Q: Identifying Transistor Terminals

[Longhair]

I have some different types of transistors (BJT, PUT) and when I was trying to follow along with a data sheet to identify the terminal layout, it was backwards because it was not the brand specific transistor for the data sheet I was using. So to make life simple for myself, I want to put together a little tester that is as simple as an LED tester - if inserted correctly, the LED lights up.

Using experiment 10 out of the Make: Electronics book as the starting point, the LED should only be lit when the button switch is pressed if the 2N2222 is inserted the right way. I took the transmitter and put it in backwards expecting nothing to happen when I pressed the button switch only to see it lit up again.

I don't have the schematic so I will have to describe it the best I can.

9v Positive - 180 Ohm Resistor - 2N2222 Collector
9v Positive - Switch - 10K Resistor - 2N2222 Base
2N2222 Emitter - 680 Ohm Resistor - LED - 9v Negative

I have replaced the 180R with a potentiometer and if the transistor is placed in one way, the LED gets will go from off to bright at a slow pace. When I put the transistor in backwards, nothing happens for about 1/2 the potentiometer's range and then it will become bright at a rapid pace (depending upon the position of the knob).

My question is why is the 2N2222 working both ways? According to everything I have read up to this point, current should only flow in one direction.

[Zero999]

The transistor will work if the emitter and collector connections are exchanged but the gain will be much lower than if they're connected correctly.

You're also lucky that the LED didn't turn on without a base signal when the transistor was connected backwards because the breakdown voltage is lower when the collector and emitter are connected backwards.

A better test would be an oscillator driving a piezo transducer or a small speaker via a buffer amplifier, which won't work when the emitter and collector connections are reversed because the gain will be too low.

[alm]

The easiest way IMO is to use the diode test of a DMM. The base-emitter and base-collector junctions both behave like diodes, in case of npn, the base is the anode (cathode for pnp). The base-emitter forward voltage drop is slightly higher than the base-collector voltage. No need to use a special tester, just use any DMM.

[Longhair]

I know this may sound very basic to you, so please bare with me...

Basic Concept: I would need to build a 555 oscillator circuit, have the output go towards the base of the transistor for the trigger and then have the emitter go to the piezo buzzer?

If you do not mind, I do not want to have the schematic posted for the finished project so I may learn by trial and error. However, any information on what required reading of subject matter would be greatly appreciated since I have only basic beginner books on hand (Make: Electronics and Getting Started in Electronics).

[Psi]

You could also get one of these for US$60

http://www.peakelec.co.uk/acatalog/jz_dca55.html

Ya just clip the 3 leads on in any order and press the button, it will tell you which is E B C or G S D for fets or A C for diodes/leds etc..
Also gives you the gain of the transistor, collector test current and any leakage current.

[Longhair]

I just seen that site (Peak Electronic Design) from reading a different thread - lol.

I haven't had time to read everything, so I just bookmarked it for later. Their products do look interesting and would come in handy.

Unfortunately, to myself it would feel like using a calculator without knowing how to solve the problem by hand first.

[Zyvek]

I just seen that site (Peak Electronic Design) from reading a different thread - lol.

I haven't had time to read everything, so I just bookmarked it for later. Their products do look interesting and would come in handy.

Unfortunately, to myself it would feel like using a calculator without knowing how to solve the problem by hand first.

I do have the Peak meter and I can confirm it does make you lazy. :-)

[Zero999]

I know this may sound very basic to you, so please bare with me...

Basic Concept: I would need to build a 555 oscillator circuit, have the output go towards the base of the transistor for the trigger and then have the emitter go to the piezo buzzer?

No a 555 timer won't help you because it'll just be doing what you've been doing manually automatically.

I'm not talking about a piezo buzzer but a transducer. A piezo buzzer is a transducer with an oscillator built-in. A piezo transducer is just a piezo speaker consisting of a copper disc with a piezo crystal wafer glued to the back of it. The transistor under test actually forms the oscillator which won't work if it's connected the wrong way round because the gain won't be high enough.

Here's an example of a circuit. The piezo transducer is a special type with a feedback lead which provides the positive feedback required for the oscillator. The part of the transducer marked E P is 0V, it normally has a black wire and is connected to the copper. M is the drive which is normally red and is connected to the large electrode on the piezo wafer. F is the feedback and is connected to the small electrode.

T can be any transistor, for PNP just reverse the power supply. Incidentally you should reduce the voltage to 6V or below when testing transistors as exceeding this voltage when the transistor is connected back-to-front can damage it.

