NPN brain teaser puzzle

[wwfieee]

yes, i've seen some of bob pease's show in the national website. strange yet funny guy. i've also read his book. very informative. I haven't had the chance to read his side column in the journal though.

does it work with pnps too? I've a couple lying around that hopefully i can tear them apart to see this. We need to be in dark room for it?

 

[nand_gates]

Gr8, Please post more such exciting things about transistors!

 

[echo47]

Yes I see similar behavior with PNP. Today I tested a 2N3906 and 2N2907A.

I cut open a Motorola 2N2907A. With 10mA of input current, I see a faint speck of greenish yellow light with my naked eye in a dark room. Through a 100x microscope I see eight rows of tiny lights. As I increase the current, more lights appear along those rows. Most curious!

The 2N3055 was the most impressive. I recall its light was yellow or amber.


I just now found a somewhat relevant paper, "Degradation of a Light Emitting Silicon Junction of a Bipolar Transistor". Does anyone have references [6] or [7]?
**broken link removed**

I found another paper that begins with this sentence, "It is well known that when a p-n-junction is biased into avalanche, light is emitted. This light is absorbed in the surrounding substrate and generates electron-hole pairs."

 

[Karthikeya]

hi echo47
it is really an amazing puzzle . here is the .pdf files u asked for.....

thank u

 

[wwfieee]

i cut off the top of a 2n2222 by ST and biased it as shown but without a resistor. it begins to glow at around 9.5Vs. Truly remarkable. Yellow and its real nice. I showed another guy and had him guess what the component is.

He never guessed it. Haha.

 

[flatulent]

I am not convinced that this is caused by the photo-electric effect. The with of the base would prevent the photons from getting to the base-collector junction. Can someone put their meter in the current mode and measure the current with different values of resistance in series with the meter. Then calculate the voltage across the BC junction and plot a voltage-current curve. If it is the photoelectric effect you should get a horizontal line of constant voltage for currents increasing up to a point and then the voltage should rapidly decline.

An alternative theory is that the current flowing around the input loop injects many electrons into the base. These have kinetic energy. Most of them are attracted by the electric field across the B-E junction and flow in that direction. A small portion use their kinetic energy to go to the B-C junction.

 

[Mazi3]

I had to try it myself. One old 2N3055 from ST was sacrified to see it current was 160mA and picture was taken at night, long exposure time.
The transistor is still good, but I didn't make any measure of his parameters, only checking PN junction. Breakdown voltage is 11.5V (thats more than 7V as stated in datasheet-but this is only maximum voltage, which can EB junction withstand). Current was tried up to 0.6A-be aware, there is a lot of heat generated..

here is picture of transistor glowing one for background...

 

[movingbait]

Forgive the question, but how do you remove the top from a TO3 package and/or a TO92 metal can.

movingbait Ü

 

[wwfieee]

you just use a clipper and clip the top end off. becareful not to clip too close to the base. 2 tiny wires are connected from the chip to the legs.

i'll measure the currents on the 2n2222 next week.

 

[nand_gates]

I also agree with you flatulent. This is exactly what I thought first.

 

[Mazi3]

use small saw and do it gentle

 

[echo47]

use small saw and do it gentle

I'm gentle! I used vise grips and a big old hacksaw. Worked great (on the third one).


Hey Mazi3, that's a nice photo.


Flatulent, I cooked my 2N3904 NPN, so here are 2N2907A PNP measurements, plotted with MATLAB.

% 2N2907A with 10mA applied to base-emitter (reverse breakdown).
% [ohms_load  volts_collector_to_base  microamps_collector_output]
results = [[9999e9 0.36 0.00]
           [2700e3 0.36 0.13]
           [1000e3 0.36 0.36]
           [ 430e3 0.35 0.80]
           [ 360e3 0.35 0.96]
           [ 220e3 0.34 1.53]
           [ 180e3 0.33 1.74]
           [ 120e3 0.31 2.47]
           [ 100e3 0.28 2.77]
           [  82e3 0.25 3.01]
           [  47e3 0.16 3.14]
           [  36e3 0.12 3.14]
           [  13e3 0.04 3.13]
           [   0e3 0.00 3.20]];
plot(results(:,3), results(:,2));
xlabel('I out, [microamps]')
ylabel('V out [volts]')