Driving a small motor - transistor issues

Discussion in 'General Electronics Chat' started by rjjenkins, Apr 26, 2015.

  1. rjjenkins

    Thread Starter Member

    Apr 16, 2011
    102
    2
    Hello

    I need to drive a little haptic vibrating motor from the output of a MCU. I'm running the circuit off 3V (a CR2032 coin cell). The motor works OK when directly connected to the battery and draws about 35mA. But I can't get enough current to it through a transistor. I first of all tried driving it through a MOSFET (a 2N7000) but it barely turns over, if at all. I tried it with a NPN transistor (2n3904) but again the motor only just works. Thinking it might be a problem of insufficient gain, I tried a Darlington, but again with no luck.

    I guess that in the case of the MOSFET, the gate voltage from the MCU is just not quite enough to switch it on, or possibly at this gate voltage, the drain-source resistance is too high. If so, can anyone recommend something better (preferably small, cheap, easily available and in SMT). In the case of the NPN, my guess is that the voltage drop across the transistor is too much to power the motor, and the drop would be even greater with the Darlington.

    I don't know much about all this or whether my guesses might be right, so would appreciate advice.
     
  2. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,800
    1,104
    I see no reason why an NPN with sufficient base current couldn't drive the motor. Bear in mind that the start-up current for that motor will be several times the running current so could be, say, 100mA or more. To switch the NPN fully on you should have a base current ~ 1/10 th of the collector current. So the base current might need to be >~10mA. Can your MCU provide that?
     
  3. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    The coin cell is struggling to provide enough current. Adding slightly more resistance in a switch between the battery and switch will prevent the motor from starting.

    Are you powering the mcu on the same coin cell?
     
  4. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,548
    2,373
    As already stated, if driving from an MCU you should be using a fet, the 2n7000 should do that no problem with the right supply, it is in fact a darlington representation in a FET package.
    What is the actual voltage rating of the motor? When fed through a semi conductor the motor supply should always be above the plate voltage of the motor, => +10% is not abnormal.
    Max.
     
  5. MikeML

    AAC Fanatic!

    Oct 2, 2009
    5,450
    1,066
    So here is the problem: During the time the motor switch is on, the current draw pulls down the coin-cell terminal voltage so low ( due to the rather high internal resistance of the coin-cell) that the NFET gate voltage is too low to turn on the NFET. If you store the coin cell voltage on a capacitor (to power the MCU while the motor is starting up), then you are likely to get beyond the start-up transient.
     
  6. rjjenkins

    Thread Starter Member

    Apr 16, 2011
    102
    2
    Thanks all. Basically the coin cell just can't deliver enough current. Even with two in parallel they struggle. The capacitor idea didn't work unfortunately. Back to the drawing board (to decide how to power this thing).
     
  7. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,548
    2,373
    Did you actually measure the voltage across the motor when it was full on?.
    RPM is equal to applied voltage.
    Max.
     
Loading...