Astro 3220 / NuVinci build

[wojtek]

hello

i am about to build the Astro 3220 / Nuvinci drive and was wondering if anyone has experience with such setup? Im thinking to use NuVinci auto option rather than manual.. I am curios about the shift map and other software setup entries for this application..

any comments are highly appreciated and i will be updating this thread along with the progress

Wojtek

[Green Machine]

HI Wojtek,

I put some research into that drive system planning to do a 3220 set up and talked to a friend of mine who actually tested the nuvinci for nuvinci with electric motorcycles.

As far as the old nuninci goes (the one that weighs 10 pounds) it will definitely handlle the power that a 3220 could dish out and then some. My friend would power etecs at 72 volts through it with no problems.


However the new nuvince that just came out is much lighter (around 5 pounds) is more expensive, and has less balls and is not as robust. So it is unclear whether or not it will take the power. I would be really interested if anyone has tested this new nuvince with an electric motor.

The good news is you can find online some of the old heavy nuvincis for pretty cheap. I picked up one for $150.

You do have the issue that the nuvince is not 100% efficient and you lose 5 percent or so of your energy. I dont know how big of a factor that is. My thought was that would be big deal for pedal cyclist but not such big deal for ebiker.

Good luck...keep us posted.

[wojtek]

yes i also got the developer's kit for 149USD

the reason i went for NuVinci was that i have a lot of hills to climb. Using HV 160 sensorLESS controller, i noticed that the drive without pedalling help [i am pedalling to charge my battery directly not as an additional drive] struggled to go uphill from standstill.
Also the high torque / top speed is limited when using just one 60T sprocket.

I am hoping that Astro will have more chance to use its potential with help of NuVinci.

So efficiency as such does not bother me as much - if i lose 5% of efficiency but have more torque available when i need it most and can ease off load on controller / motor, then im more than happy.

[Miles]

So efficiency as such does not bother me as much - if i lose 5% of efficiency.....

More like 10% to 25%, I think...

[Green Machine]

So efficiency as such does not bother me as much - if i lose 5% of efficiency.....

More like 10% to 25%, I think...

Yes thats right...i got the 5% figure from the Rohloff, getting the 2 confused. You do lose more efficiency with the Nuvinci....although i am not sure if it was ever tested how much.

Miles do you know a source of data where someone tested efficiency of Nuvinci?

Early promotional material form nuvinci reported 97% efficiency. Everyone knows that's not true. But whats the real number? Also is the older drives with more balls more or less efficient to the newer and lighter system?

Wojtek...which nuvinci did you get in your developers kit? The heavier one?

[Miles]

Miles do you know a source of data where someone tested efficiency of Nuvinci?

No, I'm afraid not.

[wojtek]

i got n171, asaik they only have this one on offer atm

[MattyCiii]

The N171 is the same as the original, except the "lightening holes" they placed on the spoke hole flange that took of a couple of grams from the 5kg weight.

[Ypedal]

it's def more than 5%... i run a 3210 with the old style nuvi and at 4kw it gets warm after a run..... i have to figure out a way to put a number on it somehow..

[amberwolf]

Need someone with a dyno to test it out. Test with and without it in the driveline, possibly using different drive/receiver sprockets on an ohterwise-identical regular wheel to emulate the ratios at which the NV itself is tested at.

Could also estimate without a dyno by running some watts-at-speed runs without the NV using different sprockets for differnet ratios, and then do the same with the NV, again in an otherwise-identical regular wheel.

Better yet, use a locking pin setup to lock the NV's input and output sides (by locking the input sprocket to the flanges?) so that you still have the same weight of wheel/etc., but are not actually sending power thru the NV internals. This method would be usable wiht or without a dyno.


I've forgotten who it was (kzolt?) that ran twin motors thru the N171 and was getting an estimated 1/3 power loss thru it, at over 1kw input, IIRC.

[MattyCiii]

Unless losses go up greatly with power/speed, 1/3 is way too high an estimate.

I ride a pedal bike with the NuVinci as my "daily driver". I'm commuting, so distance/hills/traffic lights have far more impact on average speed than acceleration and top speed. Commute is 15 miles round trip each day, so I experience whatever the loss is for at least 300 miles a month. It's worth it to have "rinse it off occasionally" maintenance plan, 360% gear range and the ability to smoothly ease the ratio down as I ride uphill, trading momentum for ratio & pedaling at constant force and velocity. I just turned 44, I can't be mashing the gears anymore.

I know viscerally it's less efficient than a traditional bike drivetrain, and heavier... I did a side-by-side comparison between two otherwise twin bikes before I took the 8 speed cluster and derailleur off bike #2 and fit that with a NuVinci. But 1/3? Nah.

[E-racer]

I'm running the dev kit on my bike w/ 50v 100amp controller. There is no way 1/3 of the power is being lost through the cvt. It simply doesn't get near hot enough to be talking that kind of numbers. The general consensus from this thread http://endless-sphere.com/forums/viewtopic.php?f=28&t=35980 is that the cvt is worst case 70% efficient at the extreme bottom and top ratio. It approaches 90+% at 1:1. Personally I think it is more than 95% efficient at 1:1. To combat the inefficiency at the extremes my bike starts in shift position 100 and finishes up shifting at 800 (as opposed to 23 and 977). For a commuter in a hilly area nothing is going to beat a Nuvinci CVT you always have the optimal gear ratio. For an all out race bike where the plan is to sustain speed a fixed gear will always be better.

[wojtek]

Isn't it like saying that without the "gearbox" like NuVinci, you have something like a car that can only have 1 gear? so what that it may be more "efficient" if you have no flexibility? you either have eg 1st gear which is great for start and going very steep hills but nothing else... or 6th gear to go really fast but you cannot start.

