New Geared Hubmotor variable-regen e-braking system that still freewheels, by ChargeBike

[MorbidlyObeseKoala]

The only safe and practical solution was to mount a small hydraulic brake (from a 1/5 rc model car) plumbed into the main brake line using the bleed port from the main caliper. The 80mm aluminum rotor is connected to the clutch planet carrier thru a hub.

That is a really cool solution and interesting test info.

How is your pad life now that you have the braking performance you require?

 

[j3tch1u]

That is a really cool solution and interesting test info.

How is your pad life now that you have the braking performance you require?

I would assume better. However, the small single piston RC brake still slips too much (even after switching to real ceramic brake pads and steel rotor) so I'm in the process of substituting it for an electromagnetic dog clutch. Due to space constraints, dual piston caliper or larger rotor are no go.

 

[Alon-ChargeBike]

Here are my findings after experimenting with freegen on a MAC geared motor (front). Regen is not strong enough on hills or slow speeds to SAFELY substitute a front disc brake. So I wrote a LISP script on VESC to activate electronic handbrake on low RPM where it sends positive current thru 2 phases to generate holding torque. Handbrake will still slip at 100A under 100 RPM so you need to send a handbrake signal continuously on the VESC (10hz is fine). This has the effect of ABS slip/hold...however the e-handbrake will overheat the motor rather quickly rendering it useless as a motor AND brake. It will also make a horrific groaning sound as it struggles to brake the motor.

The only safe and practical solution was to mount a small hydraulic brake (from a 1/5 rc model car) plumbed into the main brake line using the bleed port from the main caliper. The 80mm aluminum rotor is connected to the clutch planet carrier thru a hub. I removed the spider from a 203mm shimano ice-tech rotor and mounted that directly to the motor hub. No compromise braking and regen. Both brakes are activated by the same lever and can be modulated like normal.

Hi,
Can you share more details about your project and how you are controlling Freegen?
It's more reasonable to use Freegen on the rear wheel as a lot less braking torque is needed to lock it, on a rear 26" wheel a 12T MAC needs about 90A to lock if the rear brake is used alone, if both brakes are applied (a real life emergency brake scenario) the rear wheel will lock at about 70A.
Instead of using the e-handbrake use the VESC speed PID loop, send command for 0 RPM, that way you will use the braking current you actually need and not a constant 100A, your motor won't overheat.

 

[j3tch1u]

Hi,
Can you share more details about your project and how you are controlling Freegen?
It's more reasonable to use Freegen on the rear wheel as a lot less braking torque is needed to lock it, on a rear 26" wheel a 12T MAC needs about 90A to lock if the rear brake is used alone, if both brakes are applied (a real life emergency brake scenario) the rear wheel will lock at about 70A.
Instead of using the e-handbrake use the VESC speed PID loop, send command for 0 RPM, that way you will use the braking current you actually need and not a constant 100A, your motor won't overheat.

Hi Alon, sure. Btw thanks for sharing your great concept with the community. I'm not using any special code to track the motor or hub rpm...just sending brake signal to ADC2 and letting the VESC do its thing. Max regen current set at -100a and max battery set at -50a. While the braking is strong, it's not enough for our steep hills (I live on a mountain). I didn't try the PID loop but even the ehandbrake @100A would slip during hard braking at low RPMs. I'm on a dual motor setup and I've experimented with rear regen before but I simply cannot use any sort of regen on the rear since it is already at max temps on long uphill climbs. Front motor temps are generally much lower. For my use case (most extreme scenario), I decided that full functionality of the front mech brakes were non negotiable. My second revision used a secondary hydraulic brake/rotor but the small brake would slip at anything over 15A regen. So...I'm currently working on an electromagnetic dog clutch that will essentially lock the planet hub on brake signal. I may have to adjust the regen rampup to allow time for the eclutch to lock in place without too much shock. I've already tested this using the mini hydro brake as a dog clutch (grooves cut into the rotor) and it works effectively but has some issues with engagement at higher speeds thus the 3rd revision will use a much more robust dog clutch. In the end the effort is definitely worth it. It's so satisfying to have regen on demand for long mountain descents!

 

[Gruesome]

Very interesting. It would be very helpful to see some numbers. Then one could compare the different torques at work.
Some are easy to measure, for others one needs some knowledge about the motors.

• slope of the hill
• weight of bike + rider
• effective wheel radius for each of the wheels (measured from the center of the axle to the road surface)
  • from these three numbers one can compute the needed braking torque
• motor torque constant in Nm/A for each of the motors
• motor efficiency

Since you say the small brake (80mm rotor diameter ?) slips, it would be interesting to know more about that as well. Do you know the total lever ratio (hand lever and hydraulic piston area ratio) of your brake? Maybe there is an easy upgrade one could make (different brake lever and/or master cylinder, larger brake caliper cylinder) to increase the force at the pad.

 

[j3tch1u]

Very interesting. It would be very helpful to see some numbers. Then one could compare the different torques at work.
Some are easy to measure, for others one needs some knowledge about the motors.

