Using Reverse as an Electric Brake?

[mi7d1]

I have a 20a 36-72v clyte controller with a reverse switch. After reading several times about the hub motors having a higher rolling resistance rolling backwards with the controller turned on, I decided to try a little experiment while rolling into my parking spot at work. I flipped the switch into reverse and could feel an increased drag immediately without using the throttle. I also tried the same thing using the throttle and watching my WattsUp meter, but in the short distance I was braking I could get enough information from my meter to help. I would like to know if there is any danger to my motor or controller using it in reverse with or without the throttle while moving forward to assist in electronic braking? I was/am interested in regenerative braking not to re-charge the batteries but as an electric drag brake. My velomobile has two Sturmey-Archer drum brakes on the front wheels and nothing on the rear. It would be nice to be able to use the hub motor as an electronic brake without fear of damaging my motor or motor controller and a bonus would be to be able to use the throttle in reverse to increase the braking. Any advise or comments?

edited for better subject title. Originally posted as "Electric Reverse Brake"

 

[TylerDurden]

Stuff may not live long. It would be like trying to pull a car with your bike.

Resistance pulls big current and generates big heat. If you have any electric power-tools with electric braking, you may be familiar with how fast the braking burns up batteries and brushes.

Braking without regen can only generate heat. The kinetic energy has to go somewhere... just like your car, the energy is converted to heat by resistance: in a car through process of friction, in a motor through the process of emf.

If you want a non-regen drag brake, you might go for air resistance like a drag chute or flaps on a jet's wings... a hefty friction-dynamo and a big red light on the back of the bike could give you regen and a really bright brake light.

 

[fechter]

Tyler's right. Stuff probably won't live long. When you go to reverse, even with no throttle, basically you're shorting out the motor. In this mode, there is no current limiter, so above a certain speed (not very fast), the current will be high enough to blow the FETs, not to mention the wheel will try to lock up.

Hmm.... it does get me to thinking though, If you could pulse the reverse switch at a high frequency, like 15kHz, with a short duty cycle, you might be able to get regen out of it. I'm not sure if the commutator chip can switch that fast, but if it does, it just might work. Changing the duty cycle of the pulses would control the amount of braking.

A simple oscillator, like an LM555, could supply the pulses.

I'll have to try this on my bench tester.

 

[mi7d1]

Thanks guys,

I hadn't thought about heat. I was only thinking about the electrical side of the idea. As far as "pulse the reverse switch t at high frequency" is not only beyond my abilities at the moment, but I can barely grasp the idea. I'm glad I only tried my experiment at the speed of walking.

On one of my routes to work I have a hill about two blocks long and the velomobile can get to 30-35mph by the bottom where there is a stop. I'd like to be able to slow down without ridding the brakes all the way to the bottom of the hill.

I have seen drag chute experiments on velomobiles. Not really needed on the hills in my area however a fun idea to think about especially a retractable and re-deployable drag chute.

Are there any options or hacks with the clyte controller to implement regenerative braking or would it be best to look toward another motor controller such as an Alltrax or Curtis controller if I want to pursue this idea

 

[Rassy]

Hi mi7d1. On an earlier project I added a power trailer to a cheap mtb with poor brakes. Since I live on a steep hill, I had to creep down at about 5MPH, using the brakes all the way, to avoid any disaster. So I came up with a version of regenerative braking that has worked out pretty good, and I still use it on my trike, even though the trike has pretty decent brakes. I have a simple three position toggle on the bike. Forward position connects the motor(s) to the controller(s), via relays, for normal use. Back position connects the motors directly to several 130 Watt auto headlight bulbs, via relays, for braking.

The downside is, I still have to keep under 13MPH when braking or the bulbs will be over volted. This works okay for me, because my hill ends up on a blind corner with rough pavement. I could use pairs of bulbs in series to allow the motor to produce up to 24 volts without damaging the bulbs.

Someday I hope to improve this system so that it can be used at any speed, and even recharge the batteries, instead of lighting bulbs, but most components for this are spendy, and with 3 motors I would need 3 systems.

Note: I have brushed motors. May not work with brushless?

 

[mi7d1]

Hi mi7d1. On an earlier project I added a power trailer to a cheap mtb with poor brakes. Since I live on a steep hill, I had to creep down at about 5MPH, using the brakes all the way, to avoid any disaster. So I came up with a version of regenerative braking that has worked out pretty good, and I still use it on my trike, even though the trike has pretty decent brakes. I have a simple three position toggle on the bike. Forward position connects the motor(s) to the controller(s), via relays, for normal use. Back position connects the motors directly to several 130 Watt auto headlight bulbs, via relays, for braking.

The downside is, I still have to keep under 13MPH when braking or the bulbs will be over volted. This works okay for me, because my hill ends up on a blind corner with rough pavement. I could use pairs of bulbs in series to allow the motor to produce up to 24 volts without damaging the bulbs.

Someday I hope to improve this system so that it can be used at any speed, and even recharge the batteries, instead of lighting bulbs, but most components for this are spendy, and with 3 motors I would need 3 systems.

