Who successully runs ~10kw through a derailleur?

[lostrack]

...In need of replies, or some help on the search function!

After an age typing "3220 derailleur", "3220 shimano" etc. etc. I haven't found much on this. Any clues??


Reason : I took the beast (3220 direct drive) out for a first spin, max speed 30mph(ish) realised on a (comically big) 80T chainring, I can pedal just as fast in lycra (albeit with more effort & less EV Grins...maybe a smirk from passers by...)

Does anyone run high powered, reliable, derailleur based setups? It would fix the no gears/drivetrain headache that I've been having... save a bit of weight, "tiptronic" gears...


Motomoto currently does it on a 3210, but Mr Recumpence has reservations on drivetrain setups on 3220s and is Mr experience in this area. Just wondered if it's currently working for anyone that's not spoken up yet?

thx


*note I have tried hub gears...Shimano Alfine - even on human power (when I really race it) it goes into neutral mode to protect the gearing. Nuvincis I heard snap at around 7KW.... Rohloff too expensive to take a punt on...

 

[Warren]

Cedric Lynch rode many thousands of miles on a six speed derailleur setup with an Agni. But he was only running 1.5kW through it. When he bumped the power to 9 kW, he went to direct drive. Optibike is now running 1.1kW through their Rohloff hubs.

I think a 3-4 speed derailleur setup could be built, using wider chain and sprockets, that would handle 10 kW.

 

[motomoto]

The limiting factor is the number of teeth engaged on the derailleur sprocket and how hard you can pull
on the sprocket without the chain jumping a tooth. You could probably run 10K through a 36 tooth
rear sprocket, but as the number of teeth go down there is more of a chance of the chain jumping.
A stronger derailleur spring would help. Fresh sprockets would help. Having the derailleur wrap
the chain around the sprockets farther would help. Gearing it so that only the larger sprockets are
used would help. If the sprocket cluster could be made from between 20 to 50 teeth instead of 11 to 36,
more power could be run to the rear wheel.

 

[flathill]

If you go easy from a standing start.....

Hi,

Thank you for the advise Gary!
The engineer side of me struggles to see how an 8spd hub could could come near the torque holding ability of a 3spd hub, assuming both designs try to make use of available space in the hub for he gearing. But, if you say it's the toughest, I will get it, but I think for my gearing needs, 3 spds would be enough for me. I do want to gear for a 100mph top speed in 3rd, but I'm pretty sure that will still leave me with adequate starting torque in 1st using a pair of monster motors setup in WYE.

The 8-speed hubs are about twice the diameter, so from what I understand, the parts are beefier, even with 8 speeds.

From an earlier post by Gary:

I'm sure the Nexus hubs can handle the power. There's a guy with a bike shop down in San Diego (Rusty Spokes, in Pacific Beach...) who custom builds chopper-style motor bikes, using Etek and Mars motors. He uses these same Nexus 3-speed hubs with these beasts, and they have worked flawlessly, even pumping 15 hp through them. Instead of chain, hoever, his bikes are belt driven, using regular toothless v-belts. There's a little bit of "slippage" when starting out, which is like a clutch. Works great, he says, and these things can hit close to 60 in just 2nd gear. Anyway, he's never had a single problem with any of the Nexus hubs.

Response by Mitch:

I emailed him, quoting Gary above. The following are my questions with his replies. He didn't actually answer my question but it sounds like he has used 7 speeds without problems and believes the 8 speeds are just as strong:

Do you know if the Nexus 8 speeds are strong enough to handle 2kw or 4kw peak?

more like 68.8 mph in 1st gear.

Thanks for taking the time to answer but I'm not sure I understand your reply. Does
"more like 68.8 mph in 1st gear" mean the 8 speed is strong enough?

i have only tried 3s and 7s , but im sure it is the same...

If you prefer a 3 speed you could try one first, might be equally strong. OTH:

Yes but:

I know a guy who builds custome electric chopper-style bikes, using Mars and Etex setups, and he saythe Nexus-3 is all he'll use. I should point out, however, that he uses V-belt drives, to the rear, and that provides a bit of natural slippage, like a clutch. Even still, though , these beasts have crazy power, and he's not had a single problem with the Nexus-3 hub.

