Looking for hub motor for regen purposes

[harrisonpatm]

So you are saying with proper torque arms it should be fine?

I think yes. "Should," being the key word. It's important to note that Chalo isn't wrong about adding a mechanical issue to be aware of if one decides to use regen in their build. However, if you're aware of the issue, and take proper precautions to alleviate it, then regen can be helpful. It's also important to be aware of how much charging current your battery can take, and whether your preferred amount of regen is more/less than that.

So yeah, I agree with Chalo that regen isn't a one-size-fits-all solution for everybody that can replace mechanical brakes. There was a poster awhile back, can't remember exactly which, that was planning a scooter build with zero mechanical brake function to rely purely on regen braking for stopping. Which is of course a bad idea, what if the controller fails mid ride? He would be unable to brake. And there are plenty of examples of dropouts without torque arms failing and hub motors spinning out, or cheaply-made torque arms being used that get chewed up by steel axles reversing toque during regen.

However, I disagree with his comments about regen being inherently useless and ineffectual. For example:

Also, brakes designed as brakes work better as brakes than motors that are retrofitted into service as brakes.

Motors aren't being retrofitted as brakes. Back-EMF is present in electric motors regardless of whether regen is being used. If the effect is there anyway, and proper precautions are taken to prevent mechanical or battery damage, there's no reason why it can't be used, in addition to mechanical braking. Whether someone wants to deal with the added mechanical issues that come with using regen, and whether it is worth it for their build or not, is going to be different for every person. I think that's what this forum and DIY is all about anyway: making yourself informed about the cost/benefit of your project choices, and acting accordingly.

 

[rudivb]

I think yes. "Should," being the key word. It's important to note that Chalo isn't wrong about adding a mechanical issue to be aware of if one decides to use regen in their build. However, if you're aware of the issue, and take proper precautions to alleviate it, then regen can be helpful. It's also important to be aware of how much charging current your battery can take, and whether your preferred amount of regen is more/less than that.

So yeah, I agree with Chalo that regen isn't a one-size-fits-all solution for everybody that can replace mechanical brakes. There was a poster awhile back, can't remember exactly which, that was planning a scooter build with zero mechanical brake function to rely purely on regen braking for stopping. Which is of course a bad idea, what if the controller fails mid ride? He would be unable to brake. And there are plenty of examples of dropouts without torque arms failing and hub motors spinning out, or cheaply-made torque arms being used that get chewed up by steel axles reversing toque during regen.

However, I disagree with his comments about regen being inherently useless and ineffectual. For example:

Motors aren't being retrofitted as brakes. Back-EMF is present in electric motors regardless of whether regen is being used. If the effect is there anyway, and proper precautions are taken to prevent mechanical or battery damage, there's no reason why it can't be used, in addition to mechanical braking. Whether someone wants to deal with the added mechanical issues that come with using regen, and whether it is worth it for their build or not, is going to be different for every person. I think that's what this forum and DIY is all about anyway: making yourself informed about the cost/benefit of your project choices, and acting accordingly.

I agree we should always be cautious and he gave good input on that, but I think he underestimates the value of regen braking. That being said I have 0 personal experience with it, but from reading about it and the experience of others, it became very interesting to me especially for the purpose of braking and reducing pad wear and general brake wear (rotors/calipers/levers).

I think if you already run a hub motor it's a no-brainer to utilize its regen braking.
In my case and that of the original poster we are running a middrive and adding a hub motor solely for regen braking is debatable imo, the added weight/complexity/forces etc. not sure if it's worthy the hassle, I also wonder what the lightest option is for this purpose.
Regen braking should never "replace" the brakes imo, but "assist" them, maybe in the future when technology is ready we can rely on just regen, but not for now.

Actually now I think about it, when I was a forklift driver I NEVER used the brakes, I could rely solely on it's deceleration when releasing the throttle or reversing, which is done by regen. I drove like a madman and was one of, if not the fastest driver at that company.

 

[Chalo]

Regen braking isn't inferior to real braking because it's weak-- it's inferior to real braking because it's so poorly controllable and lacks feedback. It's analogous to how many hydraulic discs come on much stronger but max out at less braking power than mechanical rim brakes. The mismatched relationship between lever effort and braking power makes their braking less effective even if it's abundantly powerful. With the rim brake, if you want just a little drag you can get that; if you want the hardest stops with good sensitivity to prevent tipping up the bike, you can get that too. The hydraulic disc, even if it's just as powerful or more so, makes it harder to match lever effort to desired braking, so the stops are sketchier.

