2019 Motor Simulator, Mid-drives and other updates

[justin_le]

Another oddball motor from Bafang isn't even on their webpage but is something that fits an interest of ours is this geared motor designed for a wheelbarrow.



There are many times when we have customers trying to find a hub motor for really high torque and low speed applications like this, and while some of the larger ebike motors can do 50-80Nm torque levels for a long enough time it's at quite a low efficiency if we're only moving at 5-10kph speeds.

I wasn't sure at all what to expect from the sample but the side cover had the text G311 lasered on it which I found interesting, and on opening it up sure enough the inside contained the same inrunner motor as a G311 ebike hub.



The difference is in the gearing through. This one has a 16T helical gear on the motor shaft driving a 73 tooth gear on a jackshaft, while the other side of the jackshaft had a 9 tooth spur gear turning inside a 59 tooth ring gear on the side plate of the hub housing



The total gear reduction is then 73/16 * 59/9 = 29.9 : 1, compared to 11:1 for the G311 ebike hub motor. So about a 3 times higher total gearing over the familiar ebike model. It also doesn't have any freewheeling clutch, so could be used bidirectionally and with regen too which is often needed in these applications (like say a winch for pulling rope).

I've got it live on the simulator if anyone wants to play with it, shown here hauling a 220lb load up a 10% grade slope effortlessly:
https://www.ebikes.ca/tools/simulator.html?motor=MG311_WB&cont=C20&wheel=16i&hp=0&grade=10

I didn't do a thermal study yet on this motor yet, but I put it in the database using same thermal characteristics as the G311 hub motor which shouldn't be too far off.

 

[amberwolf]

Assuming hte thermal characteristics are close enough:
https://www.ebikes.ca/tools/simulator.html?motor=MG311_WB&cont=C40&wheel=29i&hp=0&grade=1&axis=mph&mass=226&batt=B7223_AC&frame=cust_1_0.005
wiht a 72v pack and 29" wheel (would need some kind of bolt-on adapter replacing the "rim" it comes with) teh simulator shows it could push me on my trike to 19mph on a 1% grade, and still stay just over boiling water temperatures. :lol:

(well, I used CdA of 1.0; I'm sure the trike isn't really *that* bad, so it could probably do better than this).

 

[justin_le]

wiht a 72v pack and 29" wheel (would need some kind of bolt-on adapter replacing the "rim" it comes with) teh simulator shows it could push me on my trike to 19mph on a 1% grade, and still stay just over boiling water temperatures. :lol:

Even if you were going downhill on a -2.9% grade where the motor produces ZERO power at 19mph it would get to nearing boiling water temperatures:
https://www.ebikes.ca/tools/simulator.html?motor=MG311_WB&cont=C40&wheel=29i&hp=0&grade=-2.9&axis=mph&mass=226&batt=B7223_AC&frame=cust_1_0.005

This is all core losses, and it's a great example why the "geared further and spinning faster is better" stops being true when half your losses are coming from the core rather than copper. The high gear reduction hub motors like the G311's are already running at pretty optimal split between copper/core loss, so spinning them faster still and gearing down further makes the motor less efficient rather than more efficient, and the simulator can be a great tool to see this. Plus the magnets would most certainly fly off at those kinds of rotor RPMs.

Run at a more sensible core RPM and it can make your crazy bike climb a 3 times steeper grade and stay below boiling water temps
https://www.ebikes.ca/tools/simulator.html?motor=MG311_WB&cont=C20&wheel=29i&hp=0&grade=3&axis=mph&mass=226&batt=B4823_AC&frame=cust_1_0.005

I should probably add a feature that shows the split between core and copper losses in the motor simulator as we've done on the trip simulator app is you can see here in the "show advanced" features on the right

 

[wil]

The Bafang G360 is just over 4.0kg, while the similar sized eZee and MAC motors are already a bit lighter at 3.8kg. That would be OK if the G360 had higher sustainable power and torque capabilities, but when you do a normalized comparison between the two, you see that across the board a MAC motor has slightly better efficiency, lower heating, and higher output torque....

