The Pusher

[51owRider]

Hi All, new user soaking up the wealth of information here, ES is a great resource.

I've been riding longish distances on my road bike for many years but my age is catching up with me and I'm thinking about adding a motor to my single-wheeled trailer. The trailer has a 16-inch wheel and 100 mm alloy dropouts. I'd like to get about 150 W of assistance and a range of at least 250 km or 10 hours.

A route that I regularly travel is 250 km, mostly flat but sometimes windy. I average 24 to 28 km/h and 150 to 200 W depending on the wind, weight of stuff in panniers or trailer, time of year etc. It would be wonderful to complete this journey a bit faster and without suffering too much the next day.

First idea was to get a direct drive hub motor for the reliability and longevity (and possibly regen braking, later on) but what I could find was either too expensive or too heavy. At the moment I'm looking at this AKM-100H geared motor at 328 RPM, am I correct in speculating that if I drive it at 48V it can deliver some power at 35 km/h on a 16" wheel?

Not sure if I'm using the motor simulator correctly but it seems a 40Ah 48V battery would be more than sufficient:


Can anyone recommend a suitable controller?

Any input is appreciated.

 

[amberwolf]

Some thoughts about pusher trailers:

If they don't have enough weight on them, there won't be enough traction for the motor to help you. How much weight is enough depends on the situation and the weight/balance of whatever the trailer is pushing, and how the pusher is connected to what it's pushing. You may have to experiment to ensure traction in all the situations.


A "heavy" motor and battery may be a benefit for those situations, depending on the other stuff the trailer has to carry.

But a DD motor will add some drag (sometimes significant) when not powered, or when your speed exceeds that of the assist capability.

A geared hub minimizes this by having a roller-clutch that only engages the wheel to the motor when the motor spins faster than the wheel.

 

[51owRider]

Thank you for the reply, Amberwolf. The trailer is a Topeak Journey, it weighs 6 kg and attaches to the rear axle. When lightly loaded it is barely noticeable but when carrying more than 25 kg the handling becomes - not unmanageable but - unpleasant as it wrestles the rear triangle. This is why I'm trying to keep the weight low.

Good point about the traction, but I'm not too worried at these low power levels, also, I might not even turn on the motor below a certain speed. Slow starts and no assistance on steep hills are not a problem.

A DD hub is not out of the question but needs a good reason to justify the added weight, and I need to actually find one that is not a full cast moped wheel.

 

[99t4]

That's a very nice trailer. However it might not work so well as a pusher trailer as the wheel is so far behind the load. Traction could be a problem. Better would be something like a kiddie trailer or utility trailer.

 

[51owRider]

It's counter-intuitve, but in my experience two-wheeled trailers are much less stable than single-wheeled ones, especially at higher speeds, one side hits a pothole or an uneven section of the surface and the trailer receives a lot of momentum on the roll axis... you need a really wide wheelbase to avoid this which is impractical. And it's much harder to find the best line on the road as the wheels are not tracking the same line as the bicycle.

They do have the advantage of higher load capacity because you can place the load directly above the axles though.

 

[portals]

It would be wonderful to complete this journey a bit faster and without suffering too much the next day.

Have you considered one of these for the bike rear wheel, might well fit the bill and it's removable.
https://www.rubbee.co.uk/product/rubbee-x/

I'm considering one for my son (14) for when we go on longer trips so he can keep up with me (1500W rear hub 52V 17.5Ah battery)!

 

[amberwolf]

When lightly loaded it is barely noticeable but when carrying more than 25 kg the handling becomes - not unmanageable but - unpleasant as it wrestles the rear triangle. This is why I'm trying to keep the weight low.

Good point about the traction, but I'm not too worried at these low power levels, also, I might not even turn on the motor below a certain speed. Slow starts and no assistance on steep hills are not a problem.

A DD hub is not out of the question but needs a good reason to justify the added weight, and I need to actually find one that is not a full cast moped wheel.

If traction isn't going to be an issue but weight is, then skip the DD hub and use a geared one, for the torque-to-weight ratio and the "freewheeling" when it is not powered.

However, before you get anything, do this test and see how you feel about it:

Load up the trailer with enough weight to "push you around" from behind when you try to brake or change directions and ride down the longest steepest hill you can. This is the closest physical simulation you can easily do to having a motor push you around under similar situations. If you don't like the way it handles this way, or it actively causes you problems, you may wish to reconsider the pusher and instead put the motor on the bike itself (front wheel is simplest, but may have traction problems under conditions where the trailer pushes "down" on the rear axle in a way that lifts the front wheel--if that doesn't happen with this hitch design then there's no worries about this potential problem).

