GoldenMotor.com Forum
General Category => General Discussions => Topic started by: pilot on August 22, 2011, 03:23:35 PM
-
Hi
3 months ago I bought a bike conversion kit for my 28" bike from the European reseller of GM.
I choose the front wheel Magic Pie for the semplicity of installation.
During the first ride with the MP, while I tried to start on a hill, the front fork was broken. I said, ok
it doesn't matter because it was a light aluminium fork. I replaced it with a strong steel fork and everything
was ok until 4 weeks ago.
I was riding as usual my bicycle on a main road of my city at about 25Km/h when suddenly the front wheel
was blocked , I flew over the bicycle and I landed on the ground. I was fortunate enough and at that moment no car was
behind me and as I always wear the helmet I didn't have any head injury although I beat on the ground.
But I broke my elbow and I will need surgery operation for full recovery.
Now during my recovery I decided, with the help of my son, try to understand what happened.
I opened the MP and I realized that the electronics of the controller were carbonized and most probably because
of the heat produced the motor was blocked considering that the spacing between the rotor and the stator is less than 1mm.
I use 28V LiPo battery thus it is not a problem of high voltage. Anyway, electronics failure can happen to any device
but if in this case the front wheel is blocked then it becomes a very dangerous device.
I would like to warn all users to avoid using the MP front wheel drive, because as it is direct drive, if it is blocked
you are down. On the geared version if the motor is blocked, the wheel still turns without problem.
I think also that GM should warn users about this dangerous possible situation and even better stop selling
front wheel MPs.
-
Actually your problem was more than likely over current.
The 25V motor is 250Watts.
25V/250W = 10 Amps.
If you draw more than 10A you are overheating
the motor. When you do, the controller tries to
push more current through (since you can't increase
the voltage.)
I believe that prior to this you must have heard noise
and probably found the motor to be hot to the touch
before this catastrophic failure.
A good rule of thumb is to run your motor at 80% of
the rated value, but even at full rating had you put
a 10A fuse in your system you would have blown the
fuse instead of doing an endo.
I purchased a resettable circuit breaker for my bike.
1000W/48V is just over 20A.
If you do get a circuit breaker be sure to check the
DC current rating, not just the 250V AC rating.
-
GM claims that the peak power of MP is 1000Watts at 48V which means that the maximum
current is 1000/48=20.8333A. This should be the maximum current in any situation and voltage.
If at 24V the power is only 250Watts is not because the current capability of the output mosfets
suddenly decreases but because of the magnets and coils.
But most important is that every good designed controller should have an efficient current limiter which
limits the current in situations of overloading. Even if overloaded doesn't mean that you
must brake your head. If an external circuit breaker is necessay, it must be
provided by GM and not improvised by the users.
As I wrote in the previous post I don't care so much about an electronics
failure, which can always happen, but about the very dangerous situation of using such
a device on a vehicle. Now I understand why many GM resellers in Europe don't sell
MP front wheel conversion kits.
-
I'm not really sure anyone should use that much power on the front wheel...lots of stress on the forks, in the wet there is lots of reduced traction, and in the case of dd hubs, any phase short will result in a complete lockup as described here...st least if its a rear, you will retain some control when stuff goes wrong
-
I am quite happy with my front wheel Magic Pie.
It is a kit. I expect to take responsibility for
using it safety and within the rated specs.
You don't expect components to limit the input signal,
You expect the applications engineer to use the device
appropriate to the application.
If you are trying to draw more power than the rated value
the motor makes noise and heats up, if you keep doing that
then you are just not paying attention.
Sorry for the rant.
I guess I resent the kind of people that make it necessary for
lawn mower manufacturers to tell you that there are spinning
knives under it and gas can makers to tell you gasoline may be
harmful or fatal if swallowed.
-
not a rant, just an exchange of opinion lol.....I guess that's what the forum is for....real world applications
-
What exactly means "responsability and within rated specs"? At 28V, much less than the
maximum power declared, is it out of the limits drive? When you drive a vehicle, on a flat
road, at moderate speed, how do you know which the limits are?
I also was really happy with my bike conversion until the accident. The purpose
of my message was not to convince people not using the kit, but warn them about
the possibly harmful front wheel direct drive conversion and also ask GM to inform
its clients about this danger.
If you make a search on the web about ready make e-bikes, you will see that all
models are back or geared front wheel.
-
I think you have a fundamental misunderstanding of the way
electric motors work.
The applied voltage means nothing.
The motor will attempt to supply enough current to feed the
power you are requesting from the motor.
It doesn't matter what voltage you are supplying.
It's the load you put on the motor.
Think of it like a car. If you have an engine and drop it into
high gear and drop the clutch, it will stall. You have to apply
the power that it takes to do the work to move the vehicle.
If you attempt to do the work per time that requires
more than 1000 Watts in an electric motor, then the
maximum current possible will flow through the windings
of the motor and they will get hot, melt, and do damage.
