Most Powerful Controller Available ?????

[John K]

G'day everyone,

I've been reading bits and pieces about the X5 motors being capable of outputting 5kW without any problems. Firstly I'd like to say that this is impressive. Secondly I'd like to know where I can find a controller that can do this reliably and continuously.

Are there any options other than the Crystalyte controllers that are capable? Or are the Crystalyte ones the best?

If I can get hold of something that isn't going to fry, I'll be putting a front and rear X5 into a custom built bike. I'm just bored with my X5 the way it is. 10kW should keep me interested for a little longer.

Can anyone help?

[BiGH]

i think you'd want an http://www.ebikes.ca crystalyte 4110 based 50amp controller.

thats going to get you some rediculous speeds on the right 5 series motor

[safe]

Power Calculations...

10kW = 10,000 watts = 13.3 hp.

The PMG 132:

72 V Performance
Peak efficiency: 91 %
Peak power @ 110 A: 9.5 hp
Peak power @ 200 A: 18 hp



...so you might be better off when you get up to that level to start thinking in motorcycle terms. Front wheel drive at that horsepower is going to produce some nasty wheelspin. Also, with the PMG 132 and a chain you can change the gearing to suit whatever conditions you are in. Go up a tooth on the front and you go faster, but climb less well. Go down a tooth and you climb better, but have a lower top speed. Aerodynamics might also make you choose gearing that matches your top speed better.

If you're going to go "all the way" you might as well do it right...

You would also shave 25lbs off your bike. (one less hub motor's weight)

[John K]

Safe,

If you're patient I'll have my other project finished in a few months. I'm running 2 of the PMG 132 motors on 84V. I'm used to 1000cc ICE bikes, but I thik this one will be fun still.

I've already built one using one of the PMG motors but it was just a little bit under powered and the chain drive makes it all too noisy.

I know the hub motors are heavier and put out a lot less power but I dig the silence.

[John K]

Big H,

Thanks for that. I'm just wondering if many poeple ave had experience with these yet and if there's been any overheating etc etc.

I've blown at least 5 of the X5 controllers by running them on 60V and 40A so I need something more robust than what I've already used.

[BiGH]

i'm running the 4310 based 35a controller atm - at 72v and up to 87 straight off the charger, i run it bloody hard, i've had it enclosed with no air and never gets more than mildly warm, the motor though does! (408)

did your controllers have the screws on the bottom of the case? or did your case have fins on the side with screws there?

do you know what mosfets your old controller had?

using the 4310 or 4110 based are the most reliable, and their internal resistance is low meaning they won't heat up as fast. I personaly wouldn't bother with controllers that ahve anything else.

[fechter]

Do the 4110 mod on the old controller and it will handle around 100 amps at 100v with a little extra beefing up on the buss bars. Otherwise you could cook the motor with one of these:

http://www.sevcon.com/pages/espac.html

There really aren't many controller on the market in this power range.
(BTW, I hate to think how much that Sevcon costs)

[John K]

Big H,

I didn't have any luck with the older type that had the screws on the bottom, but the one that I have now which has the cooling fins seems to be a lot more reliable.

I have no idea which mosfets the old one had because I read through fechters reverse engineering of the controllers and figured that for the amount of time that it would take me to pull mine apart, find the problem and change the mosfets I could just go to work, earn the money and buy a new one..... Sometimes I'm lazy.... sometimes I'm even lazier.

[John K]

Fechter,

Thanks for the tip. I'll get on to Ebikes.ca as soon as I get the chance and buy one of those suckers. Beefing up the bus bars seems easy enough even for me.

That sevcon unit looks pretty serious and I wouldn't mind cooking a motor just for fun.

Something to think about I guess.

[BiGH]

Big H,

I didn't have any luck with the older type that had the screws on the bottom, but the one that I have now which has the cooling fins seems to be a lot more reliable.

I have no idea which mosfets the old one had because I read through fechters reverse engineering of the controllers and figured that for the amount of time that it would take me to pull mine apart, find the problem and change the mosfets I could just go to work, earn the money and buy a new one..... Sometimes I'm lazy.... sometimes I'm even lazier.

not to worry - the ones with the screws on the bottom have the crappy fets.

glad to hear your current one is going well.

[John K]

Big H,

My current one is going well, but I did manage to kill one of the same model. Not sure why.... may have just been a dodgy one from the factory.

[John K]

Fecther,

Going by what you said, the motor has the potential to put out 10kW.

Just wondering how you came to that conclusion....was it though physical testing or through calculation?

I'm assuming that the motor would then need some sort of ventialtion....and I'm assuming that cross drilling the side covers would be the go.

Either way, I'd imagine there'd be a lot of heat......are the adhesives and other bits and pieces inside these motors up to that kind of punishment... I mean theoretically if I was going to climb a mountain at full throttle...would the motor cope?

Those phase wires even seem a little small.... I'm assuming those bits would need upgrading too.... yes?

[Ypedal]

Those phase wires even seem a little small.... I'm assuming those bits would need upgrading too.... yes?

I think that's going to be the limiting factor. pushing 5000w thru the stock coils means you better get up to speed within so much time.. or else Melt-Down !!

The magnets being on the outer hub spinning, will stay put, the bearings could be upgraded on the axle... but those should also hold up fairly well..

Avoid the dreaded dead-stop lock and get it on video please !!!!

[fechter]

Fecther,

Going by what you said, the motor has the potential to put out 10kW.

Just wondering how you came to that conclusion....was it though physical testing or through calculation?

Not really.
The controller could do 10kw for a time, but I'm not so sure about the motor. From user reports, the X5 motors appear to handle up to 5kw depending on air temperature Cooling holes would help and with enough air, you can about double the continuous rating of most motors. I would put the continuous output rating closer to 2.5kw in general.

You might get it to pull 10kw on acceleration, but I doubt you could get the motor to pull that much at any kind of speed, since the back EMF will limit how much current the motor will take. If you used a 5302 on a large wheel, you might get it to draw more, but the efficiency would suffer and the thing would be likely to self-destruct.

With brushless motors, it's next to impossible to run two motors off one controller, so the limiting factor will be motor heating if you have enough batteries.

There are BIGGER motors out there. The Mars motor (brushless Etek) is one example. Some of the newer full size scooters have bigger motors.

[John K]

Fecther,

[/quote]Some of the newer full size scooters have bigger motors.

Are these controllers possibley more robust than the Crystalyte units? or am I better off to stick with the Crystalyte?

Back to the subject of heating..... what kind of current can those phase wires handle?? Do you use the RMS rule or something to calculate their capacity?

Purely out of naivety and possibly stupidity..... (motor and battery capabilities aside) what would happen if I tried to run 2 controllers in to one motor. Is it possible?

If the hall sensors are delivering the same signal to each controller, would this mean that the output from the controllers would be in phase, thus doubling the power that the motor receives? or would there be some sort of issue with back emf or something??

[John K]

That wasn't supposed to be a quote BTW

[fechter]

The phase wires can probably handle more than the motor windings, since they are bigger and have better cooling. The motor windings will probably fry first. It's still not a bad idea to oversize the phase wires to minimize losses, but you can't make them much larger where they go through the axle. Using wire with a higher temperature rating would be good through the axle.

A Sevcon controller will definitely be more robust than a Crystalyte. It's probably about 5 times more expensive too. If you can afford one, I don't think you'll be sorry.

Two controllers in parallel has been done before. It's real prone to blowing up though. If one switches a few nanoseconds before the other one, you can get shoot through (instant destruction). I think you'd have to tie the inputs to the gate drivers together to get the timing close enough.