It really helps when you know the thing you are talking about completely and are enthusiastic to share that knowledge with others. Easiest public speaking for me was always about things like that and when actually scored i always got best grades from them too, so it wasn't just my opinion.
It’s nice to hear someone actually talking about the importance of regen when it comes to actual breaking. It’s not all about putting energy back in the battery well that’s absolutely a positive it’s more about the breaking feature, a lot of people overlook this.
byet ~5-15% efficiency gain is still nice "free" additional range or weight/cost saving depending how you look at it, and also more consistent range when you have more elevation (hilly route) or more payload (cargo or passenger)
@@GrinTechnologies I want a middrive that allows regen breaking. Ideally it should be as powerful as a typical hub motor so it doesn't require gears either and can use a thicker chain that doesn't wear out so fast.
San Diego is a regen city, Las Vegas is not. I live in a hillside city with about 1000 feet of vertical to work with and would like an ebike that had regen to help with the long descents. But i do not want to sacrifice climbing ability or range.
PS - its good to finally find a YT channel that is calm and knowledgable, writting some code and doing intelligent interesting things, glad i came across Grin Tech, it's one of my favourite cicites Vancouver, i used to work for a tech startup in the print game 20 years ago in Burnaby. Happy days.
I wish this talk was two hours long like your "Understanding Everything about Hub Motor Drive Systems - Spezi 2020 Live Presentation" and "Deep Dive into Ebike Motor and System Efficiency" lectures. Even the "guide to multi-motor (2WD and 3WD) ebike drive systems" talk was over 90 minutes long. There's so much to learn.
Hey Sam, don't despair, this was just a practice run. Our plan is to do a 2 or maybe even 3 part live webinar on Regen just like we did on those other topics, and that will dive much deeper. I could probably go on for 6 or 7 hours straight on this particular topic :-)
This is exciting to hear! I hope to learn about how one might go about dealing with "battery combiners" and regen as well as multi-motor regen systems. Further research on cargo bikes with electronics bay built into the front load carrying platform is the goal. My aim is to develop a frame into a good modular setup with which to test different viable ways this could be electrically engineered. In the best case scenario you'd be able to attach a sizeable supercapacitor pack in parallel to the battery system. If it were possible to have an independent bank carried in a rear single wheeled trailer similar to the Coho or ideally Mark Havran's technological monstrosity which would serve as a "solar charging loop" that would also somehow collect the Regen? with which it would then keep a constant feed going to both the supercapacitor bank as well as a Cycle-Satiator like system that is set up with a means by which the internal computer can read the voltages of each pack, track cycles and most of all; while riding it will switch which pack is being charged via regen/solar vs which is being used by the drive-train. It'd be a real funky spaghetti of wiring to do it... but I'm looking at a good concept to truly haul passengers or cargo over longer distances at a decent cruising speed, my V2 of the frame is intended to be custom fabricated from a fat tire frame. Thanks for reading, Portland Oregon here.
@@GrinTechnologies good because as a Grin DIY customer, I would welcome some more Webinar type videos to help us select and install the correct motors and accessories.
@@GrinTechnologies Missing out on few hours of Regen. I was trying to convince my old company to do it on their ebikes. Really enjoyed the systems thinking it is a shame the ebike industry is so backward technology wise. Trying to get someone to do a 240W USB C battery charger - its only electricky after all is just like pulling teeth!
THANK YOU, thank you thank you! From my experience, you've got everything figured out. It's a shame you're the only one among vendors. The main point is about brake wear. People don't realize how much of a chore, changing your break pads, is, once you don't need to do it anymore. Regen is really liberating on this point. One thing about throttle setup, is to use 2 throttles, side by side, one in reverse. In that configuration, you can simulate a 2 side handle. I tried it 10 years ago. Thumbs goes down, I accelerate, thumb goes up (pushing on the other throttle), you regen. It's pretty natural. I didn't implement it on my phaserunner, as I'm really used to a push button + accel throttle. I've been slowly developing a Bluetooth module for the cycle analyst, and I also thought about regulating power/regen with a tilt sensor, this seems so natural, it's silly nobody thought about it before. As you said, pretty sure it's the future evolution in e-bikes. And after that, you'll have power regulation based on GPS, and set destination, anticipating the profile of the road, to have an optimal power consumption. As I've been bike commuting for 20 years, e-bike for 15, my dream bike is pretty much the same as yours. I'm a commuter, I don't want maintenance on my bike!
Awesome idea there with side to side thumb throttles to mimic a single bidirectional device! That's really quite clever. We did our first tests with an accelerometer control in 2009 and it worked great then but sat on the back-burner for way too long. I suspect that in the year or two we'll start seeing this everywhere since it's just so obvious and inertial sensors are so ubiquitous and cheap. Glad to hear you are in agreement with most of the points in this video, it's reassuring when others with similarly rich firsthand experience have come to the same conclusions.
@@GrinTechnologies Hey thanks for the response. I'm in agreement for not most of your points, I'm in agreement for ALL your points. I also have a Yuba Mundo, and my dream bike is the same as yours. This is becoming very strange. Anyway, keep up the good work guys! Your products are awesome.
That's a pain them not putting analogue hall sensors in the brake levers... BUT can't you just glue a magnet to the brake lever and stick an analogue sensor on the brake lever body? That way you can get variable regen braking and mechanical brakes on one lever that is compatible with any bike..... A lot of kits have a similar thing going on... I could even d.i.y it by soldering some leads onto an analogue sensor then plastidipping it to make it waterproof You provide the support in the controller for variable regen braking that I can calibrate similar to calibrating the throttle end points on an RC car esc, and the wires to solder an analogue hall sensor on and ill do it myself.
Great presentation, Justin! I used to replace brake pads 2-3 times per year before regen. After regen, I’ve gone 36,000 km in 3.5 years with one replacement (and that was mostly because my max regen phase current was set too low for the first couple years). I used to get almost 20% regen riding in the Oakland hills. Averaged 4.4% on my 8700 km ride from Vancouver to San Diego.
This is great news to me. I was excited by the concept of regenerative braking when I first heard about it. And I then heard the negative hype about how it wasn't really worth the trouble.. That was a big disappointment for me but I kind of just accepted it. Now you have restored my enthusiasm a little. Thanks :)
The constant negative talk about regen is just so bizarre and out of touch with reality. It's was largely fueld by mid-drive enthusiasts trying to seek some justification for a major shortcoming of that topology by pretending it's no shortcoming at all. Hopefully we can start seeing a shift in perspective with more and more people singing the praises of regen.
I have variable regen on my bike with a throttle to modulate it... It stops on a dime without throwing me head over bars or locking a wheel up. In fact, I hardly need to use the normal brakes except to finalize a stop or to add extra braking force.
My bike tractor is offline at the moment, but I regularly get 15-23% regen on cargo trips to and from my hilly San Francisco neighborhood. Even on trips to flatter parts of the city and nearby cities, because of the initial descent, I'll still get 6-12%. Definitely worth it.
wow that's great. I was debating making my own custom e-bike form scratch with tools form my maker space and I was debating if regent breaking is worth it to solve the free wheeling problem I was thinking of using some sort of clutch mechanism. the motor can peacefully absorb the spin while the pedal disconnects completely from the system,
i can't thank you enough for this vid. as i am about to invest in a full Grin E_upgrade for my recumbent trike, after which i plan some serious mountain riding over the Alps, thus, reGen will be essential.
Thanks Justin, excellent presentation! Another advantage of regen is that on downhill you can still (gently) keep pedalling so you can increase easily the % of energy recuperation.
Thank you for further explaining the benefits of hub drive over mid drive. It's insane how many bike shops hate on rear drives because of made up reasons.
There are positives and negatives to both systems. In my opinion on most e bikes hub motors are perfect but the unsprung weight of a hub motor on a bike with suspension such as mountain bike is a massive issue
I don’t like to go fast down hills. I don’t feel safe (note I have well over 50,000 bike commute miles over more than 40 years of commuting, with less than 10 falls total). I love my regen, it’s a game changer for a nice relaxing rejuvenating ride. Maybe 40 years ago at 26 years old I would feel differently 😅
Yeah it's awesome!. After experimented it, it spared me a lot of brake maiontenance, disc alignement, pads replacement, caliper cleaning, bleeding etc... Both of my bikes are heavy, 165lbs-180lbs, + me 230lbs so it generates a bit but defifinitly about the brakes with all wheel drives bikes like me it's awesome and worth it.
Thanks Justin ,and Inspiring and informative presentation, much appreciated! After a year of using my own design Regenerative module I might add: 1, I found that having a multi-level regenerative settings, one for low speed pedaling that ranged from 40w to 80W from about 5 to 10mph was useful when used in conjunction with pedaling on level or slightly down hill rather than for breaking, and two others for long moderate down hill at about 200w to 300w,, and a third very aggressive mode for stopping that ranges from about 1200w above 20mph to about 500W at around 8mph. Around hilly Portland OR., I'm getting about 12 to 20% re-gen. 2,, is high battery charge current. My DIY system is capable of up to 1500w of above 15mph and I often run 500 to 700W for several blocks on moderate hills which is between 15 to 30amps of re-gen, which is well over the charge rate of most Lithium batteries, 3 constant to 10 to 15amp charging bursts might be reasonable for most eBike packs, so 300 to 500w might be a safe Regen level for such packs but for typical riding I found only delivers about 3 to 8 percent overall range improvement. To Improve range and re-gen amperage I've added a parallel 600 watt-hour LFE (Lithium Iron Phosphate pack) in parallel that adds about 20+ amps of regen charging ability. over the 6 amps (300w) I feel comfortable hitting the Li-on Pack with. As a torture test over about 800miles and several months, I hit a 48V, 500 watt-hour lithium-Ion pack that was rated at about 2amp max charging with typically 10 to 28amp of re-gen, right up to full charge voltage (NOT RECOMMENDED OR LIKELY SAFE) And the high quality industrial (Server-rack) Li-On Japanese cells I'm using appear to have suffered no measurable increase in internal resistance or decrease in capacity, but I wonder if the high level of charging could cause dendrite formation and increase the chance of shorting? A third benefit of re-gen I found is safety and stability. I use re-gen as much as possible and it helps me maintain a constant speed without lock-up as regenerative breaking automatically decreases with speed, and increases with speed matching pretty well when needed sort of like a passive anti-lock system, so I don't have to get aggressive on the brakes, As well, if the tire under re-gen breaking hits a low traction surface, by the nature of the system I've never encountered a skid under even heavy re-gen as might with mechanical breaking. Thanks again for the great presentation Justin, Regen is something I appreciate as I do your passion for it.
Hi Jack and thanks for sharing in detail your firsthand experiences too! I still plan to do a much deeper technical talk on many finer points of regen, this presentation was just a primer :-)
@GrinTechnologies Hi Justin, Do you know if there is a Chinese version of the Pinion gear for retrofitting normal bikes? I have been thinking about this for a while.
I just rode a 117 mile ride with 10,600' climbing, and got over 50% regeneration capture. The hillier the terrain, the more regeneration helps. The best part was descending a 3,000' climb and never touching the brakes, instead putting all that energy back into the battery.
That's the kind of use case I imagine and Justin's dream bike would be pretty much perfect. It's what I'd imagined would be the simplest setup but good to hear it from an actual expert...
The "Rion Curve" and the throttle from the Boosted Rev are worth looking into. They are thumb wheel inputs that physically spring to a mid-point hall sensor (so their default state is about 2.5v). Then all you do is push forward or pull backwards and you have the controller interpret that. They are AMAZING when used for throttle+regen and the ergonomics are extremely hard to beat. Sadly, as Justin has lamented, finding thing like this in the market is extremely difficult right now.
Awesome thanks for the tip on the Rion throttle, we weren't aware of that one being out there and it looks like it would be compatible with bike handlebars too.
