Importance of matching components?

[Vince_b]

I have a few questions about the importance of matching components.

1. Is it always better to match diodes? If not, how can I know when I should match them and when I shouldn't?
2. For matching transistors with the circuit by R. G. Keen on GEOFEX, do I need a DMM with a mode for transistor or any "standard" DMM will works? And is this circuit only for testing germanium transistor or is it good for silicon ones as well?

[darron]

You don't always need to match diodes. They are usually pretty well within tolerance and if they are a bit difference then you get an asymmetric clipping, which some people like. I find this becomes even less important if you are going to be driving another stage later too, for example it's going into another distortion or you are driving your tube amp, or your amp is already driven after the pedal.

As for R.G.'s transistor tester, I assume you mean the Technology of the Fuzz Face article? [url]http://www.geofex.com/Article_Folders/fuzzface/fffram.htm[/url

Which I believe only needs a voltmeter, something every multimeter will do. I think about ever $5 DMM has a transistor tester anyway, so you might have difficulty buying one without one. The transistor testers on modern DMMs however assume your transistor leakage is zero, or very close to. Good for silicon transistors and less useful for Germaniums.

[R.G.]

There are a very few places where matched components are needed. The need for matching is the exception rather than the rule.
Matched diodes are needed in some diode ring modulators. That's about it. With regard to effects, matched JFETs are needed for phasers or JFET differential amplifiers. Matched bipolars are needed for differential amplifiers and a few esoteric circuits.

Germanium transistors do not need to be matched so much as selected. The germanium matcher at my site is there to weed out transistors with too much leakage and too low a gain. Fuzz Face transistors do not need to be matched, just selected for gain greater than X and leakage less than Y. Some afficionados of the Fuzz Face will obtain a sound they like by putting specific gains in either the first or second transistor position, but this is very much selection, not matching.

[Vince_b]

Thank you for your replies, it's very helpful.

[mr.adambeck]

I'm building a passive Ring Mod (well, I plan on boosting the inputs and adding an optional oscillator), and from some things I've found online it's said having matched Ge diodes is key.  How do I go about matching them?  Also, how many do you think I will need to order to find 4 matched ones?  In other words, are they all vastly different, or usually pretty close?

[petemoore]

I'm building a passive Ring Mod (well, I plan on boosting the inputs and adding an optional oscillator), and from some things I've found online it's said having matched Ge diodes is key.   Well, yup.
  How do I go about matching them?
  I got readings of .6 and .7 from my cheap DMM, I'm not sure if that accuracy is sufficient for your application.
  Also, how many do you think I will need to order to find 4 matched ones?
  Are you feeling like it'll be a lucky draw ? Besides, the shipping'll probably kill 'ya, save on 10 or 20 in some places.
In other words, are they all vastly different, or usually pretty close?
  Hard to say what's close enough for that, or how closely clustered a hand picked 4 of 10 or 20 will be when you draw them.
  I can't even tell you if this DMM is fine-tuning enough for Hfe picking, it's only reading one or the other number when measuring Ge diodes [mostly .6's]..there are probably simpler ways than what I'm thinking to measure the diode foreward threshold.
  ..involves a small voltage adjust-ulator, DMM set to <2volts, and something that indicates when/at what voltage the diode switched.