[Miles]

Isn't it like saying that without the "gearbox" like NuVinci, you have something like a car that can only have 1 gear? so what that it may be more "efficient" if you have no flexibility? you either have eg 1st gear which is great for start and going very steep hills but nothing else... or 6th gear to go really fast but you cannot start.

It all depends on how much torque capability your motor has. Do you want to slow down on hills?

The weight of the Nu Vinci could be traded for a larger motor.....

[wojtek]

the reason i went for NuVinci was that i have a lot of hills to climb. Using HV 160 sensorLESS controller, i noticed that the drive without pedalling help [i am pedalling to charge my battery directly not as an additional drive] struggled to go uphill from standstill.
Also the high torque / top speed is limited when using just one 60T sprocket.

I am hoping that Astro will have more chance to use its potential with help of NuVinci.

sometimes i have to slow down / stop on hills. Eg got stuck in a traffic jam on a hill and had to start from standstill many time within 1km distance.
In "ideal" world, best would be to use sensored motor with Astro 3220 but it is not possible.

Bigger motor will not solve the start from standstill problem with sensorless controllers.

[Miles]

the reason i went for NuVinci was that i have a lot of hills to climb. Using HV 160 sensorLESS controller, i noticed that the drive without pedalling help [i am pedalling to charge my battery directly not as an additional drive] struggled to go uphill from standstill.

Sorry. I hadn't noted that you don't have direct pedal input....

[MakeTheNoise]

I have been seeking out someone with similar experience with a 3220 and Nuvinci.

I posted on the Nuvinci discusion forum, a 'what-do-you-think' style question and got the following reponse from nuvincisupport.

Link across here:
http://nuvinci.informe.com/forum/bicycl ... -t688.html

[MattyCiii]

Once my 3220 is built by Astro, I plan to run it through an N360. But e-power will be going to the left side, to a chainring mounted to the disc brake mounts. So I won't be sending e-power through the "gearing" - the gearing will be for pedaling only.

[wojtek]

im connecting the drive with NuVinci.. we will see if it takes the power

i expect higher top speed and better torque with such setup

[Idontwanttopedal]

Hi i have got one off these hubs and after riding losttrack bike I like the 3220 which I think would benefit from running gear so can't wait to see how your one work

[Green Machine]

Wojtek

Any updates?

[ions82]

I'm also curious about any updates, as well. I am considering the same setup (3220 & NV) for a future build. I think I read that the NV 360 is supposed to handle 98 ft-lb and be safe up to 1,000 RPM. I'm not 100% sure, but I thought that I remember it having the same specs as the 171. Of course, maybe they were both underrated and just had the same numbers slapped on them. I'm hoping that my next build will be somewhere in the 3-4 Kw range. If there are any updates for this project, please post!

[amberwolf]

Well, Fallbrook themselves recommend the 171 for powered applications as opposed to the 360. I forget where I saw that, though.

[boostjuice]

I'm also curious about any updates, as well. I am considering the same setup (3220 & NV) for a future build. I think I read that the NV 360 is supposed to handle 98 ft-lb and be safe up to 1,000 RPM. I'm not 100% sure, but I thought that I remember it having the same specs as the 171. Of course, maybe they were both underrated and just had the same numbers slapped on them. I'm hoping that my next build will be somewhere in the 3-4 Kw range. If there are any updates for this project, please post!

N171/B features 8 torque transfering planetary spheres and is rated for 65 Nm (48 lb-ft) sustained input torque, 130 Nm (96 lb-ft) instantaneous input torque rating (published Here)

N360 features 6 torque transfering planetary spheres of the same diameter as far as I can tell. Torque specs not published. Reasoning then would suggest that the N360 has at most 3/4 of the N171 torque rating (although likely lower due to the reduced positional radii of the planet spheres and their contact with the input/output discs). BUT....... the built in slippage clutch is designed to slip before the spheres & drive/driven discs to protect them from galling due to breakthrough of the friction fluid. So, we could make an assumption that the clutch slip threshold is proportionally similar, but its still only a guess until someone measures it.

My practical experience with the N360 is that I can get the clutch to slip with heavy pedalling alone in the underdrive ratios quite repeatedly which doesn't give confidence for high electrical power being added. However, being that electric motors apply smooth torque delivery rather than high peak-torque pulsating pedal strokes, there would definately be some a margin for electric assistance, just probably not more than a few hundred Watts. In any case the clutch should protect the drivetrain for the most part, so dont be afraid to find the limits and report back.

BTW though, A friend has a staton-Inc 4-stroke, N171B build pushing ~3kw that he has never had hub slippage problems with, so if your pushing high power id go with that.

[ions82]

BUT....... the built in slippage clutch is designed to slip before the spheres & drive/driven discs to protect them from galling due to breakthrough of the friction fluid. So, we could make an assumption that the clutch slip threshold is proportionally similar, but its still only a guess until someone measures it.

My practical experience with the N360 is that I can get the clutch to slip with heavy pedalling alone in the underdrive ratios quite repeatedly which doesn't give confidence for high electrical power being added. However, being that electric motors apply smooth torque delivery rather than high peak-torque pulsating pedal strokes, there would definately be some a margin for electric assistance, just probably not more than a few hundred Watts. In any case the clutch should protect the drivetrain for the most part, so dont be afraid to find the limits and report back.

BTW though, A friend has a staton-Inc 4-stroke, N171B build pushing ~3kw that he has never had hub slippage problems with, so if your pushing high power id go with that.

GREAT insight. I had read about others that got their NuVinci hubs to slip, but it wasn't specified that it was actually a clutch that was slipping and not the internal drive. If one can get a 360 clutch to slip with just pedal force, I definitely don't think it would be the better candidate for a powered system.