It's counter-intuitve, but in my experience two-wheeled trailers are much less stable than single-wheeled ones, especially at higher speeds, one side hits a pothole or an uneven section of the surface and the trailer receives a lot of momentum on the roll axis... you need a really wide wheelbase to avoid this which is impractical. And it's much harder to find the best line on the road as the wheels are not tracking the same line as the bicycle.

They do have the advantage of higher load capacity because you can place the load directly above the axles though.

Yes, the two-wheeled trailers are harder to deal with for all those reasons; the only really good ones I've ever built use wheels entirely outside the load, with axles above the deck so the deck "hangs" from the axles. Very stable, but wider than otherwise.

My next trike will be built the same way (rather than having the axles even with or under the deck); it's biggest disadvantage is that it is enough wider to become a problem passing between some things (including standard doorways, precluding taking it inside a number of places I would otherwise want to, and including some bike path bollards, though most are wide enough spaced to allow it; I ride almost always on roads so it's not too big a deal but there are rare times the paths are useful on long trips in some parts of the valley).

 

[51owRider]

Have you considered one of these for the bike rear wheel, might well fit the bill and it's removable.
https://www.rubbee.co.uk/product/rubbee-x/

I'm considering one for my son (14) for when we go on longer trips so he can keep up with me (1500W rear hub 52V 17.5Ah battery)!

Thanks for sharing, interesting concept! I like the idea of being able to easily remove it and have a clean light bike. Wondering how "hackable" it is, e.g. powering it with anything other than their proprietary batteries, controlling it by anything other than their cadence sensor... Could not find a teardown of the device, would be interesting to see the insides as well as long-term reviews. I guess it must be quite noisy and in wet weather road grime will be collected by the roller.

 

[51owRider]

If traction isn't going to be an issue but weight is, then skip the DD hub and use a geared one, for the torque-to-weight ratio and the "freewheeling" when it is not powered.

Yes, the geared ones appear to be the best candidates now, but I'm still unsure which one is able to produce 150 W at 450 RPM without magnets flying off, and at what voltage, I need to educate myself more on this.

However, before you get anything, do this test and see how you feel about it:

Load up the trailer with enough weight to "push you around" from behind when you try to brake or change directions and ride down the longest steepest hill you can. This is the closest physical simulation you can easily do to having a motor push you around under similar situations. If you don't like the way it handles this way, or it actively causes you problems, you may wish to reconsider the pusher ...

Yes, I do this test on an almost daily basis, I carry stuff around on my local hills on dirt roads. This trailer is the best way I found for carrying loads so far, panniers are okay for lighter stuff and I used a trailer which attached to the seatpost, it was much worse.

and instead put the motor on the bike itself (front wheel is simplest, but may have traction problems under conditions where the trailer pushes "down" on the rear axle in a way that lifts the front wheel--if that doesn't happen with this hitch design then there's no worries about this potential problem).

Well, I'd like to avoid modifying my bike for now. If the pusher trailer turns out to be a failure, I'll have to reconsider.

Yes, the two-wheeled trailers are harder to deal with for all those reasons; the only really good ones I've ever built use wheels entirely outside the load, with axles above the deck so the deck "hangs" from the axles. Very stable, but wider than otherwise.

Clever solution! Have you posted pics of them?

My next trike will be built the same way (rather than having the axles even with or under the deck); it's biggest disadvantage is that it is enough wider to become a problem passing between some things (including standard doorways, precluding taking it inside a number of places I would otherwise want to, and including some bike path bollards, though most are wide enough spaced to allow it; I ride almost always on roads so it's not too big a deal but there are rare times the paths are useful on long trips in some parts of the valley).

Sounds great, is this a recumbent trike or more like a cargo trike? I'm really careful in turns passing between bollards on bike paths in the cities with my narrow but long trailer, I imagine with the wide wheelbase you really need to slow down every time.

 

[amberwolf]

Thanks for sharing, interesting concept! I like the idea of being able to easily remove it and have a clean light bike. Wondering how "hackable" it is, e.g. powering it with anything other than their proprietary batteries, controlling it by anything other than their cadence sensor... Could not find a teardown of the device, would be interesting to see the insides as well as long-term reviews. I guess it must be quite noisy and in wet weather road grime will be collected by the roller.