Note: I have brushed motors. May not work with brushless?

That is an interesting idea and well within the scope of my abilities. Anybody have an idea if this would work with a brushless motor? I have noticed when rotating the wheel while the velomobile is on it's work stand that the WattsUp meter comes on and gives me a reading.

 

[The7]

Hi mi7d1. I have a simple three position toggle on the bike. Forward position connects the motor(s) to the controller(s), via relays, for normal use. Back position connects the motors directly to several 130 Watt auto headlight bulbs, via relays, for braking.

Note: I have brushed motors. May not work with brushless?

Yes, this is the regenerative braking which is also used in brushless motor without freewheeling. This regenerative energy is usually dissipated as heat in winding resistance of the motor and/or external resistors.

The 3-phase BLDC motor will generate a back 3-phase emf and this regenerative energy could be controlled by using mechanical switches (relays as your idea) or using electronic switches (power FETs and diodes).
Regenerative braking will work with brushless as well as brushed motors.

 

[monster]

if

back emf (V) - battery voltage = regen voltage

then why not have a relay switch that taps the battery down to a lower voltage for regenbraking.

i have 72v. if the relay cut this down to 24v or 12v this would then be usefull for regenerative braking.

 

[fechter]

Switching the battery to a lower voltage would work, but you might have a very high surge current when switching. This could also cause the wheel to lock or overstress the motor mount (dropouts).

I did try sending a pulsed signal to the reverse switch in a Crystalyte controller. It did result in some regen back to the power supply, so it just might work. The switching speed of the reverse circuit is pretty slow, so I suspect much of the regen energy was being absorbed by the FETs rather than being circulated back to the batteries. My test was done with a tiny 48w motor on the bench, so I'm not sure exactly what would happen with a real bike motor.

 

[rhubarb]

I wired up the little red button on the throttle to the reverse pins in my crystallite controller for fast three point turns in my SillyKart.
I would usually hit reverse before coming to a complete stop to help pull up which didn't cause any problems. I wouldn't do it at speed!

 

[monster]

what about a dynamo ?

could a regular dynamo be modified to do regenerative braking? beef it up a little?

 

[monster]

what about a car alternator? search ebay for dynamo and look at the clasic car parts

 

[fechter]

Something like that could be made to work, but the bike motor is already pretty much the same thing, so it makes sense to use it.

If you just want braking, with no charging, a giant resistor will work.

I suppose it might be possible to use a transformer with a brushless motor to boost the voltage for braking. Transformers are heavy, but then so are car alternators. You might be able to use a torroidal transformer that's not so heavy. For a 3 phase system, you really need 3 transformers.

 

[monster]

The downside is, I still have to keep under 13MPH when braking or the bulbs will be over volted. This works okay for me, because my hill ends up on a blind corner with rough pavement. I could use pairs of bulbs in series to allow the motor to produce up to 24 volts without damaging the bulbs.

a projector lamp or hydrophonics bulb can take 110v 750w or attach a kettle heating element? you can dump loads of amps across that !
http://cgi.ebay.co.uk/Projector-Lam...ryZ20706QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

i have a digital thermometer probe on my batteries to check temperature. if you were worried about reg heating the controller you could monitor this using http://cgi.ebay.co.uk/Digital-DUAL-...4QQihZ017QQcategoryZ93679QQrdZ1QQcmdZViewItem

 

[Rassy]

Monster wrote:
a projector lamp or hydrophonics bulb can take 110v 750w or attach a kettle heating element? you can dump loads of amps across that !

The problem with my "poor man's" light bulb brake system is that they don't draw a full load until they near their rating. I could put bulbs in series so that higher voltages would not hurt them, but then the corresponding braking speed would increase. Right now, with 390 Watts per motor (3 motors) on a 15% grade I coast at 12 MPH with no additional braking. If I could find light weight inexpensive variable resisters with 500 to 1000 Watt ratings (3 would be required) I would try something more sophisticated. Does such an animal exist? Of course if the system is changed the "cool" brake lights (the bulbs are in old tail light housings on the back of the trailer) would be eliminated.

 

[monster]

The problem with my "poor man's" light bulb brake system is that they don't draw a full load until they near their rating.

thats weird. i dont understand why that would be? are you saying my kettle element brake system would not work either?

i really want to try this now because i don't have a rear brake. i have good stopping power with both a disc and v-brake on the front but all my bike weight is distributed on the back. this means that in the wet i tend to pull front wheel skids if i brake too sharply going down hill! -i did this today and it scared the crap out of me.

 

[Rassy]

Monster wrote:
i really want to try this now because i don't have a rear brake. i have good stopping power with both a disc and v-brake on the front but all my bike weight is distributed on the back. this means that in the wet i tend to pull front wheel skids if i brake too sharply going down hill! -i did this today and it scared the crap out of me.

