New 1000w Geared hub MXUS XFC19

[Tommm]

Any updates on this motor? Was thinking of picking one up since the MAC motors are out of stock for an unknown amount of time at this point. The seller ChamRider on AliExpress has a kit with this motor, a 40A sinewave controller, and a 48v 24ah 13s5p hailong pack with Chinese 21700 cells for about $550 shipped and I'm tempted to pull the trigger.

Do you have a link?

chamrider aliexpress store.
https://www.aliexpress.com/store/group/500W-1000W-Geared/738588_511029183.html
I'm picking one up myself.
1.15kwh (48v 24ah) battery, charger, xfc19 laced in wheel, free maxis tire, 52v 40a 2000w controller, display, pas, and all the wires, shipping, TAX, all included, less than $590. Ridiculous value. Cells are liitokala 4800mah 9.6a discharge 1.5a charge.
Only thing you need is a pair of torque arms.

 

[qwerkus]

If this is the same motor as the mxus gdr-19 BEWARE people: it's super loud! Got one here, grinding like a coffee mill even though I use a sine wave controller. Maybe with true FOC it would be better, though one of the MXUS techs wrote me back that noise is a problem on that motor. They should have opted for helical cut gears at those power levels.

 

[Tommm]

If this is the same motor as the mxus gdr-19 BEWARE people: it's super loud! Got one here, grinding like a coffee mill even though I use a sine wave controller. Maybe with true FOC it would be better, though one of the MXUS techs wrote me back that noise is a problem on that motor. They should have opted for helical cut gears at those power levels.

Thanks for chiming in.
I saw the video by the russian guy.
[youtube]TdSJz6CVDIw[/youtube]
Noise sounds about the same as a cyclone mid drive. Deep rumble that turns into a beautiful whine at higher rpm. I'm a mid drive guy in general so I don't mind.
If I can ask, what power level do you run at, does it heat, how many miles you have on it and have you had to replace the gears? I am thinking about finding a source for metal gears instead of plastic to make it bomb proof(I know it would be louder).
In general though, repacking the motor with a bit more quality grease will quiet it down, as there is less of a reverb inside the motor casing.

 

[qwerkus]

Noise sounds about the same as a cyclone mid drive. Deep rumble. I'm a mid drive guy in general so I don't mind.
If I can ask, what power level do you run at, does it heat, how many miles you have on it and have you had to replace the gears? I am thinking about finding a source for metal gears instead of plastic to make it bomb proof(I know it would be louder).

1. Pushed up to 1400W. Heats up quite fast compared to a 9c DD hub and takes a lot longer to cool down. Much more torque though. This thing carried me up a 1mile 12% slope without flinching. I think it's slightly more powerful than a MAC though I'd love an entry for it in the ebike.ca sim for true comparison. Note that in comparison to ANY other geared motor I had, this beast can take a lot more abuse until it overheats. Guess the 4.7Kg help.

2. Not many miles - way too noisy for me. There is virtually no noise until your reach about 20km/h or about 300-400W, than it gets louder and louder.

3. I did swap gears but not because they failed but because I asked for 2x gear sets, one with locked clutch and one without. Regen worked surprisingly well with the locked clutch. Super easy: took me 25min - the most annoying part is having to glue the case back together with HT silicone glue.

4. Wouldn't recommend metal gears on that hub. Already way too noisy, and the stock gears are already pretty solid.

PS: MXUS told me it comes in two version: 500W and 1000W. I think OP got the 500W version based on the Z910 connector. Mine was the 1000W version and came with larger Z916 connector.

 

[Tommm]

Noise sounds about the same as a cyclone mid drive. Deep rumble. I'm a mid drive guy in general so I don't mind.
If I can ask, what power level do you run at, does it heat, how many miles you have on it and have you had to replace the gears? I am thinking about finding a source for metal gears instead of plastic to make it bomb proof(I know it would be louder).

1. Pushed up to 1400W. Heats up quite fast compared to a 9c DD hub and takes a lot longer to cool down. Much more torque though. This thing carried me up a 1mile 12% slope without flinching. I think it's slightly more powerful than a MAC though I'd love an entry for it in the ebike.ca sim for true comparison. Note that in comparison to ANY other geared motor I had, this beast can take a lot more abuse until it overheats. Guess the 4.7Kg help.

2. Not many miles - way too noisy for me. There is virtually no noise until your reach about 20km/h or about 300-400W, than it gets louder and louder.

3. I did swap gears but not because they failed but because I asked for 2x gear sets, one with locked clutch and one without. Regen worked surprisingly well with the locked clutch. Super easy: took me 25min - the most annoying part is having to glue the case back together with HT silicone glue.