There's only a few (70-something) posts mentioning it here on ES, but they may have relevant info for you:
https://endless-sphere.com/forums/search.php?keywords=rubbee&terms=all&author=&sc=1&sf=all&sr=posts&sk=t&sd=d&st=0&ch=-1&t=0&submit=Search

 

[amberwolf]

Yes, the geared ones appear to be the best candidates now, but I'm still unsure which one is able to produce 150 W at 450 RPM without magnets flying off, and at what voltage, I need to educate myself more on this.

For outrunner style geared hubs, the magnets are the inside of the rotor and held in by the rotor bell, so they won't fly off. (inrunners with surface-mounted (SPM) magnets (instead of IPM) might have this problem). Most of the geared hubs I've seen are designed this way. If you're not sure of a specific motor, there are often teardowns here on ES for them.

A catch with high RPM on geared hubmotors is that the motor itself inside will be running at 5x or more the external RPM, some like the G310 with dual reductions maybe 10x or more, and if it also has a high pole count it may exceed the "ERPM" of most controllers, so they may not be able to run it at the speed you want. One post mentioning this:
https://endless-sphere.com/forums/viewtopic.php?f=3&t=96950&hilit=lightest#p1420515

If you're looking for the lightest geared hub, there's a few threads with some discussions about "lightest" hubmotors in this list, though many of them discuss DD hubs rather than geared, there are some mentions of geared hubs in there. Some of the info isn't useful because it's old enough the motors discussed are no longer made, though.
https://endless-sphere.com/forums/search.php?keywords=lightest&terms=all&author=&sc=1&sf=titleonly&sr=topics&sk=t&sd=d&st=0&ch=300&t=0&submit=Search

The bafang G310 might be the still-made winner; I haven't looked around much as small light motors don't really do me much good anymore.
Main thread about it (there's a bunch of threads discussing and using and modifying it
https://endless-sphere.com/forums/viewtopic.php?f=2&t=92124&hilit=g310

Simulator link for it:
http://www.ebikes.ca/tools/simulator.html?bopen=false&motor=MG310_STD&cont=cust_18_60_0.02_V&wheel=20i&mass=140&hp=0&grade=0&axis=mph&batt=B4814_EZ&blue=Lbs

There's also the Q series, Q70, 85, 100, 128 etc.

Dapu also makes some tiny ones, like the one I fixed here:
https://endless-sphere.com/forums/viewtopic.php?f=30&t=114480&hilit=dapu

No experience with any of them myself; the only geared hubs I've had were by Fusin (no longer around) which I broke the clutch on one with a "normal" cargo bike, and an old Ezee on my brother's trike that has had very little use so no real feedback on that one.

I used a trailer which attached to the seatpost, it was much worse.

That's pretty much why I never ended up building this one into a pusher:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=21060&hilit=trailer

Clever solution! Have you posted pics of them?

This is the first MkIV version:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=76539

at the end of the thread I added a second set of wheels just outboard of the first to handle the extra load of hauling a piano, and started a new thread for modifying it into an enclosed version with air conditioning for the dogs (and groceries if necessary) as the MkV, which hasn't been built yet (probably won't be anytime soon since the two dogs it was primarly for unexpectedly died last year; the one I have left doesn't go places):
https://endless-sphere.com/forums/viewtopic.php?f=2&t=94215&hilit=trailer#p1379480

Sounds great, is this a recumbent trike or more like a cargo trike? I'm really careful in turns passing between bollards on bike paths in the cities with my narrow but long trailer, I imagine with the wide wheelbase you really need to slow down every time.

Heavy cargo trike. Originally designed to carry Tiny (a small Saint Bernard) without having to use a trailer, and larger grocery or other cargo loads (without a trailer), it's usage and capabilities have been expanded over time, to where it could carry two very friendly Saints side by side, though this was almost never needed (usualy I'd just use the trailer for the second one when necessary).