Magnetic braking is almost arbitrary in terms of braking effort versus braking torque, and it changes dramatically with changes in speed or battery state of charge. So you can have control over when you want braking, but very little control over how much braking force you get. (And then your wheel falls out and twists off its motor cable.) None for me, thanks.

 

[harrisonpatm]

Regen braking isn't inferior to real braking because it's weak-- it's inferior to real braking because it's so poorly controllable and lacks feedback. It's analogous to how many hydraulic discs come on much stronger but max out at less braking power than mechanical rim brakes. The mismatched relationship between lever effort and braking power makes their braking less effective even if it's abundantly powerful. With the rim brake, if you want just a little drag you can get that; if you want the hardest stops with good sensitivity to prevent tipping up the bike, you can get that too. The hydraulic disc, even if it's just as powerful or more so, makes it harder to match lever effort to desired braking, so the stops are sketchier.

Magnetic braking is almost arbitrary in terms of braking effort versus braking torque, and it changes dramatically with changes in speed or battery state of charge. So you can have control over when you want braking, but very little control over how much braking force you get. (And then your wheel falls out and twists off its motor cable.) None for me, thanks.

Can't argue with that. Systems and controllers with highly variable regen input are either too expensive for DIYers, or are difficult to setup. And they don't have nearly the same feedback as mechanical brakes.

I like my regen as an on/off, automatically when I release the throttle (I'm on a motorcycle), just a little bit of automatic drag that kinda feels like extra strong engine braking. Good enough to slowly drag to a stop when I'm approaching a stopsign.

 

[gromike]

I took delivery last month of a Ryvid Outset, and it's regen braking function is great. I haven't needed to use the rear brake, which is a good thing because the brake pedal is tucked a little farther away than on a regular bike.
So, from my experience, regen braking works well when done right.

 

[99t4]

Another data point:

I have used regen braking on two of my ebikes successfully for several thousands miles/years. My riding environment is mostly hilly in every direction, some acutely hilly.

Benefits:

• Welcome noticeable preservation of brake pads and rotors.
• Moderate (yet measureable) extended range (often can watch battery charge % increase on longer downhill sections).
• Quieter.

Disadvantages:

• REQUIRES high quality well designed well implemented intelligently installed TA (clamping style generally best)(see Grin TorqArm v7 for example). Cheaper TAs can work if you carefully install them in a certain fashion. Two may be called for with larger DD hubmotors/higher regen rates.
• Possible more difficult wheel removal/reinstall. TA can make flat tire repair more cumbersome.
• More thought given to control method. Cable brake levers are easy, hydraulic more complex.
• Some way to avoid overcharging the battery if first descending on a fully charged battery.

BTW, castellated nuts, Locktite, nylox nuts, Nordlock washers, etc. are not effective for this application.

There is a long pictorial TA thread here on ES, might be helpful.

 

[99zip99]

Your battery must be able to take the amount of region you're trying to throw into it.
You need to be able to program your regen.
I use norlock washers and haven't found the nut to be loose yet. But now I will be checking this.
72 volt unprogrammable P.V. 80 amp controller pushing 2.100 watts from 45mm mxus downhill I couldn't get the axle not to rock even with dual torque arms.
I will try it with rear d.d. and em3ev 36-72v 40amp controller factory set. I'll read it on my C.A.

 

[A-DamW]

Regen braking isn't inferior to real braking because it's weak-- it's inferior to real braking because it's so poorly controllable and lacks feedback. It's analogous to how many hydraulic discs come on much stronger but max out at less braking power than mechanical rim brakes. The mismatched relationship between lever effort and braking power makes their braking less effective even if it's abundantly powerful. With the rim brake, if you want just a little drag you can get that; if you want the hardest stops with good sensitivity to prevent tipping up the bike, you can get that too. The hydraulic disc, even if it's just as powerful or more so, makes it harder to match lever effort to desired braking, so the stops are sketchier.

I only have experience with VESC based controllers, but I can unequivocally say the modulation using a second thumb throttle for regen braking is every bit as good as any friction braking I have experienced.

For feedback, I don't need my fingers on a brake lever to tell me how well my brakes are working, the inertial force from dumping 150 phase amps, ~70 battery amps, tells me exactly how well it works.

Yes, there are caveats, just like there are with friction brakes.

