Why are geared motors worse than direct drive for climbing hills?
- Thread starter eecsflyer
- Start date
So here is the conundrum I see on the motor simulator. Even at low speeds, the DD motor has the same range as the geared hub motor.
https://www.ebikes.ca/tools/simulat...ot_b=45&mass=120&mass_b=120&hp=100&bopen=true
Going up at 6.5 MPH on a 7% grade, the HS3525 is at 66% efficiency and has a 13 mile range, while the G310 is at 64% efficiency and has 12 mile range. Neither is at full throttle. The G310 is using 160W while the H3525 is using 156W.
https://www.ebikes.ca/tools/simulat...ot_b=45&mass=120&mass_b=120&hp=100&bopen=true
Going up at 6.5 MPH on a 7% grade, the HS3525 is at 66% efficiency and has a 13 mile range, while the G310 is at 64% efficiency and has 12 mile range. Neither is at full throttle. The G310 is using 160W while the H3525 is using 156W.
goatman
10 MW
ive flared open the v3 torque arms, on your fork the motor slides in horizontally from the front and then the whole weight of the trike is being held up by a half inch? of metal ( is it aluminum?). so any slop in the torque arm when hitting regen could crack the fork. the phaserunner can lessen the torque ramp up for throttle and regen, easier on the torque arms and fork.eecsflyer said:
I don't know if anyone can bring that picture of the fork into the thread so people can see it, my laptop and windows 10 is new to me and for some reason I cant paste it here?
goatman said:
Interesting. I guess that might be another argument for going for a less powerful geared motor, although I wonder if it is the torque at low speeds that might flare open the dropouts? Maybe the best solution is to use two torque arms? Only problem is I don't think I can get one on the derailleur side?
You can find a picture of the dropouts here. Indeed, not a lot of metal holding the bike up..
https://photos.app.goo.gl/SzrRSrhANPfb5ReS7
dogman dan
1 PW
One good TA will be ok.
Re the sim and efficiency,, Both geared and DD hub motors are typically chosen to work efficiently at 15 mph on up. Because many want to ride at 20mph, or faster. Slow don't sell for shit, and they are in the business of selling this stuff, not using it.
Remove the hill, and 6 mph riding is perfectly safe, and very very efficient. Its just if you overload up a hill you get into melting the motor.
So despite the internal gear reduction inside the geared motor, both have windings chosen with more than 6 mph riding under heavy loads. So they get hot and melt eventually, if you ride up steep hills loaded to the point where 6 mph is your max speed.
But again, and again, and again,, if you don't overload, then you easily zip up 8% grades at 15 mph, meaning you can climb a hill that is many miles long at that speed, with pedaling I mean. Long as you can pedal up to 15 mph, you did not overload it. Sim is great, but I did the road tests, and have climbed 6-8 % grade hills 10 miles long. Rocky mountain pass shit. Shorter hills at 16% grade still go with the hubmotor, but obviously not for 10 miles. But if you get up it in 5 min, its all good on that steep part.
If you must overload, and you must climb long hills, and you cannot use a mid drive, then there is a solution that greatly improves the result. A low rpm motor, is capable of doing the same efficiency at 12mph, as the regular motor at 15 mph. Then to improve even more, make it a 20" wheel. The result will be a slow bike nobody can sell. But at least it won't melt on you.
I made one, with a DD motor. 18 mph top speed, but its max load was 400 pounds, vs the same motor in regular wind and 26" wheel, which max loads at about 300 pounds, for hills. Max weight much higher if flat riding only. I talk about max loads for the rocky mountains.
Re the sim and efficiency,, Both geared and DD hub motors are typically chosen to work efficiently at 15 mph on up. Because many want to ride at 20mph, or faster. Slow don't sell for shit, and they are in the business of selling this stuff, not using it.
Remove the hill, and 6 mph riding is perfectly safe, and very very efficient. Its just if you overload up a hill you get into melting the motor.
So despite the internal gear reduction inside the geared motor, both have windings chosen with more than 6 mph riding under heavy loads. So they get hot and melt eventually, if you ride up steep hills loaded to the point where 6 mph is your max speed.
But again, and again, and again,, if you don't overload, then you easily zip up 8% grades at 15 mph, meaning you can climb a hill that is many miles long at that speed, with pedaling I mean. Long as you can pedal up to 15 mph, you did not overload it. Sim is great, but I did the road tests, and have climbed 6-8 % grade hills 10 miles long. Rocky mountain pass shit. Shorter hills at 16% grade still go with the hubmotor, but obviously not for 10 miles. But if you get up it in 5 min, its all good on that steep part.
