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. snake0

    snake0

    54
    1
    Mar 30, 2014
    Yes side with band is connected to + voltage.

    Also tested without diode, same results.
     
  2. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    That doesn't make sense. A base current of 11 mA will saturate the transistor and it should have no trouble passing 200 mA. Vce should be 0.3V or less, and you should see at least 4.7V across the motor.
    So you measured 43.6 mA current flowing into the collector?

    That sounds like the transistor is not saturating.

    Can you upload a photo of your test setup showing all connections?
    Is it possible the transistors are crap? Where did you get them from?
     
  3. BobK

    BobK

    7,680
    1,685
    Jan 5, 2010
    The only other thing I can think of is that you cannot get enough current through the motor to start it. The start current can be quite a bit higher than the run current. Try shorting the collector to ground with everything else the same then breaking that connection. If the motor continues to run, then the start current is the problem.

    Bob
     
  4. snake0

    snake0

    54
    1
    Mar 30, 2014
    Quite possible, got them from Ebay.

    Nice idea but it didn't work

    Tried my best to take a decent picture, hope its visible..
     

    Attached Files:

  5. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    Looks OK but I can't be sure. Need to be able to see the transistor and the connections to it in more detail.
     
  6. snake0

    snake0

    54
    1
    Mar 30, 2014
    If the transistors are bad how does that explain the fact that they can run the motor if the motor takes current from the emitter?

    Edit: actually that doesnt work either if i use a resistor to join the base to positive, it was only working because i was connecting base directly to positive before
     
    Last edited: Mar 31, 2014
  7. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    If the motor "takes current from the emitter"? If you connect the negative side of the motor to the emitter, you have connected the motor straight across the battery and the transistor is not part of the circuit. That's why it would work.
     
  8. snake0

    snake0

    54
    1
    Mar 30, 2014
    No I meant positive side of motor to emitter and negative of motor to ground
     
  9. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    Oh, the emitter follower configuration? In that case, you were connecting the base directly to VCC (positive supply rail), right? So the motor current could have been flowing through the base-emitter path of the transistor, with no (or little) current flowing in the collector.

    To test for this, try the emitter follower configuration with a resistor (e.g. 330R) in series with the base. If the emitter follower stops working when you limit the base current, then that points to crappy transistors as well; if it works with no resistor, the current is mostly flowing through the base-emitter junction.

    Just out of interest, can you post a link to the eBay auction?
     
  10. snake0

    snake0

    54
    1
    Mar 30, 2014
    Yeh it's just like you say, the emitter follower stops working when base current is limited. I think I will just have to buy new transistors, which sucks because I have like 50 of these and I really wanted to find a solution.
    Unfortunately I don't have a link to the ebay page anymore, I do know they were from China though (surprise surprise).

    Next I will try to test 2 in parallel to see if there is any joy
     
  11. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    I think you might as well throw them all away. If they all have extremely low gain, paralleling them isn't going to help.
     
  12. snake0

    snake0

    54
    1
    Mar 30, 2014
    You're probably right. I did learn a little about transistors though, so I guess it wasn't a total waste. Thanks for the help!
     
  13. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    No problem. Remember to place negative feedback on the seller! Try to help others from having to go through the same trouble.
     
  14. kpatz

    kpatz

    247
    52
    Feb 24, 2014
    Another thing to consider is the maximum collector current of the 3904 is 200 mA, so you're pushing the transistor right to its limits. The motor may need more than 200 mA to start up, they always draw more until they're up to speed.

    I'd use a 2N2222 or similar transistor, which can handle collector currents of 800mA to 1A. With a minimum gain of 40, I'd set the base resistor to give a base current of 10-20 mA.

    Before leaving negative feedback, try testing the transistor with a more modest load, such as an LED. They may be OK after all.
     
    Last edited: Mar 31, 2014
  15. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    Hmm, you're right kpatz. I thought the 3904/3906 were more like 600 mA.

    He said he has used them with LEDs and found they were OK. But I'm not convinced that they can deliver 200 mA.

    snake0, there's an easy way to test them: replace the motor with a resistor that will draw 200 mA. R = V / I = ~5 / 0.2 = ~25 ohms. If the transistor can supply 4.5V or more to a 22 ohm resistor load, it's OK. Please let us know what you find.