4. Wouldn't recommend metal gears on that hub. Already way too noisy, and the stock gears are already pretty solid.

PS: MXUS told me it comes in two version: 500W and 1000W. I think OP got the 500W version based on the Z910 connector. Mine was the 1000W version and came with larger Z916 connector.

Yes, the gears look quite big, 54mm instead of 36mm used in most hubs. Where did you get the spare parts? It is good to know they are avaliable if needed.
Good catch for anyone looking to order, the seller mentioned here (old name C&D) pairs a 48v 40a controller with the 1000w listing, and generally lets you know which one you're buying. It does look like the 1000w one is a few mm wider between the spokes.
I would be suprised if a 5kg motor wouldn't handle my intended purpose, long rides between towns at about 28mph/45kph. With an occasional short offroad incline. But if so, ferro fluid and heat sinks it would be.

 

[qwerkus]

But if so, ferro fluid and heat sinks it would be.

ferrofluid is basically useless in geared hub motors. Oil cooling works better, but in nearly all case it ends up leaking, which can be very dangerous if oil reaches the brake pads.

 

[Tommm]

But if so, ferro fluid and heat sinks it would be.

ferrofluid is basically useless in geared hub motors. Oil cooling works better, but in nearly all case it ends up leaking, which can be very dangerous if oil reaches the brake pads.

I just started looking at geared hub videos. It seems like the magnets don't really touch the outside of the case. But wouldn't ferro fluid work there instead to fill the gap between the magnets and the case, instead of between the stator and the magnets like in normal hubs?
Where did you get the welded gearset?

 

[qwerkus]

I just started looking at geared hub videos. It seems like the magnets don't really touch the outside of the case. But wouldn't ferro fluid work there instead to fill the gap between the magnets and the case, instead of between the stator and the magnets like in normal hubs?
Where did you get the welded gearset?

1. The magnetic fluid will be attracted to the permanent magnets of the inner rotor, where they-d bridge the gap between rotor and stator. But then you'd still need to find a way to carry the heat from the motor shell to the outer case, where there is no ferrofluid - hence its not very useful in geared hubs. See the official FAQ: https://www.ebikes.ca/product-info/statorade.html#frequently-asked-questions

2. MXUS provided the welded clutch. Nothing fancy though: regular clutch with 3 big and ugly weld points - could have done it myself...

 

[neos]

Would this motor be suitable to run with a Baserunner and a Cycle Analyst?
I would be particularly interested in whether the internal temp sensor could
be read from a Baserunner (overheat protection) and whether the
internal speed sensor could be read by a Cycle Analyst?
Does the speed sensor send also a signal when the wheel is just rolling?
Does someone know something about the noise in combination with a Baserunner?

 

[Tommm]

Would this motor be suitable to run with a Baserunner and a Cycle Analyst?
I would be particularly interested in whether the internal temp sensor could
be read from a Baserunner (overheat protection) and whether the
internal speed sensor could be read by a Cycle Analyst?
Does the speed sensor send also a signal when the wheel is just rolling?
Does someone know something about the noise in combination with a Baserunner?

A lot of speed sensors on displays just listen to hall or phase wire. You just need to do the division in phaserunner settings, also the phaserunner is quite configurable in terms of temp sensor. I haven't wired it up to one. But it would be a good match as the motor will handle 40a battery.

 

[neos]

A lot of speed sensors on displays just listen to hall or phase wire. You just need to do the division in phaserunner settings, also the phaserunner is quite configurable in terms of temp sensor. I haven't wired it up to one.

As far as I know the speed sensor in the GMAC changes the state on the line 6 times per wheel rotation.
For a serious decision it would be helpful if someone who knows for sure whether the XF19 speed and temp
sensors are compatible with the Baserunner / CA could confirm this.

But it would be a good match as the motor will handle 40a battery.

With a GMAC 10T and the Baserunner and a 48V 30a battery, the ebikes.ca simulator
indicates a top speed of 40 - 42 kph and torque between 60 - 70 Nm.
It would be interesting to have this values also for the MXUS XF19 with
such a battery.

About the noise: In the video it seams that the noise level is even at 20 kph very high.
Why is the motor so much louder than for example a MAC? Where is the difference?

 

[Tommm]

A lot of speed sensors on displays just listen to hall or phase wire. You just need to do the division in phaserunner settings, also the phaserunner is quite configurable in terms of temp sensor. I haven't wired it up to one.

As far as I know the speed sensor in the GMAC changes the state on the line 6 times per wheel rotation.
For a serious decision it would be helpful if someone who knows for sure whether the XF19 speed and temp
sensors are compatible with the Baserunner / CA could confirm this.

But it would be a good match as the motor will handle 40a battery.