-
Sorry, but considering that I am an electronics engineer, working as a researcher in robotics I do have
a very deep knowledge how motors, brushed, brushless, step etc, work.
The power of a motor depends on the voltage applied and the current is the necessary requested by the load applied.
That's why voltage is constant and current changes according to the load.
Yes, when you apply lower voltage, the current increases, but the power remains the same(the requested) and most
important the SOA(safe operating area) of the output mosfets remains within limits.
If those limits are exceeded, it is a good designed current limiter that must bound within SOA the operation of the mosfets.
Using your example of the car, do you consider safe a car that when overloaded explodes after some time of hard operation?
All cars have a temperature gauge and in case of overheating shut down the engine or in case of high gear simlpy stop working.
But they don't block the front wheels during the ride.
Again I repeat that the purpose of my messages is not to argue on motors operation, but
warn users about a non safe product.
-
I have used dual drive Magic Pie motors on two conversions and one conversion front wheel drive Magic Pie motor only. I have never had any problems with a motor being "blocked". Im not even sure what you mean by the term blocked. Do you mean seized? If it is an internal MP2 it would have shut down for heat long before anything like this could have happined. Also after over two years as a dealer and selling hundreds of front wheel Magic Pies I have not heard of anyone with this situation. I assume it is a unique problem that only you have encountered. Also if it were a common problem I am sure it would show up on a rear wheel drive as wheel. So saying that there have been thousands of Magic Pies sold and no other report of this problem.
The only way I can think of this problem happening is if the side covers were not mounted properly and the covers slipped causing a rubbing inside. If it were simply from the heat it would have shut down and if that failed there would certainly be some warning of a squealing or rubbing sound.
Gary
-
Dear Gary
I attach a photo of the wheel after the accident. You can see that the controller is
burned and a part of it is missing. I suppose that with the high temperature
produced the rotor was dilated and because of the very small spacing, it touched the
stator with catastrofic result. Maybe it happened the opposite, because of the rubbing, due to
covers slip, the controller was destroyed. In any case I insist saying that a front wheel direct
drive kit is dangerous. During a lockup of the rear wheel at least you have some control and you
don't fly over the bicycle. By the way this has been the hardest landing of my pilot carrer and as
a friend of mine said "you better had a hard landing with the bicycle than with the airplane".
-
Wow, Thats some picture!
Looks like the controller was the last thing to fail or the bike would have stopped running alot sooner. Still out of the thousands sold I have not heard of a motor seize up before. Centainly someone would have posted it in here if it had. It looks like you have been the unlucky one. I wonder if a new controller would get it going again or if its all toast?
Gary
-
Looks like the controller was the last thing to fail or the bike would have stopped running alot sooner.
I think the controller was the first thing to fail, causing a huge current to be fed directly into the windings, causing the wheel to lock up due to the incredibly strong magnetic force that would be generated by such high current.
Had the bike been powered by a GM battery, this would not have happened due to the current limiting function of the BMS built into the battery. Because a LiPo battery pack was being used, they are perfectly capable of supplying huge amounts of current for short periods. My 25.9V 10Ah LiPo battery pack is rated at 20-30C, which is at least 300Amps! Fully charged it's around 29V, which means it could supply at least 8,700Watts of power for a short period if the resistance in the circuit was low enough.
With 3,000Watts of power, my Pie will almost wheelie off the throttle on the flat, so I would have thought that 8,700 Watts of power could stop the wheel from about 25Km/h in a very short time.
The excessively high current has caused the thick tracks on the controller board to blow, just like a large fuse, causing the damage visible in the photo.
I'm guessing that the blackening of the windings is due to the huge spark that would have been caused as the circuit board tracks simply failed under the excessive load.
I suspect that the copper windings of the motor will still be OK, because the current would have stopped before the windings became hot enough to burn through the insulation, but looking at the damage to the controller, it's difficult to say whether the hall sensors will have survived the ordeal.
Pilot, do you have a photo showing the rest of the stator windings, I'm curious as to whether they are blackened as much as the section immediately next to the burnt controller. Do you have the reverse function enabled? If so, is it possible that it might have been accidentally pressed instead of the cruise button prior to the incident?
I don't really know what has caused the controller to initially fail, but I'm pretty sure that a 30Amp fuse (or circuit breaker) on the power supply would have prevented the unfortunate incident from occurring.
Alan
-
Reverse should have no effect. At least in my situation. If I hit the reverse button by mistake when riding it just stops the motor and I have to stop and then go in reverse just for a second before it will go forward again.
Gary
-
Dear Alan
Indeed the windings are in good condition. Maybe it happened as you describe, although
I remember that just before the accident for 1-2 secs I had a very strong vibration on the
front wheel and then full lock up. The output configuration of the controller is full bridge,
which means that one burst of current of some ms could be possible, but 1-2 secs is very
long time for huge over current for the mosfets. And I would like to make clear that I was
riding on a flat road, slightly descending and with strong support from the pedals.