This is cool concept. I know electric wheelchairs in Walmart have forward backward control with one switch so it's existing techn and should be easy for a factory to make them with the right molds
I learned more from this guy than ANYONE in TH-cam when it comes to e-Bikes, thank God he chose e-Bikes as his area to work in, because with his genius, he could DO ANYTHING he put his mind to!!! 💯👍 Edit: I'd gain 20%-25% or more because I carry heavy loads, ride in stop & go traffic with hills everywhere, so what's the cheapest way to put this on a rear motor e-Bike??? 🤔
Thanks for the flattering comment about Justin! For your question, Step 1 would be get a Direct Drive motor or a Gmac... step 2 would be upgrade your controller, Baserunner or Frankenrunner so you can start doing regen. Definitely contact sales and support to get into specifics.
Thank You! I actually had no idea how regen braking really worked until I watched this video(I'm fairly new to the ebike game) I've also seen a huck cycle review video with shreddie mcskate where he locks the bike and tries to push the bike forward as if running away with it and the the Regen braking activates a drag to the bike slowing it down dramatically to wading through quicksand speeds it seemed really useful... Damm I said a lot my bad lol
The ride share scooters have this feature to help with theft although it’s only 1 wheel that locks up and they’re still light enough to pick up. It would be nice to have on my heavier dual motor scooter though
I like to salute you for this Rich information where you looked to matter from all angles giving all choses so everyone chose what's suitable for him. Thank you ones again and good luck to you.
great presentation. it's clear to me that variable regen is the quintessential application for an analog Hall sensor in the brake lever. my interest in regen goes to the design of a e-trailer with regen capability. having done a long tour with a single wheel cargo trailer, the first thing i noticed was desperate need for braking on the trailer. all that extra mass in the back puts far more stress on the brakes and also become unstable. having the trailer do the braking while regenerating is the perfect solution. your motor just went to the top of the list. i'm also reminded how similar 'marketing' is to 'propaganda.' thanks. keep smiling.
Hey stephen, yeah it doesn't take much to tie an analog hall sensor and magnet to a brake lever and get modulation that way. With a trailer you also have the option to have a force sensor on the hitch and activate regen on the trailer wheel whenever the trailer starts pushing the bike (and use reciprocal approach for power too)
@@GrinTechnologies i really like the accelerometer idea for a trailer. the problem would be averaging due to bumps in the road. i'm thinking of an e-trailer that would provide just enough torque to push the trailer's load and provide regen braking. a great design goal!
thanks for a clear and deep explanation backed with numbers of something I've always had a gut feeling for. it's exciting that ebikes are still in their infancy with so many improvements to come. blended Regen braking on the brake lever like on my ev (fiat 500e) sounds awesome. hope you'll be heard by the rest of the industry
In the meantime a good easy stop gap solution is to use a magnet glued to the brake lever, and an analogue hall sensor stuck to the brake body. Loads of ebike kits use a similar thing. Calibrate the brake end points the same way you calibrate the throttle range on an RC car ESC Some kind of input like full throttle, or button to press upon controller power connection on and boot up to enter programming mode. Where it will then periodically cycle through available settings, beeping a certain number of times (simply by using the motor as a chime, just like in RC cars) each time it switches to indicate what setting will be adjusted a step when the button is pressed or blipped with full throttle then beeping again a number of times to indicate what level the current setting has been set to. Easily navigatable with a piece of paper with numbers that correspond to individial setting, the each setting having its own numbers to indicate what steps are available for each setting) Though, when you press, or blip the throttle on the regen brake setting, it will be a little different. It will enter regen brake end point calibration mode. It will take the current signal from the brake hall sensor when you enter regen brake calibration mode as 0% brake. Then set your wheel rolling. Now start squeezing the brake lever. As soon as the wheel visibly begins slowing down. Stop squeezing. Then press the button, or blip the throttle and it will set that point as 100% regen brake. At that point you have perfectly calibrated Regen and mechanical hybrid brakes
@@GorillaZillas I'd love to test ride an ebike with strong Regen properly calibrated this way on the rear brake lever...I love my fiat 500e Regen braking on the brake pedal. I can use 100% Regen braking on my daily driving even if I brake hard and whatever Tesla one pedal driving fans can say, my brake pedal feels better than on my gas cars. the cherry on top would be to have software abs on the rear brake. front tire is way less likely to lock up when braking hard and/or downhill (although with higher consequences), it doesn't sound that hard to implement abs if you have hall effect sensors
@@geemy9675 As far as I know, Regen braking is inherently anti lock, as resistance is caused by current flow, and current flow is caused by RPM. If the RPM drops anywhere near zero, so will the resistance. You can see this effect when you drop a magnet through a copper tube. Or if you short out a motor, then turn it to feel the resistance. Even a completely shorted motor does not lock up. It might skid with a lower RPM than the speed of travel though, so there's still room for improvement.
@@GorillaZillas I don't know specifically about Regen being inherently anti lock. I don't see a reason why the wheels would stick at 0rpm but it sure seem like they could skid turn slower or maybe even backwards depending how the Regen is programmed but anyway it shouldn't be too hard to implement a feedback loop the same way electric motors can have finer traction control than ICE. I know for a fact that my 500e completely shuts down Regen below freezing temps though
@@geemy9675 Yes, actually you're right. They probably would skid while turning slower than the rate of travel, could be solved if the controller doing the regen could read RPM sensors on the wheels and reduce braking power when the regen braked wheel rpm dropped lower than the non regen braked wheel.
Justin is right. Once I got an ev car and got used to 1 pedal driving, I immediately wanted regen on my bike. I absolutely hate using brakes on any kind of vehicle now. Why not just have a set regen like ev cv cars do? How would you know if the motor is overheating? We have the same dream bike. I would love a belt drive ebike.
The engine braking feature can be implemented fairly easily using the Digi Aux regen control that's in the CA3.2 firmware. You can adjust the amount of background regen with the digi aux input, but whenever you use the throttle that overrides it.
once I got an ev car and got used to blended Regen, I immediately wanted regen on my ebike. I absolutely hate using mechanical brakes now, when I can have very strong regen and precisely controlled regen with the brake pedal, that will put back as much energy as possible back into the battery, save brake pads wear and prevent brake from overheating when going downhill. Just my personal opinion about Regen, as I think one pedal driving is far overstated and blended Regen braking far understand because of Elon Musk arbitrary preference and ignorance about blended Regen braking (which doesn't have a "weird brake feel" when implemented properly). one pedal driving and lack of blended regen limits max Regen power (a 1000hp plaid only has 75kW of Regen when a taycan has 350kW or my 111hp 3000lbs 500e has 80kW) If I was given choice I'd rather have as strong as possible controllable Regen (for instance using a brake lever position sensor) like on my car than Tesla-like one pedal Regen especially since on an ebike with pedal assist you only use the throttle when you're not pedalling
for me the dream ebike power train especially for a full suspension bike is revonte one mid drive that provides both pedal assist and infinitely variable transmission in the same package with less moving parts that a pinion or rohloff gearbox and without losses or complexity of a nuvinci cvt. basically the eCVT of Toyota/ford/Honda hybrids adapted to ebikes. They say Regen is possible in the future if the rear hub has no freewheel and the chain belt keeps turning with the rear wheel. you get the advantages of a mid drive, with a super low unsprung weight, possibility to run a belt or chain without tensioner, perfect chain line, and all the complexity and weight is located around the bottom bracket, low and close to the battery.
I added a simple horn and light switch to the regen input of my Cycle Analyst. The horn button (momentary) triggers temporary regen. The light button (on/off) adds regen for longer downhill mountain runs. Both settings allow for variable regen by twisting the throttle. Cheap and easy.
At very low rpms yes, at higher rpms its not even close. The crossover point depends on both the intensity the regen torque and the resistance of the motor and coil windings.
ok. so non permanent magnet motors can do just fine? btw the coreless motors, are they viable or will they become too big? can they be small if they go high rpm?@@GrinTechnologies
I hope that backpedal regen via crank position/rotation survives the beta and becomes a factory option. It would be my prefered option, as it seems to be the most intuitive if you can set it up right. Altough, the two throttle option dosent seem too bad, but I would have it set up differently - A twist accelerator and a thumb regen on the same handlebar sitting next to one another, far better ergonomics in my mind. I would even have the regen cutoff the accelerator for saftey. Its a bit of a hybrid with the bidirectional throttle, as you are pulling with your wrist and then pushing with your thumb.
I bought my direct drive RadCity Step-thru 3 (Canada) (low step Radcity 4) this November on the clearout sale, which has regen braking. The direct drive hub motor has "Shengyi" and "HYBRID" all over it and it is indeed a true plug-in hybrid electric vehicle. Regen braking is disabled when accelerating from a stop with human power only - PAS 0 and not using the throttle. In Hybrid and EV modes (PAS 0) when the throttle is released, low regen applies. Slightly tapping the brake levers will engage full regen braking - more than the amount by just throttle off. Too bad my direct drive Radcity is now discontinued and is essentially irreplaceable now...
Yeah it was unfortunate that the rad DD bikes were all discontinued. Glad to hear you got one at a deal. There aren't many factory bikes with regen at this point in time but at least there are many great DIY options. And in due time they will make their way way to factory ebikes in a big way.
I have variable regen on a rear DD hubmotor commuter that I built years ago. My commute included some steep downhills that would heat up the brakes to the point they were starting to fade and there was no emergency capacity left. With variable regen the speed was controlled while the brakes were cool and ready for any emergency. It was like having an adjustable parachute going down the big hill. I didn't care about the recharging of the battery, but the saving of the brakes was worth a lot. Variable regen is really nice and definitely worthwhile.
I love Regen, and I also love your passion. I look forward to buying Grin components when I can afford them. Another great 2 things about Regen, at least my theories, is 1, it's a natural Antilock Braking System, because if the wheel locks up, like on ice, then its not spinning, and the generator effect goes away. So the theory I have yet to test is, even with a torquey DD hub motor, and a high current controller with Regen. The BrakeRegen will naturally always be at the threshold of grip. But this also makes me think about how long for the induction in the coils to fade away, and if stator coils can be designed to quickly lose magnitism when magnets stop. My second Regen theory is. Wiring battery packs in parallel with capacitor packs, will greatly improve the efficiency of regen, because the charge limit of the batteries themselves is a bottleneck, but capacitors can charge & discharge quickly. The more capacitance, the better. Wire enough ultra-capacitors in parallel groups for capacity, and then wire those groups in series until their combined maximum voltage ratings cover the battery charger voltage. Or use high voltage capacitors that cover the charging voltage and just parallel those together for many Farads of capacity. My experiment & problem with my 2nd theory is. The space within & on a bicycle or motorcycle frame, to be dedicated to energy storage. What is the best ratio between batteries and capacitors? Be it ultra-caps or normal capacitors. Should the capacitor pack be as big as the battery pack? Two times as big to get the most out of Regen? It's like, be battery biased if you are traveling long distances with little to no braking. Or, fill your frame up with mostly capacitors, if you expect a lot of stop & go city use, or tight, technical routes. I don't know the exact ratio for expected use cases. I do know Capacitors have better C ratings then batteries, and can capture the most energy from regen, to dump into launching from a stop or out of a corner, or charging the battery at its own pace, if you RegenBrake to parking. I just don't have the money or time to do these experiments. And like the presenter, I hate bike maintenance. I just want to do my delivery job, with a reliable, stable, but powerful bike, then relax. I'm hoping using 2 torque biased motors, means I can phase out brake pads entirely. Using some metal shoe on the tires to act as a parking brake. Either way, thank you all at Grin, for your quality products, passion to improve EVs, and thank you for this presentation, I wish you had more time.
2 ปีที่แล้ว
As usual "classy" Justin! Very informative! Thank you Justin.
Completely agree. I ride a handcycle with e-assist and my brakes are my backup safety system. 95% of braking is done with regen. I can set a safe downhill speed on my cycle analyst and not worry about using my brakes and just enjoy the decent..
Just add a 3rd brake lever that works more like a throttle on a variable resistor. 2x levers could be on the same hinge pin so you could either pull them both together or just feather the upper one for re-gen regulation. Sometimes you will need both together anyway.
