Motor Stopped - High speed with it switched off

[benm]

Something really bad happened to my motor today! 

I was crusing down a big hill at (guessing) 60-80kms per hour.  The motor was switched off (it seems to provide a lot of resistance over 20 or 30 kms/hour if its on).  All of a sudden the motor "engaged" - it was like pulling on the brakes!

I had to push it the rest of the way to work - there was a lot of resistance and a slight shudder coming from the front wheel.

I've got a front hub 200w 24v.  I'm using pedal assist instead of a throttle.

I'm only 2 weeks old at this so any help is appreciated (I bought the motor second hand on ebay, I'm pretty sure its the MBG36F -- Mini Front Motor)

Was it wrong to cruise at 60-80kms with the motor switched off? 

Has anyone else had this problem? Anyone got any theories on what happened, or how to fix it?

Thanks

Ben

[philf]

If any two of your motor phases are shorted together, you'll get that effect.  The amount of resistance these motors can generate when they're shorted is surprising (that's what regen braking is all about).  Check each of the heavy wires going into the hub for obvious damage (particularly where they enter the hub), and if you don't SEE anything, take a multimeter and check each of the possible combinations - Green & Yellow, Green & Blue, Yellow & Blue. 

I'll wager something pops out at you - well, unless they're ALL shorted together...

The most common reason for shorts like this is when the wheel was installed with insufficient torque to hold the axle fixed.  These motors are pretty powerful, and to move the bike forward the stator and axle are twisting backwards against the forks.  If the nuts aren't REALLY tight, the whole affair turns, usually resulting in shorted/sheared wires.

Judging by the fact you were going down a very steep hill at high speed, it is likely that you used power to get UP the hill in the first place.  It's usually on such a haul (of full throttle start) that the motors twist.

Let us know what you discover!

[biohazardman]

I had a similar expereince with a regen controller, although I was not travelling as fast, when I fried a couple of diodes.  I had to disconnect the controller from the motor to be able to pedal home as just turning it off left too much drag from the motor. Original cause of the problem may have been faulty phase wire terminals similar to what Philf describes though.

[benm]

Thanks for the detailed advice.  Sounds like a fun job for the weekend!

[benm]

I just deconnected the motor completely - no resistance on the wheel.  So you're theories are looking pretty good.  thanks!  with hindsight it was probably something I should have thought to test, although...i was suprised, as the postive terminal on the battery connects straight to the on/off key, then out of the key to the controller.  That would mean there's also a short in the on/off key, right?!

Perhaps on the way down the hill the on/off key shorted, and thats what caused the motor to suddenly "engage".  Then the huge amount of force going through it has done what you've suggested, turning the whole thing and creating an additional short... hmmm.... I'll get back to you shortly...

[biohazardman]

More likely the wiring as Philf states easy to check or the controller not so easy to check if you don't have a spare to swap it out with.   Good luck

[benm]

Ok, so forget the whole on/off key theory.  With the battery completely disconnected I still get the same problem.

I followed philf's suggestion - I've checked all the heavy phase wires for obvious damage - nothing found.  Even though the wheel spins freely when completely disconnected from the controller, I decided to open up the motor just to be sure - no obvious damage to any of the connections.

Next I connected the controller back up, 1 wire at a time, to see which one is "putting the brakes on".  I tested each of the combinations - Green & Yellow, Green & Blue, Yellow & Blue.  When Green & Blue, or Green and Yellow are connected, problem.  When Blue and Yellow are connected - no problem. 

SO I focused on the green wire.  There seemed to be no damage to any wiring along the way to the controller, so I opened up the controller.  Inside I found a blown capacitor - I replaced it but no luck.  Next I disconnected the green wire from the circuit board, and the wheel ran freely again.  There was nothing else obviously blackened or shorting in the controller.  (although un-soldering and re-soldering is about the limit of my understanding of electronics so I may have missed something)

Does it sound like a problem with the controller? 

Ps. Not that I think it should make a difference, but until I had the problem, I was only using the pedal assist connection.  I have disconnected the throttle and the two brake cut offs.

Thanks

[benm]

pps.  I did check the post about the blinking light (as an indicator of what the error might be) inside the controller - http://goldenmotor.com/SMF/index.php?topic=279.0;prev_next=next#new - my controller looks very similiar to this one (the casing and the layout etc), but it doesn't have a light.  So no help there.

Thanks

Ben

[philf]

Eeek.

It sounds like you definitely have something either fried or shorted.  I observed the following in one of the controllers I received...  It was a short waiting to happen...



Dunno if you see anything similar in your controller, but if you're seeing blown parts, it's obvious that something isn't happy.  Quite likely, one or more of the FETs for that phase are shot.  If you're at least OK with a soldering iron, this isn't the end of the world - but it would be good to know what caused the issue.  No use installing new parts if there is an underlying problem that'll just put you back in the same place...

[Oliver]

Here is my guess at what happened.

