Hello,

I am trying to teach myself some beginner elecronics by going through the book "Electronic Experiments for the Evil Genius". Everything was going great till I came to the transistor part.

I think my 3904 transistor might not be working as it should... Therefore I just did a quick test on my breadboard:

I connected 'collector' to a 9V battery and 'emitter' to a LED via a 470 ohm resistor and then to ground. Is it normal that it conducts and makes the LED shine, without applying any current to the base?

I don't think so, because it is an NPN and the book says they behave more or less like a normally open pushbutton.

The weird thing is that I tested it with three different 3904 transistors and they all have the same 'problem'. I tested the transistors for shorts between the legs but there didn't seem to be any.

Any help will be greatly appreciated!
Thanks in advance.

 

I strongly suspect you have misidentified the leads. The base will show a junction to both the collector and the emitter, if you have a meter with a PN junction check. 2N3904's got made by enough outfits that the pinout can be ambiguous without experimentation. Here is a link into the Ebook on testing transistors - http://www.allaboutcircuits.com/vol_3/chpt_4/3.html.

Your results would seem to indicate you're passing current through the LED and the base-emitter junction.

It is better to apply the battery to the resistor, followed by the LED, and then the transistor's collector. The emitter will go to ground. That makes control of the transistor easier to predict.

 

Here is the datasheet of the transistor. It will help you identify the correct pins.

http://www.fairchildsemi.com/ds/2N/2N3904.pdf

 

Although you are not applying current to the base the transistor doesn't automatically show a high impedance from the collector. Put a resistor from the base to ground. A 100k should do the trick. So, when you do not apply a signal to the base, it will be pulled down to ground (hence the name "pull down resistor").

 

Although you are not applying current to the base the transistor doesn't automatically show a high impedance from the collector. Put a resistor from the base to ground. A 100k should do the trick. So, when you do not apply a signal to the base, it will be pulled down to ground (hence the name "pull down resistor").

What you have mentioned is good practice in designing transistor circuit.

But it is definitely not why his LED is shinning.

Only a tiny tiny leakage current flow will result in this case if the base of the transistor is not connected to ground. The magnitude of this current is in no way able to shine a LED.

You can do a simple experiment to prove or disprove it.

 

Thank you very much for the replies!

It was indeed a misidentified pinout problem. It's weird because in all my electronics books, the authors discribe the pins (E-B-C) with the flat side towards you. Thanks to the E-books on the forum I did some tests and found out that the pinout of my 2N3904 transistor is (E-C-B). Maybe because I live in Europe? The writing on my transistor says 2N3904 C K029. Anyway, I am extremely happy that this problem is now solved, because I used to try some B.E.A.M. circuits from the Junkbots book by Dave Hrynkyw and they didn't work, and now I know why ! (I just copied the circuits, assuming that my pinout was the same as theirs, the book didn't mention anything about different pinouts...)
While I was doing the test with the PN junction test on my multimeter, I also found out that one of my 2N3904 transistors only had one conductivity reading instead of two, and indeed, when I tested this one in my circuit it didn't work, so I guess that one is broken. Nonetheless, I read on the internet a lot of times that you shouldn't blame your components too often, and that transistors rarely get damaged... Did I mistreat mine in a way that I don't know yet?
I am a bit paranoiac about bad transistors now...

 

As I said, 2N3904's got made by many outfits, and pinouts are always worth verifying. It is a lot more standard now, but it is sometimes hard to know when your parts were made. Always check with a meter.

I wouldn't worry about mistreating transistors too much. You get dud parts now and again - the manufacturers only check for statistical failure rates. Just operate them within specs.

Back when epoxy TO-92's were new and very different (the standard can was the now-defunct TO-5), we wanted to see how good one was. We wedged it into the loop of a Weller soldering gun and held heat on it at the full rate for one minute. The transistor worked just fine after that.

 

Thanks for the explanation! Best wishes for 2009 btw.

 

Without base current the transistor will do absolutely nothing. It is an open circuit but Ico will flow in the nanoamps. that is not enough to litght a LED. LOOKING AT THE FLAT SIDE ebc