Trio Cargo e-bike mid drive

[fatty]

I probably should specify "OEM hubmotor wheels" because those are not usually built correctly or with parts that should be used together

Ah, agree completely. I was thinking in the abstract.

but even well-built hubmotor wheels can be weaker than the same wheel with a regular-sized hub in it.

If equivalently well-built (extensive caveats agreed upon below), I don't think this is the case.

And probably should also specify that the larger diameter the hub is, the more the problems tend to stand out; the closer the hub is to the more typical size used, the less the problems tend to crop up.

Though true, I would clarify that such problems result from larger hub diameter but also smaller rim diameter, or rather the ratio between the two. As the hub-rim ratio increases (larger hub and/or smaller rim), nipple angle limit will eventually be exceeded and the spoke will have to bend/curve/break at the nipple. So it's really more a systems problem, rather than just the hub. (You address this later in your post -- removed to be concise).

If you look at the angle the spokes are at, when looking at the wheel "edge on" like in the ebikes.ca spoke calculator, you may find a smaller bracing angle for a hubmotor vs a regular bicycle hub. Depends on the specific motor and the specific bike hub and the overall wheel size.

It may be possible to find a comparison where this is the case (in fact, I built my rear with a front disc hub to maximize flange width), but generally the narrower hubmotor flange width will be offset by the larger flange diameter. It's disappointing that hubmotors are built with narrow flanges and comically-large diameters, rather than wider/taller stators. I'd like to see stator height/width maxed first, before increasing diameter. That's why I recommend the XOFO DD45 or MXUS 3k 45H.

The wider the spacing of the nipple hole rows is on the rim (like fatbike rims especially), the worse this problem will be--the hubmotor flanges can actually be so close to the rim and so close together that the spokes may actually have to be laced from "drive side" at the hub to "brake side" at the rim in order to get any significant bracing angle.

I think we're generally in agreement in this discussion, but this point is most interesting -- I think this is actually an illusion or limitation in the wheel visualization. Offset spoke holes do decrease the measured spoke bracing angle, but not the effective bracing angle. The spoke-nipple-rim assembly is rigid and can be conceived as one piece, so what we care about mechanically is the triangle created by the flanges and a point in the center of the rim.
Lacing to opposite sides is a hack that further exacerbates nipple angle problems.

As for more spokes on hubmotor wheels, that's not necessarily true. There are plenty of bicycles using 36-spoke wheels. Almost every one I have ever owned, which is more than I can remember, have been 36-spoke wheels. A few wheels I've had and used (not on the bikes they came from :lol: ) were 48 spoke, and only one bike I have uses wheels with less than 36 (I think it's 32, but I can't remember ATM). All of those I listed are plain bicycle wheels. Many hubmotors use 36 spokes, but some do use less, or more.

Well, sure, but there wouldn't be any reason to with modern materials and design. I build my non-ebike wheels as 16/20. As we noted above, 36-spoke hubmotors are an artifact of poor hub/wheel design, materials, and assembly. Premium 32-spoke hubmotors are a known improvement.

Also, just having thicker spokes is not better. The wheel must have the correct rim for those thicker spokes, or they cannot be tensioned correctly and the wheel then develops problems with nipples loosening and/or spokes breaking at the elbow. If they are tensioned correctly and the rim isn't strong enough for that, then the rim will crack or deform at the nipple holes, and then the tension is lost and the other problems can develop (or worse). I've had all of those problems (even on my "normal" ebikes :lol: ) with various OEM hubmotor wheels using too-thick spokes for the rims they're built with. There's pics in some of my build and other threads showing some of the cracked-nipple-hole rims, broken nipples, broken spokes, etc., if you're curious.

If you use rims capable of handling the tension of a thicker spoke, then you can make even stronger wheels with those...but that's not typically how hubmotor wheels get built. For some reason a lot of people get it into their head that thicker is *always* better, but don't know how the wheel itself "works", and end up with wheels that fall apart because of damaged rims from too high a tension on their nipple holes.

Yep, great caveats. Just differences in abstract/theoretical design vs practical application.

 

[amberwolf]

It's disappointing that hubmotors are built with narrow flanges and comically-large diameters, rather than wider/taller stators. I'd like to see stator height/width maxed first, before increasing diameter.

