Connect with us

Transistors heating up when sinking from motor

Discussion in 'Sensors and Actuators' started by snake0, Mar 30, 2014.

Scroll to continue with content
  1. kpatz

    kpatz

    247
    52
    Feb 24, 2014
    A 330 ohm base resistor @ 5V gives you ~13 mA of base current. This is minor compared to the current draw of the motor. With a minimum gain of 40, I'd recommend a slightly smaller resistor, like 220 ohms, to ensure the transistor is fully saturated (on).

    If your transistor is controlled by an IC pin that can't source 20+ mA of current, use two transistors in a Darlington configuration and a 1K base resistor. This won't reduce the current draw of the circuit, but it will reduce the amount of current the IC needs to source to turn the transistors on.

    The example H-bridge circuits I saw on the internet that have 1K base resistors are using Darlington pair transistors such as the TIP102/TIP107 so their gain would be higher than a single transistor.

    If you're concerned about current draw into the base, you're better off using an N channel MOSFET.

    You don't need an H-bridge unless you want to be able to reverse the motor's direction electronically.
     
    Last edited: Apr 2, 2014
  2. snake0

    snake0

    54
    1
    Mar 30, 2014
    You're right, they were using TIP102/TIP107, but those transistors weren't available at my store so I opted for the 2222. I do want to be able to reverse, that's why I'm building it. Thank you for the advice regarding 330 base resistor, I will try that.
     
  3. kpatz

    kpatz

    247
    52
    Feb 24, 2014
    You can "create" your own TIP102 equivalent by making Darlington pairs with your 2222s. Then you can get away with the 1k base resistors.

    Whatever PNP transistors you're using for your bridge can also be configured the same way. PN2222, 2N2907, 2N2907A, PN2907 are good to use.
     

    Attached Files:

    Last edited: Apr 2, 2014
  4. snake0

    snake0

    54
    1
    Mar 30, 2014
    Ok, don't have enough 2222s to do that now so I'll stick with 330 at base, but if I get more 2222s do you think the darlington configuration and 1k base resistor will use less power?
     
  5. snake0

    snake0

    54
    1
    Mar 30, 2014
    Wow kpatz I just used that darlington pair setup for my 3904s with a 1k and it worked like a charm! (which is good because I burned through half my 2222s by forgetting my diodes...)
     
  6. kpatz

    kpatz

    247
    52
    Feb 24, 2014
    Cool stuff. You can make Darlingtons using a 3904 and a 2222 as well, so that you can get the higher collector current of the 2222. Put the 3904 on the input side and the 2222 on the output side.
     
  7. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,361
    2,757
    Jan 21, 2010
    Remember that when you make a darlington like this that the minimum voltage across the collector and emitter will be around 1.2V. If you're making an H bridge and rrunning the motor from a from a 5V supply, that leaves 2.6V for your motor.
     
  8. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    Yeah, what Steve said. MOSFETs are the answer to this problem. Or you can get an H-bridge IC with everything inside it. But that's no fun eh!

    I think you should try to figure out the situation with the bag of 2N3904s that you have. If they're faulty, you should throw them out to avoid confusion. Make up a test circuit with four components:

    BT1: 5V DC supply able to deliver 500 mA or more
    R1: 1k
    R2: 47 ohms
    Q1: 2N3904 under test

    Node 1 (positive rail): BT1 positive, R1 pin 1, R2 pin 1.
    Node 2 (base): R1 pin 2, Q1 base.
    Node 3 (collector): R2 pin 2, Q1 collector.
    Node 4 (negative rail): BT1 negative, Q1 emitter.

    This will feed about 4.3 mA into the base, and if the transistor saturates (which it should) the collector current will be about 100 mA.

    Measure the collector-emitter voltage; it should be less than 1V.

    I just had a look at the 2N3904's specifications. The Fairchild data sheet says the current gain is only guaranteed to be 30 at 100 mA collector current, with 1.0V collector-emitter voltage. So the 2N3904 is really not suitable for driving a motor that draws 200 mA. It's really intended to operate at a collector current around 10 mA or less.

    But I still think it would be useful to know the performance of the ones you have. Another test you could do would be to use a 10k trimpot for the base current. Start it at maximum resistance and reduce the resistance until Vce drops below say 0.3V (reasonably saturated), then either measure the base current, or measure the resistance of the trimpot and estimate the base current from that. Divide 100 mA (the collector current) by the base current to get the current gain at 100 mA collector current. If the transistor meets Fairchild's specifications, it should be at least 30. Try a few samples from the bag.

    I would be interested to know the results, but feel free to ignore these suggestions :)
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-