    You should also look into low-Vgs MOSFETs. They will do the same thing but they don't require any continuous base current. Here's a really impressive one: http://www.digikey.com/product-detail/en/NTD4906N-35G/NTD4906N-35GOS-ND/2194521 the ON Semi NTD4906N. It can switch 30V and 10A, with an ON resistance of 0.008 ohms! That means that at 200 mA load current, it would drop only 0.0016 volts, leaving 4.9984V across the load!

    It's overkill for that application, but it's a good example of what's out there now, and it's available in a package with wire leads. Many modern MOSFETs are only available in surface-mount (SMT) packages. It's USD 0.57 in one-off quantity, a lot more expensive than the transistors you got, but if you get it from Digikey it'll probably be genuine!
     
    Last edited: Mar 31, 2014
  16. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,361
    2,757
    Jan 21, 2010
    Even after reading all of this I'm not convinced that snake0's problem is the transistor.

    Having around 210mA through the motor and it working, compared to 200mA and 0.5V and it not working are a very unlikely combination.

    I question if (a) snake0 is shorting the power supply with his multimeter and measuring its short circuit current (is it a low power usb port?), or (b) is he somehow measuring the voltages incorrectly.

    If the motor has 200mA flowing through it and it still hasn't started then giving it a bit of a spin with your fingers should either start it or provide evidence that it is far more willing to operate in one direction than the other.

    Perhaps the emitter and collector have been swapped.
     
  17. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,268
    Nov 28, 2011
    I don't think that's the situation. Back in an earlier post he said:
    ... and
    I've been assuming that this 43.6 mA collector current was the actual measured current, not "calculated" as he says above, because I don't know how he would have calculated it.

    It's not the result of calculating based on the motor voltage and the motor resistance - that would give Rmotor = V / I = 5 / 0.2 = 25 ohms; with 0.7V across 25 ohms the calculated current would be I = V / R = 0.7 / 25 = 28 mA, not 43.6 mA.

    If the motor has 0.7V across it and the collector current is 43.6 mA, when there's a base current of 11 mA, that points to the transistor - unless, as you say, something isn't connected properly.
    Yes, I think that's possible. Apparently a transistor will kind-of work with them swapped (I've never tried it myself) which might explain why they worked with LEDs but not with the motor. The pinout with emitter and collector swapped matches the European pinout. It would be worth trying a new transistor from the bag with reversed orientation.
     
  18. Harald Kapp

    Harald Kapp Moderator Moderator

    9,900
    2,097
    Nov 17, 2011
    Even if in post #16 snake0 insists in having the pinout absolutely coorect, there's still achance of collector and emitter being swapped which results in a very low current gain and could explain the effects.
     
  19. snake0

    snake0

    54
    1
    Mar 30, 2014
    Back again with a bunch of 2222As, now building a H bridge.
    I found earlier that my 3904s could drive the motor to some extent by using A) 100ohm resistor at base or B) running two-three in parallel to split the current load. I don't know if that is the original spec of those transistors or maybe I just got a bad batch. Anyway they are not a complete waste, as I can use them to convert a positive signal into a negative one for my H-bridge's PNP.

    Still though I am not entirely sure of a conclusion here, despite 3 days of effort. I suppose either the motor needed too much current or the transistors sucked.
     
  20. snake0

    snake0

    54
    1
    Mar 30, 2014
    Currently with the 2222As (1k resistor at base) I am experiencing another problem, in that I have to manually spin the motor to get it to start or short the base instantaneously. The other solution is to replace the 1k resistor with a 330ohm, that starts the motor on its own. Is there another way to do this? If I have to manually spin the motor myself each time it kind of defeats the purpose of the H-bridge, which is there to control the motor's direction...

    (The reason I don't want to use 330ohm is because the book I am reading says that will cause the transistor to use up more power and drain the battery faster- plus almost all schematics for H-bridges seem to use 1k at the base of the transistors!)
     
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

-