BEST Hub Motor/controller for regen

[ddk]

In your humble opinion (with some facts preserved or presented) which off-the-shelf motor/controller combination creates excellent to adequate controlled-braking solutions. Not so interested actually using the motor as a ... ya know... "motor"
I ask this because of my intention to use e-brakes on a trailer for long-distance cruises along the mighty Hwy 101 and CA 1, where the hills are never-ending.
I am involved in other, less intellectual endeavors on the trike's construction but would like to order stuff before the winter rains.
And frankly, I've been making lousy decisions lately. (read-costly)

Thanks for any ideas.

 

[hjns]

Do you want to use a motor ONLY for braking? Or do you want to use it ONLY for regeneration of energy?

Because, if you are absolutely NOT using the motor as motor, than you may be better off using a couple of dynamos, mechanically able to engage / disengage as appropriate. If you DO want to use the motor as motor, then it makes sense to me to add a direct drive (DD) motor + controller. Go for a known controller with regen enabled. Any DD motor will do, so you may want to try to find the lightest DD motor available.

 

[ddk]

Do you want to use a motor ONLY for braking? Or do you want to use it ONLY for regeneration of energy?

only for braking but I'll be using the batteries as a sink because I have 3kw-worth on-board.

Because, if you are absolutely NOT using the motor as motor, than you may be better off using a couple of dynamos, mechanically able to engage / disengage as appropriate. If you DO want to use the motor as motor, then it makes sense to me to add a direct drive (DD) motor + controller. Go for a known controller with regen enabled. Any DD motor will do, so you may want to try to find the lightest DD motor available.

I want the trailer mechanically simple to build in this instance.
I'm thinking DD motors would be the easiest path to completion but most of experience is with geared-hub motors and brushed motors without regen or e-brakes.
I simply need a good drag brake solution for long descents.

 

[hjns]

I want the trailer mechanically simple to build in this instance.
I'm thinking DD motors would be the easiest path to completion but most of experience is with geared-hub motors and brushed motors without regen or e-brakes.
I simply need a good drag brake solution for long descents.

Well, in that case it will be easy to add two DD motors into the wheels of your trailer. However, you will need to drive them a bit while moving (or spend an enormous amount of energy dragging them along), and you will need to make sure that the speed of the DD motors is in sync with the speed of your main propulsion. This means that you will want two programmable controllers to drive the two DD motors. You will then need to couple them to a throttle or PAS sensor to enable propulsion while driving, and connect the e-brake for the requested Regen. Should be a fun project.

Note that I am not sure that the energy spent to move the additional weight of the two motors + controllers + accesoires will actually be less than the energy regained by regen.....(read: it will not be more efficient)

 

[ddk]

I simply need a good drag brake solution for long descents.

Well, in that case it will be easy to add two DD motors into the wheels of your trailer. However, you will need to drive them a bit while moving (or spend an enormous amount of energy dragging them along), and you will need to make sure that the speed of the DD motors is in sync with the speed of your main propulsion. This means that you will want two programmable controllers to drive the two DD motors. You will then need to couple them to a throttle or PAS sensor to enable propulsion while driving, and connect the e-brake for the requested Regen. Should be a fun project.

Note that I am not sure that the energy spent to move the additional weight of the two motors + controllers + accesoires will actually be less than the energy regained by regen.....(read: it will not be more efficient)

I don't expect to regain much from the regen but I simply require a good drag brake for the long, 10km @ 6% grades I'll encounter. The type of descents where my bicycle would exceed 80kph if left unchecked.
Efficiency is an issue but the trike sports two motors/controllers, a 150W solar panel and 3kw of battery and weighs in a bit over 50kg. The trailer will house an additional 150W solar panel.
Already a lost cause for efficiency lol.

 

[whatever]

i would go for a xie cheng controller with irf3077 fets if 48v , or 4110 fets if higher voltage.
A long grade under regen will need good fets, also you can adjust regen level and method to engage it,
I think the xie cheng you can use variable throttle regen ??? not quite sure on that, someone else might have to confirm it
variable throttle regen would be ideal as just full on or off regen might be tricky to keep constant speed downhill.
Heres another thought:
but complex to implement, say your running 48v system, if you can switch batteries to 24v on downhill run ( lvc needs to suit 24v)
over full throttle motor speed you get regen, at 24v that speed should be pretty low, i've found that you get more amps than using
controller regen .