I always enjoy listening to your lectures, Professor Grin! I don't know if you can answer, but I was wondering about regen's effect on battery capacity degradation. I've looked at some research papers, but some of the more applicable looking ones are behind the paywall sadly. If this is not too specific, I have some high-power 21700 cells (P42A) I always keep between 4.1-3.2V, but I've been wondering what is my regen 'C' limit for limiting long-term battery damage. I'm not limited by BMS (I use RC chargers for balancing), and the battery spec shows a chart charging at 2C (8.4A), but I know in general it's better to charge at 0.5C or less for longevity so that's what I've always done. Regen doesn't induce the same constant current for the duration of the charge cycle, so what I'm really wondering is if short bursts of 2C or higher regen rates would end up accelerating the capacity decay over time. Any insight on this?
I wanted to go into address this topic but left it out of the presentation. The quick answer is that all evidence suggests that short duration regen currents even at high rates has almost no meaningful effect on cycle life. If you do a full charge from 0 to empty at say 2C that definitely reduces cell life cycles compared to say charging at 0.5C, but just occasional bursts of 2C for a few seconds or 10's of seconds hasn't had any effect that we've been able to observe.
@@GrinTechnologies I had this question too after watching. I can regularly pull 1kW (about 1C) for 10-15s going down the hill from my house to the main road I use to go anywhere and was wondering what impact it might have on my new 52V/20Ah battery. Thx.
My Grin rear hub motor (26" wheel) regen torque seems limited by a 500 watt power ceiling it seems. How can I increase the regen braking force? My regen phase amps and battery amps are already set as high as they can go in Phaserunner suite. Is it a Cycle Analyst setting? By comparison, my other Grin front hub motor (20" wheel) is putting 1.1 Kw back into the motor and braking stronger. Thanks!
It's generally a function of how fast you are going, so if you want more regen watts you have to travel faster. To get the same max regen energy into the battery from a 26" wheel requires 30% more speed than you'd need in a 20" wheel all else being the same. That said there are many settings in place that can result in other limits kicking and this is the territory of email tech support more than YT video comments.
Although they last longer than brake pads, rim surfaces / disc rotors won't wear out as quickly either. This may be more significant for commuting due to the gritty stuff kicked up in the rain. I also wonder if mech brake cables could last another season given the reduced usage frequency and tension. Aside from cost savings, I also like that regen can reduce braking noise (particularly disc squealing).
Ah yes, good point about the brake noise too! It doesn't cost $$ directly but I've seen countless time wasted by people tweaking their pads and calipers a bit here and there to eliminate a stubborn squeal.
Great video. I would really w like regenerative motor. You gave some great information. The back pedal idea for some of who are older wouldn't be a bad idea. We all had bikes that the brakes are the pedals. We wore tires out a lot!
This is awesome.ow, I never knew about the benefit of the braking and not wearing brake pads and in wet weather, mind blown. Looking to order new wheel today I pull a trailer too.
Really great talk. I've also had the dream of an ultra low maintenance bike. But am i correct that the GMAC motor is not currently designed to attach to the front hub? Are there plans to make that possible?
All the controllers that grin currently sells (Phaserunner, Baserunner) and has sold in the past 6-7 years (Frankenrunner, Grinfineon etc). Other controllers that do good variable regen would be the Kelly controllers, the Vedder VESC, and or any industrial BLDC servo motor controller. Most inexpensive generic chinese ebike controllers do pretty crummy regen, if at all.
Hi Grin Tech, at 3:50 you explain that this pseudo regen relies on the motor spinning very fast and has a higher voltage than the battery pack and occurs at a critical speed. Why does this happen? What causes the motor to not just continue to spin faster and faster? Why does it slow down as it is putting negative current back into the battery?
In my experience this pseudo regen is happening when you have the throttle opened up all the way (so the controller has hit its max rpm output to the motor) If at that point you are still gaining momentum from the hill, the controller will be holding you back the max rpm it is programmed for by design. If you let off the throttle completely while still on the hill the controller won’t be limiting the rpm and you could potentially continue coasting faster than with the throttle engaged.
Also, once the battery and various capacitors are charged I assume the free-wheeling / direct drive will have slightly less drag? I'm hoping to build around a 48v Maxwell DuraBlue Ultracapacitor (+ Lion battery in parallel). 6.0 mΩ of ESR, rated 100A continuous, 1600A short circuit pulse!, but only storing 3 Wh over 18 cells @nominal 2.7V!
This has been a great presentation. When I was growing up we always stopped our old single speed bikes by backpedaling to activate the brakes. We didn't have rim brakes yet and certainly no disc brakes. They also coasted pretty well too so they weren't "locked" pedals. I'm pretty sure regen could be implemented by using the pedal cadence sensor? Maybe a torque sensor instead of a cadence sensor? Move it slightly it regens enough to slow you down, move it back a bit more and it regens even more effectively even all the way to a stop? With our old bikes one needed to keep some back pressure on the pedals to ensure the brakes remained on but if one wanted to release the brakes then one simply let up the pressure or began pedalling again. Another issue I'm thinking about is the lack of torque a direct drive motor has verses a geared hub drive motor. I currently have a geared hub drive motor on my bicycle and it works fine although it can be too weak sometimes. I've heard from many people about direct drive hub drives and the overall consensus is they are underpowered. Not in horsepower of course but in torque. How does one compensate for the difference between the torque from a direct drive hub motor and a geared hub drive motor? Use two direct drive motors, one in the front and one in the back? Or use a GIANT direct drive motor in the back? Of course weight becomes an issue once one reaches a certain size. Interesting problems and it's going to be interesting how it will be resolved. Thank you everyone at Grin for continuing to research and experiment for ebikes.
th-cam.com/video/gYCj4asp9pE/w-d-xo.html Direct drive motors can delivery plenty of torque for almost all applications as long as they are sized appropriately and your motor controller can give sufficient phase current. Motors with the same winding KV will give the same torque output for a given phase current whether they are geared or direct drive. Compare say a BionX D direct drive motor system to almost any geared motor ebike and it will throw your "consensus" out the door!
@@GrinTechnologies Thank you for the quick reply. Rad Power Bikes used to make their Radwagon with a direct drive motor and 26" wheels. However they made the newer one (Radwagon 4) with a geared hub motor and it's got 22 inch wheels. Their literature says they went to the geared hub drive for the increased torque and went with the 22 inch wheels to lower the centre of gravity. What you're saying is they could have kept the direct drive motor and increased the phase current to make it more powerful?
@@davemeise2192 Yes exactly. We were a bit bummed to see Rad drop the DD hub motor option too. For the same torque capability a geared motors will typically be a few pounds lighter than the DD equivalent, no disputing that. But the really well engineered DD hubs (like the Grin All Axle or BionX D series) cut that gap pretty close.
They can make locking cluches both mechanical and electrical. My regen is a thumb throttle inside my twist throttle. Has my headlight button as well. Works fine even with gloves on. The hub motor wins even on the gental down slope because you can peddle against it creating more power to regen and gravity helps you alot!
I have a Bafang BBS02 build. I like the range I get from the mid drive, but if money was no object would it make sense to add a small hub motor to the front wheel, purely to use as a regenerative brake? Almost like a dynamo hub but with much more useful power generation than the ones that just charge up your lights.
It actually makes more sense to have a decently powerful hub motor that you use all the time for motoring and for regen, and then only use the BBS02 when you need an extra kick on hill climbs.
My dream bike setup would be 42:11 with a mid-drive motor (for the integrated torque sensing+cadence sensing, and what I heard better traction in the rear wheel), and an automatic shifting rear IGH. Though not sure how much more benefit would the mid drive offer, and whether there are any technical challenges with integrating the front and rear wheel drive power. (So the front hub is mostly there for regen XD) As for the braking controls, would it be difficult to make the brake tripwire offered in your shop use some sort of hall effect + magnet setup, or some kind of pulley + encoder setup, or any other kind of a linear position sensor?
My 48V DIY bike with direct drive hub motor paired with a regen capable controller only lets regenerative braking above 16mph speeds, thereafter I have to apply mechanical brakes to slow down further. I understand the motor voltage has to be higher than controller's for regen to work, but the speaker implyed the regen should work almost down to 0mph. Am I missing anything? Those who have bikes capable of regenerative braking at what speeds does your regenerative braking cut off?
Your controller is definitely not configured for proper regenerative breaking if it stops working below 16mph. All regen controllers we have encountered will by default do regen right to a stop just fine. In some cases there is a configurable minimum regen speed setting for those who want it out a bit earlier.
@@GrinTechnologies thank you very much for this response. I'll look into the controller settings once again. Possibly I have controller settings and motor characteristics mismatch.
With hydraulic disc brakes, it seems that adding a hydraulic pressure transducer would make a lot of sense. With a transducer, the electrical signal from the transducer would be proportional to the hydraulic pressure in the brake system. The regen controller could then use that signal to proportionally vary the regen. In other words, the harder the rider squeezes the brake lever, the more regen will be applied.
Yes, this suggestion has come up as well! The slight problem with it is that you need to fully engage the mechanical brakes before the hydraulic line pressure builds up, which means you are wearing out the pads too. Perhaps with a high sensitivity transducer it would be acceptable, but still ideal is not having the brake pads touch at all.
@@GrinTechnologies You can add flow controls to hydraulic circuits, but I don't recall all the options available as I haven't done engineering work in a long time. My gut tells me it's doable though. Perhaps a flow control needle valve, that one could dial up/down to regulate the flow to the hydraulic lines with the pressure transducer upstream of that. So, you could dial up/down the regen by controlling the flow rate to the disc calipers. In other words, with the flow control valve restricting flow to the brake calipers, the transducer would send full signal for max regen to the ebike controller while the brake calipers would be only slightly modulated. If more friction braking was desired, then one would back off the needle valve to allow more flow to the calipers
Love it! I didn't know about the new feature to back pedal to turn on regen. I've always just slightly squeezed the brake lever and used the throttle to increase regen. However, even that wears out disc brake pads sooner than necessary. I'm 100% with the beta tester who suggested back pedal just to turn on regen and still use the throttle to increase it.... Looking forward to setting this up. Thanks Justin!
so you program the controller so that as soon as brake lever cut off the power, throttle is used in reverse to modulate Regen? sounds smart although it means if you are using full Regen and slowly release the brakes you are jumping from full Regen to full throttle ? why do you have break wear, isn't it possible to avoid engaging the brakes? I think the definite answer to Regen is really a brake lever with first half of the lever travel spring loaded and triggering a linear Regen, then once max Regen is reached, the second half controls mechanical/hydraulic brakes. this way you make sure that any hard braking uses 100% regen, you can still brake the same way with electronic off or Regen is disabled (with just less initial bite and more travel), you can precisely control Regen and you don't have to learn a different way to brake, just to recalibrate to the blended braking.
That larger regen in hilly areas seems odd to me since you use way more energy climbing hills in the first place and get nowhere near that energy back descending.
You get more regen in hilly areas for sure since you do more regenerative braking as you would on the flat. Your total Wh/km is still higher in a hilly area than on the flat ground, but it's much closer to flat ground wh/km figures than it is to hilly riding wh/km figures without regen.
Does a regen brake system can be installed on a bicycle with a CYC X1 Pro Gen2 mid drive motor system installed with BAC2000 controller and powered by a 72V battery?? Thank you for a detailed answer.
No it was literally mentioned at 4:40, mid drives cannot use regen. I myself am getting an X1 PRO GEN 4, there's a freewheel attached to the motor such that it'd be impossible to reversely spin the motor to generate energy even if the rear axle was fixed
I wish I could like this twice. I love every single video on your channel Justin. I'm trying to be cheap on my 3rd ebike build, but I think buying cheap components is more expensive in the long run. First 2 we're mid drive, but I think I'm sold on hub for less issues etc. Thinking about buying Grin all-axle motor and putting it on a Surly frame for the ultimate reliable ebike. And building my own 2.5-3.5+ kwh pack in the triangle. What's the best cell for longevity, even if it's at the expense of energy density. Open to other chemistries, but trying to keep battery pack under $1500 if possible (not including tools and my labor/time). I would like to have at least 50% capacity after 5,000 cycles. Considering lifepo4 even if it means less capacity than what I want
Hey Geoff, it's often 3rd times the charm on custom ebike builds! That said, a cheap direct drive motor will be plenty reliable, just heaver than the All Axle. The area not to skimp out if regen is important is on the motor controller so if there is one place to splurge a bit it's there. For your battery, the best longevity comes from using name brand cells like Panasonic more than any specific chemistry. We're seeing tons of larger capacity batteries with the Panasonic GA cells that are now reaching 7 years old under regular use and have only lost ~10-15% in capacity. There is really no place for LiFePO4 anymore when cylindrical lithium cells with almost 3x the energy density are performing this well.
