BMC V3 -- Infineon Build

[Muad'dib]

I'm bit surprised that it doesn't have the starting torque and behave exactly like a gearless motor.

I'm 2nd guy with the 1000W DD motor and I've been waiting for a geared version for more torque. By the way, how is the noise? Does motor have any gear noise? The DD is very quite, so that would be an indicator.

Oh, I'm running 48v A123 (with 40amp Crystalyte controller). Top speed is over 35mph but I stop there as its gets stupid scary.

Is it possible the starting torque thing is the controller's not having soft-start? Or does that only affect the first second or two? The motor makes some noise, but it is not loud at all. I took the chain off, and turning the bike upside down, the wheel spins freely when I turn it counter-clockwise whereas when I turn it clockwise there is resistance and a faint noise that I can imagine is the gears turning. What would I be feeling/hearing with a direct drive motor?

 

[leamcorp]

Is it possible the starting torque thing is the controller's not having soft-start? Or does that only affect the first second or two? The motor makes some noise, but it is not loud at all. I took the chain off, and turning the bike upside down, the wheel spins freely when I turn it counter-clockwise whereas when I turn it clockwise there is resistance and a faint noise that I can imagine is the gears turning. What would I be feeling/hearing with a direct drive motor?

Perhaps. I've originally started with Infinion 48v/25amp controller but wasn't satisfied so I've changed it to Clyte 24-72v 35amp controller. It took me many many tries to get the wiring combination right... so it could be a possiblity. Why did you get a different controller? It looks like your vendor have a "tested" controller.

As for resistance - on forward movement direction (with power off), it only turn once when spinned by hand. On reverse (non forward movement), it turns freely. It sounds like your wheel have same resistance on forward movement. That doesn't seem to make sense as my Bafang geared wheel will turn quite a long time in forward direction. I'm thinking... maybe its a same wheel?

I guess only way we'll know is to open the wheel ???

Edit - Icyclist, you have both 1000w DD and 600w geared. Does your 600w freewheel?

 

[Muad'dib]

I would like to settle it and open the motor. Any tips on how to open it? There are nine screws, but when I took those out it didn't seem to allow anything to budge. I need a special tool to get the chain freewheel back off right? It arrives without the chain freewheel attached, but even then it wasn't clear how to open the motor.

 

[Muad'dib]

Oh, and the reason I didn't buy a controller from this supplier is that they are a lot more expensive than any other supplier i've looked at. The Infineons have a good reputation and work with BMC hubs so I thought that was a good choice. I am also getting a Crstalyte 48v48amp 18 fet controller from Maxwell as part of the group buy.

 

[snowranger]

I would like to settle it and open the motor. Any tips on how to open it? There are nine screws, but when I took those out it didn't seem to allow anything to budge. I need a special tool to get the chain freewheel back off right? It arrives without the chain freewheel attached, but even then it wasn't clear how to open the motor.

You need to get a knife or something to pry the cover off. It's probably some adhesive or sealant. No need to remove the freewheel.

 

[leamcorp]

You'll need a gear puller. You don't really need it but I would get one as those magnets are really strong. You will need to pry open the motor case - its on there really tight and you'll scratch/gouge some surface of the hub as you try to open one. With that...

You'll most likely void the warrantee, if there's one. Only way I'll open mine is if its broken. You paid lot of money for that motor. Also, you may want to ask the vendor if their controller is matched/tested. Sometimes its worth it to have someone else match the phase/hall connectors.

We all like to learn more about each motor but for what you paid, its not really a experimental motor - just something to think about. Maybe one of the vendor could open one up?

 

[snowranger]

You only need a gear puller to pull the clutch assemby and gears from the shaft. I've done this before. You can entirely remove the inside assembly after removing that one outside cover.

To find out whether you have a geared motor, just turn the axle one way and the other. It should be easy to spin it one of the ways.

 

[Muad'dib]

You need to get a knife or something to pry the cover off. It's probably some adhesive or sealant. No need to remove the freewheel.

Got it, came off easy in fact, not a scratch. It is indeed GEARED. See photos:

 

[Sacman]

Nice job... and thanks for the pics.
I recently got a BMC motor and I got the cover off too check wiring
but didn't get a chance to pull the rest out completely to see the gears.

 

[grandmasterE]

Thanks for the work of taking it apart and taking pictures. So what's the conclusion then? The 1000W V3 hubmotor is a bit of a dog for torque/acceleration, more built for speed? Maybe built more to compete with a motor in the style of a 5303 than a 5305 (but probably not able to handle quite the same power)?

When are we going to get something with something geared towards acceleration? I think it was Drunkskunk that said it in another thread, but I want the ebike equivalent of a rockcrawker for torque!

E.

 

[Microbatman]

This is a real nice build tread here.

Motor wise I agree about accelleration.

Give me 0-30 acceleration with less weight and same RELIABILITY an X5.

