Electrical Engineering Stack Exchange is a question and answer site for electronics and electrical engineering professionals, students, and enthusiasts. It's 100% free.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm new to electronics and here's a newbie question that I would like to ask: How do I know if which leg is the emitter or the collector in a Transistor (For both PNP and NPN) using only an analog multimeter?

share|improve this question
Do you already know which is the base? – Majenko Dec 6 '11 at 12:36
up vote 13 down vote accepted

Set meter to lowish ohms range so a diode conduction can be seen - trial and error OR diode test if available.

  • With an NPN transistor the base will have two diodes facing away from it. ie with most positive meter lead on the base the other two leads will show a conducting diode when the negative lead is placed on them

  • With a PNP transistor the base will have two diodes facing towards it. ie with most positive negative (usually black) meter lead on the base the other two leads wil show a conducting diode when the positive lead is placed on them

OK - now you know NPN from PNP and which is base. Now

  • Connect positive to guessed collector for NPN and negative to guessed emitter. Set meter to 1 megohm plus range.

    -Connect base to guessed collector via a high value resistor - probably 100k to 1M. A wet finger works well. Note reading.

    • Now swap guessed emitter and collector and repeat. Again resistor is added from base to guessed collector. Note reading

One of the two above will have a much lower R_CE reading when base is forward biased. That's the correct guess.

Once you get used to this you can pick up a leaded transistor, juggle it with meter leads till you find the two diodes giving base and NPN or PNP then lick your finger and do a forward bias base test - and then declare pinout. Looks like magic to many. Works.

You can or course formalize that on a breadboard and even add (gasp) switches to swap polarity etc.

Note that you can get some idea of Beta (current gain _ from this once you learn to calibrate your wet finger.

share|improve this answer
It's a nice exercise to build a µC-controlled transistor tester that does this, and more. – starblue Dec 6 '11 at 17:45
Worth noting: some meters produce enough current during resistance tests to damage small transistors. – Robert Harvey Dec 6 '11 at 19:58
@Robert Harvey - It is ceryainly conceivable that a given meter could make enough current to damage a given transistor but I don't think I'v ever seen a combination where that is the case. Most meters with a 9V battery apply 9V max and probably less (regilated). Some use 3V or 1.5V. Some few MOSFETS would have gate-source max voltage exceeded if 9V is applies. very rare. – Russell McMahon Dec 6 '11 at 21:40

The simplest way doesn't even require a multimeter:

Download the datasheet and look at the pinout diagram.

share|improve this answer
Fine enough if it's clearly labelled and you have net access. Not so goodif doing a repair away from net access. My shown method also can be used on SOT23 transistors (where CBE is usually more obvious. – Russell McMahon Dec 6 '11 at 13:01
In an SOT23 how can you even be sure it's a transistor? Diodes, BJTs, FETs, even some linear regulators - they all look the same. – Majenko Dec 6 '11 at 13:22
re how can ... : "The sound of one dog barking" :-).ie you can but try. BUT, a pretty fair test is the 2 diodes from base to C& E for BJT. Even better, in most SOT23 BJT's the base is on SOT23 pin 1 so 2 diodes should be there | A MOSFET is almost always gDS so there should be a DS reverse diode showing it is a FET and polarity showing if N or P channel. FET diode will probably have higher Vf than a usual diode at given current. The rest may be linear regulators, or a 1 I/O pin microcontroller :-). – Russell McMahon Dec 6 '11 at 21:38
This is like a link-only answer without the link, or "just google it" without the information so obtained. Only worse. – SamB Nov 11 '14 at 22:09
While this link may answer the question, it is better to include the essential parts of the answer here and provide the link for reference. Link-only answers can become invalid if the linked page changes. – placeholder Nov 11 '14 at 22:15

Useful information to know (complements the other answers) which applies to both NPN+PNP bipolar transistors and N-channel and P-channel MOSFETs:

  • TO92 transistors are almost always pinned-out as EBC (bipolar) / SGD (MOSFET) when you are facing the flat part of the transistor package and the leads are pointing down.

  • TO220/TO247/DPAK/D2PAK transistors are almost always pinned-out as BCE (bipolar) / GDS (MOSFET) when you are facing the front part of the transistor (tab on the back side) and the leads are pointing down. This is easy to remember by the mnemonic GDS = Gosh Darn Son-of-a-gun. (Or something like that. :-)

  • Transistors with metal tabs (TO220, TO247, DPAK, D2PAK, SOT-223, etc.) almost always have the tab as the collector or drain. This has to do more with the construction of the device than from any sort of convention; the collector/drain is the part of the die that is the most thermally coupled to the metal tab, so it's a natural point of electrical attachment.

  • Surface-mount transistors with two pins on one side and the third standing alone on the other side (SOT-23, SOT-323) almost always have the collector / drain standing alone. This is because the gate-source / base-emitter voltage differential is small, whereas the collector / drain may be tens or hundreds of volts different, so it provides a larger clearance for that voltage differential to have the collector / drain off on its own. The same holds true for DPAK / D2PAK transistors, where the middle pin is cut short and sticks up in the air; it's done to provide voltage clearance and you connect electrically through the collector / drain via the tab, which is (usually) the same piece of metal as the center pin.

I suppose there are certain transistor parts that are exceptions to these rules (most likely in the SOT-23 and SOT-323 packages) but I'm not aware of any -- still, always check the datasheet.

share|improve this answer
Most of your advice is OK BUT the TO92 certainly isn't. Every variant of the 6 possible TO92 ones is probably used, and a number are common. My most used parts are pinned CBE (opposite to yours) and I've seen many ECB pinouts and, I think, some BCE. That leaves only CEB and BEC and I'd be surprised if they were not used. || TO220 - Yes || Tabs - yes (except when isolated) || SOT23 - Yes - and the left pin is most usually base or gate with 2 pins facing you. – Russell McMahon Oct 24 '12 at 6:37
? Could you show me a datasheet? Here are the parts that have EBC pinouts (again, the fine print = you are facing the flat part and the leads are pointing down): 2N3904, 2N3906, 2N2222, 2N2907, 2N4401, 2N4403, 2N5087, 2N5089, MPSA42. The only familiar bjt's I could find which have another pinout are BC546/547/548/549/550 which are CBE. – Jason S Oct 26 '12 at 17:30
Off the cuff - for jellybean transistors I almost invariably use BC337 (NPN) and BC327 (PNP) - both CBE. A transistor selector book woul;d be the easy way to check distributions as they have pinout tables which they assign a letter code to and then show the relevant code for each device. In passing: CBE - BC547, 337, 338, 557, 327, 328. | ECB - BC639, BC640, ... | More anon.| Interesting -> radiomuseum.org/forum/transistor_connections.html – Russell McMahon Oct 26 '12 at 21:19
I agree that BC337`s are all different and the only way is to test them; may be each supplier has a pet wholesaler so you know what he sells so each time you know which one to visit. A suggestion ok. David M My pet supplier sells BC337(c) version. CBE from the front pins down. RS Components Somerville road Melbourne. – user23127 Apr 27 '13 at 2:56

Simplest way is measuring forward voltage between junction BC and BE, BC junction will have lower forward voltage. In case you use normal Digital Multi Meter (DMM), with similar test current to mine 2n5551 gave me this results: Vbc=642mV Vbe=648mV if i try with resistance range Rbc=23Mohm Rbe=29Mohm For analog multi-meter test currents are little bit higher compared to the digital ones so you can expect lower resistance values (100K-1M) and it is highly non linear doing it this way, but relative forward voltage is lower for BC junction (translates relative resistance is lower for BC junction) compared to BE junction...

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.