From a performance and weight side of things, the eZee and MAC motors are slightly better, even though the G360 seems at first glace to have the best engineered inside....

It's an absolutely fine motor, but given that we already have eZee and MAC in this power/performance category it doesn't fill a new application space.

I'm sure it already comes into your testing but as the G360 has the slightly inner set magnets + spacing between them would it not have a reluctance component to the torque and therefore require field weakening to hit peak efficiency values? Or does the rotor design not cause enough for it to matter?

The main difference as you see is that the rotor uses embedded magnets rather than surface glued magnets like on the G310, which seemed like a great design upgrade given that part of the problem we've had with the G31x's has been magnets coming loose at high RPMs.

I've had my G360 up to 500 RPM, definitely a design improvement in that regard

The Bafang G360 is just over 4.0kg, while the similar sized eZee and MAC motors are already a bit lighter at 3.8kg.

My major comment on the G360 from my disassembly is it seems to have far more material than required in the case and stator casing. Probably important for strength though.

 

[uphill_fast]

This graph here shows the unloaded crank RPM as a function of the power supply voltage.
BBSxx rpm vs V Graphs.jpg

The BBS02 is a faster motor winding at about 3.27 RPM/V, compared to ~2.71 rpm/V for the BBS02.

Justin,
Am I reading this graph incorrectly? It looks like the BBSHD is the faster winding (RPM per Volt), no?

 

[DogDipstick]

This graph here shows the unloaded crank RPM as a function of the power supply voltage.
BBSxx rpm vs V Graphs.jpg

The BBS02 is a faster motor winding at about 3.27 RPM/V, compared to ~2.71 rpm/V for the BBS02.

Justin,
Am I reading this graph incorrectly? It looks like the BBSHD is the faster winding (RPM per Volt), no?

Misprint, Typo.. HD is faster winding as shown in early posts.. Should read " BBSHD is a fster motor winding ..... "

 

[la838]

Hey Justin any updates on the G370 motor? Any rough idea the performance we should expect from it and what would be other motors it would compare it to?

 

[tomjasz]

I've got other priorities now but I'm tempted to experiment to see what lightweight cooling mods and bolt on heatsink fins could be done to a BBS02 in order to make it meet or even exceed the thermal characteristics of a BBSHD.

I struggle to grasp even 50% of the data, however I got quite excited to find out if anyone has tested some cooling ideas. I’m gong full dork currently and doing a couple of folders with BBS01B 250W. I know my 5 year old 350W would benefit from cooling.

Any progress by anyone with the cooling the BBS02B?

Great stuff! Thanks!

 

[sabergo]

Hey Justin any updates on the G370 motor? Any rough idea the performance we should expect from it and what would be other motors it would compare it to?

I'm also interested in this motor. If silent its seems a like a good "assist" option.

 

[SlowCo]

Another oddball motor from Bafang isn't even on their webpage but is something that fits an interest of ours is this geared motor designed for a wheelbarrow.

What is the rim size on this wheelbarrow motor? Any more info somewhere?

 

[Alex W]

G370 is in the simulator. The hub looks awesome! When can I buy one from ebikes.ca?

Normalized for Kv it looks a lot like the G311:
https://www.ebikes.ca/tools/simulator.html?bopen=true&motor=MG310_STD&grade=0&hp=0&motor_b=MG370_10T&hp_b=0&grade_b=0&axis=mph&mass=90&mass_b=90&kv=7.29

At high loads and low RPM (this is 10% grade) it will overheat sooner, which isn't surprising given it's lower mass:
https://www.ebikes.ca/tools/simulator.html?bopen=true&motor=MG310_STD&grade=10&hp=0&motor_b=MG370_10T&hp_b=0&grade_b=10&axis=mph&mass=90&mass_b=90&kv=7.29

 

[Aquakitty]

Any plans to add other mid-drives, like the TSDZ2? For the longest time I thought the Tongxin was the Tongsheng!
Maybe we can crowd source a sample motor?