It started out specifically narrow enough to fit thru a typical house front door, but at some point had to be widened a couple inches or so which means it might fit thru the doorway but only if the door swings all the way (180 degrees) out of the way, which mine don't (considered rehanging them on the other side...too many other issues that would cause). So the trike does fit within most bollards...but there are places it might not that I haven't been to in a while so am not sure. I know the big trailer won't; it's basically about 4 feet wide, maybe a bit more. :lol:

Original version here:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=67833
which has evolved signifcantly over the years; this is almost the present version:

This is the thread for the new version, just concepts and working out ideas right now:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=84161&hilit=cruiser
With the most recent possible concept here:

 

[51owRider]

There's only a few (70-something) posts mentioning [rubbee] here on ES, but they may have relevant info for you:
https://endless-sphere.com/forums/search.php?keywords=rubbee&terms=all&author=&sc=1&sf=all&sr=posts&sk=t&sd=d&st=0&ch=-1&t=0&submit=Search

Oops, I should have searched before posting, sorry about that. Reading a bit more about this and other friction drives, I'm not too excited about them but I'm sure there are valid use cases.

 

[amberwolf]

No problem; I've been here a little while and tend to recall various discussions from long enough ago most people wouldn't think to or know of to look for, so I just post links to the stuff I know about or can find quickly enough.

(I spend a lot of time waking/dozing in bed with the computer right there, unable to sleep well most nights, so I read and reply here whenever my brain is working well enough to do so, if I'm not working on a piece of music...sometimes my replies are very short or "terse" links to stuff when I can't concentrate well enough to extract information from the stuff and distill it into something useful for the specific thread I'm replying to, and sometimes I ramble on and on and on and on..... )

 

[51owRider]

A catch with high RPM on geared hubmotors is that the motor itself inside will be running at 5x or more the external RPM, some like the G310 with dual reductions maybe 10x or more, and if it also has a high pole count it may exceed the "ERPM" of most controllers, so they may not be able to run it at the speed you want. One post mentioning this:
https://endless-sphere.com/forums/viewtopic.php?f=3&t=96950&hilit=lightest#p1420515

Thanks. On this thread they also mention "magnets being flung off the motor ring" at high RPM

Still processing all the other info you kindly provided...

 

[51owRider]

There's also the Q series, Q70, 85, 100, 128 etc.

Excellent, I found the Q100 is still being sold and is also listed in the simulator, although overheating is not modeled which may be an issue as I'd have to over-volt it to have a high enough RPM (if I understand correctly). The next question is whether or not it needs a fancy expensive controller like the frankenrunner to work at high speed.

But, what if I - instead of trying to solve this high RPM puzzle - extend the trailer's forks to accept a 20" or even a 26" wheel?



Planned extension with new dropouts is in green.

The existing dropouts are 8mm thick aluminium, I have access to a Tig welder and it doesn't look like a complicated job. Okay, maybe the 26" axle is a bit too far away from the new joint.

 

[51owRider]

Reading though your threads, Amberwolf, jaw dropped! You build really impressive vehicles and it's great that you document it all. Hats off. Really sorry for your loss of the dogs and the fire, nice to see that you're back on your feet after such an experience.

 

[amberwolf]

Thanks. On this thread they also mention "magnets being flung off the motor ring" at high RPM

Then it must be an inrunner, which makes it different than most of the geared hubmotors.

 

[amberwolf]

Excellent, I found the Q100 is still being sold and is also listed in the simulator, although overheating is not modeled which may be an issue as I'd have to over-volt it to have a high enough RPM (if I understand correctly). The next question is whether or not it needs a fancy expensive controller like the frankenrunner to work at high speed.

That, I don't know. It depends on the speed the motor ends up at inside vs teh ERPM of the controller, so if you're willing to just try stuff out, buy the Q100 along with a cheap controller capable of the voltage you want, from the same place, and see what happens.

If it doesn't work you can always resell it here on ES and try a different controller or approach.

But, what if I - instead of trying to solve this high RPM puzzle - extend the trailer's forks to accept a 20" or even a 26" wheel?

That...might change the handling. I don't know exactly what it would do, not having used much in the way of single wheel trailers I haven't compared long ones vs short ones.

If you have access to any junk bike frames you could find one wiht a rear triangle of the right size and shape to cut off and hose clamp to the trailer frame such that it puts it's bigger wheel dropouts right where you need them to be, to test the idea without cutting or modifying the trailer itself. Or a nonsuspension fork could be clamped to the frame even easier, if you have one of the right curve or shape to angle up to the right point; probably have to cut the steerer off but depends on how the bottom of the trailer is designed and the angle required.