@@GrinTechnologies thanks for your reply and for giving me the reassurance of using traditional lithium ion cells. Looking at Samsung 50E 21700s but still haven't decided exactly what I want. A lot of folks tell me to get high discharge rating, but in a 14s14p pack, I need very little amps from each cell. Hopefully that combined massively oversized pack and charging to 90%, discharge to 10%, I will have a pack that could last me a decade. Long lasting pack and durable bike means 1000s of miles riding without much worries. Thanks once again!
@@Geoff9001 The Samsung 50E2 cells would be fantastic if you are under 2C per cell discharge which would be 140 Battery Amps in your 14P application. For even better longevity, set your battery cutoff to 20% or higher as research has shown discharging lower to lower levels increases capacity degradation faster than charging to the upper limit. So something like 3.2V or even 3.3V/cell+. You'll have a ridiculous amount of range anyway with 14P high capacity cells.
@@Geoff9001 I have a 14s7p 52v rectangle Panasonic GA pack. It's pretty big ! Are you planning on 100v system or extreme range? Some Enduro frame batteries are 8kw 72v 40amp , 240 cells ... 15 kg!
Very interesting, thank you! LOVE the silent smoothness and consistency of regen! Got 10-15% when I was using RC lipos. Recently I switched to a Aliexpress ebike Battery (52V 19,2Ah 5A max charge, 21700 cells) and am scared to use regen now. Can the battery take the regen from a MagicPie5? Is it safe to use regen with a chinese battery? What would make it safe? Could the energy also be dumped as heat before the battery? If, someone knows, thank you!
Even with generic quality no-name cells you should have no problem doing regen at C/2 or even 1C. The max charge rating is usually related to the charging port mosfets of the BMS circuit (and associated connectors and wiring), and doesn't usually relate to the discharge port currents. I wouldn't worry about regen putting you at any greater risk than you are already taking with the aliexpress battery.
This also gets into the realm of antilock braking as its good to know how much or how quickly the regen braking comes on and at what point to limit it, other than human braking input, and i guess you could use spokes or brake rotars as circular encoders for wheel lock etc.
We've seen zero evidence of it. Look at the 'C' rate on say a hybrid car which does like 4C-6C or higher during regen, while on ebikes you're usually in the 2C or less realm, and virtually all decent lithium cells can handle this for short time periods without any negative conseequences. The "maximum charge current" rating that you often see published on am ebike battery pack (like 5A) is a ridiculously conservative value based more on the charge port connector limitations, not on the cell's capability.
Ideally there would be brake blending, just like with a BEV. You squeeze the brakes to apply more regen, once that's maxed the mechanical brakes are gradually added on. On my Mach-E it tells me how much of the braking energy I reclaim every time, it's usually 99-100%. I was wondering, how feasible would a middrive with regen be? I have a middrive with a one-way clutch on the pedals, so it can be done, but would it put too much strain/stretch the chain? I would hate to get chain slip though! 😱☠️⚰️
A regular chain with a derailleur doesn't work for regen even if you lock all freewheels, since the top of the chain goes slack as the derailleur gets stretched out and then the chain just falls off. You'd have to do a real custom drivertrain with both a top and bottom tensioner and have the derailleur designed to cope with full chain tension. Otherwise it's only viable with a single speed chainline and an internal gear hub that doesn't freewheel, of which the only model we are aware of is the Sturmey Archer S3X
@@GrinTechnologies IMHO "chain slack" issues not that it uses a "freewheel" should be the reason given for not being able to do regen for a mid-drive... Since I had the same thought about locking out the freewheel since the mid drive crankarm can freewheel, and that freewheeling should not affect its ability to regen. But I wondered about the tension (ie slack) on the chain, (in the derailleur's cage), being an issue? Which is the main issue to be overcome... But overall its was a very informative and thought provocting lecture on Ebike regen, thx....
Good description. On another website, I discovered that the BEMF doesn't go through the 6 drive MOSFETS, rather it flows to the battery via the 6 reverse current (MOSFET protection) diodes. So no additional diodes needed! I liked the description of how the braking modulation works. The controller looks at the throttle voltage and rpm. When the throttle is retarded below normal power off, the controller at first applies an almost equal forward voltage to match BEMF being generated at that rpm. Thus freewheeling at normal throttle power off. As the throttle is further retarded, less and less forward voltage is applied, allowing much greater braking force from the increased BEMF being applied to the battery. However, the regeneration circuit could have been made more efficient if regeneration was controlled not by forward vs. BEMF voltages, but by electronically PWM modulating any BEMF voltage generated (no forward voltage applied in regeneration mode).
Yes if the drag only happens when you are above a certain critical speed when the back emf voltage exceeds the voltage of your flat battery. But no of the drag is present at all speeds
Excellent talk. Thank you for sharing this. I have a Yuba Mundo with a StokeMonkey...and I am thinking about either the GMAC on the back (and removing the StokeMonkey) or adding a Grin All Axle Direct Drive to the front. Thanks again for all you Grin folks do for the ebiking world...and the world!
If the stokemonkey is still ticking along and running fine we'd recommend to add the All Axle on the front. The combo of a DD hub plus a mid-drive is really potent for tackling just about anything with great efficiencies. If you upgrade the controller on the stokemonkey to use a newer baserunner controller like the All Axle, then running dual systems with a single throttle control is fairly easy too.
@@GrinTechnologies Thank you. I watched your great video on multiple motor setups a couple weeks ago and I recall your positive comments about a mid motor in combination with a front DD setup. And I am really enjoying your sailboat videos too!
To add to Grin comment, if you put a rear motor, you'll loose the special reinforced wheel at the back. You won't be able to carry a full load on it. As a longtail is usually heavy and needs a lot of power, your mid drive motor must chew your whole drivetrain pretty quickly. I've got a front hub motor on my yuba mundo, it works great. As it's an already heavy bike, you won't feel the all axle in the front.
It would make sense based on your statements that a mid Drive would not be recommended when attaining a ebike if your goal is to get the most energy out of your battery. Is this correct?
Not as a general statement. There are times when a mid-drive motor will do better for battery usage (eg a steep hill climb followed by a long gentle descent), and other times when a regen capable hub motor will do better (eg a long gentle hill climb followed by a steep descent)
Yes, I personally, always Downshift my truck to slow down, so as to save brake repairs, ALWAYS! So i do want regen on my ebike: Bafang mid drive, and my etrike: ELF hub motor, how do I get it? PLEASE!
Does the re-gen switching mechanism ‘need’ to be within the brake lever mechanism? Why can’t it be in-ine, interrupting the cable at a convenient place?
Oh it can and often is. The tripwire is often installed at the caliper rather than the lever, there are sensors that sense the moving brake cable inside a sleeve, and there are hydraulic sensors that tap into and sense the fluid pressure in the brake line.
I think a great setup for electric bikes is to have the front wheel be a disk brake and have the rear brake just be a sensor which controls regen (this assumes you have a hub motor on the rear wheel)
If you mean using a dynamo in the normal way to power a light, for example, then the answer is yes. EasyMotion has ebikes with front hub dynamo powering a light. Functionally, it's an alternative way of getting battery power to the light. 💡
Have an E-MTB with 2 wheel drive and 2 batteries (17 Ah and 20 Ah) and dual suspension that weighs 90 pounds. Yeah, a cargo bike with a 400 pound load capacity, including rider weight. I got it for hunting to carry boned-out big game meat and for bike camping with my UL backpacking gear and a 14 pound solar blanket and controller. -> COULD I get regen. braking in BOTH motors with a lever braking sensor?
Yes as long as you have hub motors without a freewheel. Sadly most factory ebikes with hub drives use geared motors with a freewheeling clutch which makes them unable to do regen. If you have a direct drive motor it can do regen for sure
@@GrinTechnologies I want to REPLACE BOTH MOTORS with FREEGEN motors! Is this possible with your controller and screen? My E-Cells Super Monarch Crown has a 52 V Sine Wave 45 amp controller W/18 MOFSETS in a SEALED conductive gel and aluminum, finned housing. It's rated at IP 65 waterproof.This controller is EXTERNALLY mounted on the seat tube and easily replaced.
In practice this is just not an issue really. But if you live on the top of a very large mountain where the first thing you do after leaving is a lot of regen, then it is simple to just set your charger to only charge to 80 or 90%. It's better for the battery too
Yes but I charge to full because I do not know what way I will drive on the hill. so how can we get a resoistor to take the regen load when rolling down a hill with a full battery? I hear something about relays, but how do you activate said relay switch?@@GrinTechnologies
Don't quite follow the question, we've had kits ever since day 1 of operation nearly 20 years ago. The list of regen capable kits (inline with theme of this video) is here: ebikes.ca/shop/ready-to-roll-kits/regen-capable.html
First, what a dynamic speaker. wow!.
Thank you for covering "Regeneration" through Braking.
It really helps when you know the thing you are talking about completely and are enthusiastic to share that knowledge with others. Easiest public speaking for me was always about things like that and when actually scored i always got best grades from them too, so it wasn't just my opinion.
It’s nice to hear someone actually talking about the importance of regen when it comes to actual breaking. It’s not all about putting energy back in the battery well that’s absolutely a positive it’s more about the breaking feature, a lot of people overlook this.
Exactly. Hopefully over time we'll see more and more recognition of this fact and a general change in the regen conversation.
@@GrinTechnologies What do you recommend to use as a dump-load in cold climates so the regen will still work?
byet ~5-15% efficiency gain is still nice "free" additional range or weight/cost saving depending how you look at it, and also more consistent range when you have more elevation (hilly route) or more payload (cargo or passenger)
@@GrinTechnologies I want a middrive that allows regen breaking. Ideally it should be as powerful as a typical hub motor so it doesn't require gears either and can use a thicker chain that doesn't wear out so fast.
San Diego is a regen city, Las Vegas is not. I live in a hillside city with about 1000 feet of vertical to work with and would like an ebike that had regen to help with the long descents. But i do not want to sacrifice climbing ability or range.
PS - its good to finally find a YT channel that is calm and knowledgable, writting some code and doing intelligent interesting things, glad i came across Grin Tech, it's one of my favourite cicites Vancouver, i used to work for a tech startup in the print game 20 years ago in Burnaby. Happy days.
I wish this talk was two hours long like your "Understanding Everything about Hub Motor Drive Systems - Spezi 2020 Live Presentation" and "Deep Dive into Ebike Motor and System Efficiency" lectures. Even the "guide to multi-motor (2WD and 3WD) ebike drive systems" talk was over 90 minutes long. There's so much to learn.
Hey Sam, don't despair, this was just a practice run. Our plan is to do a 2 or maybe even 3 part live webinar on Regen just like we did on those other topics, and that will dive much deeper. I could probably go on for 6 or 7 hours straight on this particular topic :-)
This is exciting to hear! I hope to learn about how one might go about dealing with "battery combiners" and regen as well as multi-motor regen systems.
Further research on cargo bikes with electronics bay built into the front load carrying platform is the goal. My aim is to develop a frame into a good modular setup with which to test different viable ways this could be electrically engineered.
In the best case scenario you'd be able to attach a sizeable supercapacitor pack in parallel to the battery system.
If it were possible to have an independent bank carried in a rear single wheeled trailer similar to the Coho or ideally Mark Havran's technological monstrosity which would serve as a "solar charging loop" that would also somehow collect the Regen? with which it would then keep a constant feed going to both the supercapacitor bank as well as a Cycle-Satiator like system that is set up with a means by which the internal computer can read the voltages of each pack, track cycles and most of all; while riding it will switch which pack is being charged via regen/solar vs which is being used by the drive-train.