I know what you mean. On my original mtb with trailer both the front and back wheel would slide on my driveway at a real slow speed, when it was wet, and if it was dry, I could still barely get stopped at the end of the driveway. That's what prompted be to install the motor braking system, which eliminated both problems, as well as allowing me to go down the steep hill to town comfortably at 12 MPH instead of creeping at 5 MPH.

Rassy wrote:
The problem with my "poor man's" light bulb brake system is that they don't draw a full load until they near their rating.

Monster wrote:
thats weird. i dont understand why that would be? are you saying my kettle element brake system would not work either?

I don't know how a "heating" element would work as opposed to light bulbs. I tried my system with the 12V bulbs in 2 bulb series (for 24V) and the braking power was retarded to a higher speed. That test was on the old mtb. Now that I have the trike with good brakes I am considering going to the 24V series arrangement again and coasting down the hill at a higher but still controlled speed.

I did read somebody's blog where the guy just directly shorted the + and -wires from the motor and let the windings absorb all the heat generated. He said it worked for braking purposes, but the motor got very hot. I didn't think it seemed like a very good idea. If you do some tests with heating elements let us know how it works. I'm still hoping someone knows of a variable resister that could handle the voltage and amps the motors create, and I would just replace the bulb system with it.

 

[monster]

Ok I’ve done a few tests and read a bit more and here’s what I think + know now.

I did the short out motor (+ & -) thing on two bikes. First with a brushed motor and again on my crystalyte X5053. The breaking performance seemed real violent on the brushed motor so I’m now worried about locking up the back wheel but that was from a no-load condition. The brushed motor bike was chain driven on a free wheel, so I had to tie the sprocket and wheel together for the test. The wire I used got ripped up by the force!

When I did the same test on the crystalyte I unplugged the controller and shorted all three wires to the motor. The brushless crystalyte is interesting because it has three phases and generates AC! (correct me if I’m wrong on any thing here) multi-meter only registered volts when switched to AC mode.

i noticed that if i shorted only two of the three wires the braking was not as harsh. -this could give us the opportunity to have three levels of braking? It also means we would need three resistors (kettle element, light).

Here’s what I think- a variable resistor would not give you variable braking because as soon as you close the circuit you will be dumping all the amps. Weather it’s on the motor coils or the resistor doesn't matter, you will still get the same braking force. Weather you just short the wires or connect through a big resistor only makes a difference to where the heat goes and a big resistor will soak up all the heat saving your motor.

I think this will work. What type of relays are people using for electric bikes?

 

[Rassy]

Monster, pretty bold step to directly short the motors. I won't do that because I worry it may cause damage, as I said previously, the guy that set his bike up that way did report excess heat. My bulb system seems to start braking gently, even when I am first going my maximum allowable speed (12 MPH with 12V bulbs) and then throw the toggle to brake position.

As far as relays, I am using basic automotive 12V 800 Watt ones from JC Whitney. I think they are made for auxillary lights or high power sound systems.

http://www.jcwhitney.com/autoparts/Search?catalogId=10101&storeId=10101&sku=relay+800+watt

 

[dirty_d]

what about wiring some MOSFETs in parallel able to handle the current draw at stall of your motor, across your motor terminals and using a second hall throttle to control them so you can have very light braking to the amount of braking if you shorted the motor terminals.

 

[monster]

i'm not gonna short the motor im gonna use a big resistor (kettle elements probably). i just did that for a one off test.

but this is just to begin with. if all goes well i might try the three transistors and three bridge rectifiers idea to get full regenerative braking back to the batteries -not going through the controller. i am thinking of a push to make button on the handlebars, no throttle.

I have no idea what components are in the controller so i dont want my regen braking going via the controller.

 

[dirty_d]

so hows that work? i dont really get it, the backemf needs to be higher than the battery voltage to charge it and the only time that would happen is if the motor us turning over the no load speed, so what do you have to do? do you make it so when you turn on the brake it charges the batteries in parallel so if there is 20V backemf it would be like charging a 12V battery with 20V while braking? and whats the circuit look like?

 

[monster]

You use transformers to bump up the voltage to whatever you need for charging. it will only work on brushless motors cos they generate AC not brushed motors which generate DC.

transformers can be used to increase or decrease the voltage of AC. the power stays the same so the amps change in the opposite way.

then you need to change that highter voltage AC back to DC for charging so you use a bridge rectifier.

i only did electronics GCSE not A-level and that was 10 years ago so some of theis may be wrong

what you said about the series parallel switch might work as well.

 

[mi7d1]

Very interresting read. Most is beyond my understanding and ability at this time.

I think that my velomobile maybe vulnerable to high back emf due to its extreme aerodynamics. I have a 5% downhill grade for about a half a mile that I can coast from a standing start and be doing 40mph at the bottom without having peddled. I am more concerned about my controller than the batteries for the time being. When I move away from the SLA's into the Lithiums then I'll be concerned about the batteries also.

 

[dirty_d]

i suppose a boost converter would work, varying the duty cycle depending on how much brake you give it and what the backemf from the motor is.