 

[ebike4healthandfitness]

(And then your wheel falls out and twists off its motor cable.) None for me, thanks.

With single direction torque arms I can see this being a problem. Ebikes.ca describes and shows the problem in the video below:

But I haven't see anyone mention single direction torque arms for regen unless it is something like the arrangement in the video below and even then with this opposing and pre-stressed dual single direction torque arm method he only said "In a pinch sometimes you can get away with":

 

[ebike4healthandfitness]

Back-EMF is present in electric motors regardless of whether regen is being used. If the effect is there anyway, and proper precautions are taken to prevent mechanical or battery damage, there's no reason why it can't be used, in addition to mechanical braking.

I really like the idea of Back-EMF being used to limit top speed especially when a coaster brake is being used (e.g. Dutch bike). Unfortunately, all the gear hubs I see capable of locking the clutch are relatively high Kv so the point where the negative torque comes on is at a rather high speed point and thus limits the utility of such an approach.

 

[Potato011]

Alright, was offroading again and snapped 3 more spokes on the back wheel. 4/32 spokes r now broken. Given the pain of sourcing 141mm rear hubs for a new wheel, I will immediately order a 750w or 1000kw dd hub and do a wheel build from that. I've already aquired the vesc. will update later.

 

[99zip99]

I just looking at my spokes I have too many there for a DD butter in a 26 inch rim or 168 mm as I remember ??

 

[99zip99]

I don't know mine are 13 -14 ga spam spokes and when I build a motor I put a brass washer under the head.
I also have some extra 13 gauge , but you probably want those 10 or 12 gauge spokes made of steel. ?

 

[Potato011]

Hub acquired. The ride feel is a lot different. The casette has this gap so 1st gear is pretty far from the hub body. Really snug fit but there is just enough space for me to shift through all 8 gears and mount brake calipers. Riding without hub power and it feels harder to turn with that big spinning mass at the back. Or maybe that's just because I've been riding my 20" folding bike for the last 3 weeks. I'll mount the vesc in the upcoming days and see how it goes. Also need to fab torque arms

 

[hillslayer@]

Your going dual motor. Look into uping the front chainring size you want to run the bigger cogs on the cassette too take some leverage off the hub side cover. I am running a 60t on mine with a five speed freewheel keeps me in the middle cogs most the time.

 

[Potato011]

Your going dual motor. Look into uping the front chainring size you want to run the bigger cogs on the cassette too take some leverage off the hub side cover. I am running a 60t on mine with a five speed freewheel keeps me in the middle cogs most the time.

how do you divide torque between your motors?

 

[hillslayer@]

I think yes. "Should," being the key word. It's important to note that Chalo isn't wrong about adding a mechanical issue to be aware of if one decides to use regen in their build. However, if you're aware of the issue, and take proper precautions to alleviate it, then regen can be helpful. It's also important to be aware of how much charging current your battery can take, and whether your preferred amount of regen is more/less than that.

So yeah, I agree with Chalo that regen isn't a one-size-fits-all solution for everybody that can replace mechanical brakes. There was a poster awhile back, can't remember exactly which, that was planning a scooter build with zero mechanical brake function to rely purely on regen braking for stopping. Which is of course a bad idea, what if the controller fails mid ride? He would be unable to brake. And there are plenty of examples of dropouts without torque arms failing and hub motors spinning out, or cheaply-made torque arms being used that get chewed up by steel axles reversing toque during regen.

However, I disagree with his comments about regen being inherently useless and ineffectual. For example:

Motors aren't being retrofitted as brakes. Back-EMF is present in electric motors regardless of whether regen is being used. If the effect is there anyway, and proper precautions are taken to prevent mechanical or battery damage, there's no reason why it can't be used, in addition to mechanical braking. Whether someone wants to deal with the added mechanical issues that come with using regen, and whether it is worth it for their build or not, is going to be different for every person. I think that's what this forum and DIY is all about anyway: making yourself informed about the cost/benefit of your project choices, and acting accordingly.

how do you divide torque between your motors?

The hub always has a little bit of a drag so that motor needs to be on any time your moving. I pretty much just use the hub until I am climbing then pedal assist with the mid-drive. Also use the mid-drive to get up to speed. Don't bother setting up pedal assist on the hub, throttle only works best. Those direct drive hubs like to cruise around 25 mph which probably gets the best range.

