ebike REGEN, whats the record ?

Hey guys,

Just wondering how alot of you have your regen setup.

Whats the highest % you have gotten back into your battery packs?

What type of motor/controller do you have?

Do you use a throttle brake or a button/switch?

Most amps, watts going down a hill ?

I found an oddity using the CA for recording Regen: The percentage does not necessarily reflect the true efficiency. It is rather a measurement of regen relative to the total distance. (Took me a while to figure that out; I mean - when I first started using the CA, it was like wow, I'm getting 5% back! - not.)

• Example of confusion: I live on a steep hill. If I clear the CA, make my run from the top to the bottom of the hill – ebraking the whole time, my Regen percentage is wicked! Although it’s also not correct because… well, I just went downhill and nothing; everything went back to the battery except what was used to power the devices (parasitic losses).

However I like the question you ask because we can still derive useful information from the CA (if that’s the tool of choice), and by setting up some ground rules for Regen measurement we could derive some useful data.

Should we measure based upon overall trip?

• Environment: Windless day, close to 72*F/20*C near sea level (STP).
• Known Distance: 1, 5 or 10 miles or kms
• One hill: Up, then down, equidistant – same path.
• Measure Voltage, Current, Wattage, Amps-In/Out, whatever.
• Classify by Motor.
• More fun with friends! :wink:

Should we measure on a Dynamometer? Watts-in, measure Torque, heat, calculate driven efficency... Then run the motor backward as if it were going downhill, measure Watts-in, heat, watts-out...

This might be a tall challenge for consistency across the willing membership, though certainly possible. Who wants to start up an ES Standards Committee for Regen?

I like challenges, KF :lol:

KF,
To much work for only an interesting dataset.

I've got a 16rpm/volt scooter hubbie that will run 60mph on the flats. Regen is quite light in force with this high speed wind motor, so I modded the shunt of my Lyen 24fet programmable controller soley to get higher regen current. Now regen current peaks a bit above 20A with my 74V rig. I have throttle activated regen, ie I let off the throttle and regen kicks in, and I love the way it works. I let off the throttle a while before a stop or turn and once accustomed to the force I just apply a little braking with the mechanicals near the end. Going down moderate hills some touches of regen keeps speed under control. It takes a 15% grade before I have to use regen continuously with a bit of braking.

I zip around at 35-45mph cruising speeds and use regen for well over 50% of my total braking. I regularly see over 10% regen over significant rides, and just today I flipped through the CA menu and saw 21% regen when I returned from a zippy ride. I've also seen over 2kw negative power while riding. The harder I ride the greater the regen as long as I remember to let off the throttle early enough, so I'll have some fun with this thread seeing how high a regen % I can get for a round trip ride. I'd consider that the only requirement for qualified comparisons, ie round trip rides. Slight downhill grades significantly increase regen, and headwinds and uphill grades decrease it.

I can't believe anyone with a direct drive hubbie wouldn't want to use regen. I love it most because I don't have to fool with brake maintenance. I can see throttle activated regen being problematic if regen force is strong, but it's the bomb with soft force. I've actually gone entire rides without ever touching the brakes. That requires some effort, but twist the throttle to go and let off to stop is an liberating level of simplicity.

John

Percent regen is a silly statistic, but the forward amp hours is a useful number, especially when recharging. It's a good way to work out how efficient your charge-discharge cycles are.

But yeah, I love it for the brake preservation more than anything else. The slight delay is a bit annoying though.

I did some experimenting back in '08 with my 1,600ft descent on my commute and with a slow Clyte 4011 saw around 15% regen at ~42ish volts into my 10s konion test pack.. <.5ah over 3.5 miles. Since there is no way I wanted to run a 4011 at 37v nominal and the regen wasn't all that great (need more mass!) I opted instead of dealing with regen to use a motor plug brake instead. That worked really well... even more so with my subsequent Clyte 504. Crazy stopping power at speed mellowing out as you came to a stop. In fact I went an entire summer descending that mountain with just a shitty front v-brake with barely any pad left and the plug brake though I had to put a big heat sink on my rectifier cause I kept burning them up until I learned my lesson. 8)

http://endless-sphere.com/forums/viewtopic.php?f=2&t=6553

On my current setup, (Sevcon gen4 and the dualpie), on the socal track i was seeing 35% regen, ~9 ah out and ~3.14 ah back. with the maximum regen amperage being 35A. activated on throttle roll-off

i Even like the sound. Its like my ebike is powering down
So am i right to think that a HT motor will regen more that a HS motor on average?
high winding down a big ass hill will pump more amps back in than one with less turns?

I was using regen with just a switch , it was set to about 50% power, now i have a thumb hall effect throttle, Its just so much better.

jonescg said:

The slight delay is a bit annoying though.

Click to expand...

???

50% +, using Kelly Controller, Grinhill like setup with roadbike. Measured multiple times up to top of Old La Honda Road, Skyline and returning down Kings Mountain.

kevo said:

50% +, using Kelly Controller, Grinhill like setup with roadbike. Measured multiple times up to top of Old La Honda Road, Skyline and returning down Kings Mountain.

Click to expand...

Kevo,
Outstanding! I take it you pedal a lot, so some of that regen is recovering some of your human effort especially on the trip back down. Regardless, that's strong. Push for 60^ on a round trip, and then try the same route with no pedaling and let us know the difference.

I'm not sure why JonesCG and Kingfish think regen% isn't meaningful. Sure it can fluctuate wildly early in a ride, especially when starting from an elevated point. Total electric energy sent back to the battery by regen divided by the energy sent from the battery sounds like a meaningful number to me.

pwbset,
If you have a programmable controller, then regen can be tuned pretty broadly via a combination of hi/low in the pgm settings and shunt modification. A shunt mod fools the controller regen current limit just like it fools the regular current limits, and then you use program setting for current limits to get back to where you want. In your case, you'd want to try low first, and if the braking force still tool strong, you'd need to increase shunt resistance instead of the more typical reduction to get higher current.