It'd be a real funky spaghetti of wiring to do it... but I'm looking at a good concept to truly haul passengers or cargo over longer distances at a decent cruising speed, my V2 of the frame is intended to be custom fabricated from a fat tire frame.
Thanks for reading, Portland Oregon here.
@@GrinTechnologies good because as a Grin DIY customer, I would welcome some more Webinar type videos to help us select and install the correct motors and accessories.
Please address the "battery can't handle it" bullet point you skipped here due to time constraint.
@@GrinTechnologies Missing out on few hours of Regen. I was trying to convince my old company to do it on their ebikes. Really enjoyed the systems thinking it is a shame the ebike industry is so backward technology wise. Trying to get someone to do a 240W USB C battery charger - its only electricky after all is just like pulling teeth!
As is typical, Justin is 10 years ahead 😍
37:31
THANK YOU, thank you thank you!
From my experience, you've got everything figured out. It's a shame you're the only one among vendors.
The main point is about brake wear. People don't realize how much of a chore, changing your break pads, is, once you don't need to do it anymore. Regen is really liberating on this point.
One thing about throttle setup, is to use 2 throttles, side by side, one in reverse. In that configuration, you can simulate a 2 side handle. I tried it 10 years ago. Thumbs goes down, I accelerate, thumb goes up (pushing on the other throttle), you regen. It's pretty natural. I didn't implement it on my phaserunner, as I'm really used to a push button + accel throttle.
I've been slowly developing a Bluetooth module for the cycle analyst, and I also thought about regulating power/regen with a tilt sensor, this seems so natural, it's silly nobody thought about it before. As you said, pretty sure it's the future evolution in e-bikes. And after that, you'll have power regulation based on GPS, and set destination, anticipating the profile of the road, to have an optimal power consumption.
As I've been bike commuting for 20 years, e-bike for 15, my dream bike is pretty much the same as yours.
I'm a commuter, I don't want maintenance on my bike!
Awesome idea there with side to side thumb throttles to mimic a single bidirectional device! That's really quite clever.
We did our first tests with an accelerometer control in 2009 and it worked great then but sat on the back-burner for way too long. I suspect that in the year or two we'll start seeing this everywhere since it's just so obvious and inertial sensors are so ubiquitous and cheap.
Glad to hear you are in agreement with most of the points in this video, it's reassuring when others with similarly rich firsthand experience have come to the same conclusions.
@@GrinTechnologies Hey thanks for the response. I'm in agreement for not most of your points, I'm in agreement for ALL your points.
I also have a Yuba Mundo, and my dream bike is the same as yours. This is becoming very strange.
Anyway, keep up the good work guys! Your products are awesome.
That's a pain them not putting analogue hall sensors in the brake levers... BUT can't you just glue a magnet to the brake lever and stick an analogue sensor on the brake lever body? That way you can get variable regen braking and mechanical brakes on one lever that is compatible with any bike.....
A lot of kits have a similar thing going on...
I could even d.i.y it by soldering some leads onto an analogue sensor then plastidipping it to make it waterproof
You provide the support in the controller for variable regen braking that I can calibrate similar to calibrating the throttle end points on an RC car esc, and the wires to solder an analogue hall sensor on and ill do it myself.
Thanks for this. My bullshit detectors went off when I heard the "it's not worth it" line over and over again.
Great presentation, Justin! I used to replace brake pads 2-3 times per year before regen. After regen, I’ve gone 36,000 km in 3.5 years with one replacement (and that was mostly because my max regen phase current was set too low for the first couple years).
I used to get almost 20% regen riding in the Oakland hills. Averaged 4.4% on my 8700 km ride from Vancouver to San Diego.
Wonderful
As usual so much worth while info
This is great news to me. I was excited by the concept of regenerative braking when I first heard about it. And I then heard the negative hype about how it wasn't really worth the trouble.. That was a big disappointment for me but I kind of just accepted it. Now you have restored my enthusiasm a little. Thanks :)
The constant negative talk about regen is just so bizarre and out of touch with reality. It's was largely fueld by mid-drive enthusiasts trying to seek some justification for a major shortcoming of that topology by pretending it's no shortcoming at all. Hopefully we can start seeing a shift in perspective with more and more people singing the praises of regen.
I have variable regen on my bike with a throttle to modulate it... It stops on a dime without throwing me head over bars or locking a wheel up. In fact, I hardly need to use the normal brakes except to finalize a stop or to add extra braking force.
My bike tractor is offline at the moment, but I regularly get 15-23% regen on cargo trips to and from my hilly San Francisco neighborhood. Even on trips to flatter parts of the city and nearby cities, because of the initial descent, I'll still get 6-12%. Definitely worth it.
wow that's great. I was debating making my own custom e-bike form scratch with tools form my maker space and I was debating if regent breaking is worth it
to solve the free wheeling problem I was thinking of using some sort of clutch mechanism. the motor can peacefully absorb the spin while the pedal disconnects completely from the system,
i can't thank you enough for this vid. as i am about to invest in a full Grin E_upgrade for my recumbent trike, after which i plan some serious mountain riding over the Alps, thus, reGen will be essential.
Thanks Justin, excellent presentation! Another advantage of regen is that on downhill you can still (gently) keep pedalling so you can increase easily the % of energy recuperation.
@@dcktater7847 Could you elaborate better your concept ? 😎
Thank you for further explaining the benefits of hub drive over mid drive. It's insane how many bike shops hate on rear drives because of made up reasons.
Many or just Johnny Nerd Out?
@@chrisE815 many, mostly the non nerdy ones I have noticed actually, like the car salesman types.
There are positives and negatives to both systems. In my opinion on most e bikes hub motors are perfect but the unsprung weight of a hub motor on a bike with suspension such as mountain bike is a massive issue
@@chrisE815 just Johnny Nerd Out. He loves milking that hate to get more views.
@@sammiller6631 lol
I don’t like to go fast down hills. I don’t feel safe (note I have well over 50,000 bike commute miles over more than 40 years of commuting, with less than 10 falls total). I love my regen, it’s a game changer for a nice relaxing rejuvenating ride. Maybe 40 years ago at 26 years old I would feel differently 😅
Even the young folks love regen!
OMG MY DREAM BIKE JUSTIN YOUR AMAZING DUDE ZERO MAINTENENCE YES !!!!!!!!!!!!
Yeah it's awesome!. After experimented it, it spared me a lot of brake maiontenance, disc alignement, pads replacement, caliper cleaning, bleeding etc... Both of my bikes are heavy, 165lbs-180lbs, + me 230lbs so it generates a bit but defifinitly about the brakes with all wheel drives bikes like me it's awesome and worth it.
Thanks Justin ,and Inspiring and informative presentation, much appreciated!
After a year of using my own design Regenerative module I might add:
1, I found that having a multi-level regenerative settings, one for low speed pedaling that ranged from 40w to 80W from about 5 to 10mph was useful when used in conjunction with pedaling on level or slightly down hill rather than for breaking, and two others for long moderate down hill at about 200w to 300w,, and a third very aggressive mode for stopping that ranges from about 1200w above 20mph to about 500W at around 8mph. Around hilly Portland OR., I'm getting about 12 to 20% re-gen.
2,, is high battery charge current. My DIY system is capable of up to 1500w of above 15mph and I often run 500 to 700W for several blocks on moderate hills which is between 15 to 30amps of re-gen, which is well over the charge rate of most Lithium batteries, 3 constant to 10 to 15amp charging bursts might be reasonable for most eBike packs, so 300 to 500w might be a safe Regen level for such packs but for typical riding I found only delivers about 3 to 8 percent overall range improvement. To Improve range and re-gen amperage I've added a parallel 600 watt-hour LFE (Lithium Iron Phosphate pack) in parallel that adds about 20+ amps of regen charging ability. over the 6 amps (300w) I feel comfortable hitting the Li-on Pack with.
As a torture test over about 800miles and several months, I hit a 48V, 500 watt-hour lithium-Ion pack that was rated at about 2amp max charging with typically 10 to 28amp of re-gen, right up to full charge voltage (NOT RECOMMENDED OR LIKELY SAFE) And the high quality industrial (Server-rack) Li-On Japanese cells I'm using appear to have suffered no measurable increase in internal resistance or decrease in capacity, but I wonder if the high level of charging could cause dendrite formation and increase the chance of shorting?
A third benefit of re-gen I found is safety and stability. I use re-gen as much as possible and it helps me maintain a constant speed without lock-up as regenerative breaking automatically decreases with speed, and increases with speed matching pretty well when needed sort of like a passive anti-lock system, so I don't have to get aggressive on the brakes, As well, if the tire under re-gen breaking hits a low traction surface, by the nature of the system I've never encountered a skid under even heavy re-gen as might with mechanical breaking.
Thanks again for the great presentation Justin, Regen is something I appreciate as I do your passion for it.
Hi Jack and thanks for sharing in detail your firsthand experiences too! I still plan to do a much deeper technical talk on many finer points of regen, this presentation was just a primer :-)
@GrinTechnologies Hi Justin,
Do you know if there is a Chinese version of the Pinion gear for retrofitting normal bikes? I have been thinking about this for a while.
As ever, great presentation by Justin
I just rode a 117 mile ride with 10,600' climbing, and got over 50% regeneration capture. The hillier the terrain, the more regeneration helps. The best part was descending a 3,000' climb and never touching the brakes, instead putting all that energy back into the battery.
It's a great feeling indeed, and congrats on the pretty epic ride
That's the kind of use case I imagine and Justin's dream bike would be pretty much perfect. It's what I'd imagined would be the simplest setup but good to hear it from an actual expert...
The "Rion Curve" and the throttle from the Boosted Rev are worth looking into. They are thumb wheel inputs that physically spring to a mid-point hall sensor (so their default state is about 2.5v). Then all you do is push forward or pull backwards and you have the controller interpret that. They are AMAZING when used for throttle+regen and the ergonomics are extremely hard to beat.
Sadly, as Justin has lamented, finding thing like this in the market is extremely difficult right now.
Awesome thanks for the tip on the Rion throttle, we weren't aware of that one being out there and it looks like it would be compatible with bike handlebars too.
@@GrinTechnologies Yup; just wish it wasn't $200! (And, ya know, available yet)
This is cool concept. I know electric wheelchairs in Walmart have forward backward control with one switch so it's existing techn and should be easy for a factory to make them with the right molds
Sounds perfect for me as I live in San Francisco, hills and traffic are plenty. 😁
33:50 Negative pedal assist. That's what I was looking for. Are there bikes with this?
I learned more from this guy than ANYONE in TH-cam when it comes to e-Bikes, thank God he chose e-Bikes as his area to work in, because with his genius, he could DO ANYTHING he put his mind to!!! 💯👍
Edit: I'd gain 20%-25% or more because I carry heavy loads, ride in stop & go traffic with hills everywhere, so what's the cheapest way to put this on a rear motor e-Bike??? 🤔
Thanks for the flattering comment about Justin!
For your question, Step 1 would be get a Direct Drive motor or a Gmac... step 2 would be upgrade your controller, Baserunner or Frankenrunner so you can start doing regen. Definitely contact sales and support to get into specifics.
One other key point I love about regen braking on my BionX is that it is inherently anti-lock braking. This is especially nice on snow and ice.
Thank You! I actually had no idea how regen braking really worked until I watched this video(I'm fairly new to the ebike game) I've also seen a huck cycle review video with shreddie mcskate where he locks the bike and tries to push the bike forward as if running away with it and the the Regen braking activates a drag to the bike slowing it down dramatically to wading through quicksand speeds it seemed really useful... Damm I said a lot my bad lol
The ride share scooters have this feature to help with theft although it’s only 1 wheel that locks up and they’re still light enough to pick up. It would be nice to have on my heavier dual motor scooter though
I like to salute you for this Rich information where you looked to matter from all angles giving all choses so everyone chose what's suitable for him.