• slope of the hill
• weight of bike + rider
• effective wheel radius for each of the wheels (measured from the center of the axle to the road surface)
  • from these three numbers one can compute the needed braking torque
• motor torque constant in Nm/A for each of the motors
• motor efficiency

Since you say the small brake (80mm rotor diameter ?) slips, it would be interesting to know more about that as well. Do you know the total lever ratio (hand lever and hydraulic piston area ratio) of your brake? Maybe there is an easy upgrade one could make (different brake lever and/or master cylinder, larger brake caliper cylinder) to increase the force at the pad.

The hill to my house is over 30 degrees. Justin has rode it before. Here is a video approximating how steep it is (mine is steeper):

I need to max out both front/rear 4-piston brakes (203mm rotors) to go down this hill.

I'm 85kg, bike is 35kg. I won't even bother to run the other numbers---there is no way regen only front brake will safely negotiate this slope. The puny RC brake i'm using is single piston and there is a lot of flex. Larger caliper and rotor won't fit. I've thought about using twin rotors (one for mech brake, one for regen) and a modified caliper that mashes the two together but it seems janky. Dual caliper would work but my forks don't support it. Dog clutch (using the original brake bolts as pawls) seems like the way to go.

 

[Gruesome]

Thanks.

The hill to my house is over 30 degrees. Justin has rode it before. Here is a video approximating how steep it is (mine is steeper):

I need to max out both front/rear 4-piston brakes (203mm rotors) to go down this hill.

I'm 85kg, bike is 35kg. I won't even bother to run the other numbers---there is no way regen only front brake will safely negotiate this slope. The puny RC brake i'm using is single piston and there is a lot of flex. Larger caliper and rotor won't fit. I've thought about using twin rotors (one for mech brake, one for regen) and a modified caliper that mashes the two together but it seems janky. Dual caliper would work but my forks don't support it. Dog clutch (using the original brake bolts as pawls) seems like the way to go.

I know, numbers can be tiring
I'll just assume a 20" wheel then, or .25m effective radius. So, the downhill force is 0.5 * 1200 N = 600 N (sine of 30° times weight force of 85+35 kg), with a 20" wheel the torque becomes 150 Nm (600*0.25). That tells you what kind of motor you need for e-braking.
Regarding hand braking: At an effective brake caliper radius of say 33mm the 150Nm torque becomes a force at the brake pad of 150Nm/0.033m=4500 N. Assuming conservatively a friction coefficient of 0.3 (you could double that with better pads) you get a brake clamping force needed of about 15kN. With a total brake lever ratio of about 50 (just guessing, but according to google it seems to be a good number for mountain bikes) you need a hand lever force of 300 N or the equivalent of 30 kg pulling force. That's pretty hefty. If you are running a 700c wheel instead (effective radius 0.34m), the hand lever force required becomes 400N.

I see two options:
a) mount a motorcycle brake lever/master cylinder & caliper combination with a brake lever ratio of about 200
b) bulk up your forearm muscles with a fingerboard, like rock climbers use

 

[j3tch1u]

Regen test hill #1:

Regen test hill #2:

Regen test hill #3:

 

[Gruesome]

So, no problem anymore with regen braking, it seems? What solved it?

 

[j3tch1u]

So, no problem anymore with regen braking, it seems? What solved it?

These are my test hills (not hill tests). I'm still waiting for some parts to finish the e-clutch.

 

[Gruesome]

These are my test hills (not hill tests). I'm still waiting for some parts to finish the e-clutch.

Oh, sorry, I read 'regen test hill #1' as 'regen test' 'hill #1', not as 'regen' 'test hill' '#1'. Do you have videos of the failed hill tests?

 

[j3tch1u]

A quick update...dog clutch won't work. Since there is no backlash on the internal (planetary) one-way clutch, the dog plate will be too tight after engagement and won't release. I'm trying the last option...using two full size brake rotors and a modified (widened) caliper that squeezes both discs together during braking.

 

[Gruesome]

Full size (200mm diameter?) brake rotors should cut the required brake force to a third, so hopefully doable (100N pulling force instead of 300N). You could also try to mount longer brake levers. I noticed Shimano for example makes quite a few different types.

 

[lnanek]

Could always use a rim that supports v-brakes/caliper brakes as well. Use the disc for freegen and caliper for manual on the monster hill. Big beach cruiser calipers have a lot of reach so even on forks and frames without v-brake posts, they can usually reach the rim from some place you can drill a single hole.

Although seems weird to put so much effort into a front motor implementation. Normally rear motors are preferred for the hub motor and that leaves front free for pure manual breaking.

 

[j3tch1u]

Success! I modified the brake caliper by widening it and adding a floating brake pad (two thinned out pads welded to each other) between the two 203mm rotors. Stopping power is immense! Hitting 80A regen on some hills. With the twin rotors, stopping power, heat dissipation are all increased. The middle brake pad can be replaced when worn. Toughest part was getting the mid caliper plate not to leak oil under very heavy pressure.

 

[LewTwo]

Although seems weird to put so much effort into a front motor implementation. Normally rear motors are preferred for the hub motor and that leaves front free for pure manual breaking.

