introduction dates of common transistors and diodes?

Hello Eric,

You'd have to find old data sheets. The ones I have may not be the oldest but in case it helps here are some data sheet print dates:

March 1973

Dec 1973

Dec 1973

Again, there are probably older ones. Maybe one of the data sheet archive sites could be of help.

Come to think of it, I am still using the 2222 in new designs. Same with CD4000 logic, more than 30 years in production and still kicking. And cheap.

Regards, Joerg

All three are listed in my 1970 data book, but not in my 1964 ARRL handbook.

--
Cheers,
Stan Barr     stanb .at. dial .dot. pipex .dot. com
(Remove any digits from the addresses when mailing me.)

The future was never like this!

That's a good question. In principle all the 1N/2N part numbers were registered with JEDEC and they ought to have a way of getting to registration dates. As a practical matter my attempts to navigate the JEDEC website to find this information in general have failed. They do say they will sell you the original registration datasheets at 25cents a page, $10 minimum.

Registration date is not necessarily the same as production date but shouldn't be too far off.

Joerg's reply seems (to me!) nonsensical because I know all your named non-plastic parts were well into mass production in the 60's. The date you read off the top of a data sheet is not necessarily the original production date because it is entirely possible the data sheet was re-issued.

My guess as to 2N chronologies, based entirely on some 60's-era TI "bulletin" dates and done under the assumption that they are at least somewhat chronological:

2N117 is 1958 2N33x are arealy 1959 2N11xx are early 1962 2N22xx are late 1962 2N30xx are mid 1963 2N32xx are early 1965 2N44xx are mid 1966

Many of those numbers were originally issued by Motorola or Fairchild but my 60's era Motorola books aren't nearly as complete. I would guess the Moto sheets have slightly earlier dates. I don't think I've ever seen much in the way of 60's-era Fairchild databooks.

By the early 70's TI and some others have most of their new parts outside the JEDEC standard 2N series. At that point plastic packaging for consumer and non-milspec-non-aerospace stuff really takes off and the manufacturers start doing their own numbering (often riffing on the original 2N numbers but not always!)

Just to confuse things even more, the JEDEC web page says they didn't exist before 1960, so obviously the 2N11x numbers predate JEDEC, but probably have something to do with EIA. I don't know how RETMA figures into it.

And in the case of the 1N series I think any assumptions about sequential assignment of numbers are probably pure BS past the early

60's. Unfortunately my earliest TI diode datasheets mentioning your part numbers have dates of early 70's on them and those dates cannot be right for the 1N914.

I did find a 1N4148 TI bulletin dated October 1966, and that might be too late by only a couple of years :-).

The JEDEC website says the 1N3091 was registered in May 1960 and the

1N3595 was 11/5/1962. At that point I'm already very critical of the thought of sequential assignment.

Tim.

What about things sucvh as the BC107/108/109 which seemed to be ubiquitous in magazine designs in the 80's?

--
Bruce |  41.1670S | \  spoken |          -+-
Hoult | 174.8263E | /\ here.  | ----------O----------

Hello Stan,

That's what I thought. Even if the data sheets are original they often are revised and most of the time they don't publish a revision history with them (they should).

1964 were the Ge days. Luckily I kept lots of diodes and transistors from back then. A couple of weeks ago that saved our bacon when I could not buy a new 1.35V mercury battery for our older Minolta. They are now banned, for good reason. The old OA91 Ge diode was just the ticket to modify it for a 1.55V coin cell and drop those 200mV nicely.

Regards, Joerg

Hello Bruce,

Those are European numbers, mostly German. But it may be even harder to find original data sheet dates for those since Siemens (now Infineon) and others have that dreaded tendency to drop data sheets soon after something went obsolete. Then you can only use data sheet web archives and you never know whether the date on there is truly the birthday.

Regards, Joerg

Hello Tim,

Well, I did mention that the dates probably aren't the earliest. Just wanted to start it. Short of paying JEDEC the bucks the only way to get a clue here is to "peel the onion". IOW people look and post their data sheet dates, then if someone else finds an older one he or she posts that.

I bet that someone like Winfield Hill would know. He must have picked these transistors off the conveyor when they were still wet, to try them out. I just wasn't old enough, plus I certainly didn't have the connections back then.

Regards, Joerg

My Philips "Miniwatt" [the (then) tube and semiconductor manufacturing branch of Philips in Australia] shortform catalogue published in the

3rd month of 1964 lists both the BC107 and 109 but not the 108 so it is probable that these types were introduced in 1963 or thereabouts.

I've recently done exactly the same thing to keep my Sekonic spot meter running! (Altough I've had digital cameras for over 10 years, I still have a well-used collection of real film kit - used for black&white these days...)

--
Cheers,
Stan Barr     stanb .at. dial .dot. pipex .dot. com
(Remove any digits from the addresses when mailing me.)

The future was never like this!

Why not use the saturated collector drop of a silicon transistor?

• | | | o-------| | | | .-----. | | load| .-. '-----' | | | | | | '-' | | |/ ------| |>

| -

--
"If you want to post a followup via groups.google.com, don't use
 the broken "Reply" link at the bottom of the article.  Click on 
 "show options" at the top of the article, then click on the 
 "Reply" at the bottom of the article headers." - Keith Thompson
More details at: 
Also see

Because typically you only have access to one wire, not both wires of the load. Also, a diode is only one part :-)

Regards, Joerg

Also I happened to have a number of miniature, grain-of-rice size, germanium diodes that fitted easily in the limited space available.

--
Cheers,
Stan Barr     stanb .at. dial .dot. pipex .dot. com
(Remove any digits from the addresses when mailing me.)

The future was never like this!

Hello Stan,

Nice! My smallest type on hand is the OA91, still about the length of two grains of rise. But in the older Minolta SRT series there was a surprising amount of space available.

There really is not much else to adapt a 1.55V cell. Transistor saturation voltages are too low at the 100uA range of CdS sensing circuits. Unless you keep base current low but then it becomes too fickle for my taste. Schottkys are sometimes used but I didn't like them this far down the curve, too much variation in my case.

Regards, Joerg