0
\$\begingroup\$

I would like to power my Arduino and BLDC to be powered with an old computer power supply. My Arduino needs 5V and my 3 phase BLDC needs 12V.

schematic

simulate this circuit – Schematic created using CircuitLab

Details

  • I'm using a SLA5068 N-channel MOSTFET bridge. Datasheet Here.
  • My sequence is taken from the library AccelStepper--here is the homepage. The sequence has 3 steps; during each step, only one phase will be on.
  • The BLDC motor I'm using is one from an old Western Digital 500gb HDD. I can't find the specs for it, but I know it's rated for 12V because I looked up a datasheet for the motor controller chip that was on the board with it.

The power supply I'm using: enter image description here

Behaviour

  • If I use the 5V for powering my BLDC, it works just fine (except I can't go very fast).
  • If I use the 12V supply, it works for a moment and then shuts off.

Things I've tried

  • Putting a 1K resistor from the 12V to ground. Also tried with a 570 ohm
  • Putting in a capacitor to ground.
  • Putting a fan from the 5V to ground so it has a constant load.
  • Putting a flyback diode in the circuit.

The most puzzling thing

This works fine with my computer fan, which is a BLDC that has a controller circuit built inside it; this leads me to believe I am missing something in my design.

Perhaps this is an important detail, when I use the 5V supply for my 3 phase BLDC, I don't hear any clicking, but when I use the 12V supply, I hear a lot of clicking and then it shuts down quite promptly.

\$\endgroup\$
  • 1
    \$\begingroup\$ Can we have an actual schematic please - how are you driving your MOSFET gates?, what type of MOSFET are they? How is the motor wired up (schematic!). How are you commutating the motor? Is your sequencing correct. \$\endgroup\$ – Tom Carpenter Dec 4 '15 at 2:48
  • \$\begingroup\$ Sure, will do!! \$\endgroup\$ – Klik Dec 4 '15 at 2:56
  • 1
    \$\begingroup\$ Is this better? \$\endgroup\$ – Klik Dec 4 '15 at 3:17
  • \$\begingroup\$ Maybe you have not met the minimum load for the power supply so it doesn't fully come on? \$\endgroup\$ – Daniel Dec 4 '15 at 3:21
  • \$\begingroup\$ Does the supply shut off? (measure with a multimeter) or is it just the motor stalling? If you are not sequencing it correctly (because you are using an open-loop approach), it will likely just stall out because to rotor gets out of sync with the sequencing of the windings. \$\endgroup\$ – Tom Carpenter Dec 4 '15 at 3:26
0
\$\begingroup\$

I figured out what the problem was.

CLUES

  1. There would be a clicking noise when I was using the 12V output and it would run for a moment before shutting down.

    • This hinted that the power draw was not constant.
  2. My BLDC would run when connected to the 5V output and there were no clicking noises.

    • This hinted that the problem was to do with the amount of current being drawn.
  3. The power supply would stay off until I hit the switch off and then back on.

    • This hinted that it was due to a safety feature since it entered a mode where functionality was ceased completely until I restarted the supply.

Solution

One of these: enter image description here

A large capacitor connected from the positive to the negative feed of the power supply smoothed out the sudden changes in current draw and fixed the problem.

Now there is no clicking noise and the motor runs just perfectly.

I hope this helped someone else.

\$\endgroup\$
  • \$\begingroup\$ you sort of need a freewheel diode \$\endgroup\$ – JonRB Dec 4 '15 at 19:54
  • \$\begingroup\$ It's built into the MOSFET I'm using. \$\endgroup\$ – Klik Dec 5 '15 at 2:19
  • \$\begingroup\$ If the cct is exactly as you drew it... The intrinsic diode will not provide a freewheel path for the inductive current at forced commutation \$\endgroup\$ – JonRB Dec 5 '15 at 7:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.