Dahon Mu P8 that can pull 200kg up 20% grade

[hypersphere]

I live in a hilly place. I commute to work via light rail. I have two kids. It rains a lot. My favorite grocery store is a 15% grade between me and there on an otherwise pleasant ride. My driveway is 20% grade. I'm fine with a top speed around 16 mph. I require torque sensing, love regenerative braking, and hate PAS. (PAS makes me lazy, torque forces me to exercise). I can comfortably pick up 50 pounds at most.

So I did what every normal person would do and built a spreadsheet of options which allows me to tweak mass and grade to see the effect of various bike + motor combinations. I calculated the force needed cannot be achieved by the front wheel alone even at 15% grade (even assuming 1.0 coefficient of friction) on most bikes (bakfiet -- yes). I have 68 combinations of bikes and motors.

I could get a Tern GSD. It's hard to maneuver it on the train and in/out the elevators. The hills will mean lots of brake usage. I have a similar bicycle and I've dropped it a couple times on the wet hill getting out of the house, and when it's 80 lbs + kids up high you can't stop it once it starts falling.

So after a lot of thinking, I'm going with a small bicycle with lots of torque plus a weatherproof trailer. I own a Dahon Mu P8 already and it can prove the concept; or maybe be good enough forever. The trailer can be parked at the kid's school if I'm going to work or it can go on the light rail like strollers can.

Without my requirements of both high grade and large mass, there would be a range of workable solutions. Alas, there is not.

I considered the Tern Vektron. It doesn't have enough torque to get out of my driveway at maximum load. I could add a second motor, which would give regenerative braking, but it'd put the weight over 50 lbs and either a separate battery pack or deal with a proprietary one. I do generally like that they come with lots of attachment points and for just kids or just groceries it could do it. Plus it's kinda cute.

I just learned about the Bike Friday Bosch-integrated bike with belt drive in development. Only downside is no regen but less chain maintenance. Could add an eZee front hub for regen but for just regen probably isn't worth it. Given the price... I'm going to try it with the Dahon first. But I'm going to keep thinking about it .

Okay, so that gets me down to modifying the Dahon. It comes stock with a stock 52 tooth chainring and 32t largest cog. It's going to need me at the heaviest to get up the hill, so that leaves me with two feasible options for new front cranks: 64/104 BCD (3-speed mountain bike) or 130 BCD (2-speed road bike). A 64 BCD crank can get to 26T (or maybe less, but I don't need less); a 104BCD to 30T; and 130 BCD to 38T.

Some insightful person was, IIRC, writing a guide about hub motors and said something like "mid-drives can be under-powered because they always run at an efficient RPM. Hub drives must be powerful enough to keep you going fast enough they don't stay at an inefficient speed". Another insightful person, in a post about oil cooling motors IIRC, was dishing advice "if you're pushing your motor to the very limits, just pick a bigger, more appropriate motor". Prior to these revelations, I had been playing with the All-Axle, GMAC, and RH212 motors in the Grin motor simulator. I realized that, sure, the All-Axle does generate enough torque to keep me going around 4 mph up a 15% grade with 200 kg load, but it's turning 65% of its input power straight to heat. According to the simulators it won't quite overheat, but... while I'd rather my bike be 6 lbs lighter, it just needs to be below 50 lbs and the RH212 can both absorb more heat and is 10 points more efficient (33% more than All-Axle). And the GMAC 10T runs 2 mph faster than the RH212 and only 35% goes to heat. Between the motor simulator and trip simulator, I find the GMAC is going to spend most of its time in the 60-80°C range (despite the higher efficiency), while a RH212 will be 30-50°C. Would it be nice if someone made a sealed version of the GMAC that could be oil-cooled? Sure, but it can handle those temperatures, it'll save 6 lbs, and it'll use less battery.

You may have noticed I didn't choose a chainring BCD. Well, if I've chosen my motor as GMAC 10T and if I'm going to use a 48V battery with a Baserunner L10, then a cold controller would get me 8.7 mph. A 52T chainring with 32T rear at 80 RPM cadence is 7.2 mph. So I don't even need a two-speed road bike front; unless I let the controller get too hot. I have plenty of Al heatsink extrusion, so that should be preventable. But I'm solving this real time and now know I'm getting the erider 5-arm spider and not the 4-arm version, because it's a hub motor and I've got to keep the speed up. And if it broke down then I'm probably walking it up the hill (and making the kids walk too!) and on a flat I can still cycle it.

So I'm going to place an order for GMAC 10T + CA + baserunner L10 + erider 5-arm BB torque sensor + accessories, plus another for BB tools and such. Still need to pick out a BMS for 15S LiFePO4 and figure out where to buy some cells.