 

[Bullfrog]

Justin...just came across this thread. Absolutely awesome...as usual, THANK YOU for adding the mid drive option.

I have been playing with the Motor Simulator using the BBSHD motor a lot. It is great to be able to do a direct comparison of the BBSHD and the MAC/GMAC. Pretty amazing how much cooler the BBSHD runs than the MAC/GMAC if you don't gear it too tall but it is all due to the design of the motors .

Below is off the original subject but may be of interest to some...

Just FYI for all, I am currently working on some cooling experiments for the MAC. Plan to publish results on Endless Sphere in about six months. Using distilled water and Motul MoCool (5% by volume) as the cooling fluid. I tried ATF previously and it worked great but leakage was an issue for me...the tiniest bit of ATF of your brake rotor and it feels like your bike speeds up when you hit the brakes instead of slowing down. Maybe Justin can solve the leakage problem :wink: . If the MAC/GMAC can be reliably sealed, 3.5-4.0 ounces of low viscosity ATF will do wonders for keeping your MAC/GMAC motor cool.

My current experiments are using four ounces of distilled water/MoCool (5% by volume) mixture and it drops the core temp roughly from 105C to 60C for my 12T MAC in a 722mm OD wheel/tire (14s battery). The big question is how does the fluid affect the motor internals long term...and only time will tell and the reason I plan to wait before publishing any results .

 

[qwerkus]

Justin...just came across this thread. Absolutely awesome...as usual, THANK YOU for adding the mid drive option.

I have been playing with the Motor Simulator using the BBSHD motor a lot. It is great to be able to do a direct comparison of the BBSHD and the MAC/GMAC. Pretty amazing how much cooler the BBSHD runs than the MAC/GMAC if you don't gear it too tall but it is all due to the design of the motors .

Below is off the original subject but may be of interest to some...

Just FYI for all, I am currently working on some cooling experiments for the MAC. Plan to publish results on Endless Sphere in about six months. Using distilled water and Motul MoCool (5% by volume) as the cooling fluid. I tried ATF previously and it worked great but leakage was an issue for me...the tiniest bit of ATF of your brake rotor and it feels like your bike speeds up when you hit the brakes instead of slowing down. Maybe Justin can solve the leakage problem :wink: . If the MAC/GMAC can be reliably sealed, 3.5-4.0 ounces of low viscosity ATF will do wonders for keeping your MAC/GMAC motor cool.

My current experiments are using four ounces of distilled water/MoCool (5% by volume) mixture and it drops the core temp roughly from 105C to 60C for my 12T MAC in a 722mm OD wheel/tire (14s battery). The big question is how does the fluid affect the motor internals long term...and only time will tell and the reason I plan to wait before publishing any results .

Interesting experiment. But doesn't distilled water act like a super oxidant ? Aren't you worried about corrosion ?

 

[Deafcat]

Interesting experiment. But doesn't distilled water act like a super oxidant ? Aren't you worried about corrosion ?

I think you mean de-ionized water...

Distilled water is the correct choice for a soluble coolant. the MoCool he's added provides rust inhibition, among other important attributes.

 

[qwerkus]

Interesting experiment. But doesn't distilled water act like a super oxidant ? Aren't you worried about corrosion ?

I think you mean de-ionized water...

Distilled water is the correct choice for a soluble coolant. the MoCool he's added provides rust inhibition, among other important attributes.

Thanks for the clarification. How about gas forming once the motor reaches boiling temps ?

 

[1N4001]

Deionized/distilled/purified water will leech salts and metals out of its surroundings and become conductive and corrosive. There's no way around that, even antifreeze or other corrosion inhibitors will only work for a limited time.

Water also will wash out the gear grease while itself not providing comparable lubrication. It boils at a mere 100°C, but then again, nothing compares to its cooling capacity.