I think the next mfg that comes up with this will really have something.

 

[Muad'dib]

This is a real nice build tread here.

Motor wise I agree about accelleration.

Give me 0-30 acceleration with less weight and same RELIABILITY an X5.

I think the next mfg that comes up with this will really have something.

I guess it remains to be seen how this V3 geared motor performs in the low speeds with a powerful Crystalyte controller. In a couple of weeks I should be able to report on that. The 600w Torque model was also very tempting to me, and probably is a great choice for 0-30 acceleration. The general reliability of the BMC motors is good, so I hear, and supposedly these gears are made of some kind of stronger composite plastic.

 

[snowranger]

That gear cluster looks exactly like the one BMC sent me, which is what is also on the 600 watt version. The difference I see in your pics is the cut-outs to see the windings from the gear side. Better cooling I presume.

 

[leamcorp]

Thanks for the pics.

I went on a ride yesterday to do some testing and my top speed came up to 44-45mph before running out of road. From 0 to 12mph, I had to pedal as motor struggles and makes an awful noise. From 12 to 20mph, its decent for acceleration but m most of the road distance is taken at this point. From 20mph to xxx it speeds up fast. With this motor, I pass lot of cars on local roads.

I've also noticed that motor is not quiet as I thought. Its loud enough for people to noticed as I'm passing by (not just looking but looking at the wheel). This motor does suck lot of juice - at max throttle, it draws 44+amp. Only thing stopping that is the controller, which also get very hot when pushed. For average ride, the controller is warm. It doesn't help that its in a bag with no circulation.

 

[lcyclist]

Thanks for the pics.

I went on a ride yesterday to do some testing and my top speed came up to 44-45mph before running out of road. From 0 to 12mph, I had to pedal as motor struggles and makes an awful noise. From 12 to 20mph, its decent for acceleration but m most of the road distance is taken at this point. From 20mph to xxx it speeds up fast. With this motor, I pass lot of cars on local roads.

I've also noticed that motor is not quiet as I thought. Its loud enough for people to noticed as I'm passing by (not just looking but looking at the wheel). This motor does suck lot of juice - at max throttle, it draws 44+amp. Only thing stopping that is the controller, which also get very hot when pushed. For average ride, the controller is warm. It doesn't help that its in a bag with no circulation.

Leamcorp,

Is this the Direct Drive 1000W motor or the new geared 1000W?
I remembered you weren't satisfied by the claims by the seller regarding the speed potential of the direct drive motor.
What change on your setup? Just curious as I own the other 1000W direct drive and it behaves exactly as you described.

lcyclist.

 

[leamcorp]

Thanks for the pics.

I went on a ride yesterday to do some testing and my top speed came up to 44-45mph before running out of road. From 0 to 12mph, I had to pedal as motor struggles and makes an awful noise. From 12 to 20mph, its decent for acceleration but m most of the road distance is taken at this point. From 20mph to xxx it speeds up fast. With this motor, I pass lot of cars on local roads.

I've also noticed that motor is not quiet as I thought. Its loud enough for people to noticed as I'm passing by (not just looking but looking at the wheel). This motor does suck lot of juice - at max throttle, it draws 44+amp. Only thing stopping that is the controller, which also get very hot when pushed. For average ride, the controller is warm. It doesn't help that its in a bag with no circulation.

Leamcorp,

Is this the Direct Drive 1000W motor or the new geared 1000W?
I remembered you weren't satisfied by the claims by the seller regarding the speed potential of the direct drive motor.
What change on your setup? Just curious as I own the other 1000W direct drive and it behaves exactly as you described.

lcyclist.

A lot has changed. I went from 48v Mr. Lau to A123 48v 20ah, new bike with slicks (this helped a lot), new Clyte 35analog/50amp digital (Infinion was dog for me for some reason), etc. What perturb me then was vendors claim about high speed. Sure it could be done but at very high cost - on top of the motor, which was a small part. At average setup (ping/20amp controller, etc) won't get you anywhere near the speed I'm seeing now. In fact its efficiency is much lower than other motors, if they could get up that high.

I'm now questioning if our wheels are direct drive. I'm thinking its same motor 1000w geared motor. I may open up the motor during the holiday. It sounds like BMC isn't coming out with DD model after all?

 

[lcyclist]

I'm now questioning if our wheels are direct drive. I'm thinking its same motor 1000w geared motor. I may open up the motor during the holiday. It sounds like BMC isn't coming out with DD model after all?

Don't need to open.
The BMC 1000W I have is definitely a direct drive. How do I know? It spins backwards in certain combination of the phase wires; the 600W motor I have will not spin backwards--you can hear it spin internally but the clutch prevents it from spinning backwards.

Although I have not taken mine above 40 mph, it is very efficient around 20-25 mph (10-15 Wh/mile), similar to my geared 600W and if I pedal assist, this goes even less 5-9 Wh/mile, which allows me to use 6AH, 36V battery for 20 miles. This is, of course, on flat ground.