With a GMAC 10T and the Baserunner and a 48V 30a battery, the ebikes.ca simulator
indicates a top speed of 40 - 42 kph and torque between 60 - 70 Nm.
It would be interesting to have this values also for the MXUS XF19 with
such a battery.

About the noise: In the video it seams that the noise level is even at 20 kph very high.
Why is the motor so much louder as for example a MAC? Where is the difference?

Maybe less/thinner grease used, or controller is square wave.
Also a motor accelerating while doing 20 will be much louder than cruising at 20.

 

[AHicks]

A lot of speed sensors on displays just listen to hall or phase wire. You just need to do the division in phaserunner settings, also the phaserunner is quite configurable in terms of temp sensor. I haven't wired it up to one.

As far as I know the speed sensor in the GMAC changes the state on the line 6 times per wheel rotation.
For a serious decision it would be helpful if someone who knows for sure whether the XF19 speed and temp
sensors are compatible with the Baserunner / CA could confirm this.

But it would be a good match as the motor will handle 40a battery.

With a GMAC 10T and the Baserunner and a 48V 30a battery, the ebikes.ca simulator
indicates a top speed of 40 - 42 kph and torque between 60 - 70 Nm.
It would be interesting to have this values also for the MXUS XF19 with
such a battery.

About the noise: In the video it seams that the noise level is even at 20 kph very high.
Why is the motor so much louder as for example a MAC? Where is the difference?

Re: speed sensor, I can share that the white wire from a MAC will NOT supply a speed signal while coasting (my MAC 12t doesn't anyway). The motor stops when coasting/not under power, causing the indicated speed to drop to -0- when using the white wire for a speed sensor. I had to install an external sensor to get a full time speedo.

Regarding the Mxus noise, I would want to know about the controller. Is it a sine wave, or a square wave? If a square wave, wondering if a sine wave would quiet the drive, or if there is something in the actual design causing it.

 

[Tommm]

Regarding the Mxus noise, I would want to know about the controller. Is it a sine wave, or a square wave? If a square wave, wondering if a sine wave would quiet the drive, or if there is something in the actual design causing it.

2. MXUS provided the welded clutch. Nothing fancy though: regular clutch with 3 big and ugly weld points - could have done it myself...

You know if its square wave?

 

[AHicks]

Regarding the Mxus noise, I would want to know about the controller. Is it a sine wave, or a square wave? If a square wave, wondering if a sine wave would quiet the drive, or if there is something in the actual design causing it.

2. MXUS provided the welded clutch. Nothing fancy though: regular clutch with 3 big and ugly weld points - could have done it myself...

You know if its square wave?

No, I do not know. That's my question as well..
In the youtube demo posted earlier, that's an LCD3 display, which is used with a KT based controller. What I do know is that controller design is available either sine wave or square wave, with some models being able to switch between them. My MAC 12t is using a KT w/sine wave, and is nearly silent, accelerating or not, at any speed.

 

[neos]

MXUS states 100 Nm for the 1000W XFC19, does that mean that the 1000W MXUS has a higher torque than a
10T MAC? What rpm has the 1000W XFC19? What would be the top speed with 48V or 52V?

 

[Tommm]

MXUS states 100 Nm for the 1000W XFC19, does that mean that the 1000W MXUS has a higher torque as a
10T MAC? What rpm has the 1000W XFC19? What would be the top speed with 48V or 52V?

A russian guy who bought it wrote max speed on 52v is between 55-59kph. So for 48v it would be 50-55kph. For that low of a top speed at 2000w, it will pull hard.

 

[qwerkus]

My MAC 12t is using a KT w/sine wave, and is nearly silent, accelerating or not, at any speed.

You run a MAC on a KT SVPRL controller ? Do you have a picture of that setup ? Until now nobody managed to run a MAC on a KT controller.

I did run my Mxus GDR-19 on a KT sine wave controller however, and the noise was awful. Maybe a better controller like an EBS3xx or an expensive one that can do FOC would yield better results, but I seriously doubt it. The noise comes from the gear design - they should have gone for helical cut gears.

 

[Tommm]

The noise comes from the gear design - they should have gone for helical cut gears.

Helical gears produce a lot of thrust, straight gears are more difficult to get wrong, and these hubs are made of $30-40 worth of material, they will cut the worst corners... I'm fine with them not increasing complexity.

From what I've seen, it is louder than a bafang but similar to a cyclone and nothing like a cyc or tangent.

A hub guy's deafening loud is a mid drive guy's quiet. :lol:

 

[neos]

A russian guy who bought it wrote max speed on 52v is between 55-59kph. So for 48v it would be 50-55kph. For that low of a top speed at 2000w, it will pull hard.

Then the XFC19 seems to be more comparable to a 8T or 6T MAC? But what about the torque? 8T or 6T have low torque, therefore
100Nm torque for the XFC19 seems hard to believe? Because with the MAC you can have high top speed (8T) or high torque (10T), but not both. The XFC19 has both?