For those with variable regen,
Is that controller specific, or something we can do with more common controllers. Somehow I missed how to accomplish that, and while I love my soft smooth regen, if I could vary it in a controlled manner at the handlebars that would be even better. I'd love to use the mechanical brakes even less than I do.

John

I'm looking forward to testing what my Siberian Husky can produce

John in CR wrote: Try the same route with no pedaling and let us know the difference.

Click to expand...

Probably not as much difference as you expect

Justin in 2008 thread on Regen wrote:
There is a pretty wide range ... of regen current where half of the original kinetic energy was recovered. I was rather surprised that the result was this high and over such a broad range

Click to expand...

That range is broader now with the improved C charge rates of Lipos.

The regen of interest is potential gravitational energy back into the battery (top to bottom of hill).

John in CR wrote: For those with variable regen, Is that controller specific, or something we can do with more common controllers?

Click to expand...

Certain Kelly controllers have variable regen, not sure if or how well it works in the more common controllers.

kevo said:

The regen of interest is potential gravitational energy back into the battery (top to bottom of hill).

Click to expand...

Start blasting around in the 40mph range and all regen braking is of interest.

John you're absolutely right, all regen is of interest! Both kinetic and gravitational E recapture.

40mph speeds give energy to be recaptured by braking as Justin's kinetic graph showed


Justin's Regen thread (2008 no less!) http://endless-sphere.com/forums/viewtopic.php?f=16&t=9045 is just that (only 40kph). He shows the high efficiency range vs max amps.

Max downhill regen requires slow to avoid wind losses, but fast enough to avoid copper losses.
My Astro gets better regen at 20mph downhill than the typical faster speeds most descend at.

Like ebikes performance design, the design choices of regen can be incredibly varied.
Which keeps it interesting

Kevo,

Thanks for that last post. I didn't realize anyone was running an RC motor with regen. Now I can eliminate the expensive freewheels and more expensive fancy brakes and get on with several more bike-like plans.

John

variable regen is great, With my old lipo pack i was limited to a 2 c charge, like 20 amps...
now i have better batteries, i can allow max motor current back into packs

on my kelly controller i have 3 options for braking
-switch/button regen for a preset % of total controller motor current
- variable with a 5k pot or hall effect
-regen on throttle release ( used for forklifts )


The only thing that has ever annoyed me with my kelly controllers, The software it comes with is limited to % not actual amp settings
but they never ever fail on me so ill keep using them.


regen % is a good thing for my own personal observations.

John in CR wrote: I didn't realize anyone was running an RC motor with regen.

Click to expand...

Sure there are many doing regen w/ RC motors. Although many RC folks at ES go for the very fast speeds.

It gets weirder (as always). By choosing the gearing, the KV of the motor and the battery voltage regen can occur whenever the rider exceeds a threshold speed (including on the flats w/o braking).

Easy to measure that threshold by lifting the rear wheel and measuring the no load speed at max throttle.
Doesn't apply to the speedy folks here, but for efficiency 18 mph is a nice threshold for max regen.

Probably only confirming how weird ones choices can be

that's weird.
sounds like a segway setup

Iv ran regen to engage on a forklift the moment the throttle is released.

I run throttle release regen on my cargo bike and it's a gentle enough force that it works great. The nature of my riding is that very rarely do I hold the throttle at a fixed position for more than a few seconds, so I very rarely experience the one detriment of throttle regen. That's an oscillation between on and off trying to keep the bike at a fixed speed at around 50%.

My best regen record is 8% but i have to say it depend ALOT on how you drive and how efficient is your ebike.. depending on teh tire pressure, tire type, motor winding. I gained twice more regen power just by keeping my tire pressure at the optimum lebel instead of soft for confort...

i'm also apprecaiting the variable regen of my Kelly on my Giant DH comp.

Actually it is using a 5305 on a 22" wheel and 30s 15Ah lipo. The highest regen power i had was 5.6kW durig few second.

What i like is the way i installed the regen throttle grip!.. i have the motor throttle on the right as all ebike are and also have the regen throttle grip installed on the left. So when i want to regen, i just have to twist the regen throttle on the opposite direction compared to the throttle grip.

Regen throttle is nice because it allow you to reduce the speed while going down a hill but to keep a great speed too and not to just stop the ebike.

Kelly have very nice designed options fo rthe regen!

Doc

I must admit, i do like kelly, they are cheap, reliable and easy to set up.



That's a good idea doctorbass, im going to change the direction of my regen throttle, I am going to do it so i push forward!

This reminds me that all my thumb throttles are operated that way (and placed on the left), since my right thumb got sore of operating it the 'correct' way.

Anyone tried the dual drive regen?

cwah said:

Anyone tried the dual drive regen?

Click to expand...

Mine worked for a while, but then the primary controller began to hold the slaved controller high and not allowing it to see –EBS go to ground. So for a short time with exactly the same pair of controllers, I had regen on both wheels. Later I stiffened up the force and it was almost unbearable until I mounted the trailer – in which case there was enough mass to reduce the jarring force. One of the controller went TU (my fault) and I used a spare 12FET to slave, and in the process, lost dual regen. Through conversation with the supplier, we think that a diode in-line on –EBS will prevent the slave (or master) from being hijacked by the other controller. The theory has yet to be tested, although I will do shortly with the month as I culminate the like-to-have mods on the present ebike.

Long-winded, sorry. KF