 

[Lebowski]

if you only want braking... what I would do is to build hub motors into the wheels of the trailer. Use 6 diodes per
motor to convert the motor's AC into DC. The voltage you get this way will be speed dependent and
can be higher or lower than the battery you're charging. I would build a buck-boost converter:

http://en.wikipedia.org/wiki/Buck-Boost_Converter

to adapt the generated voltage to the battery voltage. I would control the switching duty cycle by regulating
the currents delivered by the motors, this will give you control over the braking force.

Easy, right ?

You can also use proper motor controllers with variable strength regen but I can imagine this'll be more expensive.

I'm assuming you have a freewheel somewhere in the trike (MAC motors ?) 'cause else why not use regen
on the trike motors ?

 

[ddk]

i would go for a xie cheng controller with irf3077 fets if 48v , or 4110 fets if higher voltage.
A long grade under regen will need good fets, also you can adjust regen level and method to engage it,
I think the xie cheng you can use variable throttle regen ??? not quite sure on that, someone else might have to confirm it
variable throttle regen would be ideal as just full on or off regen might be tricky to keep constant speed downhill.
Heres another thought:
but complex to implement, say your running 48v system, if you can switch batteries to 24v on downhill run ( lvc needs to suit 24v)
over full throttle motor speed you get regen, at 24v that speed should be pretty low, i've found that you get more amps than using
controller regen .

Thanks for the info, I'll look into the xie cheng controller. My system is 36V (10S LiPo). I doubt I'll want to re-arrange batteries for braking

if you only want braking... what I would do is to build hub motors into the wheels of the trailer. Use 6 diodes per
motor to convert the motor's AC into DC. The voltage you get this way will be speed dependent and
can be higher or lower than the battery you're charging. I would build a buck-boost converter:

http://en.wikipedia.org/wiki/Buck-Boost_Converter

to adapt the generated voltage to the battery voltage. I would control the switching duty cycle by regulating
the currents delivered by the motors, this will give you control over the braking force.

Easy, right ?

:lol: makes me want to go with the space heater approach instead! (or lots of high-power resisters heating the air)

 

[ddk]

gonna re-do the subject
I already determined a DD hub motor and controller with ebs or regen would net me the easiest way of getting a good drag brake.
Now I'm looking for the best motor for this application and my google-fu is weakened by my concentration on mechanical issues.

i.e Looking for the motor with the least amount of iron inside, that can stand long-ish durations of power generation (10-15 minutes or so)
-before meltdown occurs.
and a controller that would serve the motor's needs

whatever suggested the xie cheng controller

Anyone else have an idea to suit this situation? As in motors because I could just rectify the motor's output and load it into a resister bank or nichrome wire or whatever to absorb the power, but I guess using the power for regen might be useful.
My battery banks are all 10S LiPo and I want my speeds controlled to 20 mph downhill.
The trike has 2 disc brakes in the rear and a rim brake up front.
The trike with 4 battery packs and solar panel weighs over 50kg (est. 120 pounds)
The motor-brake system would be built as a trailer.

 

[ddk]

ahha
now I gets it
what I want doesn't exist but KF is working on it
http://www.endless-sphere.com/forums/viewtopic.php?f=30&t=42666&start=120
and since my project is still in the "add motor #2" stage I might have a working tricycle about the time KF has a working model
The only difference is I am willing to use a third (and forth) means of activating the trailer brakes, possibly by using hall sensors on the brake cables but being an old guy I avoid having to invent or build electronics stuff.

if kF reads this: simply use a bipolar transistor in the FB path of an opamp to change slopes as needed but that's as much as I care to think about it. I prefer to not remember much about my past life.

 

[neptronix]

Don't use a magic pie, those cog like crazy

Anyway, why go for multiple motors? the DD will just drag on your system continuously, so just use one big DD motor for the whole bike..

 

[ddk]

Don't use a magic pie, those cog like crazy

Anyway, why go for multiple motors? the DD will just drag on your system continuously, so just use one big DD motor for the whole bike..

lol-noted:no pie for me!