[my underlining]

 

[neptronix]

Interesting stuff.

I was at optibike's HQ today and was told that the 1.1kW motor can dump out a peak of about 3kW from a standstill. That's a pretty major torque load. In 3rd gear, the optibike felt like it was going to do a power wheelie like my golden motor setup.

I think they are using the rohloff speed hub.

10kW on the other hand..

 

[flyinmonkie]

I haven't hit 10kw, but I have seen peaks of 5kw through a dérailleur with no issues. I only have 3 sprockets for the motor to rive and have limited the dérailleur so I can not shift off either end.

It seems to me that chain skip would be the issue. The dérailleur should not see any of the torque, especially if you let off to shift (which just seems like a good idea no matter what you are running in my mind). I don't get the obsession to shift under power.

Clay

 

[Miles]

I was at optibike's HQ today and was told that the 1.1kW motor can dump out a peak of about 3kW from a standstill.

Whatever that means........ :lol:

 

[neptronix]

I was at optibike's HQ today and was told that the 1.1kW motor can dump out a peak of about 3kW from a standstill.

Whatever that means........ :lol:

Erm.. mean's what i said I visited Optibike's engineering center and showroom today, and that's exactly what they told me.

 

[Miles]

Perhaps you should have asked them why they were talking BS?

 

[neptronix]

Eh, they were being nice, showing me around the production and engineering areas, didn't find a need to start a fight. Maybe it's a fib; perhaps the controller just has a hard time limiting the current from the get go; definitely felt like it, as at higher gears, it would produce a gnarly jolt of torque, sorta like my MAC motor on high voltage on the EB2 controller, which had the throttle response of an on/off switch..

 

[Miles]

Must have been interesting. Did you take any pics?

 

[lostrack]

The limiting factor is the number of teeth engaged on the derailleur gears and how hard you can pull
on the sprockets without the chain jumping a tooth. You could probably run 10K through a 36 tooth
rear sprocket but as the number of teeth get less and less the more chance of the chain jumping. It
is just physics. A stronger derailleur spring would help. Fresh sprockets would help. Having the derailleur
wrap the chain around the sprockets farther would help. Most importantly, gearing that causes you
to use only the sprockets with most teeth. If the cluster could be made from 20 teeth to 50 teeth
instead of 11t to 36t, more power could be run to the rear wheel.

Yes, we'll made point. I'd like to agree that the defining factor here is chain tension completely.
I have a dynamic chain setup on the direct drive that provides a small amount of tension. It pulls tight under suspension compression but not tight enough under drive. This is on an 80 tooth!!

Under power, it skips very easily, and it's the skipping that snapped the chain (it's rated for a ton, but continuous power)
More tension will fix the problem, although it starts to become a trade off when there is so much tension it's causing lots of drag...

I have written to a spring manufacturer, by good fortune the MD is a fellow cyclist - so I popped e spring in the post and they are making me an updated spring that will slot back in and provide much more tension. Hope it works!

 

[neptronix]

Must have been interesting. Did you take any pics?

No :/.. i could though, as i may move to the area later. They had a lot of their parts on display in the front.

They have a really nice building downtown, right next to a shop that converts Priuses to plugins.. right next to a shop that does EV conversions. Never seen anything like that in America, kinda blew my mind.

I usually bash on prebuilt eBikes here, but the construction of this bike is just slick enough to earn my respect even though these things are stupidly expensive. 3-5 pound mid-drive motor, little controller about the size of a 9FET but totally custom, battery packs made of some secret formula battery with a custom BMS that had way better size/weight than my RC Lipo ( GRRR!!! ), and an aluminum frame that is amazingly light that hides all the wiring.

I'm a little shocked that the guy was willing to show me almost everything in the building, including the areas where they dyno test the bikes, etc. I guess that's what you get for coming in their and knowing some shit? eBike geeks always love to talk shop..

( then proceeded to talk to the prius guy next door for over an hour, he had an eLithion BMS hooked up to a 80-some odd cell lithium pack, he had a laptop with a friggin' unix console hooked up to the BMS showing the voltage of all the cells, delta voltage, max and minimum cell voltage, all sorts of data.. BTW, eLithion is based out of here as well. )

 

[Tom Tom]

Interesting stuff.

