Is Regen Braking bad for Lipos?

[EBJ]

I've been wondering if regen-braking while using Lipo batteries would be a good idea.
I know you aren't supposed to over-charge a Lipo, so I thought it would work to have regen so long as you only charge the batteries to 90% capacity.

*because I imagine it would be bad to have the pack close to 100% full then go ride down a long hill w/ regen filling the pack even more.
(I ask because I live on top of a hill)

I then started thinking that maybe regen would over-volt a Lipo pack regardless of its current charge...
I'm considering a 36V system w/ about 1500W motor power. i imagine the more powerful the motor the more powerful the regen?

So, Is regen braking bad for lipos?

Thanks.

 

[neptronix]

If you're charging lipos, it's a good idea to charge to 4.15v/cell to give you headroom against disbalance in the first place.
By doing so, you also sacrifice about 5% of your capacity.

Do this and see what your voltage per cell is like at the bottom of the hill. My guess is that it will not be anywhere close to 4.2v/cell. Regen doesn't take a lot of energy in.

Lipos can take up to about 5 volts before blowing up. You don't want them going over 4.2v/cell but they can do it without anything tragic happening.

If 4.15v/cell max charge isn't good enough, lower it to 4.125 or 4.10v per cell in your charger.

 

[el_walto]

You can use Regen with lipo. My Regen cuts out once it hits 57V, set like that in the controller

 

[EBJ]

You can use Regen with lipo. My Regen cuts out once it hits 57V, set like that in the controller

Thats where I'm a little confused. So if your system (i'm guessing it's a 48V system you are running) ever hits 57V then your controller shuts down regen.
But to what effect do the batteries actually see when you hit that cut-off (57V) ? Because if it were your 48V pack going to 57V I imagine your lipos would go bye-bye.
correct?

I guess what I'm wondering is: How much voltage do the batteries actually see when you are re-gen braking ? lets say on a steep hill at full regen power... hum.

 

[EBJ]

If you're charging lipos, it's a good idea to charge to 4.15v/cell to give you headroom against disbalance in the first place.
By doing so, you also sacrifice about 5% of your capacity.

Do this and see what your voltage per cell is like at the bottom of the hill. My guess is that it will not be anywhere close to 4.2v/cell. Regen doesn't take a lot of energy in.

Lipos can take up to about 5 volts before blowing up. You don't want them going over 4.2v/cell but they can do it without anything tragic happening.

If 4.15v/cell max charge isn't good enough, lower it to 4.125 or 4.10v per cell in your charger.

Thanks for the info. That's what I thought would be a good solution. I just question how much the battery voltage "spikes" when you get the regen going.
I've also read that people had problems w/ charging (when using the Hyperion) when trying to limit the cell V... instead they just resulted in limiting the % capacity.
(I'm still not sure how you can get limited capacity without a lower cell voltage....? maybe I mis-read something)

 

[knoxie]

never liked regen its pretty much worthless on an ebike IMHO as you have such little mass to generate any meaningful power back in the batteries, you can gain better range by careful throttle use and controlling your speed, if you are worried about it get an ebike that freehweels well and do just that, its much more efficient on an ebike to just freewheel as long as your drive doesnt cog the motor too much.

And yes I have thought for a long time that regen inst good for batteries, so many folks on the Tidal force forum had issues with it, I have never and will never implement it on a push bike, electric car maybe but not a bike, Justin made that mistake on his Canada trip, charged up at the top of a hill then had 15 miles down hill where he couldnt effectively brake as he would have overcharged his pack, he was in the rocky mountains at the time.

You are also in essence bulk charging your pack and as most dont run active BMS boards on their kits this could be a problem

I am sure others will pitch in with some figures but i think that it doesnt really give you back anything at all, the figures you see on the CA are just effectively your free-wheeling watts.

 

[neptronix]

Cell voltage and state of charge are one in the same actually.
Hyperions are a bit weird, you have to set them to like 95% to get the 4.15v/cell.

The cool part is that you save yourself some energy charging by doing this, lol..

BTW if you had a voltage limit on your regen, that would suck, because regen would cut out and you'd get no braking effect, which you definitely need on a heavy ass DD motor going downhill!

Don't worry about voltage spikes, That's how regen works and it's fine on every other battery, so lipo isn't some special exception. There are a lot of people here using regen with lipos

 

[liveforphysics]

You put a voltage maximum limit on your regen function of the controller to cut-out regen if the pack voltage gets too high.

If you're starting at the top of a big hill though, and you charge your pack to full before you leave your house, then you shouldn't be trying to store more energy by regen'ing, because your pack is all ready full. No room to capture that energy.


