transistor equivalents...

[Branimir]

greets!

i posted this kinda post on an another thread, but i felt like posting a thread of its own, since lots of people won't read the other one (this Q came as an offtopic)....

is there an online transistor database, so i can find transistor equivalents for those that i cannot buy around here..

for instance BS250 used for the obisidian...

thanks!!!

[petemoore]

I used 2n7000 for Obsidian, I can't remember how many of them, I used other type...hybrid.

[Aharon]

You can use the NTE cross-reference search,of course it references to NTE product but you get data sheets and you can always compare to a known product.You get the idea.
Aharon

http://nte01.nteinc.com/nte/NTExRefSemiProd.nsf/$$Search?OpenForm

[R.G.]

I posted this on a discussion about transistors equivalents, but it seems appropriate here.

===========start of duplicate post============

Let me cool off the 2N5088 discussion some more.

I write this because none of the discussions I'm seeing has much basis in the things I think matters in picking transistors for the highly limited class of circuits most effects employ. Since opinions are like noses, I'm going to stick mine into the argument.

Keen's Second Law says "When in doubt, use a 2N5088."

There are actually some reasons behind that law. First and foremost, for 99% of all small signal bipolar circuits, it doesn't matter what transistor you use, as long as you have "enough".

"Enough" comes in several flavors. Enough BVceo keeps it from breaking down under power supply voltage. Enough current capability keeps the bonding wires from fusing. Enough power dissipation keeps the device from frying under the heat. Those are kind of trivial enoughs for 9V effects circuits.

The "enoughs" that tend to matter in effects are (a) enough gain (b) low enough noise. High gain is not particularly a virtue once you get enough. Diminishing returns set in quickly, as anything over a gain of 100-200 satisfies most circuits. Still, there are circumstances which beg for a high gain, particularly because in one of her magnanimous moments, Mother Nature seems to have decreed that high gain is correlated to some degree with lower noise in bipolar transistors.

As I keep saying, classical good bipolar transistor design is largely the science of designing the circuit to be tolerant of any amplifying hunk of silicon you stick in there. This practice was developed because early devices had huge parameter spreads, and rather than instantly improving transistors - which Mother Nature didn't let us do, it took decades - we instantly improved the design techniques.

Over a few decades, I came to prefer certain transistors for circuits. These transistors had the advantages of highish gain, low noise, good linearity, reliability, and ... (drum roll, please)... availability. The devices that hit those optima for me were made by Motorola (now passed on to Onsemi and Freescale) and showed up typified by the 2N5088. The 5088's were good enough for most things, and great for small signal amplification. They're marginal for relay drivers, where I use 2N3904s, but they do work there as well. They're better than the other stock industry answers for general purpose small signal amplifiers, the 2N4125 and the 2N4401. They suffer from not having a true complementary device like the 2N4401 does (2N4402) or the 2N3904 (2N3906). But complementarity is not a big deal in effects.

The 2N5089 started as a high gain selection of the 2N5089; some wafers came out with higher gain, and these were sold with a different type number, at a premium price, before modern silicon processing made literally any nano-acre of silicon the same price.

The MPS-A18 is different. It is Moto's attempt to design the ultimate high gain audio input transistor. As such, it has even higher gain, and is specified for super low noise for the range of signal impedances that you'd get in, say, a cassette deck head or a moving coil phono cartridge. The funny thing is, as Moto released this, National released the LM381 low noise preamp IC, and shortly afterwards, cassette and phono died. So it's kind of an orphan device, but it's still around, and because all silicon costs the same, it's cheap.

The Moto devices and generic devices are not the only ones to meet the criteria. There are absolutely Euro (BC series) and Japanese (2S...) devices that are good. Here's what to look for:

Bvceo = more than 9V, preferably 30V
Icmax = 100ma
Pd = 100mW or more
Hfe = more than 200; I like 350-500. They get fragile over that; see "When good opamps go bad" at GEO
Noise figure = 4db or less in the audio band.

Anything that fits inside those hoops will do 99% of all effects jobs. The real message? Go find out what you can get, then look up and read the datasheets to know what you have to work with.

It strikes me that it may be better to pick a transistor that you can remember the pinout for than to have the technically best one for the circuit. You'll have a better success ratio.

People being people (i.e. inherently pattern-matching devices) they will extract a pattern, even if there is none. People see patterns in the show on vacant TV channels. People pick and stay with transistors with which they have had good luck (in the true sense of that word) and avoid the ones with which they have had bad luck. The transistors don't care. They are what they are.

So go look up the data sheets - especially the pinouts - and pick any good-enough device that you can get, and make your own second law.

I like mine, and it works for me.  

=========end of duplicate post===========

[LinuxMan]

Hehe,

RG busy-as-a-bee today.

Told you you'dd have to put it in a FAQ.

Cheers
LM