So there won't be much to add here until I order all those things and they get delivered.

 

[neptronix]

Yeah GMAC will pull like hell.

If your bike has 20" wheels then the all axle will climb much better if you don't starve it for amps with a baserunner.



The GMAC 10T overheats about the same time in this same scenario despite the better efficiency.. the reason is that it has 1/2-1/3rd the heat shedding capability of the DD.

RH212 would be the best climber of the grin lineup and in the same scenario takes 4 minutes to overheat vs the all axle's 2.3 minutes. it would also have the nicest regen on the way down.

If your bike has bigger than a 20" wheels then i'd really suggest a mid drive for these kinds of hills.

I'd strongly recommend a DD for this purpose.

 

[Papa]

A 40" wheelbase +GMac +20" wheels +15% hill =flying front wheel. And that's without cargo or, gawd forbid...a trailer.

 

[offGridDownUnder]

I put an All-Axle on the front wheel of a Tern S8i belt drive. It was quite capable of breaking tire traction going up hill. Given your requirements, I would not rely on a front drive.

The second link in my signature is to my electric Cruzbike. I put an All-Axle in the rear - the pedals and drive are in the front. A person I know here put a Gates belt and Rohloff hub in the front of their Cruzbike and rode it around Europe - you could have everything. But it is a semi-recumbent front drive bicycle. They do tend to have all the hill climbing records due to the geometry (without motors) - it's just not what you might be used to.

But mechanically it's perhaps the best platform for you.

 

[hypersphere]

A 40" wheelbase +GMac +20" wheels +15% hill =flying front wheel. And that's without cargo or, gawd forbid...a trailer.

GMAC 10T + baserunner should have 78 Nm of torque at the rear axle should only be able to apply roughly 78 Newtons of upward force at the front axle. I think I have 240 N downward at 20% grade. I plan to install the battery and other cargo in the front carrier which should grant another 85 Newtons downward to the front axle. Also I fully anticipate having to lean forward a bit here.

 

[hypersphere]

I put an All-Axle on the front wheel of a Tern S8i belt drive. It was quite capable of breaking tire traction going up hill. Given your requirements, I would not rely on a front drive.

Definitely, I calculated 380N of force just to keep from sliding backward and there's nowhere near enough front traction for that.

The second link in my signature is to my electric Cruzbike. I put an All-Axle in the rear - the pedals and drive are in the front. A person I know here put a Gates belt and Rohloff hub in the front of their Cruzbike and rode it around Europe - you could have everything. But it is a semi-recumbent front drive bicycle.

I'm considering some type of recumbent for longer trips, but something much larger with the pedals out front isn't going to be fun to maneuver in the crowded train (even my tiny folder sticks out into walkway because of the poor train design) and it doesn't fit in my car's trunk.

 

[hypersphere]

Yeah GMAC will pull like hell.

If your bike has 20" wheels then the all axle will climb much better if you don't starve it for amps with a baserunner.

The GMAC 10T overheats about the same time in this same scenario despite the better efficiency.. the reason is that it has 1/2-1/3rd the heat shedding capability of the DD.

RH212 would be the best climber of the grin lineup and in the same scenario takes 4 minutes to overheat vs the all axle's 2.3 minutes. it would also have the nicest regen on the way down.

I'd strongly recommend a DD for this purpose.

The All-Axle+baserunner is already a couple hundred $$$ more than the GMAC 10T; phaserunner or other controller is going to be even more. Right now I'd prefer to buy from one vendor with parts known to work with each other, which limits me to the Phaserunner, which is only 10A more. So if I compare them:



The All-Axle is wasting twice the battery. Yeah, I know the GMAC is going to stay hot, but it shouldn't be too hot. I'd also need a battery capable of more power too.

For example, on a continuous 15% grade with maximum speed of 23 kph and an All-Axle SLW+Statorade with a monster 180A continuous controller would reach 125°C in 1.8km:


but GMAC 10T with baserunner L10 would go 2.3 km with far cheaper components and 30% less battery:



Also the 15% grade I need to traverse is just a short section. A couple hundred meters at most. It's mostly single-digit-grade rolling hills for the rest and any of these options can handle it.

I agree the RH212 would be better, maybe I'll find gear noise annoying? But 6 extra pounds doesn't seem worth it because I mostly only need the power for short bursts, which sufficiently makes up for the poor heat dissipation of the GMAC. But if I wanted to sustain faster speeds than 16mph then hands-down I'd go for the RH212 with the better heat dissipation and higher torque than All-Axle.

But a bit of gear noise instead of either +$$$ or 6 more lbs mass seems a fair trade.

 

[hypersphere]

If your bike has bigger than a 20" wheels then i'd really suggest a mid drive for these kinds of hills.