I don't recall the thread name, but some years back there was discussion and math (that I don't follow) that appeared to show a larger diameter hubmotor would get more torque for the same mass of copper and laminations and whatnot, vs a smaller diameter but wider version. It's around here somewhere, rpobably in the motor technology section. I don't know if the conclusions were correct or not, or even that I am remembering all of the details.

That's why I recommend the XOFO DD45 or MXUS 3k 45H.

I wouldn't recommend that MXUS 3k 45H simply because they are generally very poorly built. I have two different versions here, I think a 1 and a 3 (might be a 2 and a 3, can't remember). One of htem is a little better than the other, but both of them have problems with the winding (one has stator shorts from insufficient or no insulation between laminations and winding wires; I fixed it with some coronadope drizzled down in there for a while, but after some months something else went wrong that I haven't opened it up to find yet) and axle materials and manufacturing, runouts on covers / rotor, etc. I've probably got a dozen hubmotors of various kinds and these are the lowest quality of any of them (I'd thought that title would always be held by the "9C" types I have, but these win hands down).

According to others, they're also misrated. Unless you're using it in a 20" wheel, it's not likely to handle the 3000w that it's advertised as. I couldn't verify this as mine are not used heavily, power-wise (they get about 2kw+ bursts for a few seconds at a time, then a few hundred continuously).

I'm not totally certain, as I don't have anything but a butt-dyno but AFAICT the little Ultramotor (formerly from a Stromer) gives more torque for the same power than the 4503 (might've been the 4504, cant' remember) did before I had to replace the MXUS with the UM after the MXUS failed again (and appears to have taken out the controller, but haven't had a chance to test that, either). THe UM gets a lot hotter doing it, as it's much smaller...but it is also MUCH MUCH better build and materials quality.

Never used the XOFO, so can't say anything about that one.

Lacing to opposite sides is a hack that further exacerbates nipple angle problems.

Yes, but there have been people that laced up hubmotors into fatbike rims with widely-spaced nipple holes that couldn't get the wheel to be usable until they did that; otherwise the spokes were basically vertical from flanges to holes, and IIRC had some form of sideloading issue from it. THere's at least one thread around here somewhere about such a build, I think from a few years ago. Wish I could remember more details about it.

 

[fatty]

I don't recall the thread name, but some years back there was discussion and math (that I don't follow) that appeared to show a larger diameter hubmotor would get more torque for the same mass of copper and laminations and whatnot, vs a smaller diameter but wider version. It's around here somewhere, rpobably in the motor technology section. I don't know if the conclusions were correct or not, or even that I am remembering all of the details.

Interesting. I don't know enough about design for motor constants, so I'd love to read it if you happen across it.

I wouldn't recommend that MXUS 3k 45H simply because they are generally very poorly built. I have two different versions here, I think a 1 and a 3 (might be a 2 and a 3, can't remember). One of htem is a little better than the other, but both of them have problems with the winding (one has stator shorts from insufficient or no insulation between laminations and winding wires; I fixed it with some coronadope drizzled down in there for a while, but after some months something else went wrong that I haven't opened it up to find yet) and axle materials and manufacturing, runouts on covers / rotor, etc. I've probably got a dozen hubmotors of various kinds and these are the lowest quality of any of them (I'd thought that title would always be held by the "9C" types I have, but these win hands down).

According to others, they're also misrated. Unless you're using it in a 20" wheel, it's not likely to handle the 3000w that it's advertised as. I couldn't verify this as mine are not used heavily, power-wise (they get about 2kw+ bursts for a few seconds at a time, then a few hundred continuously).

I'm not totally certain, as I don't have anything but a butt-dyno but AFAICT the little Ultramotor (formerly from a Stromer) gives more torque for the same power than the 4503 (might've been the 4504, cant' remember) did before I had to replace the MXUS with the UM after the MXUS failed again (and appears to have taken out the controller, but haven't had a chance to test that, either). THe UM gets a lot hotter doing it, as it's much smaller...but it is also MUCH MUCH better build and materials quality.

Great info -- much appreciated. I think I've only directly recommended the DD45 because I have full specs and faith in Justin/Grin, and assumed (incorrectly) the MXUS was equivalent due to popularity. I'll keep this in mind.

