[sdiy] Analog MAT12 matched NPN pairs

Matthias Herrmann matthias.herrmann at fonik.de
Sun Jun 1 13:54:56 CEST 2014

> -----Ursprüngliche Nachricht-----
> Von: synth-diy-bounces at dropmix.xs4all.nl [mailto:synth-diy-
> bounces at dropmix.xs4all.nl] Im Auftrag von synth
> Gesendet: Sonntag, 1. Juni 2014 03:57
> An: john slee
> Cc: Synth DIY
> Betreff: Re: [sdiy] Analog MAT12 matched NPN pairs
> I don't think so.  Tapped 3904s are very closely matched and if that
> is not enough for you, you can use Ian's matching method.  You can
> google "ian fritz transistor matching" for plenty of info.

not necessarily. i rencently matched quite a few 2n3906. all from the same
batch, and while most pairs had less than 2mV Vbe difference (the value moog
once specified), one transistor had been way off (7mV Vbe for the pair). i don't
have exact numberes, but as a rough estimate i would say 1 out of 30? anyways, i
you pick randomly you might get the worst pair from your batch, no matter how
large your batch is. why risc this while matching could be done easily?

BTW i did use ian fritz' method, which uses very few parts. one could easily
build it on breadboard or a small piece of proto:

just my 2 pence.


> On May 31, 2014, at 8:48 PM, john slee wrote:
> > I think my MFOS modular PCBs arrived at work last week. I was
> > originally planning to match 2n3904s and cuddle them up in goo
> > and heatshrink, but these caught my eye as I was browsing Mouser
> > for VCO parts...
> >
> > http://www.mouser.com/ds/2/609/MAT12-246314.pdf
> >
> > Comments? They're quite expensive ($30+ each even in /10 qty)
> > but OTOH I don't need that many, and they're in nice shiny metal
> > cans, they're available, and they're already matched. Worth it?
> >
> > John

More information about the Synth-diy mailing list