll just use anything at hand. Ordering 4 transitors from Small Bear with ~$12 postage is out of the Q.

Does anyone know how to measure transistors' V with a MM?

[Steve Conner]

Wait, are these FETs or regular transistors? The hfe function on a multimeter doesn't work on FETs. It's probably best to use one of R.G. Keen's JFET matching fixtures described on his site. If those readings you quoted really are millivolts, it doesn't sound like a very good matching procedure.

When it comes to matching FETs, every one is unique and temperature has quite an effect on them. So it's kind of like trying to match snowflakes

[R.G.]

You're battling several monsters.
First, the P90 uses JFETs, not bipolar transistors like NPN and PNP. They are quite different. JFETs don't have a current gain/hfe in the same sense as bipolar transistors. Bipolars are the kind that cheap multimeters can test; they can't test JFETs at all.

The JFET matcher at GEO is the simplest way to test and match JFETs for a phaser.

Far better than doing nothing, construct this:
9V battery + connects to drain of device under test.
Source of device under test connects to ground through a 2.2K resistor.
Gate of device under test connects to ground.
Ground connects to - terminal of the 9V.

Measure the voltage across the 2.2K resistor with your meter and select the closest possible devices.

This does not measure what you really need to measure, but it is far better than just plugging in random, unmatched devices.

Transistors have no voltage, only what appears when you put them into a circuit.

Thanks but I cant understand a word of that its so complicated, or how to actually measure them bar this way: with the hfe thing on the DMM- some of mine read around 515 (515 'hfe'?) although the reading steadily creeps up from say 510 to 520 over 5 mins which is rather unhelpful. Whether these are a 'match' or hopelessly unmatched, Ive no idea though. I guess I'll just use anything at hand. Ordering 4 transitors from Small Bear with ~$12 postage is out of the Q.

Does anyone know how to measure transistors' V with a MM?