I use one motor 99.1% of the time... until I visit my sister who lives at the end of a 30% grade driveway.
All of a sudden I need MORE POWER than necessary the other 99.1% of the time... and those times I want to climb a 6+ mile 6-11% grade climbs. One (10S-powered) motor is hard-pressed to accomplish these tasks.
I like using two motors over one motor for my application because one motor gets er done most the time and I'd have to have one BIG motor/controller to handle the weight of the trike, me, and my declensions.
The trailer/brake is for long-distance mountainous descents, carrying tents, stuffs and to prop up a solar panel.
The other motor (on the trike) is for those abnormal times where even my 454 motor has problems with. (of course, the 454 is pulling 6 tons instead of a measly 1/4 ton)
http://endless-sphere.com/forums/viewtopic.php?f=3&t=41572&start=105

 

[spinningmagnets]

You may find it useful to research a "plug brake", its a coil of resistor wire. By tapping into its length at several points, it can be a variable resistor. None of the regenerated electricity goes back into the battery, but...also none of the generated heat goes back through the controller either. They are cheap to make. It may be possible to salvage the resistor wire from an electric clothes heater or electric water heater thats being thrown away. Also toaster ovens, perhaps an electric room heater?

 

[John in CR]

And frankly, I've been making lousy decisions lately. (read-costly)

It sounds like you are intent on making another one by using a motor as only a brake, because they work just fine as both.

You'll want a programmable controller so you can adjust the braking force. My most pleasant regen braking is on my cargo bike, which had a regen force so soft even on the highest setting that I did a shunt mod to allow higher current. That's a speedy wind motor too, so the resulting 20A+ of regen current is a soft braking force. What's nice about soft is that it's not jerky upon engagement and going down hills it maintains a decent speed. It takes some getting accustomed to the distances to know when to let off the throttle (which engages regen on that bike) for a given speed, load, and distance. Everyone who rides that bike loves the twist the throttle to go and let off the throttle to slow down. It's downfall is trying to cruise at a steady state speed below about 30-40% of top speed where it can oscillate between powered running and braking.

A speed wind motor creates more regen current just like it uses more current during acceleration and higher current with soft braking force can result in greater overall energy recovery, but it does require batteries that can handle healthy charge current. In errand riding (somewhat stop and go) I regularly achieve a 15% or better energy recovery using regen, and that's with a lot of wasteful accelerations and giving a lot of energy up to the wind at 35-40mph cruising speeds, so if I geared the motor lower to make acceleration and cruising more efficient I could easily see 20-25% overall energy recovery without lowering speeds.

A speed wind motor as a mid-drive so it can be geared down to handle the heavy load it sounds like you'll be hauling would be my recommendation. Use a fixed gearing with no freewheel so you still have regen. That gives you higher rpm for better cooling due to the faster more turbulent air flow in the motor, and gives you the capability of higher power without going to extreme voltage where the controllers get more expensive. The minimum motor would be a 9C 2805...too bad the price is now double the sale price Ebikes.ca had them at last year, which was the best hubmotor deal around at the time. The next step up would be a scooter motor with a Kv of 15-17rpm/volt. Motor selection depends on the max load (weight and hill grade) that you need...another money saving tip for you, which is that you have to build for your maximum loads so the system can handle it without breaking and you run at higher efficiency at all lower loads.

 

[ddk]

You may find it useful to research a "plug brake", its a coil of resistor wire. By tapping into its length at several points, it can be a variable resistor. None of the regenerated electricity goes back into the battery, but...also none of the generated heat goes back through the controller either. They are cheap to make. It may be possible to salvage the resistor wire from an electric clothes heater or electric water heater thats being thrown away. Also toaster ovens, perhaps an electric room heater?

lol
earlier in This Thread I alluded to that being one possibility of a brake control
But
It misses the use of a low-level actuator, instead requiring as many relay/commutators as taps.
Not impossible, just clumsy

 

[ddk]

sorry I missed your reply before my nap (long one because it started raining)

It sounds like you are intent on making another one by using a motor as only a brake, because they work just fine as both.

however, this trike, if it handles like my other trike, won't handle well when being pushed from behind (tried it, didn't much like it) So I would rather not use the motors as a push vehicle.

You'll want a programmable controller so you can adjust the braking force. My most pleasant regen braking is on my cargo bike, which had a regen force so soft even on the highest setting that I did a shunt mod to allow higher current. That's a speedy wind motor too, so the resulting 20A+ of regen current is a soft braking force. What's nice about soft is that it's not jerky upon engagement and going down hills it maintains a decent speed. It takes some getting accustomed to the distances to know when to let off the throttle (which engages regen on that bike) for a given speed, load, and distance. Everyone who rides that bike loves the twist the throttle to go and let off the throttle to slow down. It's downfall is trying to cruise at a steady state speed below about 30-40% of top speed where it can oscillate between powered running and braking.

And which programmable controller do you suggest?