I suspect that you had too much regenerative voltage going into the controller and you toasted the flyback diodes and or the FETs.  Explanation: In electronics that use FETs, it is very common to use flyback diodes to prevent ESD (Electrostatic Discharge) from burning up the semiconductor junctions.  In devices with inductive loads such as motor controllers, diodes (or similar semiconductor devices) are also used to shunt the back EMF (Electromotive Force) to prevent the FETs from being burned up due to excessive source to drain voltage.  Your 24V motor controller has FETs that are probably designed to operate up to a maximum voltage that is somewhere in the 30's--for example 35V max.  They probably have zener diodes across their drain to source junctions that are designed to conduct just below 35V--for example 34V.  These diodes don't turn on (conduct) until they see a voltage that is greater than 34V.  This is the first half of the equation.  BTW, FETs are the electronic switches used in motor controllers to connect the battery to the motor windings.

The second half of the equation is your motor.  Motors as I’m sure you are well aware convert electrical energy into mechanical energy.  The opposite is a generator, which converts mechanical energy into electrical.  It just so happens that motors and generators are really the same types of devices, just optimized for their specific task.  So, motors can and are used as generators.  Your motor on your bike, when not pulling you, is acting as a generator.  It has the ability to generate voltage as high as its max operating voltage (24V) when spinning at it its designed speed, and much higher when spinning much faster than its max designed speed.  I suspect that your 60kph to 80kph decent generated sufficient voltage to cause the controller’s zeners to conduct.  When zeners conduct, they are like a piece of wire with a very small voltage drop.  These diodes were shorting the windings of the motor, which produces a large braking force. 

However, the small diodes that are in the motor controller can only dissipate the braking energy for a short period of time before they overheat and burn up.  They can fail to a shorted state, as opposed to an open state but they can fail open too.  Long story short, (no pun intended ;-)), they may not have protected the FETs. 

You’re probably asking why the motor controller failed when it was turned off.  The answer is that turning off the motor controller does not disconnect the electronics from the motor; it only disconnects the battery from the controller so that that controller is no longer trying to commutate the motor and the FETs are not trying to conduct.  However, the motor is still generating voltage and the diodes and FETs are still electrically connected to the motor. 

For speedy descents without motor braking it is necessary to disconnect the motor for the controller.  Simply unplugging the motor from the controller is the fastest solution with the least amount of work but is not necessarily the most elegent.
 
That’s my guess.

[benm]

Thanks for your detailed reply.  I'm new to electronics and quite interested in how it all works. 

After reading the two posts I did the quick check suggested by philf - always good to check and recheck the easy stuff first!  Unfortunately still no luck.

So, I re-read Oliver's post a couple of times, and my layman's understanding is basically the motor fed back power which blew up the controller!  (I assume the FETs are in the controller and not in the motor) - I'm not sure what FETs or flyback diodes look like - but would I notice black burn marks if they were blown, by looking at the controller circuit board?  Assuming I shouldn't expect to see anything obvious, everything about the theoy fits...

...which means my next step is a new controller. 

Once I get a new controller, it sounds like a cut off switch between the controller and the motor is the go.  Which wires would be the ones feeding back the power?  ie.  which wires would need the cut off switch?  the 3 heavy phase wires?  Would I need to worry about the 5 smaller ones (I think these are the hall sensors)

Any tips on best/quickest place to get a controller (I'm in Sydney, Australia)?

Thanks

Ben

[e-lmer]

I do not recommend that you put a switch anywhere
between the speed controller and the hub motor.

I 'm not an expert either, but from what I have
read here, the regen switch should not ever
be switched while the bike is in motion. 
I suppose it belongs someplace other
than on the handlebars conveniently,
right beside the thumb throttle.

When you switch it off, the voltage that was
supposed to be charging the battery is now
suddenly shunted back into the controler.

I think that flyback diode is used to bleed
off high voltage across a relay or motor
because when you shut off current to a
coil the collapsing magnetic field tries
to jack up the voltage high enough
to keep the current flowing.

[benm]

I'm not sure that I follow.  Are you saying that some set ups have a regen switch (usually located on the handle bars)?  I'm really new to this, but I understand their are some controllers out there that can do regen.  They have additional wires that feedback power from the motor to the controller - is that right?

I don't think mine is the regen type.  It only has 3 thick wires and 5 thin wires between the motor and the controller.  Here's a diagram of the cut off switch I was proposing.  I'm really keen to understand this so I appreciate all feedback.

[Leslie]

My controllers I buy on the cheap will do exactly this when switched off when running the bike.

Not sure, but I think the controller needs to be on to limit current and or volts.  When regen is enabled it doesnt deactivate just switching the controller off unless  both the battery in and regen voltage = 0.  No controll over regan it just goes your fets too hard.

Not 100% sure..

[Oliver]

Ben, I'm sorry to have been out of the loop on this.  As far as disconnecting the motor wires, they are the three large power wires coming from the motor.  The sensor wires do not need to be disconnected.  The motor wires should never be disconnected while the bike is in motion.  The bike should be stopped, the power disconnected from the battery, and then the motor wires disconnected.  Once disconnected there should be no problem spinning the motor up to speed.  The motor will still generate voltage but there won't be any current.  It's kind of like the AC receptical in your home.  There is potential (voltage) there but because it is not connected to anything, no power flows from it.

As for controllers, I'd recommend the new "Cruise" controller from Golden Motor.  I don't have any experience yet as I just received mine and I have not tested it yet.  The reviews seem to be good.  Unfortunately, like most of the GM products, there is little information to make an informed decision.  I'll post something later as to my experience.