Yes, but there have been people that laced up hubmotors into fatbike rims with widely-spaced nipple holes that couldn't get the wheel to be usable until they did that; otherwise the spokes were basically vertical from flanges to holes, and IIRC had some form of sideloading issue from it. THere's at least one thread around here somewhere about such a build, I think from a few years ago. Wish I could remember more details about it.

I think you mean radial, which would indeed be problematic, but not directly resolved by lacing to the opposite side.
My only guess would be that the nipple holes were oriented inwards, whereby lacing to the same side with a large diameter, widely-spaced hub would exceed the nipple swivel angle, and as discussed above, cause the spoke to bend at the nipple entrance. In this case, nipple holes on the opposite side of the rim would actually be oriented better. But my offset rim is profiled such that the nipple holes are (properly) oriented outwards, so this is just conjecture.
As usual, I'd be curious to review the thread if found.

 

[LewTwo]

Fatty and Amberwolf: You folks ever look at a Model "A" wire wheel?

 

[amberwolf]

I think you mean radial, which would indeed be problematic, but not directly resolved by lacing to the opposite side.
My only guess would be that the nipple holes were oriented inwards, whereby lacing to the same side with a large diameter, widely-spaced hub would exceed the nipple swivel angle, and as discussed above, cause the spoke to bend at the nipple entrance. In this case, nipple holes on the opposite side of the rim would actually be oriented better. But my rims are profiled such that the nipple holes are (properly) oriented outwards, so this is just conjecture.
As usual, I'd be curious to review the thread if found.

No, I don't mean radial, I mean literally vertical from the flange to the nipple, if you were able to look at the wheel directly from the top (or front or rear) thru the tire and rim.

Something like the way this one looks in the spoke calculator
https://ebikes.ca/tools/spoke-calc.html?hub=cust_d191_s32_o0_n36_l135_h3_p33.3&pair=false&rim=cust_dia559_e474_lo13_ro13_w67

If I find any of the threads I'll link them, but I wouldn't hold my breath. :lol:

 

[amberwolf]

Fatty and Amberwolf: You folks ever look at a Model "A" wire wheel?

Yes. That is what Justin referenced, IIRC, when he wanted me to try the bent-spoke 1x lacing on one flange, and radial on the other. (the bent spokes broke from flexing; went full radial and haven't had problems since then; there's too little space between motor and rim to use anything else without dimpling the nipple holes or otherwise modifying them to allow angling the nipples, or using rims already made like this).

However, the model A wheels were thick welded spokes, AFAIK, and not tensioned.

 

[LewTwo]

However, the model A wheels were thick welded spokes, AFAIK, and not tensioned.

Darn ... I forgot that bit of Trivia. Maybe I should have reference MGA spoked wheels but the pattern is not nearly as radical.

Roger Musson covers Radial wheel lacing in his book "Professional Guide to Wheel Building" (page: 79). Short excerpt:

Radial lacing can be used on any side of the wheel that doesn’t transmit torque (drive torque or
brake torque via disc and hub brakes). For wheels using rim brakes the options for radial are both
sides of the front, and the rear non drive side. For wheels with hub or disc brakes the only option
for radial is on the front wheel on the opposite side to the brake, but since the front brake provides
the majority of the stopping power then it’s really better to have the wheel cross laced on both sides
because on a well designed hub there is torque transfer to the non brake side with both sides
sharing the brake torque. There are no performance benefits from using radial lacing but they do
look different.

 

[Voltron]

Like this maybe?

This isn't one of mine, but had to do a few builds like this, where all the spokes had to go to one side of a fat tire rim instead of staggered to get the dishing right, leaving the drive side spokes nearly vertical....

Re. performance benefits of radial lacing, there are some, but unless you're a pedal powered human on a race bike you won't really feel any.

 

[fatty]

No, I don't mean radial, I mean literally vertical from the flange to the nipple, if you were able to look at the wheel directly from the top (or front or rear) thru the tire and rim.

Something like the way this one looks in the spoke calculator
https://ebikes.ca/tools/spoke-calc.html?hub=cust_d191_s32_o0_n36_l135_h3_p33.3&pair=false&rim=cust_dia559_e474_lo13_ro13_w67

Ah, I see. I would think offset spoke holes in this context would still be stronger, especially given the concerns with spoke-nipple angle. I wonder what the problem was.

 

[fatty]

Roger Musson covers Radial wheel lacing in his book "Professional Guide to Wheel Building" (page: 79). Short excerpt:
There are no performance benefits from using radial lacing but they do look different.