 

[Potato011]

The hub always has a little bit of a drag so that motor needs to be on any time your moving. I pretty much just use the hub until I am climbing then pedal assist with the mid-drive. Also use the mid-drive to get up to speed. Don't bother setting up pedal assist on the hub, throttle only works best. Those direct drive hubs like to cruise around 25 mph which probably gets the best range.

I intend to implement virtual freewheeling via the vesc. My original idea was to use the hub motor exclusively for regen braking. I would hook up a left hand twist throttle and use that for proportional regen braking. Should the middrive die, id have a switch that switches the right hand twist throttle from middrive to hub drive.

An alternative config if i want to use both motors regularly is hooking the left twist throttle to hub power. then i add a third thumb throttle to the right side for regen braking. only issue is under max braking, squeezing the hydraulic brakes while thumbing the regen may be clunky.

last config is same as second option but instead of thumb throttle, i implement some kind of hall effect switch on my brake levers that gives proportional regen.

do you find that the hub motor is more efficient than middrive at lets say 20-25mph on flat ground?

 

[hillslayer@]

I intend to implement virtual freewheeling via the vesc. My original idea was to use the hub motor exclusively for regen braking. I would hook up a left hand twist throttle and use that for proportional regen braking. Should the middrive die, id have a switch that switches the right hand twist throttle from middrive to hub drive.

An alternative config if i want to use both motors regularly is hooking the left twist throttle to hub power. then i add a third thumb throttle to the right side for regen braking. only issue is under max braking, squeezing the hydraulic brakes while thumbing the regen may be clunky.

last config is same as second option but instead of thumb throttle, i implement some kind of hall effect switch on my brake levers that gives proportional regen.

do you find that the hub motor is more efficient than middrive at lets say 20-25mph on flat ground?

If you have the hub it will add ten mph to your uphill speed no reason not to use both for climbing. On flat ground or slight hills the hub is more efficient just let the mid-drive freewheel, as a bonus less wear on the drivetrain.

 

[ebike4healthandfitness]

I intend to implement virtual freewheeling via the vesc. My original idea was to use the hub motor exclusively for regen braking. I would hook up a left hand twist throttle and use that for proportional regen braking. Should the middrive die, id have a switch that switches the right hand twist throttle from middrive to hub drive.

An alternative config if i want to use both motors regularly is hooking the left twist throttle to hub power. then i add a third thumb throttle to the right side for regen braking. only issue is under max braking, squeezing the hydraulic brakes while thumbing the regen may be clunky.

last config is same as second option but instead of thumb throttle, i implement some kind of hall effect switch on my brake levers that gives proportional regen.

do you find that the hub motor is more efficient than middrive at lets say 20-25mph on flat ground?

Don't forget about backpedal Regen. This where your Regen is activated by how much you pedal backwards. Then you can use your left hand twist throttle for hub and your right hand twist throttle for the mid drive and not need a thumb throttle or brake levers for Regen.

 

[Potato011]

Don't forget about backpedal Regen. This where your Regen is activated by how much you pedal backwards. Then you can use your left hand twist throttle for hub and your right hand twist throttle for the mid drive and not need a thumb throttle or brake levers for Regen.

I dont think i can do backpedal regen. The bbshd has a freewheels so pedaling backwards doesn't move the chainring. Closest thing i can think of is some kind of disc connected to the cranks with a hall sensor.

 

[Potato011]

If you have the hub it will add ten mph to your uphill speed no reason not to use both for climbing. On flat ground or slight hills the hub is more efficient just let the mid-drive freewheel, as a bonus less wear on the drivetrain.

using both motors at high throttle will require an upgrade to my batteries. Rn my largest pack is 52v and outputs 40A but most of my other packs can only do 30A. The bbshd alone can draw almost all that power

 

[hillslayer@]

If you have the hub it will add ten mph to your uphill speed no reason not to use both for climbing. On flat ground or slight hills the hub is more efficient just let the mid-drive freewheel, as a bonus less wear on the drivetrain.

I have not done regen braking but have watched some YouTube video of a guy using the vesc controller, he just had it activate regen as soon as he let off the throttle I think and would use the app to adjust how much he wanted.

 

[hillslayer@]

using both motors at high throttle will require an upgrade to my batteries. Rn my largest pack is 52v and outputs 40A but most of my other packs can only do 30A. The bbshd alone can draw almost all that power

Dual batteries is what I use one for each motor.

 

[Potato011]

Dual batteries is what I use one for each motor.

That is a very good point. I forgot I could do that. With some kind of power distribution system, i could probably safely absorb more current for regen that way.