Well at least I know whats going on. When I first got the bike I had the controller fail in teh first 20 miles and got a new one from EVT (6 months later) but got one, swapped it out and it has run fine for 3 years and 2700 miles, until now. I ran it in hot weather and as fast as it could go for about 10 miles and the next day it started acting up, jumping, lurching and growling on starts. It sort of moves but this is exactly what it did right before the controller failed last time.
So my question is what controller should I put in it now?
Which one and from where?
I would suggest a 150A kelly or larger
wiring it in the first time is a little bit of a pain, but they are reliable in this application.
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
I am guessing I need a 72v one?
So either
http://kellycontroller.com/keb7245024v-72v150a45kwe-bike-brushless-controller-p-275.html
or
http://kellycontroller.com/keb7260024v-72v200a6kwe-bike-brushless-controller-p-278.html
I see in the options waterproofing? I don't think I have ever had mine "wet" but it might have condensation issues?
I also see optional high speed 70k or 100k What are those? I don't think I need the "stick shift" option, don't need to go backwards at 40 mph ;)
a 72v controller would work fine, and would probably give you more headroom to raise the voltage (and top speed).
What is your present voltage?
You just need the basic version, plus perhaps the waterproof option.
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
To be honest I don't know whats in there, even the replacement controller had no real information on it. The scooter has 5 38ah batteries in series so it runs in the 60's. One odd thing that got brought up a bunch of times initially was to turn off the main breaker during charging because it could damage the controller. I often wondered if it was a 48v controller being pushed to hard. I do charge the batteries individuality and used the stock charger in a pinch, the stock charger is really noisy (electrically), it makes the garage light flicker and they are not on the same circuit :)
Does the 150 amp or 200 amp really matter to the motor? Or is as long as the controller is above the rated motor it doesn't matter.
In other words would a 200 amp controller actually push more power in to the motor than the 150 amp version?
Yeh, I'd get a 72V controller :)
The 200A controller will allow you to apply more power to the motor at low speed, but the main reason you would get a slightly larger controller is there is more margin between the rated continuous current, and what you actually use.
That way on a hot day, you are less likely to encounter thermal shutdown.
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
Sounds good.
Thanks again, is the best place to get the controller from Kelly?
http://kellycontroller.com/keb7260024v-72v200a6kwe-bike-brushless-controller-p-278.html
sure is, they are the manufacturers and will give you the best price
I highly recommend getting a usb-serial cable if you don't already own one.
That way you can set the various limits with a computer.
If you don't have one, it will still work.
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
DANG
Swapped out the controller, same problem. I took a video of what it is doing. It starts out really rough, then smooths out eventually and once it's running smooth it runs fine, until it decides to quit again.
The good thing when it does work it is much more smooth and seems to have more power and a slightly faster top end. But it is so inconsistent I can't take it to work or in traffic. Is it the motor? I would swear the phasing is off but why or how could it run smooth sometimes with full power? I checked the big power lines and everything is tight.
Anybody know if this is how one or more marginal Hall sensors would behave...?
My Thunder has had the slightest tad of such behaviour at one very certain RPM. If I controlled the throttle very smoothly to keep the motor spinning at that RPM it ran roughly and made noises that were clearly induced by the coils, as letting the motor rev down currentless over that RPM ran smooth as silk. What caused it I have not been able to find out up till now. Kelly swears it has nothing to with the controller switching modes or anything along those lines. I haven't contacted the motor manufacturer yet. However, having a new stronger Kelly on board now I have not yet had the time to check if the anomaly still exists. But it in no way really infuluenced driving the scooter, it was just a small twitch during acceleration, hardly noticable.
But this motor clearly has some kind of mechanical problem with a clicking sound at every round, no matter if under current or not. That should be looked into also...
I would also diagnose severe commutation trouble due to partially malfunctioning hall sensor(s) or a loose connection somewhere. Clearly the wiring connections are all correct, else the motor would never even start spinning, and it DOES run smoothly towards the end of the video over the full RPM range. But the controller seems to be disoriented at times at how rotor and stator are alligned. The halls need checking, as well as their wiring. Someone with a 6-channel oscilloscope would be handy in this case...
My rides:
2017 Zero S ZF6.5 11kW, erider Thunder 5kW
The clicking on every revolution is in the brakes. I know that isn't good, but its the brakes and not the motor anyway.
On the plus side with the new Kelly controller I see a .... .... error when it is what I call grinding. Once it smooths out it is fine, the error is a "locked rotor error". I did check all the hall wires coming back and they all appear fine all the way to the motor. I have never opened up the motor/wheel but I suppose at this point whats it going to hurt?
It does this pretty much ever time I start it "cold" after a day or so, sometimes I have to spin the rear wheel and then hit the throttle to get it to start, but again once it's smoothed out it runs ok for a while.
