[sdiy] newbie alert - transistor question
Roy J. Tellason
rtellason at blazenet.net
Wed Jan 5 08:20:47 CET 2005
On Tuesday 04 January 2005 09:02 pm, groovyshaman wrote:
> Hopefully this isn't too brutal of a common knowledge question!
>
> Ok, I'm in the process of ordering some 2N2222A transistors for a few
> projects I working on (one of them is a Blacet dark star - finally). And
> so I have come upon a strange question: is there any advantage to buying
> them in TO-18 cans over TO-92 plastic?
Yes. It depends on how close to the device's limits you're going to push it.
The cans will dissipate heat better than a plastic package will. Look at
data sheets for thermal resistance numbers...
> Why do they still make cans?
So we have a choice?
> Should I even be concerned with this?
Depends on the design of the circuit involved.
> I've been reading lies.. err.. datasheets, googling, etc. but other than the
> cans handling slightly higher junction temps, I just don't see the benefit
> in spending 30 bucks more for 100 pieces O' TO-18. (Besides, if they're
> anywhere near 150C I've likely got a much bigger problem!) I appreciate any
> thoughts, strange or otherwise! :)
You may be surprised at how high the junction temperature can get without the
actual case warming up all that much. Aside from that, it doesn't have to
be that high to cause problems. The lifespan of semiconductor devices is
adversely affected by thermal cycling, among other things.
The way I design stuff, I wouldn't even be using a 2222 for anything over a
few hundred mA, I'd go to something in a medium-power package like TO220 or
something, which will deal with half an amp to an amp quite nicely without a
heatsink in some instances. I wouldn't go out of my way to get 'em in a can,
but if I had them I'd use them (and I do have some, salvaged).
BTW, the part number I have on the TO92 equivalent is PN2222A, _not_
2N2222A, which I guess _is_ the part number for something in a metal can.
Biggest mistake I ever made was buying one of those bags of transistors that
radio shack sells, which purported to be "equivalent" to a 2N2222 -- they
weren't. Not only were they in TO92 packages, but they didn't come near to
meeting the spec, and often had something as absurd as a 6-8V collector
breakdown voltage.
<...>
> and a 76477 noisemaker I built 20 years ago.
I bought one back around then, and *still* haven't gotten around to doing
anything with it yet!
Got any good schematics? :-)
More information about the Synth-diy
mailing list