My feelings on regen are the same as Knoxie's. On paper it seems like a good idea. In practice, it's generally worse/neutral/or an extremely mild range increase that isn't worth the free feeling of being able to coast when desired.

 

[curious]

In a programmable controller regen cutoff voltage should be set at or slightly below full pack voltage. Unless you live at the top of large hill this is not really an issue.

so many folks on the Tidal force forum had issues with it

I've used proportional regen for two seasons with zero problems. It does not offer huge efficiency increase in my riding conditions (mild hills). I typically see 4-7% regenerated capacity. If you live in a mountain terrain that efficiency increase will be much higher. If you already have a DD hub there is no efficiency loss from regen option. Choosing freewheeling hub vs DD with regen in terms of efficiency is more complex, it really depends on terrain, rider weight and special needs like loaded touring.

But besides efficiency regen gives you a very powerful and linear "boosted" brake action with no pad wear, no adjustments, no squeaking etc. Very, very nice feel really. You should try it...

 

[biohazardman]

If you have batteries that will accept the regen, Lipo should be fine, and if the system is set up correctly it is great. I have done the regen thing and because my batteries were not up to the task had some controller problems. Seldom had to use my brakes though so a big plus there. Still an 6-14+% return is great but it depends on the stop and go as well as hills and how you ride how much regen will do for you. Yes, regen set up poorly can do damage to your batts especially if you top off your charge and go down a long hill. So don't top them off or run a heating coil or something that can be used to burn off the extra when you need.
Having had both regen and non I have to say both have positives and negatives. Me I would prefer it but don't like the weight of a DD motor so do without for now. My a123 batts seem to be very tolerant of overcharge and over discharge 4.3v-.8V , they survived both, so might be another option.
In the end like most things if you do it right all will be good if not then maybe not so much.

 

[itchynackers]

The coasting issue is fixed by holding the throttle open just a tad, so you are only putting 0.xx amps out. Then regen isn't activated, and you coast when you want to.

 

[curious]

The coasting issue is fixed by holding the throttle open just a tad, so you are only putting 0.xx amps out. Then regen isn't activated, and you coast when you want to.

This is controller specific. Kelly does not require throttle management for coasting (although it can be programmed this way), on my setup regen is controlled by brake lever with ratiometric hall sensor.

 

[CrazyJerry]

Regen is bad for any battery if the battery can't take what the regen is dishing out. I use an external regen and it makes a nice speed control down the hills here with some added juice returned. Without it I would destroy brake pads, warp drums, etc. It's not possible to freewheel down a couple of these hills unless you have some baby-wipes handy.....
- CrazyJerry

 

[curious]

Regen is bad for any battery if the battery can't take what the regen is dishing out.

Actually it is not that bad. Controller/motor combos that we use are fairly lossy, typical efficiency is around 70% and it works both ways: for a 1W mech. power at the wheel we need like 1.4W of battery power during acceleration and get back about 0.7W during regen deceleration. Simplifying a bit, braking power at say 20A regen current back to the battery feels similar to 40A current during acceleration. Example - I weight around 240lb and my x5 bike is fairly heavy and I average around 8A regen current on a long twisty downhill on my favorite 25mi loop that otherwise requires continuous braking for safety (because of few turns).

Modern lipos can handle over 2C continuous charge and I am fairly certain they will handle even more in short bursts without long term effects. Heat kills lipos, charging lipo is endotermic process I beleive. So I would not be too much worried about slightly exceeding continuous charge rating during regen bursts as long as proper LVC/HVC monitoring is used.

 

[EBJ]

Thanks everyone for the replies.

I was afraid this would turn into a regen vs no-regen debate. Glad it didn't.
In that last scenario Curious explained. I think I feel safe with just charging my Lipos to 90%....
it seems like it would take a hell of a long time on the regen before I over-charged the lipos if they were started at 90%.
Especially a 20Ah pack.

I'll b going ahead with the lipos.
Thanks for the info.

 

[dogman dan]

You could just have just one brake handle that activates regen. I'm assuming that you would engage regen with the brake handle, not just letting off throttle.
So you could have the front brake work with no regen, and use that going down the first hill. Once you ride a short way, then your regen would be fine.

Or if there is more hill, ride uphill 1/4 mile at the start of your ride, then turn back down the hill.

Lotsa ways to make some room in the battery so you could regen that hill. Just charging to 4.1v of course could be the easiest.

 

[Harold in CR]

I have the same conditions as EBJ, living on top of a hill.

