car alternator to motor pwm

Discussion in 'The Projects Forum' started by yourownfree, Dec 30, 2012.

  1. yourownfree

    Thread Starter Active Member

    Jul 16, 2008
    I am looking for a circuit that would work to turn an automobile car alternator to run like a motor. I might have already had a circuit to work but stopped short of trying it with a car battery. Would it have worked. I dont have the diagram in front of me but it used a cd4017 ring counter hooked to a 555 timer.I had the 555 timer set up to go from ablout 1Hz to 14 KHz, measured with a frequency counter. The output of cd4017, one-two-three went to a high power mosfet capable of 54 amps each continuous. I had one mosfet for each coil. at the base of the mosfet was a resistor to help the mosfet shut off. I also had the mosfet protected at the load with a rectifier 1n4007. Using high side switching on the mosfet. I even inserted a transistor between the cd4017 and mosfet to make sure the input went from v+ to zero, because I was uncertain if the mosfet was going on all the way and shutting down all the way.. I dont recall the mosfet number as it has been almost a year.

    My question is basically this: will it work (assuming the mosfets are big enough) or do you have a better circuit or idea that will make an alternator spin like a motor. Oh and what frequency range and duty cycle?
    Something simple or just set me on my way with your ideas.
    Thanks. I enjoy your ideas here.
  2. AfdhalAtiffTan

    Active Member

    Nov 20, 2010
    I've done that before and I couldn't make it work as I want it to be...

    I use 4017 chip too, configured to give out six-commutation phase.


    There are H, L, and S labelled means High side, Low side, and Sense.

    At first I use 555 as my clock source, it worked fine; I use MOSFETs with HB-driver.
    I could make it run fast, but, the torque is so low, rendering it useless...

    Once it gets fast enough, it began to slips, once it slips, it fall to stall.

    After researching on the net, I found out that what I do is an open-loop control of BLDC and it is ineffective (for me at that time).

    So I tried closed-loop ones.
    I construct another circuit to detect zero-crosses between all three phases, this means rather than the 555 give the clock pulse, I let the motor generate its own clock.

    It worked better than before, but all the MOSFETs getting hot when get loaded; maybe due to instability of supply rail.

    So I introduced PWM, but this mess up the zero crossing detector signal.

    Burnt all the MOSFETs, replace with BJTs. That's a costly experience... :)
    Being current controlled device, I make it to a Darlington pair, all failed because of my weak snubber...

    Then, I give there, up till now, still waiting for solution. :)
    Maybe there's someone could help us?
    Last edited: Dec 31, 2012
  3. shortbus

    AAC Fanatic!

    Sep 30, 2009
  4. AfdhalAtiffTan

    Active Member

    Nov 20, 2010
    I prefer sensorless bldc design...
    How could I install hall sensor with precision? It's quite hard to apply (to me).

    I read some appnote on sensorless bldc, but most of them utilises mcu.
  5. NFA Fabrication


    Aug 12, 2012
    What is intended to be driven by the alternator? I'd imagine an alternator will end up being a very weak motor as it was not intentionally designed with the magnets and windings around being a motor.
  6. AfdhalAtiffTan

    Active Member

    Nov 20, 2010
    How could that be? Common alternator could handle power close to 1kW...
    It might be strong, but inefficient; due to core design and heaving cogging.

    The rotor electromagnet could be simply powered, even better, the field could be adjusted for every each load requirement. But this of course, needs some kind of intelligent control.

    I'm pretty weak in MCU at present time, hence limiting my capability...
  7. rccubed

    New Member

    Feb 7, 2010
    When you control the DC voltage on the rings you can greatly increase the motor power or decrease it's power by lowering the voltage/current going to the slip rings of the alternator.
    The problem is getting the phases going through the windings in the housing.