How does regen work?

[ryan]

I'm testing out the regen option on Lyen's controller (just connect the jumpers), but I'm a bit confused at how it works.

I climbed a big hill (600ft) and looked at my current amps used (1.800). I tried to regen as much as I could on the descent and barely got to 1.790. But with one quick flick of the throttle I'm up to 1.820A.

The quick display metrics show that as soon as I cut the throttle I my current amps jump to -5 or -7, but then quickly go up to ~ -0.01. So it looks like I only really get the benefit of regen for the first few seconds after cutting the throttle.

I know that regen won't change my life, but I was hoping to see some 5% return on a the downhill portion of a climb. Do I need to configure my CA to take advantage of this? Any other tips? How does this work?

 

[liveforphysics]

Regen in a controller as a max voltage cut-off. If your headways packs have BMS's, they also have some highvoltage cut-off point in the charge circuit.

If either of those limits is reached, it will stop regen to prevent what it thinks is the risk of over-charging a battery.

I don't know if this is what is happening in your situation or not, but perhaps as a test, give the same hill a decent while your pack was perhaps below half-way discharged, and see if it maintains regen for you.

 

[DrkAngel]

Regen does sound like an impressive, complex technology, however, due to inefficiency and cost, your money might be better spent on a larger battery. Unless you ride extremely "hilly" areas, where additional braking is advisable, prolonged or steep downhills etc.

 

[Alan B]

Regen is useful for saving your brakes on those descents. There are settings in the controller for how much regen is applied, and there is a maximum voltage it can produce. It also doesn't work much at lower speed. What speed were you going, and what voltage are your batteries. Have you corresponded with Lyen? It should produce significant braking effect if it is working right. It does not sound like yours is working yet.

 

[Jeremy Harris]

Regen does sound like an impressive, complex technology, however, due to inefficiency and cost, your money might be better spent on a larger battery. Unless you ride extremely "hilly" areas, where additional braking is advisable, prolonged or steep downhills etc.

Nothing really complex about regen, it just means switching the FETs in the controller in a different way. It doesn't need any extra components, either, apart from a brake switch to activate it and doesn't add any cost to the controller. In terms of efficiency it's about the same as the controller working normally, so probably better than 95%.

The energy gain is tiny, though, and very dependent on total weight and terrain. Even a heavy vehicle like a car can only recover around 8% of the energy it uses at best, with a bike I think you would be lucky to recover half of that, due to the different balance between the energy needed to overcome aerodynamic drag and the energy needed to accelerate and climb gradients.

It's a good way of saving the brakes though, and can provide progressive braking that naturally fades as speed decays.

Jeremy

 

[Philistine]

My understanding is that on the Infineon controllers regen has an effective max regen voltage of 60v. Your signature says you are running 72v, so I presume the regen isn't working because of your higher voltage.

I am running 72v as well (20s lipo, so 84v hot), and like you, when I tried throttle regen I could see minor regen on my CA, but I couldn't even feel any regen braking. But on my friends bike running 48V I could really feel the throttle regen braking.

My understanding is that because the Infineon's have a max regen voltage of 60V, if you want to use regen over this voltage, then you have to mod the R12 resistor, such that you effectively trick the controller's perception of voltage. This will also mean your LVC is effected as well.

I have been looking at this, because I would like regen not for the battery regen effect, but for its braking capability. Basically I get no braking effect from regen either.

If you want to see a thread on this issue, this one is good:http://endless-sphere.com/forums/viewtopic.php?f=2&t=8317&hilit=regen

 

[Lessss]

How does regen work?

First it slows you down, then it cuts out unexpectedly. When you re-pull the brake it then blows up your controller.

 

[Spacey]

Regen is great for ripping apart your bike if you do not have torque arms

 

[LegendLength]

I am running 72v as well (20s lipo, so 84v hot), and like you, when I tried throttle regen I could see minor regen on my CA, but I couldn't even feel any regen braking. But on my friends bike running 48V I could really feel the throttle regen braking.

I'm running at 72v (lipo 66.6v nominal) on a lyen 12 fet and I can feel the regen braking, although it doesn't stop me completely like a brake. Did you have the white cables connected to each other when you tested, which enable the throttle regen? I removed mine and set 'fake indicate' in the controller and now it coasts (no throttle regen) but I do have brake regen. I was just thinking maybe the throttle regen is masking the brake regen for you and that's why you can't feel it.

 

[Philistine]

Philistine wrote:
I am running 72v as well (20s lipo, so 84v hot), and like you, when I tried throttle regen I could see minor regen on my CA, but I couldn't even feel any regen braking. But on my friends bike running 48V I could really feel the throttle regen braking.

I'm running at 72v (lipo 66.6v nominal) on a lyen 12 fet and I can feel the regen braking, although it doesn't stop me completely like a brake. Did you have the white cables connected to each other when you tested, which enable the throttle regen? I removed mine and set 'fake indicate' in the controller and now it coasts (no throttle regen) but I do have brake regen. I was just thinking maybe the throttle regen is masking the brake regen for you and that's why you can't feel it.