IMO this is a bad idea, but I'm nonetheless curious about the results.

 

[Deafcat]

How about gas forming once the motor reaches boiling temps ?

Well that's steam, and it's powerful stuff, and probably do a number on the internals...

I'm also leaning towards IN4's deduction, that it will not end will having water coolant circulating inside an electric motor... watercooling motors is typically done thru a jacket/extrusion/etc not in contact with the moving parts.

 

[Bullfrog]

Just a little scoop on what/why I am doing with water cooling the MAC...

All good concerns above .

My temps don't get anywhere close to boiling...max temp I have seen so far is 59C and that was when the ambient temp was 95F/35C. My test loop has a combination of hills and flat and before adding the water/MoCool mixture, I would run around 105C-110C once I reached steady state.

Somebody mentioned deionized (relatively pure water)...yep it will leach ions from whatever it comes into contact with over time. I thought about using it with and without the MoCool but seemed sort of pointless since over time it will contain ions...mainly from the corrosion products inside the motor. And if I use MoCool it is highly ionic (conductive) and would be sort of pointless to used deionized water.

Washing the grease out...well yea but there isn't any grease in my motor :lol: . I carefully cleaned everything as best I could to remove all the grease prior to buttoning up the motor and adding the fluid. The water/MoCool does act as a lubricant...although not nearly as good as the grease I removed.

Disassembly and inspection in a few months will be interesting for two reasons:
A. How do things like the composite planetary gears hold up with water as the only lubricant?
B. Corrosion...how much do I get and is it a problem. Originally used distilled water only...but after running it for a while I drained the water and added the four ounces of distilled water/MoCool (5% by volume) mixture.

One suggestion...if anybody does anything similar, I'd recommend positioning the motor so the printed circuit board with the Hall sensors is above horizontal. The board can be in two positions because the axle flats can be positioned in one of two positions which are 180 degrees apart. By positioning the board above horizontal, it doesn't sit in the cooling fluid.

I still like Automatic Transmission Fluid the best as far as potential cooling fluids...for a number of reasons, but ANY leakage becomes extremely dangerous because if it gets on your brakes...they don't work.

Just FYI, I am running a 12T MAC in a 722mm OD wheel/tire. The tire is a Schwalbe Super Moto-X 27.5x2.8 and is ECE-R75 certified for 50 KPH...it also has the highest load carrying capacity of any ECE-R75 tire I could find so IMO the best "bicycle" tire made for an ebike.

 

[Bullfrog]

I've got other priorities now but I'm tempted to experiment to see what lightweight cooling mods and bolt on heatsink fins could be done to a BBS02 in order to make it meet or even exceed the thermal characteristics of a BBSHD.

I struggle to grasp even 50% of the data, however I got quite excited to find out if anyone has tested some cooling ideas. I’m gong full dork currently and doing a couple of folders with BBS01B 250W. I know my 5 year old 350W would benefit from cooling.

Any progress by anyone with the cooling the BBS02B?

Great stuff! Thanks!

Guys...something you can do to monitor the motor temps is use a "Temp Plate". They are made by Wahl and similar products by other companies. They are very accurate and respond quickly to measure the surface temp of whatever you stick them on. Of course the surface temp is going to be different tan the motor core temp but you can use the "Motor Simulator" to get your core temp for a given set of riding conditions and then correlate the Temp Plate measurements. Here is a link to the ones I use: https://www.wahltempplates.com/product/441-110f/

 

[qwerkus]

The tire is a Schwalbe Super Moto-X 27.5x2.8 and is ECE-R75 certified for 50 KPH...it also has the highest load carrying capacity of any ECE-R75 tire I could find so IMO the best "bicycle" tire made for an ebike.

I agree that super moto are great tires. Though the 2.4" version is stronger than the 2.8". If only they'd make 20x3" tires...

 

[Bullfrog]

The 27.5x2.8 has the highest load carrying capacity of any of the Super Moto-X tires...not sure exactly what you mean by "strong'.

