2SC1583F (Paired transistors) relpacement?

[ballooneater]

Hello all,

As my first post on this forum, I'd like to ask: what kind of replacement I should use for this transistor pair?

Would any 2 "paired" NPN transistors work?

Thanks for your time!

The bottom view of the transistor in question:


How it looks in the schematic:


And a link to the schematic I'm working off of (Ibanez FL-301 Flanger)
http://matthew.kloiber.googlepages.com/schematic.GIF

[R.G.]

They're using the exponential relationship of IC to Vbe in a bipolar to give an exponential characteristic to the modulation. The ear likes that in many cases.

If you have no dual transistors (and they're just about extinct) then you can match a pair of NPNs. Get same type number and same lot number to start with. That gives you a chance that you will come up with transistors that were actually on the same wafer. Generally it's better if you buy a number of them. If you look, you can find NPNs for maybe $6 - $10 per hundred, which is not going to break the bank.

First number a bunch of them with little dot labels, then test them for Hfe in any kind of hfe meter, taking care to wear cloth gloves or better using pliers or tweezers to not let your hands heat them up. Just churn through a batch of them, writing down hfe as you go. Then sort the list by HFE in word or excel. Pick a group with almost identical hfes, then set up a test rig with a 9V source (regulated supply if you have it, fresh battery if not) and measure Vbe with the base fed through a resistor of perhaps 100k. If you can get Vbe within 15-20mV in a close HFE, they ought to work well enough for many synth circuits.

Or you could just stick in random devices and hope. 

[ballooneater]

Thanks for the tips! 

Any specific NPNs I should lean towards or will the standard 2N5088/9s do the trick?

[notchboy]

AMS still has 2SC1583 for around $4 each:

http://www.americanmicrosemi.com

[ballooneater]

Thanks for the info, but $4.00 per transistor pair is a bit more than I'm willing to spend.  I'll try R.G.'s method and if that doesn't work, I'll order from AMS. 

[Paul Perry (Frostwave)]

I doubt if it matters much in that particular application anyway.
And i do not know, but I suspect that american micro semi has a hefty minimum order. Anyone know?

[JDoyle]

And i do not know, but I suspect that american micro semi has a hefty minimum order. Anyone know?

$50

[Paul Perry (Frostwave)]

And i do not know, but I suspect that american micro semi has a hefty minimum order. Anyone know?

$50

Thanks! that's not too bad - I think ballooneater is out, though..

[Mark Hammer]

Is the thermal matching that comes along with being in the same package (i.e., 2SC1583 vs a pair of matched NPNs) critical in our sorts of applications, or is that really only a concern for the DIY synth folks when making oscillators?

[ballooneater]

And i do not know, but I suspect that american micro semi has a hefty minimum order. Anyone know?

$50

Thanks! that's not too bad - I think ballooneater is out, though..

Not out yet!  If thermal matching was important, would it help to put the two transistors together with some thermal adhesive between them?

[R.G.]

Is the thermal matching that comes along with being in the same package (i.e., 2SC1583 vs a pair of matched NPNs) critical in our sorts of applications, or is that really only a concern for the DIY synth folks when making oscillators?

I would guess that thermal tracking is only critical when you want a musically-stable volt-per-octave system. For LFOs (which is what I was thinking from the circuit) then the bulk exponential is more important than tracking.