I was at optibike's HQ today and was told that the 1.1kW motor can dump out a peak of about 3kW from a standstill.
.

Thats interesting.

So their "1.1kw" motor can produce nearly 3kw all day long (ok, maybe not all day)
just like other hub motors(9c,mac,etc.). Seems their motor is conservatively "rated".........hmmmm

Explains how their "top" rider(who is an mtb racer btw) could "pedal" to 38mph near the top of pikes peak.

 

[pdf]

The limiting factor is the number of teeth engaged on the derailleur sprocket and how hard you can pull
on the sprocket without the chain jumping a tooth. You could probably run 10K through a 36 tooth
rear sprocket, but as the number of teeth go down there is more of a chance of the chain jumping.
A stronger derailleur spring would help. Fresh sprockets would help. Having the derailleur wrap
the chain around the sprockets farther would help. Gearing it so that only the larger sprockets are
used would help. If the sprocket cluster could be made from between 20 to 50 teeth instead of 11 to 36,
more power could be run to the rear wheel.

I would like to second all these points and am glad to see someone state these observations. Even at relatively low power, pulling on the small sprocket in the back can cause the chain to pull out of the sprocket, even a fresh sprocket. With a little age on the rear sprocket, I have to be careful to downshift before I stop. I replaced my rear sprocket and chain at around 4000 miles because it was jumping pretty badly when even close to cross-shifted.

I can get a lot of chain jumping even with 500kW if I am in the smaller sprockets but cross shifted into the smaller rings in front (this would be from a stop). Manual bike people obsess about bent derailleurs if you talk about chain jump; they assume the derailleur causing the chain to shift improperly. and that it is not actually jumping out of the teeth. But few people can put the amount of torque into a rear sprocket that a motor can. I am very glad to hear someone comment that a stronger derailleur spring would help; I've thought this many times but whenever I've mentioned it to someone else, they discount it. I'd have thought perhaps the long bike community would talk about this due to chain sag, but I've yet to hear it brought up.

I notice a lot more chain jumping when the weather is wet; this is another issue that makes people assume I am off my rocker, but it is very significant. The only explanation I can come up with is that the oil-water emulsion on a wet chain makes is more slippery so it jumps easier. Can anyone confirm this?

Finally, since the chain is pulling across the top run, I don't think a derailleur set-up (on the return leg) is intrinsically failure prone at high power. Obviously, you have to power off before you shift or you will bend the heck out of both a front or rear derailleur. Otherwise, I don't see a problem, except that a fixie can't chain jump. Chain suck might be a problem at higher speeds, although I've never had a problem with it. I've heard mtn bike people talk about it but never experienced it myself.

 

[fizzit]

I have not put 10kw through my derailleur but I would say that I have definitely applied 5, and it has never slipped.

If you go easy from a standing start.....
internal gear hubs

Wow that is really cool! I thought I was good with my derailleur but now I am looking at Shimano 3 speed hubs. Does anyone know if these are better than derailleurs at shifting under power ? Do they have any components inside that would slip from the high torque of an RC motor?

 

[lostrack]

I have not put 10kw through my derailleur but I would say that I have definitely applied 5, and it has never slipped.

If you go easy from a standing start.....
internal gear hubs

Wow that is really cool! I thought I was good with my derailleur but now I am looking at Shimano 3 speed hubs. Does anyone know if these are better than derailleurs at shifting under power ? Do they have any components inside that would slip from the high torque of an RC motor?

Yes - forget 8spd Alfine hubs. They have a protection system in which shifts to neutral if you pull to hard. Bear in mind this happens when you are pedalling as hard as you can, and >90% of the time ends in a colossal FAIL, facegrind on tarmac. I hate them for this. Don't even think of doing motored power. (can you tell this has happened more than once?? )

The 3 speed nexuses apparently don't suffer from this, which makes them cool.

I have a 12mm hardcore rear axle though, so they don't make hubs to fit that.