Some definite changes that will happen with a bigger wheel:
--rides better over nonsmooth surfaces.
--torque from the motor will be lower but with proportinally higher speed (essentially changing the gear ratio with the bigger wheel)
--easier to lace the motor into a stronger wheel (or buy one already laced, if you don't mind the often-poorly-built OEM wheels with too-thick spokes for the rim used)
--easier to find a wider variety of tires for the wheel that might be more suitable for your usages

 

[amberwolf]

Reading though your threads, Amberwolf, jaw dropped! You build really impressive vehicles and it's great that you document it all. Hats off.

Thanks--I mostly experiment until stuff works (the same way I do most things, including life, music, etc), and I learn schtuff. Sometimes things don't work out as expected and I learn schtuff (the hard way).

If I had money I'd pay someone to build what I want, but I don't think that will ever be an option, so....

Really sorry for your loss of the dogs and the fire, nice to see that you're back on your feet after such an experience.

More or less...nothing will ever be quite the same (including me), but life goes on, things change (whether we want them to or not), so I adapt. I definitely recommend against the experience.

At least for now I still have the Perfectly Normal Schmoo (but a Very Strange Dog), Jelly Bean:

 

[51owRider]

I tested a large wheel exactly as you suggested. Rigged up an unused carbon fork to the trailer using some pieces of wood and rubber and tightening everything with straps. The whole thing turned out to be much more solid than what I expected. Popped in a 700 wheel, loaded up the trailer with 30 kg of metal and water and went for a short ride. Except for being a little flexy due to the stretching of the straps the handling was excellent, maybe even nicer than with the stock wheel, but this might be just my imagination. The center of mass is now a little bit lower compared to the axle height, I'm sure this helps. So, in conclusion, big success, thanks again!

 

[gobi]

Op,
I have no experience in a pusher trailer, my comments below is from a point of simplicity of design:

Add a small gear hub on the rear wheel, battery in the trailer (check out kingfish's posts).

How are you planning on charging the batt?
I have thought of modular design, where I can take off the batt and take with me to charge indoor overnight.
In Michigan, we have freezing temps nearly 9 months of the year, so leaving a batt outdoor is not a option sometimes.

 

[51owRider]

Hi Gobi, yes, a small geared hub is the current plan, it's just that I haven't found a motor that fits all the requirements with the 16" wheel, so I'll probably modify the trailer to use a larger diameter wheel. 2x ~1 kWh removable batteries and a controller on the trailer bed, wireless connection to the head unit. I keep all the bikes inside the house, charging in freezing temps isn't going to be an issue.

 

[torker]

If you could find a small dd . I love my dd regenerative. It is my only rear brakes. I only run 16 mph though. It will throw me forward a little. You may have traction issues though. Unless you can adjust it. Trailer brakes would be the bomb...

 

[amberwolf]

2x ~1 kWh removable batteries and a controller on the trailer bed, wireless connection to the head unit.

Wireless...can be done, but test to ensure LOS = system shutdown. Don't want uncontrolled runaway operation because the head control unit battery died or someone with an overly powerful CB/ham/etc radio CQd as they passed you.

Eskate (Esk8) stuff (like VESC) can do wireless controls, and there are a number of VESC variants. Flipsky seems the most popular one here on ES for OEM-built rather than custom, that is also capable of driving ebike stuff vs just RC-motor stuff at lower voltages like some esk8 is limited to. There are a number of variations (like Cheap-FOCer, etc) here on ES with different feature sets than available in OEM-built versions.

 

[amberwolf]

I tested a large wheel exactly as you suggested. Rigged up an unused carbon fork to the trailer using some pieces of wood and rubber and tightening everything with straps. The whole thing turned out to be much more solid than what I expected. Popped in a 700 wheel, loaded up the trailer with 30 kg of metal and water and went for a short ride. Except for being a little flexy due to the stretching of the straps the handling was excellent, maybe even nicer than with the stock wheel, but this might be just my imagination. The center of mass is now a little bit lower compared to the axle height, I'm sure this helps. So, in conclusion, big success, thanks again!

That's how my brain works....

Just remember if you end up TIGing the extensions to the frame, you may need to re-heat-treat the frame and extensions to regain the original strength.

Optionally you could build a secondary frame with dropouts on it, including it's own triangulation, that is shaped to match the top of the trailer frame's dropout and rear frame section, then you can use clamps of the appropriate type to secure it to the frame. If correctly triangulated from the dropouts to the forward part of the extension frame, it could be strong enough even without heat treating, or it could be made of cromoly steel (old bike frames, for instance) and not need to be heat treated like the aluminum alloy would.

Something like this: