E.S. bottom bracket drive.

[d8veh]

I've yet to see wh/mi figures, or even watts per continuous speed on the flats for the GNG kit despite asking for it :/
If you could provide some data on that, that would be wonderful.
Maybe post it up on the GNG kit thread too.

I took my GNG bike on my regular hilly 20 mile circuit and went at the same speed as I do on my other bikes. I pedal all the time and my speed spends nost of the time between 15 and 20mph. Surprise, surprise, I used exactly the same amount of power as my other bikes, which are powered by Bafang BPM hub-motors. All my bikes are 36v. The GNG is run at 25 amps and my regular BPM, 30 amps. The consumption for these journeys is approximately 15wH/mile. So, overall efficiencies are about the same. Maybe the situation changes if you drive to the max qll the time, but I don't bekieve that there will be a big difference. Perhaps the GNG is more efficient when climbing and the hub motor wins on the flat. The GNG has so much torque that I rarely change out of top gear except for steep hills, so I probably don't get the maximum efficiency available from it.

 

[deecanio]

If you are happy with the compromise of only being able to contribute meaningful pedal input at low duty cycle.

hi Miles,

this is the bit i can't compute :lol: , as long as the throttle control is good and the motor is happy to play at low duty cycle without burning, i dont see any need to limit the cadence at the crank? you'll be able to emulate the system that matches human cadence with throttle alone but also have a lot of scope for added yeehaah, what am i missing here?

D

 

[Miles]

as long as the throttle control is good and the motor is happy to play at low duty cycle without burning, i dont see any need to limit the cadence at the crank? you'll be able to emulate the system that matches human cadence with throttle alone but also have a lot of scope for added yeehaah, what am i missing here?

Hi D There's the efficiency hit, that's all. I'm not arguing against it.

 

[neptronix]

BTW, if i haven't mentioned this yet.. i can get the MAC versions of these motors. The cost on them is in the low $100 figures.
This would help bring the cost down quite a bit.

If anyone is interested in performance data on these motors, i can provide.

 

[Thud]

this is the bit i can't compute , as long as the throttle control is good and the motor is happy to play at low duty cycle without burning, i dont see any need to limit the cadence at the crank? you'll be able to emulate the system that matches human cadence with throttle alone but also have a lot of scope for added yeehaah, what am i missing here?

I have some experiance w/regard to this.....the varient is I was using a very small outrunner (turnigy 6374) the motor had no issues turning the cranks at 130 rpm.....but it was worthless trying to pedal along with the motor. Any time you would add motor in a lower gear the pedals would just drop out from under your feet & the bike would wheelie away....once re-geared to the max comfortable human cadance, the throttle became a dial to add as much boost as you needed...& the entire gear range came into play....bike still did 25+ mph & became a joy to stand up & pedal on the trail & up any hill we could find.
Hard to put into words, but It is my favorite e-bike drive I have experianced.....a close 2nd was extreamgreenmachines Surly " Pugsly" with a geared front wheel hub drive paired with a rolloff speed hub in the back wheel. It was an awsome bike under pedal & with motor assist (& I hate front wheel drive bikes! :lol: ) but his set up is perfect for crusing the beach & pedaling along at any speed.
I say match the human....you can allways use road chain rings to get the speed your looking for.

Nep, that Mac motor does not include any planitary reductions IIRC?

 

[crossbreak]

I say match the human....you can allways use road chain rings to get the speed your looking for.

I want to do 6:1 with decent power (I only want 35mph @ 80 cadence) , so I need 78/13 gears. even if the 78T is available it will almost touch ground This is obviously NOT the case.

The rohloff's price is about what I will spend on a whole bike without battery :lol:

The 6374 motor has the power which would be perfect for a bike if reduced right. A 6374 is actually more like 90mm + reduction gear. It will need a bevel gear set or a 200mm bottom bracket we would not talk about this if it was so simple. The bevel gear is sophisticated but expensive.

What we need is a NEW design. The GNG is not sophisticated AT ALL. That's why I do research on this one, it's not perfectly sophisticated like the rohloff (which you can buy with 2000W+ drive here in germany for more than 2 years yet, only 6500€ at my local store). Plz thud, give us some thoughts about it, your are the one with most experience on gearboxes on the sphere i know

 

[neptronix]

Nep, that Mac motor does not include any planitary reductions IIRC?

Nope. And they spin at 2000-3000rpm , much like the 2kW MAC motor i ordered.
Peak efficiency is about 85-88% though.

 

[Cyclebutt]

I've yet to see wh/mi figures, or even watts per continuous speed on the flats for the GNG kit despite asking for it :/
If you could provide some data on that, that would be wonderful.
Maybe post it up on the GNG kit thread too.

So far, my data shows a range of 35-36.1w/mile as a single speed (top speed down hill was 27.3mph), testing mostly on flats. 29.9-33.1w/mile as an 8 speed (top speed downhill was 31.7mph), mixed trail and flats. Efficiency of the stock kit is better than expected, IMHO. I'm a BIG guy at 230# and my battery pack is 20# on a Cannondale F5, your results should be better. 8)

 

[neptronix]

I'm surprised at the efficiency.
Too bad we can't buy the motor alone and build a better system around it :]
The small MAC and BMC motors are the next best thing IMHO.

Just thinking - we read that Timma2500's planetary gearboxes failed pretty spectacularly back in 2010-2011, however there was an awful lot of torque on the output shaft, i am sure. He was running a 16:1 and blew it up with a large Turnigy motor - go figure! even 3ft-lb of torque becomes 48ft-lb at the shaft, in a reduction box like that.

I think that running one of the lower reduction factor units would suffice for the first stage of a crank drive reduction system.

Attached is some performance data of the M12500-2 ( small 1.5KG MAC motor ) producing around 600 watts on 48v.
At 609 watts, it is producing 1.87-nm at 3106rpm on 48v.

http://banebots.com/pc/P80K-S4/P80K-33-0005

Now run that motor through the 9:1 P80 gearbox..

3106rpm / 9 = 345rpm.

9 * 1.87 n-m = 16.83nm ( 12 ft-lbs )
Remember, the P80 gearbox is limited to 85 FT-LB. Yeah, i think it should survive! :lol:

Then you only need a 3.45:1 ratio to properly gear it down for the chainring :]. You can then do a 15T BMX freewheel to a 50T chainring for about 105rpm at the cranks.
Or go with a 13T micro freewheel to a 45T crank for a more normal look.


Okay, let's talk about the slightly larger 3kg? 2.5kg? MAC motor;
It makes 3.7NM at 2609rpm at 1011w / 48v.

Run that through the 9:1 and you get 286rpm and 33.3 nm ( 24 ft-lb )

I think the P80 gearbox can be used as a simple and clean gear reduction for a crank drive - what do you think?

Another avenue is to use the P60 gearbox for the smaller 600W MAC motor.

16:1 P60 would make that 3160rpm / 1.87nm into 197.5rpm / 29.92nm ( 22ft-lb )
This might be pushing the P60 a little because the stated limit is 35ft-lb. You could gimp the motor down to about 500w by reducing the amps though and have an amazing lightweight setup.

I think the secret to getting them to survive is limiting the phase amp to battery amp ratio so that low end torque doesn't shock the gears too bad from a stall / low rpm, producing a flatter torque curve in the motor.

This is how i have got the plastic gears in my MAC motor to last so long, despite running over 2000w into it for long periods of time.

 

[neptronix]

While i'm on a roll..

http://www.ffrtrikes.com/Micro-Stainless-Steel-Freewheel-Adapter-p/11ssmicro.htm

FFR trikes sells this micro freewheel adapter that hooks up to a 1/2 shaft ( hey how convenient, that's exactly what comes out of the P60 and P80 gearboxes :lol: )

http://www.ffrtrikes.com/Odyssey-13T-Freewheel-p/13todyssey.htm

And here's a 13T freewheel from Odyssey

ACS makes a 13T freewheel but i'm sure it's for the smaller thread as well.

And that pain in the ass 17 shaft coming off the larger MAC motor? yeah, staton makes a freewheel thread adapter for that.

http://www.staton-inc.com/store/pro...n_Adaptor_for_freewheel_sprocket-1308-50.html

I don't know how you would get the power into the P60/P80 gearbox from the MAC motors though. Not sure what the input shaft size is on those.

 

[Cyclebutt]

@neptronics: You wouldn't happen to have some good links (just bookmarked all your links above, thanks!) for free wheel crank sets, would you? I see a reduction hub mid drive in my near future, want a good quality (and/or low cost) free wheel crankset for it. Thanks!

 

[crossbreak]

You can get 14T BMX freewheel inluding adapter from cyclone for a reasonable price. You can also use a 13T freewheel with the same adapter.

Edit: got one from here: http://eclipsebikes.com/tooth-cyclone-dicta-motor-freewheel-p-1001.html

 

[spinningmagnets]

Non hub resources thread (stickied)
http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=7192

24-page "freewheel crank resources" thread (linked in the above thread, for when you want to find it again in the future)
http://www.endless-sphere.com/forums/viewtopic.php?f=28&t=7641ZA

I haven't read any feedback yet on the flanged ACS Cossfire FW, but it looks like it's worth trying (five of the peripheral holes are threaded, and five of them are through-holes), believed to be stronger than the cheaper Dicta, costs less than the best-one (ENO):
http://www.ebay.com/itm/ACS-Crossfi...Cycling_Parts_Accessories&hash=item4abe8ce650

 

[neptronix]

@neptronics: You wouldn't happen to have some good links (just bookmarked all your links above, thanks!) for free wheel crank sets, would you? I see a reduction hub mid drive in my near future, want a good quality (and/or low cost) free wheel crankset for it. Thanks!

Sorry, i don't. I would prefer not to use one.

 

[crossbreak]

It's the low end torque that kills these gears / chain and belts. I bet the standard controller isn't tuned for a proper battery to phase amp ratio. I'll share you a little trick that saved the planetary reduction gears in hub motors before we had those fancy composite gears ..

If you have a lyen/infineon controller, the default phase amp to battery ratio is something like 3:1, or 60 amp phase to 20 amp battery, for example.

This allows you to put out really fantastic low end torque from a stall, but it does produce a very large peak of torque, as you know electric motors are characteristically capable of producing.

If you dial back the phase amp to battery amp ratio to say, 2.5:1, the low end torque will be much less intense, but the mid-high torque will be just the same. Venture into 2:1 ratios and you will see in some motors - torque actually being really low from 0rpm, and gradually ramping up, much like an internal combustion engine where all your torque is made near the RPM limit of the motor.

Give it a shot guys. I bet your springs and belts will last longer if you do this. It will be less exciting, it will be harder to do a wheelie, but the kit should last a lot longer.

Thanks a lot for sharing this. Already thought about if this was possible. It might be the holy grail of middrives. Hope I can push my tiny Bafang 250W to 1200W this way

 

[Miles]

Could you take a look at this recent application, crossbreak?

After a quick look I can't see any innovatory step.

 

[crossbreak]

This is just the common electric drill drive. The converted hubmotor we already discussed in this thread but with 2-stage reduction.

claimed advantages are:
-it's a middrive with all benefits (and short comings)
-the reduction is sealed
-it looks better than the GNG

 

[spinningmagnets]

Choosing the motor is key to the success of an "open source" BB-drive. The shape of the motor completely controls the options concerning the shape and mounting of the bracket.

If someone were happy with 1000W, its almost not worth the effort to make a new kit, but rather to simply add a few upgrades to the GNG (in much the same way the AFT is a Cyclone kit with a more robust bracket). The problem seemed to be a slight flex in the modest bracket when using 72V X 30A = 2100W, plus the primary belts dying an early death at 2100W, and an inability to get much more gear-down than the stock 21:1 (using off-the-shelf parts).

Benefits we want to keep: motor weight is low and centrally located, allows rear suspension when off-road (High-watt rear hubs have a lot of unsprung weight). Giving the motor some gears allows awesome slow speed crawling on steep uphills (whether a road-commute or off-road) without high motor/controller heat.

After much thought, I keep coming back to the cell_man geared MAC motor. My only remaining concern is the width of the axles after trimming to the minimum width. I'd suggest experiments using the slow 12-turn motor. If a 12T ends up too slow for a particular builder, the drive cog is easily enlarged (getting a drive cog as small as possible seems to be the constant concern). For those who wish to keep the smallest drive-cog, cell_man has faster windings, 10T, 8T, 7T, and 6T...lots of options.

Possible arguments against this:

A: geared hubs at 2100+ watts don't shed heat well.

By being able to downshift and keep the RPMs up, we won't be making excess heat in the first place. Also, the oil-cooling/air-cooling thread showed there is significant potential to add active cooling for those who want even more than 2100W.

B: The GNG was quieter due to a belted primary, the MAC gears are known to be loud.

When the hub is in the wheel, it has been reported by many that under hard acceleration from low speeds, it growls until you get near top speed. If we keep the RPMs up by down-shifting, the hub gear-noise should be near the same as when a common wheel-hub is in cruise.

C: "X"...fill in the blank

Here are my inspirations, bzhwindtalker and whiplash (take note the top one is using the larger diameter DD-hubs).

 

[Cyclebutt]

Here are my inspirations, bzhwindtalker and whiplash (take note the top one is using the larger diameter DD-hubs).

!00% with you, Spinningmagnets. These guys (bzhwindtalker and whiplash) are going the direction I've been leaning toward, watching and learning about. BTW, sickbikeparts.com has the freewheel crank set to go with their White bros. freewheel. All someone needs to do is market the BB mounted brackets, and a universal kit is born. Frame/bike recommendation will be in order. It gets kind of old (and expensive) replacing perfectly good components to make a kit work on your bike.

 

[animal]

+1 on the mac

 

[Whiplash]

If I can get all my MANY projects in order I plan to offer bracket only and complete kits like mine with a few small modifications to allow some chain adjustment and easier install. I need only to find the time to design the small changes and get the brackets drawn up so they can be laser cut. I'm planning on going thinner and using chro-moly to add rigidity and save some weight. I'll keep you guys posted. I'm almost done with my recent Heads/Cam/Intake project on my 91LX Mustang daily driven soon to be 11 second street car (with any luck!). Once I get that done I'll get things buttoned up...

 

[bandaro]

Hmmm, I see a very strong point for the hub motor drives, they are very simple, reliable, and easy to use systems that run controllers and complementary right off the shelf. However, they do have a significant weight to them, that a brushless motor does not have, even with the reduction. I think with some testing and engineering it is possible to get a brushless system running just as well. I do not have the engineering to get such a design first time, but can make it easy to use and look pretty, that sort of stuff, also make and test it.

For the record, what is the weight and size of a geared and dd hubs that people are using?

 

[snath]

Here's a 12T MAC, that I got from Ilia married, to a crank/freewheel/BB/chainrings that I got from Sickbike Parts and, ultimately, to a Nuvinci 360.

I think the motor weighs about 6lbs. The bike, with the Nuvinci, 10s 6p lipo, the 1.5in. square tubing battery frame, and the battery box weighs about 75lbs.

I just finished it and I don't have many miles on it yet; but so far, I'm pleased.

Gearing is as follows:

Motor= 24t
Outside crank= 44t Constrained by clearance to battery box (probably should reduce that to ~36t anyway)
Inside crank= 30t Constrained by clearance on OD of motor.
Nuvinci= 24t

The fastest I've had it to is about 15mph. I think I calculated it to 20 mph. I will probably go to a smaller cog on the Nuvinci and experiment with 30t and 36t on the outside crank gear. Low gearing is pretty much useless currently but it will climb anything.


View attachment 1

I like the concept well enough that I will pull the BMC 2T out of the back wheel of my coffee roasting cart when Green Market season is over in a couple of weeks and convert it to a similar setup. I could use more low end on Atlanta hills for that thing. Fully loaded and with me on it, it weighs almost 500 lbs.


Probably the world's only coffee roasting cart; complete with 20 note street organ and animated monkey ("cranks" the organ and tips his hat). Runs on the same battery that I use on the bike; 8 motors: two blower motors and two stirring motors on the roaster, a tiny gear motor for the monkey's left (cranking) hand, a RC servo motor in his right elbow, a blower to operate the organ (music is MIDI on an SD card), and the BMC in the rear wheel. The battery will operate everything for the 4 hour duration of the market and make the 6 mile round trip (me pedaling) with no problems.

 

[spinningmagnets]

snath! its been a year since you last posted, I'm glad you are keeping busy. I always liked your work in the past.

Just for the record, could you please list the tooth-counts on the hub-cog, the two chainrings, and the NuVinci sprocket? Also the top speed at 10S (26-inch tires?)?

I don't want to imply that I feel we must all agree on a single open-source ES BB-drive. there can be several, especially since the common frames might have enough variation to require a variety of bracket shapes. The MAC/Ariens lawnmower motors would definitely require a different bracket compared to the MAC. It is thinner, but may require an external belted primary, and works best @1600-RPM max.

 

[crossbreak]

A: geared hubs at 2100+ watts don't shed heat well.

By being able to downshift and keep the RPMs up, we won't be making excess heat in the first place. Also, the oil-cooling/air-cooling thread showed there is significant potential to add active cooling for those who want even more than 2100W.

Yeah, well, that's just, like, your opinion. If we want power, there will be heat. A hub has an efficiency of ~80% if geared right, if you push in 2000Watts, 400Watts of that will come out as heat. No gearing can prefent this, it can just offset the point where efficiency drops below 70% and heat gets excessive.

BTW first test results show that my bafang converted (with passive cooling heat bridge) can get rid off about 400Watts continuous, while staying below 60°C, using a strong fan blowing on it. Unmodified I could not push in 150Watts without overheating it, even with the strong fan. I just push current into 2 of the phases (while it's standing still, making sure it has an efficiency of 0%), but IMO that should not make a difference.
Have no yet no good thermo probe. I will open a new thread with my test results. Plz feel free to benchmark your MAC motors as well!

Started a new thread for these tests here: http://endless-sphere.com/forums/viewtopic.php?f=30&t=45871