Thank you ones again and good luck to you.
Thanks Justin, Really like the last picture of a low maintenance Ebike with the front wheel having a motor. Hi from Seattle
Right on! Take Care!
great presentation. it's clear to me that variable regen is the quintessential application for an analog Hall sensor in the brake lever.
my interest in regen goes to the design of a e-trailer with regen capability. having done a long tour with a single wheel cargo trailer, the first thing i noticed was desperate need for braking on the trailer. all that extra mass in the back puts far more stress on the brakes and also become unstable.
having the trailer do the braking while regenerating is the perfect solution. your motor just went to the top of the list. i'm also reminded how similar 'marketing' is to 'propaganda.' thanks. keep smiling.
Hey stephen, yeah it doesn't take much to tie an analog hall sensor and magnet to a brake lever and get modulation that way. With a trailer you also have the option to have a force sensor on the hitch and activate regen on the trailer wheel whenever the trailer starts pushing the bike (and use reciprocal approach for power too)
@@GrinTechnologies i really like the accelerometer idea for a trailer. the problem would be averaging due to bumps in the road. i'm thinking of an e-trailer that would provide just enough torque to push the trailer's load and provide regen braking.
a great design goal!
thanks for a clear and deep explanation backed with numbers of something I've always had a gut feeling for. it's exciting that ebikes are still in their infancy with so many improvements to come. blended Regen braking on the brake lever like on my ev (fiat 500e) sounds awesome.
hope you'll be heard by the rest of the industry
In the meantime a good easy stop gap solution is to use a magnet glued to the brake lever, and an analogue hall sensor stuck to the brake body. Loads of ebike kits use a similar thing.
Calibrate the brake end points the same way you calibrate the throttle range on an RC car ESC
Some kind of input like full throttle, or button to press upon controller power connection on and boot up to enter programming mode.
Where it will then periodically cycle through available settings, beeping a certain number of times (simply by using the motor as a chime, just like in RC cars) each time it switches to indicate what setting will be adjusted a step when the button is pressed or blipped with full throttle then beeping again a number of times to indicate what level the current setting has been set to.
Easily navigatable with a piece of paper with numbers that correspond to individial setting, the each setting having its own numbers to indicate what steps are available for each setting)
Though, when you press,
or blip the throttle on the regen brake setting, it will be a little different. It will enter regen brake end point calibration mode.
It will take the current signal from the brake hall sensor when you enter regen brake calibration mode as 0% brake. Then set your wheel rolling. Now start squeezing the brake lever. As soon as the wheel visibly begins slowing down. Stop squeezing.
Then press the button, or blip the throttle and it will set that point as 100% regen brake.
At that point you have perfectly calibrated Regen and mechanical hybrid brakes
@@GorillaZillas I'd love to test ride an ebike with strong Regen properly calibrated this way on the rear brake lever...I love my fiat 500e Regen braking on the brake pedal. I can use 100% Regen braking on my daily driving even if I brake hard and whatever Tesla one pedal driving fans can say, my brake pedal feels better than on my gas cars.
the cherry on top would be to have software abs on the rear brake. front tire is way less likely to lock up when braking hard and/or downhill (although with higher consequences), it doesn't sound that hard to implement abs if you have hall effect sensors
@@geemy9675 As far as I know, Regen braking is inherently anti lock, as resistance is caused by current flow, and current flow is caused by RPM. If the RPM drops anywhere near zero, so will the resistance.
You can see this effect when you drop a magnet through a copper tube.
Or if you short out a motor, then turn it to feel the resistance. Even a completely shorted motor does not lock up.
It might skid with a lower RPM than the speed of travel though, so there's still room for improvement.
@@GorillaZillas I don't know specifically about Regen being inherently anti lock. I don't see a reason why the wheels would stick at 0rpm but it sure seem like they could skid turn slower or maybe even backwards depending how the Regen is programmed but anyway it shouldn't be too hard to implement a feedback loop the same way electric motors can have finer traction control than ICE.
I know for a fact that my 500e completely shuts down Regen below freezing temps though
@@geemy9675 Yes, actually you're right. They probably would skid while turning slower than the rate of travel, could be solved if the controller doing the regen could read RPM sensors on the wheels and reduce braking power when the regen braked wheel rpm dropped lower than the non regen braked wheel.
This is how i broke my hip. Steep hill, forgot to turn off regen rear wheel locked on some gravel while i was carving then i slid over potholes.
Yeah...must be the regen's fault. 😒
i pop popcorn for their videos and his staff are funnier thats refreshing
Justin is right. Once I got an ev car and got used to 1 pedal driving, I immediately wanted regen on my bike. I absolutely hate using brakes on any kind of vehicle now. Why not just have a set regen like ev cv cars do? How would you know if the motor is overheating?
We have the same dream bike. I would love a belt drive ebike.
The engine braking feature can be implemented fairly easily using the Digi Aux regen control that's in the CA3.2 firmware. You can adjust the amount of background regen with the digi aux input, but whenever you use the throttle that overrides it.
once I got an ev car and got used to blended Regen, I immediately wanted regen on my ebike. I absolutely hate using mechanical brakes now, when I can have very strong regen and precisely controlled regen with the brake pedal, that will put back as much energy as possible back into the battery, save brake pads wear and prevent brake from overheating when going downhill. Just my personal opinion about Regen, as I think one pedal driving is far overstated and blended Regen braking far understand because of Elon Musk arbitrary preference and ignorance about blended Regen braking (which doesn't have a "weird brake feel" when implemented properly). one pedal driving and lack of blended regen limits max Regen power (a 1000hp plaid only has 75kW of Regen when a taycan has 350kW or my 111hp 3000lbs 500e has 80kW)
If I was given choice I'd rather have as strong as possible controllable Regen (for instance using a brake lever position sensor) like on my car than Tesla-like one pedal Regen especially since on an ebike with pedal assist you only use the throttle when you're not pedalling
for me the dream ebike power train especially for a full suspension bike is revonte one mid drive that provides both pedal assist and infinitely variable transmission in the same package with less moving parts that a pinion or rohloff gearbox and without losses or complexity of a nuvinci cvt. basically the eCVT of Toyota/ford/Honda hybrids adapted to ebikes. They say Regen is possible in the future if the rear hub has no freewheel and the chain belt keeps turning with the rear wheel.
you get the advantages of a mid drive, with a super low unsprung weight, possibility to run a belt or chain without tensioner, perfect chain line, and all the complexity and weight is located around the bottom bracket, low and close to the battery.
I added a simple horn and light switch to the regen input of my Cycle Analyst. The horn button (momentary) triggers temporary regen. The light button (on/off) adds regen for longer downhill mountain runs. Both settings allow for variable regen by twisting the throttle. Cheap and easy.
wow, that's beautiful.
Marvelous, I wanted to see the talk since I knew it was happening.
Question about non permanent magnet moroes and regen. Does one use more power to create the fields than what one gets back from the subsequent regen?
At very low rpms yes, at higher rpms its not even close. The crossover point depends on both the intensity the regen torque and the resistance of the motor and coil windings.
ok. so non permanent magnet motors can do just fine? btw the coreless motors, are they viable or will they become too big? can they be small if they go high rpm?@@GrinTechnologies
I hope that backpedal regen via crank position/rotation survives the beta and becomes a factory option. It would be my prefered option, as it seems to be the most intuitive if you can set it up right. Altough, the two throttle option dosent seem too bad, but I would have it set up differently - A twist accelerator and a thumb regen on the same handlebar sitting next to one another, far better ergonomics in my mind. I would even have the regen cutoff the accelerator for saftey. Its a bit of a hybrid with the bidirectional throttle, as you are pulling with your wrist and then pushing with your thumb.
I just love how regenerative braking feels
I bought my direct drive RadCity Step-thru 3 (Canada) (low step Radcity 4) this November on the clearout sale, which has regen braking. The direct drive hub motor has "Shengyi" and "HYBRID" all over it and it is indeed a true plug-in hybrid electric vehicle.
Regen braking is disabled when accelerating from a stop with human power only - PAS 0 and not using the throttle.
In Hybrid and EV modes (PAS 0) when the throttle is released, low regen applies.
Slightly tapping the brake levers will engage full regen braking - more than the amount by just throttle off.
Too bad my direct drive Radcity is now discontinued and is essentially irreplaceable now...
Yeah it was unfortunate that the rad DD bikes were all discontinued. Glad to hear you got one at a deal. There aren't many factory bikes with regen at this point in time but at least there are many great DIY options. And in due time they will make their way way to factory ebikes in a big way.
I have variable regen on a rear DD hubmotor commuter that I built years ago. My commute included some steep downhills that would heat up the brakes to the point they were starting to fade and there was no emergency capacity left. With variable regen the speed was controlled while the brakes were cool and ready for any emergency. It was like having an adjustable parachute going down the big hill. I didn't care about the recharging of the battery, but the saving of the brakes was worth a lot. Variable regen is really nice and definitely worthwhile.
I love Regen, and I also love your passion. I look forward to buying Grin components when I can afford them.
Another great 2 things about Regen, at least my theories, is 1, it's a natural Antilock Braking System, because if the wheel locks up, like on ice, then its not spinning, and the generator effect goes away. So the theory I have yet to test is, even with a torquey DD hub motor, and a high current controller with Regen. The BrakeRegen will naturally always be at the threshold of grip. But this also makes me think about how long for the induction in the coils to fade away, and if stator coils can be designed to quickly lose magnitism when magnets stop.
My second Regen theory is.
Wiring battery packs in parallel with capacitor packs, will greatly improve the efficiency of regen, because the charge limit of the batteries themselves is a bottleneck, but capacitors can charge & discharge quickly. The more capacitance, the better. Wire enough ultra-capacitors in parallel groups for capacity, and then wire those groups in series until their combined maximum voltage ratings cover the battery charger voltage. Or use high voltage capacitors that cover the charging voltage and just parallel those together for many Farads of capacity.
My experiment & problem with my 2nd theory is.
The space within & on a bicycle or motorcycle frame, to be dedicated to energy storage. What is the best ratio between batteries and capacitors? Be it ultra-caps or normal capacitors. Should the capacitor pack be as big as the battery pack?
Two times as big to get the most out of Regen?
It's like, be battery biased if you are traveling long distances with little to no braking. Or, fill your frame up with mostly capacitors, if you expect a lot of stop & go city use, or tight, technical routes.
I don't know the exact ratio for expected use cases.
I do know Capacitors have better C ratings then batteries, and can capture the most energy from regen, to dump into launching from a stop or out of a corner, or charging the battery at its own pace, if you RegenBrake to parking.
I just don't have the money or time to do these experiments.
And like the presenter, I hate bike maintenance. I just want to do my delivery job, with a reliable, stable, but powerful bike, then relax. I'm hoping using 2 torque biased motors, means I can phase out brake pads entirely. Using some metal shoe on the tires to act as a parking brake.
Either way, thank you all at Grin, for your quality products, passion to improve EVs, and thank you for this presentation, I wish you had more time.
As usual "classy" Justin! Very informative! Thank you Justin.
Completely agree. I ride a handcycle with e-assist and my brakes are my backup safety system. 95% of braking is done with regen. I can set a safe downhill speed on my cycle analyst and not worry about using my brakes and just enjoy the decent..
Just add a 3rd brake lever that works more like a throttle on a variable resistor. 2x levers could be on the same hinge pin so you could either pull them both together or just feather the upper one for re-gen regulation. Sometimes you will need both together anyway.
I always enjoy listening to your lectures, Professor Grin!
I don't know if you can answer, but I was wondering about regen's effect on battery capacity degradation. I've looked at some research papers, but some of the more applicable looking ones are behind the paywall sadly. If this is not too specific, I have some high-power 21700 cells (P42A) I always keep between 4.1-3.2V, but I've been wondering what is my regen 'C' limit for limiting long-term battery damage.
I'm not limited by BMS (I use RC chargers for balancing), and the battery spec shows a chart charging at 2C (8.4A), but I know in general it's better to charge at 0.5C or less for longevity so that's what I've always done. Regen doesn't induce the same constant current for the duration of the charge cycle, so what I'm really wondering is if short bursts of 2C or higher regen rates would end up accelerating the capacity decay over time. Any insight on this?
