Go Down

Topic: 48V e-bike motor control (Read 1 time) previous topic - next topic


I have been working on an e-bike conversion project for a while now and decided that I would like to make my own motor controller. I have played around using an ArduinoUno to control some small BLDC motors somewhat successfully. I have a good feel for the programing aspect of the motor controller but do not understand the power control that goes along with it. I am using a 48V BLDC hub motor that I got from Golden Motors (MagicPiex2 for use with external motor controller). I have seen some pictures of the inside of the corresponding motor controller but cannot make heads or tails of some of the power electronics of the circuit inside.

This link provides lots of great views of the capacitors in question.

I understand the use of the transistors as a bridge to drive the different phases of the motor, but I cannot understand how they are using the capacitors and how it is all wired together.

Could anyone provide a schematic of the circuitry or some explanation of how I will need to implement the use of power electronics (and why)?


It takes many, many years of study and hard knock experience to successfully design a high power motor controller. Buy a good one and appreciate the fruits of someone else's hard work!

See this thread for just one consideration: http://forum.arduino.cc/index.php?topic=237033.0


The capacitors are simply decoupling, just rather more is needed for a high-current
motor controller.

A three-phase MOSFET bridge is used to drive BLDCs.  The power electronics issues
are mainly those of getting the protection circuitry right so you don't blow-up MOSFETs
(yes, they explode if you get it wrong - eye-protection is needed).

At a minimum you need a proper MOSFET high-low drivers, overvolt protection on
the gate-source side, transient spike suppression on the main power bus, attention
to low-inductance signal routing, and most importantly rapid over-current shutdown.

With 48V at dozens of amps every part of the circuit is subject to intense noise pick-up
if you get the layout wrong.

Its a big project, you'll need to do your research and be prepared for failures,
and you will need an oscilloscope and a large bench PSU.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]


Thank you for your feedback. I was aware it was a large task but did not quite understand the intricacies of the circuit requirements. It sounds like I have a lot more research to do on the protection circuitry.


Well, that is a huge adventure and will involve lots of smoke and flames for sure.  Snubbers, charge induced into Cgs by high output slew rates and stray inductance will do things you would never have imagined, all of which result in destruction.  I recommend buying a big pile of very cheap mosfets, e.g. IRFZ44 or whatever you can pick up in bulk for <50c each; try Tayda Electronics (they seem to sell genuine semis) or pick up like 50 on eBay and hope they're relatively good fakes.

To get you started, read the datasheet for the IR2130 and all three of its appnotes (AN-985, DT04-04, DT99-7); they contain so much gold regarding the subtleties of what you need to get right in order to not have it burn down. 


One piece of advice - always start with a lower voltage and low current limit and
work upwards, checking everything is OK (waveforms, shoot-through, thermal
dissipation) before increasing current limit or voltage step by step - much less
likely to blow anything up.

Just connecting 48V lead-acid battery and hoping will most likely fail so
spectacularly that you have no evidence of what happened first...
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

Go Up