I expect the efficiency to drop at high speeds (due to air resistance) since the power required is proportional to cube of the speed.

Anyway, it is an impressive for a motor half the size and weigh of a big Crystalyte motor.

 

[Muad'dib]

I got the new Ecrazyman controller, hooked it up, and nothing happened. For the first try I did the wiring as with the first controller, but with a 10amp fuse in the battery line. The fuse didn't blow, but turning the throttle gave no response at all. We are kind of stumped. Opened up both the old and the new controller and the wiring colors connect to the exact same spots on the board in the same order. Any ideas? Is there a way to localize what is happening?

Here's some data, I don't know if this is normal or not:

Throttle wires measured with ohm-meter (it is a half-twist throttle from ebikes.ca):

1.8kohm measured between two wires (red and yellow) and no continuity between any of the other 3 wires. In no case did turning the throttle change the resistance.

One more thing to report back on is that on closer inspection of the first controller that "burned", it seems that one of the solder lines on the bottom of the board, just beneath the fuses, melted, and that drop of solder connected with the chassis shorting something else.

 

[Mike1]

My crazyman controller did exactly the same thing in the same spot on the board by the look of your photo. The solder melted at the bottom end of the shunt. The blob then shorted to the case and then the track melted at the weakest point which was at the other end. The tracks are reinforced to a certain extent but only on the long sections, the corners have been missed, this is where it melted. Its worth adding some extra copper to the main traces once you get it working.

Those throttle measurements sound about right. You normally see about 2K across the 0 and 5v connections. You really need to check the voltages on the throttle at closed and fully open. You need to open the controller and check the values of the regulator power resistors, that will help confirm that it is configured for the correct voltage. Which board did you get EB812 or EB212?

 

[Muad'dib]

Thanks for the tips. What was the context of your controller failure? Was it with a BMC motor? Were you pushing the limits of the controller? I'm still wondering if it was because of the weird effects of the new BMC or if it is just a controller flaw. The thing about the throttle is that the resistance doesn't change at all between no-throttle and WOT. That's not right is it? I'll have to wait till the weekend to check the version number.

 

[Mike1]

The throttle isn't a resistor, its a hall effect device. The 2k you measure is a coil that creates a magnetic field. When you twist the throttle a sensor moves in that field and gives out a voltage between 1v and 4.3V (approx) depending on position. You can only really test it with it powered up.

Yes I was pushing the controller past its designed limits but I expected the current limit to cut in before it melted.

 

[leamcorp]

Man I'm really watching this and the other 1000w thread as this will be my next purchase.

If you need a another Infinion controller quickly, I have couple of 48v 20amp's that I don't use. You could borrow that and a matching throttle is you wish (just don't send me a burned one back )

 

[Muad'dib]

Leamcorp,

Wow, thanks. There shouldn't be anything wrong with this 2nd controller, I mean it looks new on the inside and I assume Keywin is trying to keep customers happy. Any day now I should be getting a Crystalyte 48v48amp 18 fet controller. I'm going to try hooking that up this weekend and if it works, then it works and I'll be happy. Thing is, this controller coming is "digital" and supposedly not compatible with BMC. But Hi-power cycles sells a digital Clyte that they say works with BMC: http://www.hi-powercycles.com/product.sc?categoryId=9&productId=46. In talking to my dad, a seasoned electronics and programming guy, he couldn't understand why digital would be a problem, since the ecrazyman controller runs on a microprocessor and is therefore "digital". Is he right?

 

[BVH]

I'm watching the two threads also. My new bike with BMC V3 1000 watt is supposed to be ready for pick up this week. It's equipped with a Crystalite 48 -72 V, 40 amp controller. Not sure if digital or analog. I'll be using a 48 V, 30 AH Lipo pack.

 

[grandmasterE]

Leamcorp,

Wow, thanks. There shouldn't be anything wrong with this 2nd controller, I mean it looks new on the inside and I assume Keywin is trying to keep customers happy. Any day now I should be getting a Crystalyte 48v48amp 18 fet controller. I'm going to try hooking that up this weekend and if it works, then it works and I'll be happy. Thing is, this controller coming is "digital" and supposedly not compatible with BMC. But Hi-power cycles sells a digital Clyte that they say works with BMC: http://www.hi-powercycles.com/product.sc?categoryId=9&productId=46. In talking to my dad, a seasoned electronics and programming guy, he couldn't understand why digital would be a problem, since the ecrazyman controller runs on a microprocessor and is therefore "digital". Is he right?

If I've read the other threads correctly, I think the difference in the digital and analog controllers was the voltage supplied to the hall sensors. If I remember correctly, the digital controller wasn't able to raise the hall state quick enough at the high frequencies of the geared motor, but I think someone (fechter?) was using pull up resistors to help speed up the state change? I'll see if I can dig it up....

E.