NPN transister switch question

Discussion in 'The Projects Forum' started by hensle, Jan 1, 2012.

  1. hensle

    Thread Starter Member

    Dec 31, 2011
    39
    0
    [​IMG]

    This is an NPN switching circuit. If I understand it correctly, unless the input is connected to the +voltage source, the switch is off and no current should flow.

    I tried this exact circuit with three different NPN transistors....each of which I successfully used as voltage amplifiers, and each one, put in this circuit, failed to restrict the current flow, unless I switched the load to between the emitter and ground.

    Does somebody know why this is the case?

    I am making an H-bridge and the NPN transistor must completely block the current flow when the load is on the collector side or it just can't work.

    The transistors I tried were
    TIP31C, TIP31AG, MPS3704
     
  2. MrChips

    Moderator

    Oct 2, 2009
    12,447
    3,363
    Are you certain that you had the leads of the transistor identified correctly? Show us what you did.
     
  3. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    Please also describe your load device. The circuit as shown should work correctly if is used to switch a resistive load of not much less than 1000 ohms. Much lower loads night not be turned on fully (the transistor may not saturate), and significant power may be dissipated in the transistor. With a low enough load a small transistor could overheat and may go short-circuit.

    If the load was inductive (a relay or a motor, for example), with no protection against back-emfs, a voltage surge could cause failure at switch-off.
     
  4. wmodavis

    Well-Known Member

    Oct 23, 2010
    737
    150
    " failed to restrict the current flow "

    Do you mean when the transistor switch was turned OFF the current still flowed in the load? If that is not what you meant then the information you supplied and your question need to be expressed more clearly and precisely.
     
  5. hensle

    Thread Starter Member

    Dec 31, 2011
    39
    0
    I did it exactly as shown in the schematic. That is why I provided it, to show you what I did. (is the schematic visible in your browser?)
     
  6. hensle

    Thread Starter Member

    Dec 31, 2011
    39
    0
    Oh...OK I did not specify the load. It was just a 50 ohm resistor and in separate tests, a 10k ohm resistor. The issue is not saturation. The issue is, according to everything I have read, in books and on the internet, when the input is not attached to the +voltage source, the current should not flow (at all), but in these tests there was a current of 180mA at 9 volts. In one place I read that this tested the transistor and if current flowed (when the base was open or grounded), the transister was bad. So are all the transistors I tested bad? (I tested 10 transistors, 6 of them brand new) And if they were bad, how was I able to use them to amplify voltage? If this is the normal functioning of the NPN* transistors, then an H-bridge would be impossible to build....and yet I know H-bridges exist and function....you see...it is just a bizarre conundrum.


    *I tested some PNP transistors with the appropriate circuit and they functioned as expected. i.e. with open base, no current flowed.

    Thanks to all who replied.
     
    Last edited: Jan 2, 2012
  7. MrChips

    Moderator

    Oct 2, 2009
    12,447
    3,363
    I can see the schematic. What you need to post is a drawing of the physical pin-out of the transistor.
    Something like this:


    [​IMG]

    I betcha you wired the transistor incorrectly.
     
    Last edited: Jan 2, 2012
  8. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    This is about as basic a transistor circuit as you can get, and it will be very much worth your while finding out what the problem is before going any further. It is not uncommon for beginners to convince themselves that such circuits do not work: the answer is usually a simple repeated mistake, which when made once is far easier to repeat than to find.

    It is possible, but really quite unlikely that you might get six bad transistors from new (perhaps all of the same type, from a bad batch). Unless, that is, you were buying manufacturers' reject devices - we used to do that back in the day when I was a teenager, but surely nowadays that kind of junk has gone the way of the chamber pot and the paddle steamer? But I digress...

    Referring to the diagram in Mr. Chips' last posting, note that transistor pin-outs vary between different types, and although devices in similar encapsulations often follow the same pattern, this is not always so. Here, the data-sheet is your friend - consult it!

    Other problems are possible, including reversed supply voltages: the collector and base supplies must both be positive for NPN transistors, but negative for PNP devices.

    Another possibility is that you are having problems building the circuit. Connections to solder-less breadboards can be confusing, or if you are soldering the circuit together it is possible to damage components, either by bad technique or faulty equipment such as a non-grounded iron with AC leakage. If possible, please post a photograph of your circuit - this may allow somebody to spot the problem.
     
  9. jimkeith

    Active Member

    Oct 26, 2011
    539
    99
    Recheck your work--If a circuit does not work as expected, I recheck--still have to do this after 40 years experience!
     
  10. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,392
    1,606
    A transistor has three leads, meaning there are six different variations on how they can be wired. On a bad day it can take me all six variations until I get it right.

    180mA at 9 volts indicates the 50 ohm load is essentially shorted to ground. One way that could happen with a good transistor is if the collector and base connections are swapped.
     
  11. hensle

    Thread Starter Member

    Dec 31, 2011
    39
    0
    Thanks for the replies.

    I discovered the problem and its a bit odd so it might be worth mentioning.

    It turns out that the transistor was wired correctly and the transistors were functioning correctly...However,
    I was measuring the voltage accross the resistor with an oscilloscope. The current was flowing to the oscilliscope ground, bypassing the transistor all together.

    It was a kind of zen-quantum thing. I was disturbing the experiment in the act of measurement.
     
  12. Adjuster

    Well-Known Member

    Dec 26, 2010
    2,147
    300
    That certainly was worth mentioning, both as a reminder to others of the underlying problem, and to relieve anyone who might still be scratching their heads. It's quite common for things to get shorted out by grounded connections to oscilloscopes (and other instruments like signal generators). This is something that is best kept in mind, as in some cases damage can result, or personal injury, up to and including loss of life (!)

    The ground side of any instrument should normally only be connected to a line which is already at ground potential, or can be forced to ground potential without altering the circuit operation.

    The simplest approach when using an oscilloscope is to reference all measurements to ground. In cases where it is necessary to measure the difference between two points, neither of which is at ground potential, some oscilloscopes with two or more inputs may have differential mode facilities which allow this.

    Whatever you do, please do not be tempted to remove the mains ground from the oscilloscope. This increases the risk of somebody getting a lethal electric shock, and results obtained this way are more likely to be corrupted by AC mains interference.
     
Loading...