If you must overload, and you must climb long hills, and you cannot use a mid drive, then there is a solution that greatly improves the result. A low rpm motor, is capable of doing the same efficiency at 12mph, as the regular motor at 15 mph. Then to improve even more, make it a 20" wheel. The result will be a slow bike nobody can sell. But at least it won't melt on you.
I made one, with a DD motor. 18 mph top speed, but its max load was 400 pounds, vs the same motor in regular wind and 26" wheel, which max loads at about 300 pounds, for hills. Max weight much higher if flat riding only. I talk about max loads for the rocky mountains.
goatman
10 MW
doctor bass still selling torque arms?
https://endless-sphere.com/forums/viewtopic.php?f=31&t=29129&start=525
https://endless-sphere.com/forums/viewtopic.php?f=31&t=29129&start=525
dogman dan
1 PW
Pretty easy to make custom torque arms, if you have a few basic tools. Drill, file, angle grinder, and a pair of vise grips.
dogman dan said:
Can't entirely agree with that statement. If all you need is a one-way piece of metal keeping from stripping the dropouts during peak accelerations of geared hubs, than yes, it's pretty easy to do yourself.
Now if you move to DD hubs with Regen, you need to have a perfect fit between the TA and the axle, otherwise the axle will wiggle back and forth in the torque arm and create more an more play until the TA is basically useless - or the axle destroyed.
In my experience such a fit is quite hard to achieve with hand tools. I made around 2 dozen TA myself, and none of them was tight enough to achieve 0 play, especially on cheaper motor with lax machining tolerances of the axle. Smart designs have some sort of clamping mechanism, which hold the axle perfectly in place but this usually requires a very wide piece of metal, and some drilling skill.
So what I recommend now is simply ordering the parts at some online laser cutting website. Surprisingly cheap, and perfect tolerances.
Draw the parts around 0.05mm narrower than your axle width (usually around 9.85mm total, depending on your motor) and than (after reception) slightly grind away the sharp edges on both sides of the torque arm opening. Once everything has been double checked to see if it fits, use a light hammer and nail the TA slowly down the axle. Now you have a torque arm with garanteed 0 play.
DogDipstick
100 kW
dogman dan said:
One thing alot of peeps forget... to mention... is the hardness.
IDKY but Im a metalhed, and I always gotta mention.
Just carbonize it ( the steel.. ).. with a acetylene torch. Make it 100X harder ( case hardened too, so stiff core and hard strong shell).... heat it up red and shoot burnt acetylene ( No O2, turn the O2 off for a carbonizing flame) over it, and let the red hot metal absorb the burnt carbon from the torch ( acetylene burning so rich it produces carbon in atmosphere without regulated O2 feeding the non-stoich lambda of the flame.. creating a carbon shell of hardened metal... ) and you wont ever drill, marr, file,, cut, or break that thing again. Or use the powder case hardening compound: works the same way.
I dont know ( read: highly doubt) anything "lasercyut" is strong enough ( you know the power laser take and what they cost? Anyone who can caters to ebikers has a cheap laser , let me tell you. Lol. No $250,000 machine working for us ebikers, we are to cheap... or poor... ) or any grade of good steel.
Link to the "chep" ones? I only know of the ones wanting alot of money, esp for one offs.qwerkus said:
I personally have Mennonite kids who will plasma CNC anything for me.
Grin tq arms are what, stamped, stainless, or shat steel? Anybody?
On emachineshop you can of course choose you metal and whatnot.. : you might even be able to request tool steel and have that lasered.. but the cost of lasering stuff on emachineshop is silly. Plus that edge.
Cracks become the only danger. , but it held...
You will be able to anneal it if you have the time and some more Oxy/ Acetylene for the annealing.
i just welded a big washer to a steel open dropout to a wrench. Then hardened. Holds 8Kw and 50mph impact loading. In aluminum open dropouts. Lol.
It must be cause I can weld that I also think the simple fabrication of a Tq. arm is easy peasy, and if you understand the fundamentals, should be easy to beat the sellers options that we have ..... better and stronger.
MadRhino
100 GW
You don’t need hard steel to make custom dropouts or torque plates. Just use the steel of 4 inch angle stock, that is 1/4 thick. Not very hard, easy to cut and file. 1/4 each side makes 1/2’’ to hold torque, enough for a big hub fed lots of power. The important factor is not hardness, but precision. The fit need to be tight.