I concur on this point for sure. Because I want regen, I mostly evaluated hubs at lots of different sizes and they just don't make sense with the larger wheels because it's niche. Someone could make a GMAC with twice the internal gear ratio, but the only people who would need it are lightweight cargo haulers that are irrationally in love with regen and we're a very small club. The Tern GSD is just so good IMO (belt-drive has 19 wheel-inches in the lowest gear! with a mid-drive!) there's just not a lot of market left for anybody else .

 

[neptronix]

The All-Axle+baserunner is already a couple hundred $$$ more than the GMAC 10T; phaserunner or other controller is going to be even more. Right now I'd prefer to buy from one vendor with parts known to work with each other, which limits me to the Phaserunner, which is only 10A more. So if I compare them:

View attachment 359200

The All-Axle is wasting twice the battery. Yeah, I know the GMAC is going to stay hot, but it shouldn't be too hot. I'd also need a battery capable of more power too.

Just one note..
In the above scenario, both motors are being starved for amps, moreso the All Axle. This will reduce efficiency a lot and contribute to rapid overheating. It unfairly skews the comparison, too. Notice that the amp draw is pegged to the controller's limit for both motors.



52v seems to help... but we need an extremely stout battery.

But a bit of gear noise instead of either +$$$ or 6 more lbs mass seems a fair trade.

Hey if your 15% grade sections are short & not sustained.. you are right that the GMAC would be a good choice.

Unfortunately the phaserunner is gonna choke both motors up hill though.. the phaserunner is also not great at running a MAC at high RPMs. You will probably not hit this high RPM problem with this configuration though.

I haven't tried this yet, but i believe a VESC would be much better at controlling such a high speed motor ( VESC is primarily designed to drive motors that have >10x the RPM ). A typical VESC ( example: spintend's ubox ) will be rated for '100A'... so, should be able to handle these big bursts of power.

 

[neptronix]

Someone could make a GMAC with twice the internal gear ratio, but the only people who would need it are lightweight cargo haulers that are irrationally in love with regen and we're a very small club. The Tern GSD is just so good IMO (belt-drive has 19 wheel-inches in the lowest gear! with a mid-drive!) there's just not a lot of market left for anybody else .

There is such a motor, but it's very heavy and only a bit more powerful than the MAC.. it also has poor efficiency on the flats.. i wouldn't recommend it!

 

[Chalo]

To be fair, I think this is a good case for two hub motors. That would make it easier and cheaper to double the torque and double the controller amps.

 

[neptronix]

Yeah.. but a pair of 4lbs/1.7kg hub motors with dual reductions is producing insane amounts of heat from input power at a 20% grade..



Can't reach the target speed though, we need another kW or so!

On the other hand if you pedal like a madman, you can make it up the hill with such a configuration.



Impressive showing honestly, we can climb 12% grade at a meager 12.8mph and actually get away with it.

 

[hypersphere]

Impressive showing honestly, we can climb 12% grade at a meager 12.8mph and actually get away with it.

Those two for my scenario, 200kg@15%, is surprisingly comparable to the single-motor GMAC:



But I'd have to lock the clutches, and I'm not sure any are still made, can't find rears, plus I'll need a new front fork (Dahon fork is 75mm OLD).

 

[hypersphere]

To be fair, I think this is a good case for two hub motors. That would make it easier and cheaper to double the torque and double the controller amps.

If I did two I think it'd be mid-drive performing the majority of the energy and a front hub at max power + regen.

 

[neptronix]

Those two for my scenario, 200kg@15%, is surprisingly comparable to the single-motor GMAC:

Yeah. Those dual gear reductions motors respond well to having their load split.

You should see two shengyi SX2 fast winds.
Puts a single MAC to shame.. and has nicer efficiency!

Makes me question the necessity for mid drives, honestly!

 

[crunchysteve]

(PAS makes me lazy, torque forces me to exercise). I can comfortably pick up 50 pounds at most.

I love my PAS, but I dig what you're saying. From a technical perspective it makes sense. If I ever upgrade the rig I have, I think I'll give torque sensing a try, at least before I decide on sit or change.

The lazy thing is more a choice of how to use PAS, though, I have to reach "strain goals" before I allow myself a level-up. "Laziness" was always my excuse for resisting ebikes altogether, I was wrong, and PAS was what proved to me I was wrong.

 

[crunchysteve]

No judgement, "Ride what you bring" is my motto, but the bike industry has always oversold mid-drive. The problems for me are the expense, the nonstandard frames (although the Chinese sidestep this one, thank you Bafang) and the elephant in the room... lower centre of gravity?!