Radial laced spokes are shorter, and thus lighter and more aero. I lace it on the front of rim-brake wheels.

 

[Chalo]

Radial laced spokes are shorter, and thus lighter and more aero. I lace it on the front of rim-brake wheels.

Radially laced spokes (and one-cross spokes) can be laced all on the same side of the flange, which is a convenient way to reduce imbalances in spoke tension. Most of the time I use radial lacing anymore is to move non-drive side rear spokes closer to the hub center, so I can turn them up a little tighter.

When I use radial lacing on front wheels, I usually put all the elbows outside the hub flanges, to increase bracing angle for a stronger wheel. That's not as aero as putting them all inside the flange, though.

 

[fatty]

When I use radial lacing on front wheels, I usually put all the elbows outside the hub flanges, to increase bracing angle for a stronger wheel. That's not as aero as putting them all inside the flange, though.

Same. But I expect this is more than compensated by the resulting ability to use even thinner bladed spokes.

 

[yabert]

Begs the question .... Just what kind of cargo are planning to carry.

Kids and/or food most of the time.
But you can expect us to put many thing in this cargo: mower, another bike, wife, lead acid batt ship to the recycling, steel, dirt and others

 

[dmcd]

Hi yabert,
Could you provide details as to which Bafang motor you used and if you needed to do any frame modifications?
Any info/photos would be appreciated as I'm thinking of converting my older Triobike Cargo from a BionX hub drive to a mid drive system.
Thanks

 

[yabert]

Hi yabert,
Could you provide details as to which Bafang motor you used and if you needed to do any frame modifications?
Any info/photos would be appreciated as I'm thinking of converting my older Triobike Cargo from a BionX hub drive to a mid drive system.
Thanks

Hi
As you can see first page, I used a Bafang BBS02.
I had to cut the frame, quite a bit, to fit the motor.

 

[Vbruun]

Hi
As you can see first page, I used a Bafang BBS02.
I had to cut the frame, quite a bit, to fit the motor.

How is the bike holding up?

 

[yabert]

How is the bike holding up?

Very well! At least the aluminum cut frame.
The rest of the bike need regular maintenance as it is use 365 days/year and winter in Québec is not soft with bike components.
We are now on the second bafang controller and it start to have some random Error 8 time to time... not good.

 

[Vbruun]

I did the same to my bike but used the toseven motor. Fingers crossed that its good.

 

[offGridDownUnder]

a larger diameter hubmotor would get more torque for the same mass of copper and laminations

Given that the parameters are fixed, except for the larger diameter, I would note the larger diameter provides a longer lever arm for the magnetic field attraction. I would expect more torque.

 

[yabert]

I did the same to my bike but used the toseven motor.

Wow, the timing... our motor is now dead (almost). Water intrusion everywhere.
Corrosion on the hall sensor. I will try to add solder, but it's like corrosion destroy many things.

The Toseven motor seem like a nice replacement for us. What model do you have and how do you like it?
Pictures of your set-up?

 

[amberwolf]

Salted roads?

 

[yabert]

Salted roads?

Absolutely.

it is use 365 days/year and winter in Québec is not soft with bike components.

 

[Vbruun]

I have the DM02. The quality seems pretty good and I really like the way they have made the motor easy to configure.
The torque sensing works well but the settings need a lot of tweaking so you need to be comfortable with that.

The engine pulls OK but is a bit noisy. If I were to do it again I would use the dm01, which I hear should be more quiet and is more powerful. The q factor is annoying to me but should be similar to other retrofit mid drives.

I use the Shimano linkglide chain and casettes 10 speed. Highly recommended!
The chainring on the dm02 is not narrow wide and the chain will fall off in the two lowest gears. The tsdz2 narrow wide chainrings fit and I have ordered one of those from AliExpress.

Feel free to ask any questions.

 

[yabert]

I have the DM02

Thanks for the details.
It look like our display (SW102), throttle and brake sensor are not compatible with Toseven stuff
I'm really torn about choosing between: New complete BBS02 (600$cad), new hall sensor (25$), Toseven DM02+stuff (800$) or even TSDZ2B (is it work at 52V?).

 

[Vbruun]

Throttle and display were included in my package though. I think all 3 options are viable and its probably mostly a matter of personal preference what you should get