This circuit is actually used inside most piezo buzzers, instead of building it, you could just rip open a buzzer, remove the transistor and replace it with short test leads which connect to the transistor being tested. I wouldn't recommend this to someone new to electronics because the components could be SMT and it will require some reverse engineering.

[Bored@Work]

I just seen that site (Peak Electronic Design) from reading a different thread - lol.

I haven't had time to read everything, so I just bookmarked it for later. Their products do look interesting and would come in handy.

Unfortunately, to myself it would feel like using a calculator without knowing how to solve the problem by hand first.

I do have the Peak meter and I can confirm it does make you lazy. :-)

Build your own http://coremelt.net/files/software/repository-tarballs/semiconductor_tester.tar.gz

[Zero999]

I have another: you could build a Joule Thief.

A Joule Thief is a blocking oscillator which boosts the voltage of a 1.5V battery to over 3.5V so it can power a blue or white LED. It almost certainly won't work if the transistor is connected backwards and is pretty easy to make. If you want to test PNP transistors you need to reverse both the power supply and the LED but you can connect two LEDs in reverse parallel to make it easier for you.
http://blog.makezine.com/archive/2007/11/make_a_joule_thief_weeken.html

[Chasm]

Build your own http://coremelt.net/files/software/repository-tarballs/semiconductor_tester.tar.gz

The DIY version is actually nice. The firmware is not only open but also still gets improved.
It is really cheap, the biggest investment is usually the LCD.

I'm currently working on one myself, trying to cram some more stuff on the board "because I can", ot at the moment rather "because I want" (too much, to fit without a lot of work).

[Longhair]

Hero999 - I see where I went wrong by wanting to build a 555 oscillator circuit first. My google searches are coming up with more advanced topics (for my needs at the moment) than what I really want. I took a step back and googled "transducer" and found this video: that helped me understand better what you were talking about in your first reply. Thank you for the additional information. I do not have the parts to put together the two different schematics that you have provided yet. I will put them on my list of things to order because what good is theory without practical experience?

BoredAtWork - I poked around the files of the link after extracting it with WinRAR. The instructions are in German (I think) so it may be a while before I get the time to fully understand everything in there. It does look to be more advanced but that pushes me forward. Thank you for the files.

[Zero999]

Great video.

The Joule Thief oscillator works on exactly the same principle. The difference is that feedback is taken via the transformer formed by putting two windings on a ferrite bead.

[Chasm]

BoredAtWork - I poked around the files of the link after extracting it with WinRAR. The instructions are in German (I think) so it may be a while before I get the time to fully understand everything in there. It does look to be more advanced but that pushes me forward. Thank you for the files.

The project was posted in one if the German electronics forums, mikrocontroller.net, after a while the firmware development moved into the SVN.
There is a long topic about it, some parts of it are in English.
There is also a German language wiki article which describes the principles used. Google translate seems to work more or less.

If you have a specific question about the project, posting without registration is possible at mikrocontroller.net and widely used, even posting in English is encouraged. But better try to read or at least skim through (english parts of) the Thread before posting, or someone will tell you to RTFM.

[FreeThinker]

Build your own http://coremelt.net/files/software/repository-tarballs/semiconductor_tester.tar.gz

The DIY version is actually nice. The firmware is not only open but also still gets improved.
It is really cheap, the biggest investment is usually the LCD.

I think Dave has a few of these to get rid of 

[mikeselectricstuff]

The Peak is a handy box, especially when doing repairs. However if you don't have one, you can easily identify the pinouts of standard (working!) BJTs using the diode test function of a DMM and a wet finger :

1) Connect probes to pins at random til you see a voltage  around 0.6-0.7V
2) Swap the -ve lead to the unconnected pin. If you get 0.6-0.7V again, it's NPN and the +ve lead is base. If not, it's PNP and the lead you originally had the -ve connected to is base. Confirm this by returning the -ve lead to the original pin and swapping the +ve to the other pin - you should again see 0.6 to 0.7v. 
You now know if it's NPN or PNP, and which lead is base.
3) Connect the leads to the emitter and collector (i.e. the ones that are not the base identified above).
4) Lick your finger (you  want slightly damp, not dribbling with drool). If it's an NPN, apply wet finger between base and the + lead, if PNP, between base and -ve lead.
5) Swap leads and repeat step 4.
In the orientation that gave the lowest reading (<1V), or required the least finger pressure, the + lead is collector if NPN, Emitter if PNP.

[williefleete]

go back to basics

this is plastic board and screws basic and similar to Dick smiths funway into electronics volume 1 "transistor tester"
attach a "reversing switch" and inverse parallel LEDs so you can test PNP's and test them reversed or make two of them