Re: speed sensor, I can share that the white wire from a MAC will NOT supply a speed signal while coasting (my MAC 12t doesn't anyway). The motor stops when coasting/not under power, causing the indicated speed to drop to -0- when using the white wire for a speed sensor. I had to install an external sensor to get a full time speedo.

Yes, that applies to the older MAC motors. As far as I know there also MAC motors with speed sensors that can also send data while coasting.
But how does the XFC19 speed sensor work? The MXUS XF15 for example can also send speed data at idle.

 

[AHicks]

My MAC 12t is using a KT w/sine wave, and is nearly silent, accelerating or not, at any speed.

You run a MAC on a KT SVPRL controller ? Do you have a picture of that setup ? Until now nobody managed to run a MAC on a KT controller.

I did run my Mxus GDR-19 on a KT sine wave controller however, and the noise was awful. Maybe a better controller like an EBS3xx or an expensive one that can do FOC would yield better results, but I seriously doubt it. The noise comes from the gear design - they should have gone for helical cut gears.

This is the controller I'm using. I've found the "torque simulation" = sine wave.
http://www.pswpower.com/ven.php?cargo.2016-3f-28ym

Previous to the MAC, this same controller was used on a 1500w direct drive. A lot of people don't know that the KT will work just fine with the MAC, but there's a suspicion that the fact it's a 12t and not a faster wind might have something to do with it. Not sure, but I do see that the manual for the new color display has some "L" parameters, one of which enables a "high speed" function, though mine is the older LCD3 display..... -Al

 

[AHicks]

A russian guy who bought it wrote max speed on 52v is between 55-59kph. So for 48v it would be 50-55kph. For that low of a top speed at 2000w, it will pull hard.

Then the XFC19 seems to be more comparable to a 8T or 6T MAC? But what about the torque? 8T or 6T have low torque, therefore
100Nm torque for the XFC19 seems hard to believe? Because with the MAC you can have high top speed (8T) or high torque (10T), but not both. The XFC19 has both?

Re: speed sensor, I can share that the white wire from a MAC will NOT supply a speed signal while coasting (my MAC 12t doesn't anyway). The motor stops when coasting/not under power, causing the indicated speed to drop to -0- when using the white wire for a speed sensor. I had to install an external sensor to get a full time speedo.

Yes, that applies to the older MAC motors. As far as I know there also MAC motors with speed sensors that can also send data while coasting.
But how does the XFC19 speed sensor work? The MXUS XF15 for example can also send speed data at idle.

The MAC 12t I have is about a year old... I did wire it up hoping to save myself the trouble of an external sensor, and it worked as noted previously - with the speedo going to -0- while coasting. If you look at internal pictures of where that sensor is mounted, I think you'll agree it's impossible for that sensor to indicate speed while coasting.

 

[qwerkus]

This is the controller I'm using. I've found the "torque simulation" = sine wave.
http://www.pswpower.com/ven.php?cargo.2016-3f-28ym

Previous to the MAC, this same controller was used on a 1500w direct drive. A lot of people don't know that the KT will work just fine with the MAC, but there's a suspicion that the fact it's a 12t and not a faster wind might have something to do with it. Not sure, but I do see that the manual for the new color display has some "L" parameters, one of which enables a "high speed" function, though mine is the older LCD3 display..... -Al

Thanks for your reply. Do you have a picture of the sticker on your controller ? Would be helpful, because on the pswpower website, it says "square wave". Even more definitive would be a picture of the controller pcb. If there is a hall sensor on the green phase wire, it's sine wave. Otherwise it's not. That sensor is required to find out the 0 moment of the motor signal, which is than used to calculate the angle correction required for a nice sine signal. A truely FOC capable controller would have a sensor on each phase, and use it to calculate the exact rotor position for even smoother results.

 

[AHicks]

Picture not possible. Sticker is not visible without disassembly (up against the seat tube), same with the PCB. Bike is used almost daily.

I looked more closely at the controller in the link I sent. It's a sensorless controller. The one I am using is sensored - but other than that is identical to this controller. Sorry for any confusion. Either they've change the specs on this site, or I got it somewhere else, possibly on amazon. There is no doubt that the controller is a KT 12fet, 35a max., sensored, and using an LCD3 display.

The biggest feature for me is the "imitation torque sensor" done with a cadence sensor only. Each PAS level determines a voltage range available to the motor - and NOTHING is based on speed. It does simulate a torque sensor setup nicely. A very well done software feature......

 

[1boris]

mine is not loud with 48volt 22 a sinewave controller,it is louder than the newest Mac motor but better than the bafang swx02 and bpm motors I have