I would like to ask about a way to switch on-off the regen, mostly for the downhill braking effect. Would it be possible to have a switch hooked up to the motor leads, so it would be able to switch from regen to the Batteries, to NO regen, but, shorted to each other, to get more "drag", so you still get resistance braking ?? Hope I stated this correctly ?? Maybe even a 3 way switch, so you could cut off ALL resistance for free wheeling, if desired ?? Thanks

 

[999zip999]

Wouldn't Dogman's way work for that if you remenber to use the rear brake at top charge or just 4.1v -lipo ? What are the other possible problems ? Great info. Thanks

 

[CrazyJerry]

Regen is bad for any battery if the battery can't take what the regen is dishing out.

Actually it is not that bad. Controller/motor combos that we use are fairly lossy, typical efficiency is around 70% and it works both ways: for a 1W mech. power at the wheel we need like 1.4W of battery power during acceleration and get back about 0.7W during regen deceleration. Simplifying a bit, braking power at say 20A regen current back to the battery feels similar to 40A current during acceleration. Example - I weight around 240lb and my x5 bike is fairly heavy and I average around 8A regen current on a long twisty downhill on my favorite 25mi loop that otherwise requires continuous braking for safety (because of few turns).

Modern lipos can handle over 2C continuous charge and I am fairly certain they will handle even more in short bursts without long term effects. Heat kills lipos, charging lipo is endotermic process I beleive. So I would not be too much worried about slightly exceeding continuous charge rating during regen bursts as long as proper LVC/HVC monitoring is used.

Actually I am thinking more of the weak battery in the string. Your blanket statement is well taken if all the batteries are healthy but in the case of a weak cell a very different outcome may happen... Even though a battery string may be able to take the amps, what might happen in the voltage on cell-x that were to rise beyond what it should? Nice if the bms is there to correct. I agree that your 8 amp figure is probably not very harmful but I've seen considerably larger spikes and sustained currents on a few hills here... Just my 2 cents.
~CrazyJerry

 

[CrazyJerry]

I have the same conditions as EBJ, living on top of a hill.

I would like to ask about a way to switch on-off the regen, mostly for the downhill braking effect. Would it be possible to have a switch hooked up to the motor leads, so it would be able to switch from regen to the Batteries, to NO regen, but, shorted to each other, to get more "drag", so you still get resistance braking ?? Hope I stated this correctly ?? Maybe even a 3 way switch, so you could cut off ALL resistance for free wheeling, if desired ?? Thanks

Yes. If you're using a DD hub (3 phase) you can easily hook up an automotive alternator rectifier to those leads, then use the Pos+ and Neg- from the rectifier straight to your switch. Use a heavy dc switch. Some of the heat from this type of "brake" will accumulate in in the aluminum heat fins of the rectifier but it works well. You must be on the ball and remember to disconnect (turn off) the switch before you hit your normal controller throttle though... Using the rectifier and something like a momentary DPDT switch you could charge battery pairs in a 48 volt system... I have successfully done this on a 48 volt system (24 volts on each side of bike). This allows for some "engine braking" so to speak down long grades without having to cycle a purely on-off dead short type of setup. If you have a regen enabled controller than all of this is mute, but if you want regen external of the controller and watch what you are doing this works well and is super simple and robust. Should one side of the batteries be low, you can bring up that side if necessary on-the-fly... How whacked is that?
~ CrazyJerry
Pic of rectifier and switch: http://diesel-bike.com/Aurora/aurora_files/Bargain_Brand_Regen.JPG
Caution: Use at your own risk!

 

[auraslip]

bah, regen haters gonna hate.

You'll go through mad brake pads without it. For that reason alone it's indispensable IMHO.

Two options for disabling regen:

Use one e-brake lever and one normal brake lever as I've done on all my bike.

Run a spst switch from the BR pad to GR so you can disable regen when you feel like it.

Honestly though, where do you live that has hills large enough for regen to overcharge your batteries? And if you do live in that area, regen would freakin be a goldmine!

 

[Harold in CR]

If y'all are posting to me, that is very good info. I live about 1500' elevation, and, it's nearly ALL downhill, in a 2.5 mile stretch. I have the occasional flat between hills. I need all I can get for the ride back up the hill to the house.

At the moment, my initial controller will have NO features. It's not a bike controller, so, I NEED to stop at the bottom, or, run out into a 2 lane road. :lol:

Thanks for the info. Harold

 

[curious]

I live about 1500' elevation, and, it's nearly ALL downhill, in a 2.5 mile stretch.

At the moment, my initial controller will have NO features.

In your case rheostatic braking makes perfect sense. There was a large thread on it a while ago.