LegendLength (How long is a legend these days BTW? Awesome username..) didn't mean to cause confusion. I have only tried the throttle effected regen (I am waiting on ebrake levers to arrive). I have never tried the actual brake lever regen (the OP referred to throttle regen, and that definately doesn't work for me - and then when I read the threads about regen requiring modding to the R12 resistor for over 60V I presumed that would mean ebraking with a lever wouldn't work for me either). But from what you are saying I am excited.

Just like your setup I wanted coasting (no throttle regen), and then just my ebrake lever to work. Look forward to my lever turning up now....

 

[Lessss]

Even at 20 mph top speed I'd rather have regen than not to supplement brakes.

 

[LegendLength]

I have only tried the throttle effected regen (I am waiting on ebrake levers to arrive). I have never tried the actual brake lever regen (the OP referred to throttle regen, and that definately doesn't work for me - and then when I read the threads about regen requiring modding to the R12 resistor for over 60V I presumed that would mean ebraking with a lever wouldn't work for me either).

Ahh ok. Maybe it's bad though because I actually could feel the throttle regen on my system. It felt like engine braking in a manual car although it didn't bring me to a complete halt. But definitely noticable, not subtle.

I saw you have an infineon controller but for some reason thought you wrote a lyen controller. I know they are similar but perhaps my lyen 12 fet handles regen differently? The 'EBSLimVoltage' field in the controller software has 3 values for mine: 55, 60 and 75.

 

[John in CR]

DrkAngel
Regen doesn't cost anything, and even if it did, it would be worth every penny. What makes regen great is for braking, and it tremendously reduces your brake maintenance, as well as increasing safety down big hills because it eliminates the potential for brake fade. What it adds back to the pack is just icing on the cake. When I first got mine my consumption went down almost 10%, but I was riding hard using most of the energy for acceleration and throwing it away with hard braking, so I was wearing out brake pads every 4-6 weeks.

I use the throttle regen, because it's just so easy...twist to go, let off to brake. There is that one small range of medium slow speed that the controller can't decide between slow or go, but that's the only inconvenience I've had after learning how to manipulate the throttle for coasting and developing a good feel for regen braking distances to end up using the mechanical brakes as little as possible.

 

[spinningmagnets]

Its my understanding that if ANYthing gets warm while discharging the battery, part of the measured power coming from the battery is being converted to heat (part of the reason its better to up the volts instead of the amps for more power whenever possible). Lost heat on the uphill, can never be re-absorbed on the downhill.

Also, its seems most battery chemistries can discharge faster than they can be recharged. By that I mean, on a long downhill, your hub may be producing more regen amps than the battery can swallow. This brings up an interesting question in regards to LiPo batteries,...can they absorb more of the available regen watts than LiFePO4? (I don't know)

I agree with John in that the best part of having regen would likely be that your brakes stay cooler for an unexpected emergency stop at the bottom of a long downhill. I have rolled through an intersection with my hot brakes on full (luckily no accident!). I have heard of simple "plug brakes" being installed to add the E-braking effect to a bike with a non-regen controller.

If the harvesting of available regen watts is important to you, it may be useful to investigate paralleling a super-capacitor with your battery. I don't know what's needed for that, but I recall reading about an E-scooter that had it, and sounded like it was working OK....

 

[LI-ghtcycle]

So, if I understand correctly, typically regen is actuated with either backing off the throttle, or using E-Bike brakes? I have what I believe is a 9 FET Infinion controller, and I was thinking that the "on/off" switch on the throttle might be the way to turn on the regen, (didn't come with E-brakes) just curious what everyone thinks, I will be riding soon than I thought! The roads have cleared up!

Also, if I do have the throttle style regen, is it all or nothing or progressive? And is there not a "coasting" range in-between? Wouldn't make sense to have the throttle either causing braking or accelerating?

 

[John in CR]

I think the max current Lyen's controllers do on regen is 10A, so think of what 10A feels like during acceleration and that's the force you get of deceleration. 10A shouldn't be generating much heat either, though I guess some of the lower power and low capacity packs may not like it.

 

[amberwolf]

So, if I understand correctly, typically regen is actuated with either backing off the throttle, or using E-Bike brakes?

Depending on the controller design, yes. I have one brushless and one brushed that do the former, and all the rest use the latter.

I have what I believe is a 9 FET Infinion controller, and I was thinking that the "on/off" switch on the throttle might be the way to turn on the regen, (didn't come with E-brakes)

You can use that, yes, but I'd personally use a momentary switch instead. If it has a horn button, use that. Using the on/off switch may be problematic if you flip it off to brake and then need to accelerate hard due to traffic or something, you won't be able to....

Also, if I do have the throttle style regen, is it all or nothing or progressive? And is there not a "coasting" range in-between? Wouldn't make sense to have the throttle either causing braking or accelerating?

