I want to do some experiments developing brushless DC motor control hardware, and maybe wiring my own motors for my own edification. LiPo batteries aren't convenient enough, and I want current-limiting protection against shorts or failure in my electronics or motors.
The intention is to use commercial 'Hobby' RC BLDC motors, and my own hardware to drive BLDC motors. (I will try to use commercial RC ESCs as well, for testing)
Almost all of the BLDC motors I am interested in are 3 LiPo, i.e. a nominal 11.1V, or less.
I would like a bench power supply which lets me set the current limit, in order to reduce the chance of damaging my electronics or motors. The aim is to emulate LiPo batteries, but with a user-defined current limit. For the sort of things I expect to do, 40A should be plenty.
Unfortunately, the sort of equipment sold by electronics distributors is unacceptably expensive. 40A Bench PSU's seem to start at 500GBP, which is too much cost for a 'hobby' project when motors and electronics are under 50GBP. e.g. Farnell
So I am looking for a cheaper solution which would provide over-curret tripping (like an electronic fuse).
There are lots of switch mode power supplies which claim 12V at 40A or more, for around 50GBP. 12V is close enough to 3LiPo's that it is an acceptable compromise.
A real bench power supply limits the current, and that would be ideal. However, I don't see a way to achieve that at sub-100GBP. Something which rapidly turns off power when it detects over-current, would be an acceptable compromise (unless I am missing something).
I could make an over-current sensor using an off-the-shelf current shunt measuring chip, and a microcontroller. I would 'dial in' my required current limit using a suitably tasteful human interface. It would measure the actual current, and signal an over current measurement, which needs to disable the power supply.
I am not sure how fast switch-off should be. I expect, say, 20ms (1 cycle of UK mains) would be good enough (500W for 50th second sounds survivable). I would like faster, but I am assuming I might need to switch the mains input power off. I would much prefer a solution which does not require me to control the mains input.
I am happy to put fuses in the line too, but I do not believe they are flexible enough, or convenient enough to be the primary current limit
I had imagined some switched power supplies would have an "enable" input, but I haven't found any so far.
Edit: Q1: is no longer relevant
The proposal is to 'disconnect' the PSU power from the BLDC driver after the PSU, using a suitable implementation of an electronic fuse.
Q1a. Will a rapid (couple of ms) disconnect of the PSU from the reactive load create a problem that might damage a switched PSU?
Q2: There is going to be a lot of quite high frequency modulation from the motor control PWM, as it switches power to the motors (I am guessing in 100's of kHz). Is it likely that this will significantly effect a switched power supply, and maybe even make the whole idea unworkable? Might a simple fix, be several mF of capacitors, or is it more complex than that (i.e. it'd need a tuned circuit, Edit: or adjust motor PWM to work around the power supply)?
Edit Q3: was "I assume the current shunt should be on the high-side. Is that correct?"
Commercial RC ESC's don't care because they are designed to work off batteries.
However, using a high-side shunt should make experiments and test-rig measurement a bit easier, by maintaining a ground reference.