Shop OBEX P1 Docs P2 Docs Learn Events
2N2222 as phototransistor - not? — Parallax Forums

2N2222 as phototransistor - not?

john_sjohn_s Posts: 369
edited 2015-04-02 11:47 in General Discussion
Many years ago I was desperate to try a "novel" opto switch circuit.
But it was late Saturday and I had no photo-transistor in my box.
So I've decided to make it out of 2N2222 by opening its metal T0-18 case.
The photoefect was week but enough to proof a concept, and It worked as expected... nearly.

However, it did not last for more then few trials after which my home made "opto's" kept dying,
as I kept cutting the tops of few more transistors to replace the dead ones :)

So my question was - and still is - why the "optotransistors" made out of exposed to light 2N2222 did not last?

Comments

  • CuriousOneCuriousOne Posts: 931
    edited 2015-03-11 13:56
    oxygen in air kills crystal. you needed to seal crystal with laquer.
  • Mike GreenMike Green Posts: 23,101
    edited 2015-03-11 14:05
    Oxygen in air, moisture in air, pollutants in air ... all rapidly shorten the life of exposed devices. That's why phototransistors, LEDs, erasable EPROMs all have had glass windowed or clear plastic packages. The EPROMs had quartz windows which transmits UV light which is more energetic than visible light and can more easily and quickly erase the cell contents.
  • ercoerco Posts: 20,256
    edited 2015-03-11 15:45
    I had some LEDs in a breadboard circuit this week, hooked up to a microprocessor, checking out some new features on a new editor. I found that some displayed variables were changing as the LEDs were exposed to light, basically my arm casting a shadow on the LEDs. An accidental experience with the "Mims Effect", whereby LEDs act as light sensors. But of course I am repeating the findings of others: http://forums.parallax.com/showthread.php/153403-Using-an-LED-as-a-button
  • $WMc%$WMc% Posts: 1,884
    edited 2015-03-11 16:47
    Mr. Mims
    '
    Was this one busy dude or what!!!
    '
    Go Forrest, Go.
  • john_sjohn_s Posts: 369
    edited 2015-03-11 20:19
    erco wrote: »
    I had some LEDs ...[/url]
    That sensor thingy was in need of photo-transistor i.e. E-B-C contraption...
  • john_sjohn_s Posts: 369
    edited 2015-03-11 20:21
    Mike Green wrote: »
    Oxygen in air, moisture in air, pollutants in air ... all rapidly shorten the life of exposed devices. ...

    I was really surprised that an open transistor junction lasted no longer than few minutes.
    Google for pictures of the very first experimental transistor with clearly visible and exposed plates/wires touching germanium.

    https://www.google.ca/search?q=first+transistor&biw=1280&bih=871&tbm=isch&tbo=u&source=univ&sa=X&ei=mAgBVZiyNYH8oQSUmoHoAQ&sqi=2&ved=0CBwQsAQ#imgdii=_

    No glass or else around it. Also, the very first crystal radios used to have a diode made out of pointy metal spike simply touching a crystal blob...

    point transistor.png
    firsttransistor.gif
    point transistor.png
    400 x 273 - 31K
    firsttransistor.gif
    359 x 359 - 42K
  • LeonLeon Posts: 7,620
    edited 2015-03-12 03:36
    When I was at school, I and my older brother used a metal-cased transistor as a photo-transistor, by cutting off part of the case. We didn't have any problems with it failing. It was filled with something like grease, which would have protected it.

    A couple of years before that Mullard made an early Ge transistor, the OC71, that was encased in glass with black paint on the outside. They also sold a much more expensive photo-transistor, the OCP71, which was identical, but without the black paint. Lots of people simply converted the cheaper device by scraping off the paint. Mullard got wise to this, and put the paint on the inside of the glass.
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2015-03-12 06:20
    I call the conclusion you come to concerning rapid failure of the "phototransistor" a premature assumption or just jumping to conclusions. I have no idea what this circuit is that you have, perhaps it is responsible for frying the transistor itself and in fact there may be a fault in it. Have you tried a real phototransistor in the circuit since? Does it continue to work? Anyway, when I need a phototransistor and don't have one handy I just use an LED.
  • Heater.Heater. Posts: 21,230
    edited 2015-03-12 12:58
    Just a few weeks ago Dave Jones on his EEVBlog did exactly this. He cut the top off an old metal can transistor and demonstrated it being light sensitive.

    Now perhaps he did not continue his experiment long enough but he is the kind of guy who might notice if it fails after a while.

    I find it hard to believe such an old style transistor would succumb to exposure to the atmosphere so quickly.

    Back in uni one of our experiments was to make a silicon diode. A plate of n-type silicon with a blob of p-type fused on top. Or was it vice versa? Anyway, point is it lasted long enough to perform measurements on and was still working as a diode many months later.

    I think I'm going to have to find an old 2N2222 and try this experiment.
  • kwinnkwinn Posts: 8,697
    edited 2015-03-12 19:14
    Heater. wrote: »
    Just a few weeks ago Dave Jones on his EEVBlog did exactly this. He cut the top off an old metal can transistor and demonstrated it being light sensitive.

    Now perhaps he did not continue his experiment long enough but he is the kind of guy who might notice if it fails after a while.

    I find it hard to believe such an old style transistor would succumb to exposure to the atmosphere so quickly.

    Back in uni one of our experiments was to make a silicon diode. A plate of n-type silicon with a blob of p-type fused on top. Or was it vice versa? Anyway, point is it lasted long enough to perform measurements on and was still working as a diode many months later.

    I think I'm going to have to find an old 2N2222 and try this experiment.

    The fact that it has an old part number does not necessarily mean it uses the old style manufacturing process. It could have been changed to fit a more modern process.production line.
  • Heater.Heater. Posts: 21,230
    edited 2015-03-12 22:58
    kwinn,

    Quite possibly true. That's why I say it's an experiment I'm going to have to have to try with an old transistor.
  • john_sjohn_s Posts: 369
    edited 2015-03-16 20:30
    @Heater - I did this experiment approx.20 years ago and it was pretty much the same as in Dave Jones experiment mentioned elsewhere ...
    The 2N2222 setup was very simple - E to GND, and C via R=1k to +5V. The output (taken from C) was then directly DC-coupled to the base of another transistor (p-n-p this time).

    I forced the photo-effect by shining a 60W 110VAC lamp directly into the TO-18 opening from a very close distance (perhaps not more than 10inches) .

    Every time the transistor lasted no longer than few trials.
  • john_sjohn_s Posts: 369
    edited 2015-04-02 11:47
    Leon wrote: »
    When I was at school, I and my older brother used a metal-cased transistor as a photo-transistor, by cutting off part of the case. ...

    ... 2N3055 as photovoltaic cells :)

    https://www.youtube.com/watch?v=gRY7Re-Q0WI

    https://www.youtube.com/watch?v=bGO5KIikKWA
Sign In or Register to comment.