 

[fizzit]

Yeah, I would think that 3-speed would be much more robust than 8-speed. That's a bummer about the axle.
I also just saw that SRAM makes a 3-speed hub which is very similar, but can use 6-bolt discs and is a little cheaper. I don't see anything about the torque ratings for either of these hubs though

 

[Thud]

the history on internal 3-speeds is bad.
they have all failed so far as i know.
Matt in aus. (1000w) fail..
Ausie Jester.....Fail.

still havn't heard of any one useing a Sram dual drive...(they look tougher to me)

 

[fizzit]

Hey thud, do you think that SRAM's 3-speed:

is closer to the Shimano 3-speed or the SRAM DualDrive in quality/torque handling?

 

[flathill]

Another
endless-sphere.com/forums/viewtopic.php?f=28&t=12267&hilit=nexus+3220

I think having it adjusted proper is key
And no full throttle shifts...
I'd like to see a 3220 put 10kw to the ground
I'm guessing 10kw in 7.5kw out once it heats up
The efficiency rating is at low power short term
I think

 

[flyinmonkie]

Pretty sure Deec killed a sram with an astro trying to shift under power. Again, I ask what is the fascination with this? You wouldn't try to shift your car without letting up on the throttle. I let up a bit to properly shift a pedal bike with no motor on it. Otherwise you shock the drive train and the chain skips.

If you can shift without power and ease the throttle back on till the slack is taken up, then you can probably put a lot of power through most stuff. If you are really worried about shock loading your drive train, then check out Matt's (recumpence) torque limiter. I'd like to see a sram or SA 3 speed run with one in the system.

With standard gears if the chain is skipping, then the sprockets are worn or there isn't enough wrap/tension. Or, it is just too much for standard bike stuff. Everything has a limit.

Clay

 

[fizzit]

Pretty sure Deec killed a sram with an astro trying to shift under power. Again, I ask what is the fascination with this? You wouldn't try to shift your car without letting up on the throttle. I let up a bit to properly shift a pedal bike with no motor on it. Otherwise you shock the drive train and the chain skips.

If you can shift without power and ease the throttle back on till the slack is taken up, then you can probably put a lot of power through most stuff. If you are really worried about shock loading your drive train, then check out Matt's (recumpence) torque limiter. I'd like to see a sram or SA 3 speed run with one in the system.

With standard gears if the chain is skipping, then the sprockets are worn or there isn't enough wrap/tension. Or, it is just too much for standard bike stuff. Everything has a limit.

Clay

This is true, shifting under power is a good way to quickly kill a gearbox, but from looking back at older threads it seems that some of these transmissions have been killed just from the torque of the motor alone. It makes sense that a torque limiter would save a 3-speed from certain death.

 

[amberwolf]

usually it is not the torque directly, but the shock load partly due to the lash in the gearing. what usually happens from the ones people have dissected after destruction is something like this:

bike is stopped, left in lowest gear.
throttle is slapped hard to start, motor pulls as hard as it can really suddenly.
chain or belt drive to the igh tenses up, pulls hard.
igh gearing is driven, and usually was in a relaxed state, with perhaps some small gap in the teeth. the teeth are slammed together and the full force of that is transferred to the root of the teeth, which often are cut at a sharp angle, and provide a perfect stress riser to shear off at.

any kind of "cush" drive, where a rubber pad or something that provides even very minimal shock absorption between the drive input to the igh adn the igh's gearing, would probably prevent this.

ramping up the motor input over a second or less would also probably prevent it.

pre-applying throttle just a touch to start the forward motion of the drivetrain just before applying real power would also probably prevent it.


one of the most common lash problems is likely when sitting at intersections and stops, becaues here in the usa, even here in flat phoenix, they seem to always try to put the center of the intersection uphill from all incoming traffic. thus when you ahve to stop you are always going to "roll back downhill" behind you if you don't hold yoru brake, though it's often not steep enough to actually roll back very far, it is noticeable to me on my crazybike2, and i do have to hold my brake, since at a stop my feet are on the ground not the pedals.

if this were to happen with a freewheeling igh, the internal gearing could "unmesh" just a little bit so it has just that bit of lash-gap to then slam into the teeth hard when power is suddenly reapplied.


the problem could be somethign else entirely, but the above is what i've gathered from the threads about it so far.

 

[flathill]

There is much truth to wolfamber
http://motoczysz.com/
These guys apply power to the motor to take up the slack but not enough to move the bike

You wanna see something real cool see wrightspeeds syncroless gearbox shift instantlyno by syncing electro
I think lebowski could pull this off with his new controller brain