While I am handy enough to swap a controller, I don't have access to an oscilloscope nor would I know what to look for.
The thing I really don't get is if it is a hall sensor why would it work sometimes and not others?
The brake disk looks as if it was running smoothly in your vid, but if it's just that it has nothing to do with your problem. If a Kelly Controller morses "motor locked rotor" it apparently cannot see any changes in the hall sensor signals and thus assumes the motor is no longer turning despite the controller sending current through the three motor phases. That would fully explain the rough going because the controller temporarily keeps polarity constant on two phases, though with only 90% of the previous current. This will brake the motor in two positions and accelerate it in only one.
I would say this is a very clear indication that there is some thermal problem with your hall sensors. Either they temporarily shut up when still cold and only start to send their signals after heating up above a certain threshold, or it is some contact problem that is thermally fixed when things get little warmer.
What you COULD maybe do even without an oscilloscope is to check one sensor out at a time while manually turning the cold motor - with power on. That would mean connecting one lead of a multimeter (in a range capable of measuring 5V DC) with pin 14 which is hall sensor ground on the Kellys, and the other lead successively with pins 11...13 which are the hall return signals. You should get regular 5V pulses while turning the wheel on every one of the three sensor, usually every 120°. If you like you could mark your motor or tire to help in getting the timing. If one of sensors sends no signal you have the culprit.
Now if you have never taken apart such a motor before I would not recommend doing it. Better to leave this to a seasoned mechanic of electric bikes, hoping you have one those around?
My rides:
2017 Zero S ZF6.5 11kW, erider Thunder 5kW
Well that makes sense. I would say when it starts it is running about 1/3 speed then jumps to 2/3 speed then finally smoothing out. I didn't realize the controller would send any power if the sensors were off or not returning a matching signal.
I know which three wires are the hall sensor returns to the controller so I will try as you suggested and see if one isn't signaling back while spinning the wheel.
The error this time was .... .. which is Hall sensor error, totally makes sense.
Took it apart again and spun the wheel. The yellow was the first one I checked. It was consistent as I turned the wheel, then the green, the same thing, slight off from the yellow and then the blue, nothing no matter where I had the wheel. So then I tried the throttle and ran rough, then it eventually smoothed out and checked the blue again, it then behaved just like the green and yellow.
So the blue hall sensor is damaged in some way or the wire leading from it. How do the wires leave the motor? Is it a brush thing or ? Again if open the motor will parts pop out all over the place? Or could it be something as simple as a bad brush / contact / broken loose wire?
Now what to do, I don't think there is anywhere near me that would have a clue what to do with it. I could box it up and send it somewhere? What would a new motor / wheel cost?
Thoughts?
Excellent, you have at least found the offending party! Next you need to find the offender himself: As you appear to have a brushless motor it is the stator which has the coil windings and hall sensors, meaning nothing needs to be brushed, it is just cables coming out of a hollow axle end. The spot where the cables actually come out is one of the possible places where wires could be damaged. I do not know how exactly things look on an EVT scooter, so it might help to have another EVT owner show up in this thread to help out with the details.
Generally speaking, nothing will pop OUT if you open the lid of the motor, rather things might pop IN, as running around the stator are very powerful permanent magnets that attract any magnetic material readily and you might never get it out of there again. What you might loose when you open the cover of the motor is a sealing. The lid should be completely sealed towards the wheel itself with some kind of sealant that you would have to reaply after opening.
When dealing with hall sensors it should not be necessary to completely take the motor apart, as that would be the hard part which should not be done by the casual do-it-yourselfer. You have a set of strong permanent magnets surrounding the stator with a very small airgap in between, and you may be aware of how the magnetic force increases exponentially when the airgap is reduced. Therefore you should have some specialized tooling for compeletly taking apart and especially re-inserting the stator into the rotor.
But then you shouldn't need to do that anyway, first check how the cables look where they come out of the axle. Sometimes an incorrectly fixed axle can come loose and turn in the swing arm, thus additionally damaging or even severing the cables...
My rides:
2017 Zero S ZF6.5 11kW, erider Thunder 5kW
I checked all the leads and everything looks good, no rubbing or fraying anywhere.
So can or should I send this to someone to get fixed or buy a new one, if so from who?
Indeed you are now faced with a more severe decision. Are you otherwise content with your ride? If yes, and/or if finances prevent ordering a new scooter, it would surely be the better idea to send at least the motor, but possibly the whole scooter to someone knowledgeable to check out why that hall sensor conks out at times and repair that. To whom I cannot say as I reside on another continent.
As to what ride could serve as a replacment I think you should find enough leads on VisforVoltage to point you in interesting directions :-)
My rides:
2017 Zero S ZF6.5 11kW, erider Thunder 5kW