Curse you semiconductor manufactureres!

[hex65000]

Okay, so I'm setting up a little experiment at home and I got some 2n2222 transistors to essentially act as solid state switches. Not a problem right?

 

Well... Here's just a little something goofy for you to chew on... I'm using a really common part to do this experiment and behold two different data sheets (in terms of pinout) based on the manufacturer.

 

Mind you, the Fairchild datasheet says PN2222 v/s the 2N2222 from ON semiconductor. There is a cross reference table that points at the PN2222 part. Still, a classic WTF moment. The good news is that I can experimentally sort this out, but this is just silly and an important caveat when dealing with "mature generic parts". Which is what I thought I was dealing with and thought one MFGR's datasheet will be as good as any other MFGR.

 

Nope.

 

Lesson learned folks.

Fairchild Semiconductor link

ON Semiconductor link

 

Hex.

[ Just an interesting discovery, and I thought I'd share... ]

[+Stephen Moss]

That is annoying, it is rare but I have a dim recollection of it happening to me once, although it should not as manufactures should follow guidlines that prevent this.

It may be that this happens on really old parts that were manufactured before the guidelines were produce although it is not beyond the with on man to make them match so the manufactures really have no excuse for it. I tried looking it up in Towers but the latest copy I have is 14 year old and shows the 2N2222 with different package type.

 

I depends on your application but generally if you want solid state swithces FETs work better than bipolar because they are transimpedance and to my knowledge all analogue switch ICs use FETs.

[hex65000]

You know, that was a great suggestion! I did a little poking and found that the 74hct4066/cd4066 will do exactly what I want. Not sure if that's what you were driving at, but thanks for the tip.

 

Hex.

[ gets to stack computers today... ]