Soft braking would be the cake. Having the means to control the level of braking would be the icing on the cake.
~1.5kW is what I calculated I need to reduce speed down a greater-than 15% decline.
That's why I'm thinking two motors each capable of 1kW would be what I need (including safety margins) and there's the bit where I might take wrong assumptions in my calculations (happens regularly)

A speed wind motor as a mid-drive so it can be geared down to handle the heavy load it sounds like you'll be hauling would be my recommendation. Use a fixed gearing with no freewheel so you still have regen.

can't, as I'm running the rear motor thru the NuVinci hub. However the NV hub sports a disc brake and drives a fixed-gear on the rear axle, which also sports (or will sport) a disc brake

That gives you higher rpm for better cooling due to the faster more turbulent air flow in the motor, and gives you the capability of higher power without going to extreme voltage where the controllers get more expensive. The minimum motor would be a 9C 2805...too bad the price is now double the sale price Ebikes.ca had them at last year, which was the best hubmotor deal around at the time. The next step up would be a scooter motor with a Kv of 15-17rpm/volt. Motor selection depends on the max load (weight and hill grade) that you need...another money saving tip for you, which is that you have to build for your maximum loads so the system can handle it without breaking and you run at higher efficiency at all lower loads.

I'm thinking a couple of the whatever-24v-motors-yescomusa-sells mounted in 20" rims might be workable, as it's been alluded in the forums that these are high-speed winds
I tested a couple of scooter motors as e-brakes. The results were hilarious... but I wasn't using a controller. Just shorting the windings via a relay. Instant skid-stop. My funds were *very* limited at that time so I couldn't investigate different controllers for regen.
I then attempted the use of the scooter motors' drum brakes, which was at best, difficult to control and keep in alignment with one brake lever. I've since sold those motors to someone who wanted to build a scooter.
-note: After the fact I worked out a Better Way to control two brakes via one lever (better mechanically) re-working old trailer designs of yore and applying standard brake cable parts. I might use this approach to apply rim brakes as a parking brake on the trailer and a passable emergency brake (never have enough brakes for me)
Drag brakes make lots of heat/power that needs dissipated and the mechanical versions need to be biggerer than your average bicycle brake to shed the heat.
Using only one brake-at-a-time I have successfully created brake failures with ALL available bicycle brakes when keeping the speed of my trike(s) below 20mph on one particularly long, steep decline

I've looked into mechanical brakes for minibikes/gocarts/bigger scooters and new, they're a bit more expensive than a couple of hub motors and controllers once you add in the mechanical work and the fiddly bits. Ya know... in the age of washers costing 50 cents each.

When I get to the point of building the trailer and I still haven't seen a working throttle-controlled brake I'll go with mechanical brakes ...but until then here's hoping...

 

[John in CR]

Yeah I'm sorry, I have a bad habit of that, get too caught up in the response to a question that I forget all about previous knowledge of your trike. That's how I got ripped off by that Sanges character on the group battery buy. I forgot all about his bike, and had I linked the name and the bike no way I would have participated.

Yours is a tough nut to crack, because you've got an unstable trike and now a trailer in the equation. Add in steep descents and it's a recipe for disaster if you don't keep the speeds well under control. I agree with you that you definitely need braking on the trailer. You'll want the trailer hitch as close the the rear axle as possible for the least effect on steering, and with it there you may also be able to have the trailer help drive too since it's little different from driving the rear wheels.

It's going to be interesting no matter what.

 

[ddk]

Yeah I'm sorry, I have a bad habit of that, get too caught up in the response to a question that I forget all about previous knowledge of your trike. That's how I got ripped off by that Sanges character on the group battery buy. I forgot all about his bike, and had I linked the name and the bike no way I would have participated.

group buys are always risky imo. I just buy hobby king lipo and so far I've yet to receive a dud, even with buying 31 battery packs.

Yours is a tough nut to crack, because you've got an unstable trike and now a trailer in the equation. Add in steep descents and it's a recipe for disaster if you don't keep the speeds well under control. I agree with you that you definitely need braking on the trailer. You'll want the trailer hitch as close the the rear axle as possible for the least effect on steering, and with it there you may also be able to have the trailer help drive too since it's little different from driving the rear wheels.

It's going to be interesting no matter what.

Actually the trike is only unstable if you let go the handlebars.
The steering is twitchy because of the fork angles but not unmanageable.
The reasons I went with a delta trike design is two-fold:
-The number one reason is me, getting me off and on the trike without pain.
-The second reason is changing flats. Delta trikes with single-ended rear axles offer the absolute easiest wheel and tire removal.