For both of mine with "demand speed" throttle control (which actively brakes when throttle demand is less than actual speed), braking is not progressive. THey both brake as hard as possible given the speed they're at and the battery voltage/current.

But the brushless one can coast, if you let the throttle snap to zero instead of reducing it. The brushed one doesn't--it is a pure analog controller and doesnt' know the diffference.

So it depends on how it'd designed.

 

[LI-ghtcycle]

Kewl thanks for the info! As it turns out, what ever kind of controller I have, it's a 12 FET, and it doesn't play well with my 9 x 7 for some reason, so it's out of the equation for now.

I will just stick with getting one from Lyen eventually once I decide what motor to use. Thanks!

 

[DrkAngel]

How does regen work? - Poorly!

Even the best systems, seem to recover, less than, 20%.

Given the lack of efficiency, combined with the increased complexity, (multiplying the possibility of problems, or failure), it just doesn't seem worth the extra cost.

 

[MadRhino]

Regen feed back my Lipo about 5%, nothing important as energy saving, but it makes very efficient braking "compression", true lifesaver at high speed on slippery surface.

 

[LegendLength]

How does regen work? - Poorly!

Even the best systems, seem to recover, less than, 20%.

Given the lack of efficiency, combined with the increased complexity, (multiplying the possibility of problems, or failure), it just doesn't seem worth the extra cost.

I don't think there's much complexity as long as the controller supports it (e.g. a Lyen controller). The only issue I have, and have seen on my own bike, is the torque arms get massive strain on them if not secured well.

The upside isn't so much the 5% extra range you get, it's more about saving your brake pads from heat and wear.

Nice cat by the way I've got the same type but he honestly does look very handsome for a russian blue.

 

[Bubba]

If the harvesting of available regen watts is important to you, it may be useful to investigate paralleling a super-capacitor with your battery. I don't know what's needed for that, but I recall reading about an E-scooter that had it, and sounded like it was working OK....

Interesting idea about the "super-capacitor".... I used to work field service for a company (MEPPI) who made the electric propulsion and dynamic braking systems for the Bombardier M7 rail car. They all came with the regen option, but the LIRR and MNRR in NY City didn't use that feature. Instead, there were huge capacitors (system ran at approx 650 vdc) and several resistor banks to bleed off what was created while using dynamic braking to slow the train, since it's wasn't being fed back into the third rail. I'll have to do a little experimenting on this subject.... Thanks for the idea! Could help to protect the batteries in some configurations.

Todd

 

[ZOMGVTEK]

To me, the more or less 'solid state' braking is worth it alone. The power that gets back into the pack isn't really noticeable.

However, being able to just squeeze a brake lever partially and have very smooth braking is nice. I don't even have pads on the rear, I just use a switched brake lever to kick in regen. The front brake isn't switched due to being hydraulic, and I almost never use it. The regen at 40+ MPH works very well, and its nice when you're coming into a hill at 40 and don't want to ride the brakes the whole way down. The majority of my braking is regen only, which more or less fades out around 10 MPH, down to nothing at 6 MPH. I just put my feet down or tap the front brake to come to a stop.

How effective is it? At 40+ my regen only stop is essentially the same braking force as my friends cheap Schwinn with V-Pulls, and my brothers WalMart front disc bike. No noise, never fades, never wears. If I jam on the front brake and rear regen brake lever, the rear tire locks up and skids as I feather the front 6 piston gatorbrake to not flip over.

 

[sn0wchyld]

so if controllers like lyens 12 fet are limited to 10ams max regen breaking, what happens if you exceed this? or will it simply provide < 750W of breaking power (for a 75v pack)?

I'm guessing that you need a battery pack to bleed off the extra energy generated by breaking, otherwise that energy has nowhere to go but heat? (or noise, but 750w of noise is going to be LOUD). Ie there is no practical way to have your hub motor provide breaking without having some kind of 'sink' for the generated energy?

 

[ZOMGVTEK]

I dont know about this 10A limit... Im fairly sure my Lyen 12 FET will do over 1KW back into the pack down a hill. But this is at 95-100V. You can never 'exceed' the 10A regen controller limit, the controller limits it.

The regen power will always go back into the pack. There is nowhere for it to magically dissipate 1KW of heat. Previously people mentioned that if the voltage exceeds the controllers regen limit, it will cut off regen. There is no mystery power going anywhere, or regen braking, it just will not regen until voltage drops to what the controller considers a 'safe' limit. My controller has been modded to allow for 100V regen. A fresh pack hot off the charger will still regen, even though its about 100V. Also, if your battery has a BMS that limits charging current, or charging voltage, and you have a fresh pack, its probably going to not allow the power to flow into the pack, killing the regen. The controller will still try, but the power wont go anywhere, and that will simply result in no regen braking and regenerated power.

1KW is quite a bit of power, even momentarily. Its like running 10 100W incandescent bulbs... Not something the controller can just dump on its casing or something like that. If you are slowing down, the power is going somewhere. Hopefully it isnt heat, since that would be quite a bit of heat.