I have installed a power meter on the bicycle and although at that moment I didn't
watch it, I suppose from past readings that current was between 3 to 5A.
It wasn't a problem of overcurrent or something similar. Just a simple electronics
failure which in this configuration could be extremely dangerous.
-
It's difficult to say what actually caused the failure, but I suspect that either two (or more) of the six MOSFETs have somehow short-circuited internally, or the chip that controls them has caused them to stay permanently closed simultaneously.
If one was on the positive switching side A+ and another on the negative switching side C-, this would result in the full battery voltage being put directly through two of the three sets of windings. This would produce enough electromagnetic force to quickly stop the wheel, especially if B+ or B- was also switched.
(http://www.arhservices.co.uk/GoldenMotor/Mosfetcircuit.JPG)
Without any form of battery protection (Fuse, trip switch or BMS etc.), if two MOSFETs on the same phase wire were somehow switched (or short-circuited) at the same time (for example A+ and A-), this would instantly produce a complete short-circuit directly across the battery wires, drawing maximum current from the LiPo pack, which would instantly burn out the tracks on the circuit board, presumably accompanied by a very loud bang from the controller. :o
I don't know if a faulty hall sensor could cause this sort of failure, but I wouldn't completely rule out the possibility.
Reverse should have no effect. At least in my situation. If I hit the reverse button by mistake when riding it just stops the motor and I have to stop and then go in reverse just for a second before it will go forward again.
If reverse was enabled and the "Immediate" option had been selected from within the programmable options, I would have thought that there is a strong chance that some components could be put under tremendous strain if reverse was accidentally selected at speed with the throttle open. It would be like rolling backwards downhill at speed and then trying to stop by opening the throttle instead of using your brakes.
I would strongly recommend that anyone using LiPo packs, or similar packs without any form of inbuilt current regulation, should include either a simple fuse, or some form of current sensed circuit breaking device, to protect the battery and controller from severe damage, should a short-circuit fault occur, regardless of how it may have been caused!
Once a short-circuit has occurred, if your controller doesn't blow itself up (creating an open circuit), there's a very real chance that your LiPo pack can actually explode into flames! :o
(http://www.mikrocontroller.net/attachment/57020/lipobrennt.jpg) (http://www.mikrocontroller.net/attachment/57020/lipobrennt.jpg)
Click the picture to see it full size.
Please notice that the above pictures are of one single cell, so just imagine what a whole pack exploding would be like. :o :o :o :o :o :o :o
Now ask yourself which you would prefer; a blown fuse? or a blown controller and/or an exploding battery pack? The choice is yours!
Alan
-
Mr Imer
You say applied voltage doesn't matter? I'll be damned, I've been doing it wrong for 40 years.
I too had a controller fire. No one was hurt, the motor simply burned up. Obviously a poor design.
-
Hi all
maybe a good quailty RCCD with DC fault sensing could have all this ! (and no magic smoke )
Anyone ever though to use one ? on this application ?
-
Reverse was not connected and it was impossible to be selected by accident or on purpose.
I insist saying that it was just a
simple and clear electronics failure that may happen. I am wondering why GM doesn't use a simple
mechanical unlock mechanism for the direct drive MP. In the case of a motor lock up it could release
the wheel, save the riding guy and maybe a lot of money from GM.
I am not sure even if a fuse could be a solution for this problem. For sure it is a good protection
for the batteries but if a current burst caused the lock up of the wheel then I don't know if the
fuse could be enough fast to avoid it. I suppose that first a burst caused an initial lock up and
then the sides of the motor were moved and caused the permanent lock up.
-
Alan;
in response to your theory of possible mosfet malfunction, would this not result in phase imbalance wich should trip an overload or is that only found on large industrial 3 phase motors. I have personally experienced phase imbalance on a 500hp and a 2000hp motor and they do begin to shake if the trip is slow to engage. Given the outcome for this poor gentleman should such a trip not be done by the controller? After all "Safety First"
Just a thought;
Andy
-
Not to push in your post Alan :D But in that case, there would be a serious amount of outstanding claims to various manufacturers.....
Like if you had a voltage spike from your power company/lightning etc. and your TV was runied, I don't think you would blame the TV manufacturer for not having in-built MOV's/Surge protection etc.
Same for a Car amplifier, if the MOSFETS short inside due to an issue with the input, speakers etc - and you did not have a fuse connected with your amplifer from the battery then you can damage your engine ECU etc. and I don't think the Car Audio manufacturer would be to blame (although many have in-built fuses anyway to protect this)
But you can get the gist... I'm not running a BMS with one of my GM batteries, and I have an in-line fuse connected to my wheel. It was around $5 with 10AWG wire I just soldered in-line with the battery. My main reason for this was to protect the battery, not so much the wheel but both are equally important.
Sorry if my post sounds like a bummer!
:D peace out