I wanted to go into address this topic but left it out of the presentation. The quick answer is that all evidence suggests that short duration regen currents even at high rates has almost no meaningful effect on cycle life. If you do a full charge from 0 to empty at say 2C that definitely reduces cell life cycles compared to say charging at 0.5C, but just occasional bursts of 2C for a few seconds or 10's of seconds hasn't had any effect that we've been able to observe.
@@GrinTechnologies Nice, thanks for the response :)
@@GrinTechnologies I had this question too after watching. I can regularly pull 1kW (about 1C) for 10-15s going down the hill from my house to the main road I use to go anywhere and was wondering what impact it might have on my new 52V/20Ah battery. Thx.
My Grin rear hub motor (26" wheel) regen torque seems limited by a 500 watt power ceiling it seems. How can I increase the regen braking force? My regen phase amps and battery amps are already set as high as they can go in Phaserunner suite. Is it a Cycle Analyst setting? By comparison, my other Grin front hub motor (20" wheel) is putting 1.1 Kw back into the motor and braking stronger. Thanks!
It's generally a function of how fast you are going, so if you want more regen watts you have to travel faster. To get the same max regen energy into the battery from a 26" wheel requires 30% more speed than you'd need in a 20" wheel all else being the same. That said there are many settings in place that can result in other limits kicking and this is the territory of email tech support more than YT video comments.
@@GrinTechnologies Thank you :)
Although they last longer than brake pads, rim surfaces / disc rotors won't wear out as quickly either. This may be more significant for commuting due to the gritty stuff kicked up in the rain. I also wonder if mech brake cables could last another season given the reduced usage frequency and tension.
Aside from cost savings, I also like that regen can reduce braking noise (particularly disc squealing).
Ah yes, good point about the brake noise too! It doesn't cost $$ directly but I've seen countless time wasted by people tweaking their pads and calipers a bit here and there to eliminate a stubborn squeal.
If nothing else, it is extremely convenient to save wear on the mechanical brakes, not having to change brakepads or shoes all the time.
Regen braking is the best thing ever, my GMAC brings me to a complete stop. It does take a little getting used to though.
thank you!
Nice work Justin
Great video. I would really w like regenerative motor. You gave some great information.
The back pedal idea for some of who are older wouldn't be a bad idea. We all had bikes that the brakes are the pedals. We wore tires out a lot!
This is awesome.ow, I never knew about the benefit of the braking and not wearing brake pads and in wet weather, mind blown. Looking to order new wheel today I pull a trailer too.
Really great talk. I've also had the dream of an ultra low maintenance bike.
But am i correct that the GMAC motor is not currently designed to attach to the front hub? Are there plans to make that possible?
Super interesting! Justin's dream bike sounds kinda like my dream bike.
Gimme dream bike: th-cam.com/users/clipUgkxn_EHFPyE-pMGEJ7dDorWl8cih_NMsS1Z
Which controllers can do active electronic braking to come to a complete stop?
All the controllers that grin currently sells (Phaserunner, Baserunner) and has sold in the past 6-7 years (Frankenrunner, Grinfineon etc). Other controllers that do good variable regen would be the Kelly controllers, the Vedder VESC, and or any industrial BLDC servo motor controller. Most inexpensive generic chinese ebike controllers do pretty crummy regen, if at all.
Hi Grin Tech, at 3:50 you explain that this pseudo regen relies on the motor spinning very fast and has a higher voltage than the battery pack and occurs at a critical speed. Why does this happen? What causes the motor to not just continue to spin faster and faster? Why does it slow down as it is putting negative current back into the battery?
In my experience this pseudo regen is happening when you have the throttle opened up all the way (so the controller has hit its max rpm output to the motor) If at that point you are still gaining momentum from the hill, the controller will be holding you back the max rpm it is programmed for by design. If you let off the throttle completely while still on the hill the controller won’t be limiting the rpm and you could potentially continue coasting faster than with the throttle engaged.
Very good presentation
Lots of great information here. I learned a lot. Thank you.
this will be fantastic if it catches on its the future for sure
Also, once the battery and various capacitors are charged I assume the free-wheeling / direct drive will have slightly less drag? I'm hoping to build around a 48v Maxwell DuraBlue Ultracapacitor (+ Lion battery in parallel). 6.0 mΩ of ESR, rated 100A continuous, 1600A short circuit pulse!, but only storing 3 Wh over 18 cells @nominal 2.7V!
Well done. Thank you!
This has been a great presentation. When I was growing up we always stopped our old single speed bikes by backpedaling to activate the brakes. We didn't have rim brakes yet and certainly no disc brakes. They also coasted pretty well too so they weren't "locked" pedals. I'm pretty sure regen could be implemented by using the pedal cadence sensor? Maybe a torque sensor instead of a cadence sensor? Move it slightly it regens enough to slow you down, move it back a bit more and it regens even more effectively even all the way to a stop? With our old bikes one needed to keep some back pressure on the pedals to ensure the brakes remained on but if one wanted to release the brakes then one simply let up the pressure or began pedalling again.
Another issue I'm thinking about is the lack of torque a direct drive motor has verses a geared hub drive motor. I currently have a geared hub drive motor on my bicycle and it works fine although it can be too weak sometimes. I've heard from many people about direct drive hub drives and the overall consensus is they are underpowered. Not in horsepower of course but in torque. How does one compensate for the difference between the torque from a direct drive hub motor and a geared hub drive motor? Use two direct drive motors, one in the front and one in the back? Or use a GIANT direct drive motor in the back? Of course weight becomes an issue once one reaches a certain size. Interesting problems and it's going to be interesting how it will be resolved.
Thank you everyone at Grin for continuing to research and experiment for ebikes.
th-cam.com/video/gYCj4asp9pE/w-d-xo.html
Direct drive motors can delivery plenty of torque for almost all applications as long as they are sized appropriately and your motor controller can give sufficient phase current. Motors with the same winding KV will give the same torque output for a given phase current whether they are geared or direct drive. Compare say a BionX D direct drive motor system to almost any geared motor ebike and it will throw your "consensus" out the door!
@@GrinTechnologies Thank you for the quick reply.
Rad Power Bikes used to make their Radwagon with a direct drive motor and 26" wheels. However they made the newer one (Radwagon 4) with a geared hub motor and it's got 22 inch wheels. Their literature says they went to the geared hub drive for the increased torque and went with the 22 inch wheels to lower the centre of gravity. What you're saying is they could have kept the direct drive motor and increased the phase current to make it more powerful?
@@davemeise2192 Yes exactly. We were a bit bummed to see Rad drop the DD hub motor option too. For the same torque capability a geared motors will typically be a few pounds lighter than the DD equivalent, no disputing that. But the really well engineered DD hubs (like the Grin All Axle or BionX D series) cut that gap pretty close.
They can make locking cluches both mechanical and electrical.
My regen is a thumb throttle inside my twist throttle. Has my headlight button as well. Works fine even with gloves on.
The hub motor wins even on the gental down slope because you can peddle against it creating more power to regen and gravity helps you alot!
I have a Bafang BBS02 build. I like the range I get from the mid drive, but if money was no object would it make sense to add a small hub motor to the front wheel, purely to use as a regenerative brake?
Almost like a dynamo hub but with much more useful power generation than the ones that just charge up your lights.
It actually makes more sense to have a decently powerful hub motor that you use all the time for motoring and for regen, and then only use the BBS02 when you need an extra kick on hill climbs.
Deep stuff! The stuff of dreams 😍
My dream bike setup would be 42:11 with a mid-drive motor (for the integrated torque sensing+cadence sensing, and what I heard better traction in the rear wheel), and an automatic shifting rear IGH. Though not sure how much more benefit would the mid drive offer, and whether there are any technical challenges with integrating the front and rear wheel drive power. (So the front hub is mostly there for regen XD)
As for the braking controls, would it be difficult to make the brake tripwire offered in your shop use some sort of hall effect + magnet setup, or some kind of pulley + encoder setup, or any other kind of a linear position sensor?
My 48V DIY bike with direct drive hub motor paired with a regen capable controller only lets regenerative braking above 16mph speeds, thereafter I have to apply mechanical brakes to slow down further. I understand the motor voltage has to be higher than controller's for regen to work, but the speaker implyed the regen should work almost down to 0mph. Am I missing anything? Those who have bikes capable of regenerative braking at what speeds does your regenerative braking cut off?
Your controller is definitely not configured for proper regenerative breaking if it stops working below 16mph. All regen controllers we have encountered will by default do regen right to a stop just fine. In some cases there is a configurable minimum regen speed setting for those who want it out a bit earlier.
@@GrinTechnologies thank you very much for this response. I'll look into the controller settings once again. Possibly I have controller settings and motor characteristics mismatch.
With hydraulic disc brakes, it seems that adding a hydraulic pressure transducer would make a lot of sense. With a transducer, the electrical signal from the transducer would be proportional to the hydraulic pressure in the brake system. The regen controller could then use that signal to proportionally vary the regen. In other words, the harder the rider squeezes the brake lever, the more regen will be applied.
Yes, this suggestion has come up as well! The slight problem with it is that you need to fully engage the mechanical brakes before the hydraulic line pressure builds up, which means you are wearing out the pads too. Perhaps with a high sensitivity transducer it would be acceptable, but still ideal is not having the brake pads touch at all.
@@GrinTechnologies You can add flow controls to hydraulic circuits, but I don't recall all the options available as I haven't done engineering work in a long time. My gut tells me it's doable though. Perhaps a flow control needle valve, that one could dial up/down to regulate the flow to the hydraulic lines with the pressure transducer upstream of that. So, you could dial up/down the regen by controlling the flow rate to the disc calipers. In other words, with the flow control valve restricting flow to the brake calipers, the transducer would send full signal for max regen to the ebike controller while the brake calipers would be only slightly modulated. If more friction braking was desired, then one would back off the needle valve to allow more flow to the calipers
Just found this, pretty similar to what I was thinking: th-cam.com/video/vMoQwY5WrwQ/w-d-xo.html
@@GrinTechnologies What about using potentiometer, for cable power brake?
Random question but with the prevalence of Hydraulic brakes, why not have a 0-5v brake sensor in line to activate the regen?
Love it! I didn't know about the new feature to back pedal to turn on regen. I've always just slightly squeezed the brake lever and used the throttle to increase regen. However, even that wears out disc brake pads sooner than necessary. I'm 100% with the beta tester who suggested back pedal just to turn on regen and still use the throttle to increase it.... Looking forward to setting this up. Thanks Justin!
so you program the controller so that as soon as brake lever cut off the power, throttle is used in reverse to modulate Regen? sounds smart although it means if you are using full Regen and slowly release the brakes you are jumping from full Regen to full throttle ?
why do you have break wear, isn't it possible to avoid engaging the brakes?
I think the definite answer to Regen is really a brake lever with first half of the lever travel spring loaded and triggering a linear Regen, then once max Regen is reached, the second half controls mechanical/hydraulic brakes. this way you make sure that any hard braking uses 100% regen, you can still brake the same way with electronic off or Regen is disabled (with just less initial bite and more travel), you can precisely control Regen and you don't have to learn a different way to brake, just to recalibrate to the blended braking.
This should of been done ages ago when ebikes became so popular.
That larger regen in hilly areas seems odd to me since you use way more energy climbing hills in the first place and get nowhere near that energy back descending.
You get more regen in hilly areas for sure since you do more regenerative braking as you would on the flat. Your total Wh/km is still higher in a hilly area than on the flat ground, but it's much closer to flat ground wh/km figures than it is to hilly riding wh/km figures without regen.
Does a regen brake system can be installed on a bicycle with a CYC X1 Pro Gen2 mid drive motor system installed with BAC2000 controller and powered by a 72V battery?? Thank you for a detailed answer.
