All testing of faulty controllers should be done on a current-limited bench power supply: testing a faulty controller from a battery will almost certainly exacerbate the fault, causing yet more damage.
The power supply should be current limited, say between 5 and 10 amps. However, be aware that, if a motor is connected, then, when the controller regenerates, it may pump up the power supply voltage to 50 or 60 volts or more (depends on voltage of controller) so make sure your bench supply can take this. The controller itself is protected.
Beware also of operating the motor with the controller ramps set to very fast: this will cause rapid acceleration, shorting out the power supply completely. Loss and re-appear of the power supply is not a designed-in operating state of the controller and nothing can be guaranteed.
It is possible to test the operation of the bridge without connecting a motor and without having the top board in place.
With the top board removed, the base board is arranged so that all four bridge legs are biased off – a safe condition. The appropriate input pins of the 14 way connector can be pulled individually high or low (as appropriate) to activate each bridge leg in turn for testing, as shown in the diagram.
|Pin||Bridge Leg||To activate|
This is a method of testing the bridge with MOSFETs still connected and without the top board in place without a motor connected.
With all four of the legs off, the motor outputs should drift low – this means that it is easy to see hiside operation, but not so easy to see loside operation. To make lowside operation easy to see, use a resistive load connected from motor terminal to +Vbat. Or you could use two resistors (470 Ohm – or thereabouts – power resistors, they will dissipate nearly 3W at 36v) as a potential divider to hold the bridge output (motor terminal) at around half battery voltage.
Now activating the hiside will pull this half voltage up to full supply and activating the lowside drive will pull the motor output down to zero volts. Any fault in high side or low side either causes a large current drain (which is why the supply current must be limited) or the output will not move properly.
Since the MOSFETs do most of the work in a motor controller, most faults are likely to result in their destruction. This is usually visibly apparent – the MOSFETs either split wide open or they erupt in a cloud of flame and dense smoke.
In either case, you will have to dismantle the baseboard assembly and test the MOSFETs individually.
If any MOSFETs have failed, you must then test the drive circuitry. In particular you should check, for each bridge leg where MOSFETs have failed:
- All the 10R gate resistors (subminiature size, as these generally fail open-circuit under fault conditions).
- The 12v gate transient clamp zener
- The three driver transistors.
It is rare for a single MOSFET to fail: usually there is a chain reaction and other components follow the initial failure. The internal current limit and shutdown will however often switch the controller off within the failure chain. This can be a valuable aid to checking where the problem started. For instance, if a wiring fault (such as a short from a motor terminal to battery negative) were to blow the high side, all the hiside MOSFETs would fail but the current limit could well restrict damage to the corresponding low side, so that only one or two MOSFETs might fail there.
These are detailed on a separate page.
You can also buy spare base boards (without MOSFETs) and control boards for the controllers. Email for prices.