The smaller you go with the diameter the smaller the footprint is for a given width...so I tend to stay away from 20" tires but I could use a 24x3.00 Super Moto-X .

 

[qwerkus]

The 27.5x2.8 has the highest load carrying capacity of any of the Super Moto-X tires...not sure exactly what you mean by "strong'.

The smaller you go with the diameter the smaller the footprint is for a given width...so I tend to stay away from 20" tires but I could use a 24x3.00 Super Moto-X .

https://www.schwalbe.com/en/tour-reader/super-moto-x.html

2.8" = raceguard structure
2.4" = greenguard structure

Both 67 EPI (textile density) but the greenguard comes with an additional 3mm anti puncture layer. Hence protection level 5 instead of 4 (="stronger"). Load carrying capacity depends a lot on the rim you use. Don't know if there is any major difference here between the 2.8 and 2.4" versions, but I doubt it, even if schwalbe lists the 2.8" for 155kg and the 2.4" for 145kg. I've seen both and I can tell you from experience that the heavier 2.4" will last you longer but the 2.8" feels somehow smoother.

In my case, it boils down to a different matter: if chainstays offer enough space, 27.5x2.4" are a perfect replacement for 28x1.5-1.75" tires mounted on most city bikes where I live. No change in bike geometry and a lot more suspension. Did it for 2 conversion bikes now, and it's amazing how well it fits. Of course it only works on bikes with disc brakes.

 

[Bullfrog]

Load carrying capacity is for the tire only and has nothing to do with the rim.

Puncture protection is for puncture protection only and has nothing to do with Load Carrying Capacity.

If you want to define "strength" as puncture protection...the Green Guard is stronger .

As far as how long they will last...Schwalbe rates all of the Super Moto-X tires level "5" for Durability.

I believe what you are feeling is the "Green Guard" strip that is just a layer between the tread and the cord of the tire so that if something punctures the tread, it hopefully won't puncture the cord which is the structure ultimately holds the air in the tire. In this case, thicker doesn't equate to stronger or longer lasting...just able to be punctured with a longer object before losing the air. In reality the Green Guard tire won't last as long because the Green Guard layer will flex and the tread will wear faster because it flexes more...but I am getting a lot deeper into tire design than anybody probably really cares about.

 

[Bullfrog]

OK so back on the subject of this thread :lol: , I just love the graph with the MAC and the G360...I can very easily understand data when it is presented in such a fashion as opposed to just words. You can visualize the rates of change (derivative) sooo much easier...at least I can.

Just for giggles, I went through the motor simulator (greatest tool ever IMO) and did a run for every single motor available in the drop down menu when you select "Show All". The 12T MAC produced more "peak" torque per amp than any motor except the Bafang G60 750w...and the G60 only produced a small amount more...142 Nm vs 131 Nm. You wouldn't want to run a motor very long at those conditions, I was just using the peak torque for comparison purposes. Just for those out there not familiar with the G60...it is 175mm OLD (wide) so only fits in dropouts on the wider end of the spectrum...if you have 135mm dropouts, it is not an option.

As most know, the drawback of the MAC is keeping it cool...that is why I am working on ways to cool it (I am by no means the first) and heat sinks/forced air across the exterior don't help much because the windings (where the heat is produced) have a relatively large air gap between the outer shell (where you want the heat to go so it can be rejected to the atmosphere) and the windings and air is a pretty good insulator so basically the winding are wrapped in a blanket and the heat can't get out. IMO liquid cooling is the best option...by liquid cooling, I mean using a fluid to thermally connect the windings to the shell. Tough to get better heat transfer than convective cooling with a liquid.

So with the MAC we have the motor that produces the most torque per amp (12T) and now it is just a matter of figuring out how to keep it from overheating. Can you tell I am a big MAC fan :lol: .

Edit...ooops, the wheelbarrow motor DOES produce more torque but the rpm isn't fast enough for most people looking for a bike motor. Maybe when I get a little older .