No it was literally mentioned at 4:40, mid drives cannot use regen. I myself am getting an X1 PRO GEN 4, there's a freewheel attached to the motor such that it'd be impossible to reversely spin the motor to generate energy even if the rear axle was fixed
I wish I could like this twice. I love every single video on your channel Justin.
I'm trying to be cheap on my 3rd ebike build, but I think buying cheap components is more expensive in the long run. First 2 we're mid drive, but I think I'm sold on hub for less issues etc. Thinking about buying Grin all-axle motor and putting it on a Surly frame for the ultimate reliable ebike. And building my own 2.5-3.5+ kwh pack in the triangle. What's the best cell for longevity, even if it's at the expense of energy density. Open to other chemistries, but trying to keep battery pack under $1500 if possible (not including tools and my labor/time). I would like to have at least 50% capacity after 5,000 cycles. Considering lifepo4 even if it means less capacity than what I want
Hey Geoff, it's often 3rd times the charm on custom ebike builds! That said, a cheap direct drive motor will be plenty reliable, just heaver than the All Axle. The area not to skimp out if regen is important is on the motor controller so if there is one place to splurge a bit it's there. For your battery, the best longevity comes from using name brand cells like Panasonic more than any specific chemistry. We're seeing tons of larger capacity batteries with the Panasonic GA cells that are now reaching 7 years old under regular use and have only lost ~10-15% in capacity. There is really no place for LiFePO4 anymore when cylindrical lithium cells with almost 3x the energy density are performing this well.
@@GrinTechnologies thanks for your reply and for giving me the reassurance of using traditional lithium ion cells. Looking at Samsung 50E 21700s but still haven't decided exactly what I want. A lot of folks tell me to get high discharge rating, but in a 14s14p pack, I need very little amps from each cell. Hopefully that combined massively oversized pack and charging to 90%, discharge to 10%, I will have a pack that could last me a decade. Long lasting pack and durable bike means 1000s of miles riding without much worries. Thanks once again!
xD I checked out the company in that article, Nano One, turns out they're from BC! Burnaby
@@Geoff9001 The Samsung 50E2 cells would be fantastic if you are under 2C per cell discharge which would be 140 Battery Amps in your 14P application.
For even better longevity, set your battery cutoff to 20% or higher as research has shown discharging lower to lower levels increases capacity degradation faster than charging to the upper limit. So something like 3.2V or even 3.3V/cell+. You'll have a ridiculous amount of range anyway with 14P high capacity cells.
@@Geoff9001 I have a 14s7p 52v rectangle Panasonic GA pack. It's pretty big ! Are you planning on 100v system or extreme range? Some Enduro frame batteries are 8kw 72v 40amp , 240 cells ... 15 kg!
Very interesting, thank you!
LOVE the silent smoothness and consistency of regen! Got 10-15% when I was using RC lipos. Recently I switched to a Aliexpress ebike Battery (52V 19,2Ah 5A max charge, 21700 cells) and am scared to use regen now. Can the battery take the regen from a MagicPie5?
Is it safe to use regen with a chinese battery? What would make it safe? Could the energy also be dumped as heat before the battery? If, someone knows, thank you!
Just use a fuse with the same value as your battery voltage . This is probably the cheapest way to solve the issue
Even with generic quality no-name cells you should have no problem doing regen at C/2 or even 1C. The max charge rating is usually related to the charging port mosfets of the BMS circuit (and associated connectors and wiring), and doesn't usually relate to the discharge port currents. I wouldn't worry about regen putting you at any greater risk than you are already taking with the aliexpress battery.
@@GrinTechnologies Awesome! Thank you
This also gets into the realm of antilock braking as its good to know how much or how quickly the regen braking comes on and at what point to limit it, other than human braking input, and i guess you could use spokes or brake rotars as circular encoders for wheel lock etc.
Great video, thanks!
are there no concerns with the battery charging rate? i.e. the high current flowing into the battery during regenerative braking
We've seen zero evidence of it. Look at the 'C' rate on say a hybrid car which does like 4C-6C or higher during regen, while on ebikes you're usually in the 2C or less realm, and virtually all decent lithium cells can handle this for short time periods without any negative conseequences. The "maximum charge current" rating that you often see published on am ebike battery pack (like 5A) is a ridiculously conservative value based more on the charge port connector limitations, not on the cell's capability.
@@GrinTechnologies thanks a lot for your answer! it is good to know that in practice you haven't found any problems regarding the charging rate.
Ideally there would be brake blending, just like with a BEV. You squeeze the brakes to apply more regen, once that's maxed the mechanical brakes are gradually added on. On my Mach-E it tells me how much of the braking energy I reclaim every time, it's usually 99-100%.
I was wondering, how feasible would a middrive with regen be? I have a middrive with a one-way clutch on the pedals, so it can be done, but would it put too much strain/stretch the chain? I would hate to get chain slip though! 😱☠️⚰️
A regular chain with a derailleur doesn't work for regen even if you lock all freewheels, since the top of the chain goes slack as the derailleur gets stretched out and then the chain just falls off. You'd have to do a real custom drivertrain with both a top and bottom tensioner and have the derailleur designed to cope with full chain tension. Otherwise it's only viable with a single speed chainline and an internal gear hub that doesn't freewheel, of which the only model we are aware of is the Sturmey Archer S3X
@Grin Technologies I didn't even think of that!
@@GrinTechnologies IMHO "chain slack" issues not that it uses a "freewheel" should be the reason given for not being able to do regen for a mid-drive... Since I had the same thought about locking out the freewheel since the mid drive crankarm can freewheel, and that freewheeling should not affect its ability to regen. But I wondered about the tension (ie slack) on the chain, (in the derailleur's cage), being an issue? Which is the main issue to be overcome...
But overall its was a very informative and thought provocting lecture on Ebike regen, thx....
I’ve been looking for a regen bike for the last two years,the Rad one isn’t available in 🇬🇧 uk
You should build your own! It's really not that difficult, and has many advantages over factory ebikes: ebikes.ca/getting-started/kits-vs-turnkey.html
@@GrinTechnologies thanks, I’m up for that,I’ll give it a go.
Good description. On another website, I discovered that the BEMF doesn't go through the 6 drive MOSFETS, rather it flows to the battery via the 6 reverse current (MOSFET protection) diodes. So no additional diodes needed! I liked the description of how the braking modulation works. The controller looks at the throttle voltage and rpm. When the throttle is retarded below normal power off, the controller at first applies an almost equal forward voltage to match BEMF being generated at that rpm. Thus freewheeling at normal throttle power off. As the throttle is further retarded, less and less forward voltage is applied, allowing much greater braking force from the increased BEMF being applied to the battery. However, the regeneration circuit could have been made more efficient if regeneration was controlled not by forward vs. BEMF voltages, but by electronically PWM modulating any BEMF voltage generated (no forward voltage applied in regeneration mode).
The thing I love about regen is that I have saved tens of hours and thousands of dollars in brake pads over the last years.
What? Thousand of dollars in pads? How?
haha
Will disconnecting my bike’s 3 motor terminals remove or reduce the drag I’m experiencing when pedalling with a discharged or flat battery?
Yes if the drag only happens when you are above a certain critical speed when the back emf voltage exceeds the voltage of your flat battery. But no of the drag is present at all speeds
Excellent talk. Thank you for sharing this. I have a Yuba Mundo with a StokeMonkey...and I am thinking about either the GMAC on the back (and removing the StokeMonkey) or adding a Grin All Axle Direct Drive to the front. Thanks again for all you Grin folks do for the ebiking world...and the world!
If the stokemonkey is still ticking along and running fine we'd recommend to add the All Axle on the front. The combo of a DD hub plus a mid-drive is really potent for tackling just about anything with great efficiencies. If you upgrade the controller on the stokemonkey to use a newer baserunner controller like the All Axle, then running dual systems with a single throttle control is fairly easy too.
@@GrinTechnologies Thank you. I watched your great video on multiple motor setups a couple weeks ago and I recall your positive comments about a mid motor in combination with a front DD setup. And I am really enjoying your sailboat videos too!
To add to Grin comment, if you put a rear motor, you'll loose the special reinforced wheel at the back.
You won't be able to carry a full load on it.
As a longtail is usually heavy and needs a lot of power, your mid drive motor must chew your whole drivetrain pretty quickly.
I've got a front hub motor on my yuba mundo, it works great. As it's an already heavy bike, you won't feel the all axle in the front.
@@xmtxx Cool...thank you very much!
What does 'Grin' stand for?
It describes what happens the moment you ride your first ebike.
It would make sense based on your statements that a mid Drive would not be recommended when attaining a ebike if your goal is to get the most energy out of your battery. Is this correct?
Not as a general statement. There are times when a mid-drive motor will do better for battery usage (eg a steep hill climb followed by a long gentle descent), and other times when a regen capable hub motor will do better (eg a long gentle hill climb followed by a steep descent)
Yes, I personally, always Downshift my truck to slow down, so as to save brake repairs, ALWAYS!
So i do want regen on my ebike: Bafang mid drive, and my etrike: ELF hub motor, how do I get it? PLEASE!
"JUSTIN" You are just a super genius.
Does the re-gen switching mechanism ‘need’ to be within the brake lever mechanism?
Why can’t it be in-ine, interrupting the cable at a convenient place?
Oh it can and often is. The tripwire is often installed at the caliper rather than the lever, there are sensors that sense the moving brake cable inside a sleeve, and there are hydraulic sensors that tap into and sense the fluid pressure in the brake line.
I think a great setup for electric bikes is to have the front wheel be a disk brake and have the rear brake just be a sensor which controls regen (this assumes you have a hub motor on the rear wheel)
what about dynamo would that work on standard Ebikes
If you mean using a dynamo in the normal way to power a light, for example, then the answer is yes. EasyMotion has ebikes with front hub dynamo powering a light. Functionally, it's an alternative way of getting battery power to the light. 💡
Have an E-MTB with 2 wheel drive and 2 batteries (17 Ah and 20 Ah) and dual suspension that weighs 90 pounds. Yeah, a cargo bike with a 400 pound load capacity, including rider weight. I got it for hunting to carry boned-out big game meat and for bike camping with my UL backpacking gear and a 14 pound solar blanket and controller.
-> COULD I get regen. braking in BOTH motors with a lever braking sensor?
Yes as long as you have hub motors without a freewheel. Sadly most factory ebikes with hub drives use geared motors with a freewheeling clutch which makes them unable to do regen.
If you have a direct drive motor it can do regen for sure
@@GrinTechnologies I want to REPLACE BOTH MOTORS with FREEGEN motors! Is this possible with your controller and screen? My E-Cells Super Monarch Crown has a 52 V Sine Wave 45 amp controller W/18 MOFSETS in a SEALED conductive gel and aluminum, finned housing. It's rated at IP 65 waterproof.This controller is EXTERNALLY mounted on the seat tube and easily replaced.
How about regen solar and wind turbine windmill generator in the other wheel without hub motor or crank generator ????
How can we get a resistor to take the regen load when rolling down a hill with a full battery?
In practice this is just not an issue really. But if you live on the top of a very large mountain where the first thing you do after leaving is a lot of regen, then it is simple to just set your charger to only charge to 80 or 90%. It's better for the battery too
Yes but I charge to full because I do not know what way I will drive on the hill. so how can we get a resoistor to take the regen load when rolling down a hill with a full battery? I hear something about relays, but how do you activate said relay switch?@@GrinTechnologies
what is the concept or model of the controller between battery and hub motor ?
It is a 3 phase bldc motor controller. You can't really run the motor without one
Do you have a complete kit yet?????
Don't quite follow the question, we've had kits ever since day 1 of operation nearly 20 years ago. The list of regen capable kits (inline with theme of this video) is here:
ebikes.ca/shop/ready-to-roll-kits/regen-capable.html
I wonder if solar panels could be integrated into the fenders and cargo box lids to aid in recharging the battery?
Yes of course, but you do need a fairly large area to get in a meaningful charge:
th-cam.com/video/14yliWlykfg/w-d-xo.html
Thank you. I said this in 2003 on Yahoo Power Assist and the mid drive folks lost a nut.😂
I knew it!
Love you guys 👍 👌
It's mutual :)