A problem with that is some people prefer IGH (Internally Geared Hub) to derailleur chain drives. Belt drives are also getting very popular. With the exception of the IGH motor Hub that Grin had for a while, rear hub motors preclude the use of both IGH or Belt drives.

I thought it was interesting that Justin mentioned:

" ... The EU rules for EN15194 specify that the rear brake needs to be able to apply a 220 Newton stopping force, ... "

Notice he says "rear" brake (some of us old codgers survived many of our teenage years with nothing more than rear coaster brakes ... and yes, we had long steep hills back in ancient times as well). Brakes are covered in section 4.3.5. Is there a requirement for front brakes as well ? As an aside: These are the same people that determined 250 watts should be the maximum power allowed for an E-Bike and the maximum speed should be 25 KPH ....

So I present the alternative scenario: A front hub motor leaves the rear free for pure manual breaking.

 

[crunchysteve]

That potentially puts the hub ahead of the midmount from a range and market perspective, and even though regen is a small gain (my e-scooter in back-of-an-envelope calcs gets between 5 and 7% more range using regen than not), eliminating drag with a friction closer on a freewheel eliminates any disadvantages to regen on bicycles. I'm impressed.

 

[crunchysteve]

So I present the alternative scenario: A front hub motor leaves the rear free for pure manual breaking.

This, plus discussion earlier in this thread about thru axle, the latter makes front safer from regen tearout, just like it prevents front disk tearout. (Tearout's a thing in mountain biking apparently, due to the griping force being behind the front axle on front wheels, but only with riders doing crazier speeds on more bonkers trails than I've ever attempted.) I suspect regen tearout is less of a thing (but not impossible) because the vectors are all axial rather than off-axis like disk brakes.

 

[MorbidlyObeseKoala]

Success! I modified the brake caliper by widening it and adding a floating brake pad (two thinned out pads welded to each other) between the two 203mm rotors. Stopping power is immense! Hitting 80A regen on some hills. With the twin rotors, stopping power, heat dissipation are all increased. The middle brake pad can be replaced when worn. Toughest part was getting the mid caliper plate not to leak oil under very heavy pressure.

Feel like I need a picture to understand the further modifications here....

 

[Gruesome]

[...] because the vectors are all axial rather than off-axis like disk brakes.

Vectors as in force vectors, and axial as in along or in the same direction as the wheel axis? The brake force vector has to be in the wheel rotation plane, not perpendicular to it, otherwise you are not going to slow the wheel (and the bike) down.

 

[LewTwo]

Vectors as in force vectors, and axial as in along or in the same direction as the wheel axis? The brake force vector has to be in the wheel rotation plane, not perpendicular to it, otherwise you are not going to slow the wheel (and the bike) down.

If one is accelerating or decelerating the bicycle there is a rotational force vector at the point where the tire patch is on the riding surface (i.e. the pavement). The lever arm is the distance from the tire patch to the center of the axle. There is an equal and opposite force between the rotating wheel and the frame of the bike.

In Justin’s video , “Freegen Explained”, there is a section he calls “Torque Arm Simplicity”. It begins at 12 minutes 38 seconds into the video. In this section Justin recounts the experience that they had when someone forgot to fasten the Toqure Arm to the frame. He points out that all the braking resistance force was transferred to the frame via the Brake Caliper mount rather than the Torque Arm.

I believe this is what crunchysteve was referring to.

Note (In my opinion): Cheap, flimsy, bolt on, FleaBay brake caliper mounts are definitely NOT recommended.

 

[Gruesome]

Well, then he shouldn't have written 'axial', if he doesn't mean it.

 

[MorbidlyObeseKoala]

Success! I modified the brake caliper by widening it and adding a floating brake pad (two thinned out pads welded to each other) between the two 203mm rotors. Stopping power is immense! Hitting 80A regen on some hills. With the twin rotors, stopping power, heat dissipation are all increased. The middle brake pad can be replaced when worn. Toughest part was getting the mid caliper plate not to leak oil under very heavy pressure.

Are you saying you reduced the street bike strategy of dual disc brakes (2 rotors, 2 calipers, opposite sides) on the front wheel to a one sided set up with dual 203mm rotors, 2 sets of brake pads and a single modified caliper? Did you remove the mini disc brake set up or is that still part of the equation there?

 

[j3tch1u]

Are you saying you reduced the street bike strategy of dual disc brakes (2 rotors, 2 calipers, opposite sides) on the front wheel to a one sided set up with dual 203mm rotors, 2 sets of brake pads and a single modified caliper? Did you remove the mini disc brake set up or is that still part of the equation there?

I deleted the mini brake. Now: 203mm rotor is attached to hub + 203mm rotor attached to clutch planet carrier. When you squeeze the brake, the dual brake caliper stops both rotors. These are 4-piston brakes so...2 pistons on each side. The double-sided brake pad in the center floats (can move sideways) so you are not losing any hydraulic power flexing the rotors. Works like a charm.

 

[Gruesome]

So as the brake pads wear down, the two disks have to move closer together, otherwise part of your clamping force is lost to bending the disks. It doesn't look like either of them is on a floating mount.