But no one has addressed the question of which DD hub motor cogs the least.
The Aotema motor I recently installed exhibited greater cogging than the scooter motors. -who'd a thunk?

 

[hjns]

But no one has addressed the question of which DD hub motor cogs the least.
The Aotema motor I recently installed exhibited greater cogging than the scooter motors. -who'd a thunk?

I am not sure you will get the answer you are looking for. All DD motors cog when not driven. If you go for a DD "brake" solution, you will have to drive them and control their speed while driving. Otherwise you are going to see a very big performance hit. I have driven both the 9C and HT motors without electric assist and wouldn't want to do that for more than 50 meters. Estimation is that at 15km/h, I needed at least 250W to overcome cogging.

 

[nacke]

Does this mean that building a budget ebike with regen is almost out of the question? Where does that 250 watts of power go when an DD hub is rolled?

 

[wineboyrider]

FWIW I have a yescom 1000w motor coupled with a lyen 6fet infineon programmed for variable regen and it works great! I have a horn button separate so I can use regen only when slowing to stop signs etc. and can apply my caliper together. Bear in mind I have a steel specialized hard rock frame and only regen at 48v or less. I love the ability to slow down as I approach curves or stop signs it's just easy as pushing a button...

 

[dogman dan]

I'm almost afraid to speak up in this thread, but I see it as a fairly simple problem.

You want a dd motor to help you brake down the hills, great, do so, it's as easy as wineboy made it sound. You just need some kind of controller that is regen enabled. Or you could do the plug brake thing.

I have the horn button on my bike now too, and it works great when you want extra braking. I need to make mine softer though, it slows me too much down a hill.

Then to avoid drag while you run a motor 99% of the time, run the dd motor. If you need more to get up a certain hill, that's when you engage a planetary gear motor that's been freewheeling for miles.

Or just run both motors at all times, with each motor pulling less. Seems fairly simple to me. I'd say if you need it on a budget, just snag one of those yes (GM) kits, or a muxus/9c dd.

 

[Drunkskunk]

Does this mean that building a budget ebike with regen is almost out of the question? Where does that 250 watts of power go when an DD hub is rolled?

Welcome to the forum.

If you mean the 250 watts your bike is consuming when you cruse along at a constant speed, it's being consumed by overcoming wind and rolling resistance. None of that energy is recoverable.

It works like this.

imagine you have a 20 mile stretch of super flat, level roadway, and no wind at all.
You and your bike weight 250 pounds combined. You start out your bike and accelerate up to 20mph using 450 watts for 10 seconds. That is 45watt seconds, or 0.125 watt hours. That energy was converted from electric potential energy in the battery into kinetic energy, Your 250lbs in motion

Then you continue down that road for the next 20 miles at 20mph using 250 watts to maintain the 20mph. (normally an ebike motor would use ~350w, but we'll use your number since this is just theory) That would be 250 watt hours. That energy is used to overcome the wind resistance and rolling resistance.

At the end of the 20 mile track you stop by using regen braking. Your 250lbs of mass at 20mph's energy is converted through the motor from kinetic back to electrical potential. That total kinetic energy is the same as what you put in at the beginning. 0.125 watt hours.

It sucks, but you can only get out of a system what you initially put into it. The 250 watts you put into the bike over those 20 miles was just maintenance to keep the kinetic energy you initially put in, and isn't recoverable.

 

[nacke]

Welcome to the forum.

If you mean the 250 watts your bike is consuming when you cruse along at a constant speed, it's being consumed by overcoming wind and rolling resistance. None of that energy is recoverable.

Thanks for the reply, I understand the concept of overall (wind+mechanical+tyre RR etc.) resistance and kinetic energy storaged.

but hjns wrote "Estimation is that at 15km/h, I needed at least 250W to overcome cogging."

That can't be right can it? If the rolling resistance of a hub is so big, it would overheat and die after few minutes. (If you had the strength to pedal with such a resistance).

 

[Drunkskunk]

Ah, ok. Yeah, that sounds wrong. Normal human output is ~100 watts. And can I ride my huge monster bike with no power going to its low speed monster sized motor (Clyte 5404). I can pedal it up to 30KPH and hold that for a quarter mile... what ever that is in metric... with no power, and full motor cogging.

The 5404 has got to be one the absolute worst cogging motors made, and I suspect it takes 25 watts, maybe, to overcome it's effects... at 30kph