My Reid Urban X2 has a BB drop of 50mm (2"), lets say a Bafang puts a little over half its mass another 25mm below that, that's a 75mm drop on axle height. The largest contributor to COG on a bicycle is the rider. Their mass is centred around 300mm (1') above the saddle. A sporty male, 70kg on a 1m (and a bit?) moment arm? Compared to, ah, I dunno, 5kg on a 3 inch shorter moment arm with a mid mount?! 10" inches/200mm off the ground? Compared to 5x the mass at 5x the height above ground. Give me a lever, Archimedes.

In 50 actual years of riding on the road, I've rarely seen a less honest claim by the bike industry... better centre of gravity? Only when the bike is parked!

The non-standard frame mounts may already have an ISO number and be a standard, I have read the industry wants this. The industry likes standards, that's why BSA won't die, the venerable pommie BB shell is just too easy, hence Bafang use it for even their mid mounts. Torque sensing is easier to do with a middie. The expense, when I look at the market, was my biggest obstacle to accepting eBikes in the first place.

But that lower COG thing? Ask a physics teacher. "A kilo off the rider's belly is worth 10 off the frame," as my grade 9 science teach would say to "fat kid Crunchy" as I locked my "steed" to the hitch rails in the school's bike shed.

 

[crunchysteve]

And, I wasn't known as "Crunchy" that far back.

 

[hypersphere]

The Q factor is what I dislike about standard BB mounts.

I have made some progress. I bought two used 4s8p A123 packs from Battery Hookup, have cut the tabs down the middle to make is 8x 4p which I'll weld some nickel tabs on to make 2x 4s4p and solder on a few wires to make 8s4p.

However, I rented a Tern Quick Haul recently in addition to commuting without my kids with the Dahon unpowered.

So, I live in Seattle, I hate taking the elevators because downtown because they always reek of piss and occasionally shit. I've liked just carrying my Dahon (at ~35 lbs) up the escalator instead but then tweaked my shoulder a bit . I'm considering only electrifying the kid wagon.

The other revelation is I may not need torque control. With the Tern I leave it on Eco mode, except for steep hills, when I (have to/want to) crank it up. So why can't I do that automatically? Well, many of the VESC controllers have a builtin IMU; how about I program them to generate the appropriate torque to countermand gravitational forces and otherwise let me just bike? Torque is constant per slope; without input torque measurement I'll still *accelerate* slower up-hill but that's like a real bicycle and that's kinda what I want. I can add weak PAS (which'll still require I get exercise). Then I don't care about my torque input at all and can save money on the torque sensor. I'm considering the Spintend controllers know for best balance of price and capabilities (notably builtin IMU). The other reason is there are only 2 analog inputs on most VESC and if using variable brake handle and throttle that doesn't leave room for the torque sensor...

 

[Guerney]

Dahon Mu P8 that can pull 200kg up 20% grade

I don't visit this forum often, I'm usually on pedelecs.co.uk, and I've only recently seen this thread. My "250W" 36V BBS01B converted 20" wheeled Dahon Helios P8 (manufactured April 2006, electrified November 2019) regularly pulled over 180kg (rider+converted bike+ trailer+cargo) up 12% grade hills for two summers with no problem, and that was when the controller was (legally) limited to 15A = 540W. I would test to see what weight it can haul up those hills, now that I have increased the controller's amperage limit to 20A = 720W, if the handling of the bike while hauling heavy weight wasn't so hairy. Also, as I discovered while dragging over 100kg of water cargo across a golf course (shortcut, they'd built a golf course into the middle of woods [I'm outraged this was allowed], and I had to empty out some water onto the golf course while watching out for ballistic golf balls - this caused golfers watching to angrily mutter into cellphones) too heavy a trailer causes the rear QR axle to be yanked out by the trailer. Small wheels are mightier than larger. My lowest gear is 52T>32T = 32.5 gear inches. I could fit a 42T chainwheel= 26.25 gear inches, but then I wouldn't be able zoom along the flats fast enough with 52T>11T = 94.5 gear inches, when I am not hauling a bike trailer. Unfortunately, 20% hills are far too far far away for testing. I do sometimes worry too much motor torque could cause my folding bike to acquire a permanent fold. A couple of years ago I kept hearing a mysterious creaking whenever the motor applied high torque. I eventually ascertained it was emanating from within the folding mechanism of the middle fold. It hadn't been lubricated since leaving the Dahon factory - fixed with a couple of drops of 4 in 1 sewing machine mineral oil. Personally, I don't think it'd be wise to fit a more powerful mid drive to my folding bike. I noticed Tongsheng and Bosch lower the torque of their motors installed on 20" wheeled folding bikes. For example, the Tongsheng on the Dahon Uno is only 200W. The shrunk by 20% Bosch motor on the Tern Vektron P7i only outputs 40Nm.