DogDipstick
100 kW
Balmorhea said:
MadRhino said:
ok Karens.
yall do you, I'll do me. Lol. To each their own, but I just told you a very easy, cheap way to toughen the metal. Might not NEED it.. .. but I also dont "NEED " more than 750w and 20mph according to some, others, that I also consider, ... Dead wrong.
MadRhino
100 GW
DogDipstick said:Balmorhea said:
MadRhino said:
ok Karens.
yall do you, I'll do me. Lol. To each their own, but I just told you a very easy, cheap way to toughen the metal. Might not NEED it.. .. but I also dont "NEED " more than 750w and 20mph according to some, others, that I also consider, ... Dead wrong.
I ride powerful bikes for a decade. The first had a damaged axle because the torque plates were harder than the axle steel. I you believe in hard torque arms, just wait, you will understand some day that you push your motor a tad too hard.
Then, I don’t know Karen but somehow it does sound like an insult. English is not my native language so, for now, you have the benefit of the doubt.
serious_sam
10 kW
100% agree.MadRhino said:
No point having a high strength torque arms if the axle is generic Chinese steel. Minimising stress in the axle is the goal. Minimise stress by maximising the area over which the force is applied => tight fit and maximum surface contact.
Even aluminium works well if there is adequate surface area and tight fit. Absolutely no need for hardened steel.
JackFlorey
100 kW
Right. What about that seems surprising?eecsflyer said:
A given motor has a given motor constant, that lets you determine no-load RPM at a given voltage. If your no-load speed is somewhat higher (say 30-50%) than your cruising speed you have a good motor to voltage match.
If you use a DD motor then that cruising speed is going to mean a very low RPM. To get enough torque you need powerful magnets, heavy steel and lots of turns.
If you use a geared motor then that cruising speed is going to mean a higher RPM. That means much less requirement for torque for a given power (remember power = RPM x torque.) The gearing allows you to use a much smaller, faster spinning motor.
Having the same power DD vs (fixed) geared hub is going to give you very similar performance. The geared hub will just be lighter. However, if you have a VARIABLE gear ratio then you can get the best of both worlds - a light motor that performs well when you are cruising, and that can also provide much higher torque for climbing. An example of such a hub is the Xionga hub; drive it forward and you get "high" gear, drive it backwards and you get "low" gear.
Sunder
10 MW
serious_sam said:
Maybe this is beyond the scope of the comment since I am moving from physics to manufacturing, but right now, I am prototyping a torque arm for my Ninja. The process I have previously used is measure, print in plastic, adjust, and then send the file for cnc machining in steel when I am ready.
I can 3D Print down to 0.1mm precision, and measure with calipers down to 0.01mm. I find if I print too close to measurements, I may need a small amount of force to fit it, but no part prior to this has needed that much precision. Apparently the plastics I use has up to 1-3% warpage depending on the axis of print.
Given the criticality of fit in this application, do you think it would be better to make to hole very slightly too small, and hand file it to fit? I am an inexperienced metalworker, but have patience and can follow instructions.
LeftieBiker
100 kW
- Joined
- Nov 5, 2012
- Messages
- 1,107
It's always better to adjust to fit by slowly filing than to have to shim a too-loose part - especially one that is under load.
MadRhino
100 GW
Yep. I drill smaller, and shape with files. You know it is a tight fit when it is difficult to push the axle through and you need a mallet to sit it completely.
Sunder
10 MW
Okay, thanks guys. I may make the plans something like 0.5mm too small, and then spend a few hours filing it to a perfect fit.
I don't think I want to use a mallet - Feels like there's a decent risk of damaging the motor's axles, and it's a QS273, a bit expensive to lose such a cheap part.
I don't think I want to use a mallet - Feels like there's a decent risk of damaging the motor's axles, and it's a QS273, a bit expensive to lose such a cheap part.
Sunder
10 MW
Great idea! Thanks!
MadRhino
100 GW
All of mines were fitting with a few small hits at first. But with all the tire changes, they are now fitting by hand. None of them are loose.
The big QS motors have a bigger axle, giving extra surface contact for torque plates.
The big QS motors have a bigger axle, giving extra surface contact for torque plates.
serious_sam
10 kW
That's basically what I did. I also found that each side of the axle was slightly different across flats and machined rough (something like 9.9mm on LHS and 9.7mm RHS), so I made my mounts 9.6mm, and sanded the axle to fit. I also got my mounts anodised, so sanding the axle was a better option anyway.Sunder said:
