Using RC motors on E-bikes [Archive]

[johnrobholmes]

The face of the motor is stationary. The can of the motor spins, and is where you would have to bolt the chainring.

 

[MitchJi]

The only negative with those motors is the lack of an output shaft. They are designed with a prop adaptor on the end of the can, but not an output shaft.

Matt

Hi Matt,

Thank you very much!

You might have just saved me $275 or $515.

I just assumed every electric motor had an output shaft that spins separately from the motor mount!

 

[MitchJi]

Hi,

Even one of the most powerful bikes on the forum would have much better low speed performance with gears. With two or three speeds it could have great low end performance and great high speed performance.

http://www.endless-sphere.com/forums/viewtopic.php?p=98906#p99074

I know what you mean.

Even though I have my bike setup for 4,000 watts, it is geared for 40mph. So,
the accelleration (though powerful) is not nearly as powerful as it could be.
If I geared this thing for 30mph, it would probably wheelie. :wink:
Actually, when I tighten the clutch to pull 7,000 watts, the bike just starts
to feel truely quick. Tall gearing really hampers accelleration.
Matt

One of the fastest bikes on the forum (top speed) with about 2.5kw has pretty bad low end performance that could be drastically improved with gears. With two or three speeds it could have great low end performance and great high speed performance.

http://endless-sphere.com/forums/viewtopic.php?f=6&t=5168&st=0&sk=t&sd=a&start=165#p94568

You have to bear in mind it doesn't have any real accel at less than 20mph. As it is
now on 24V it has good fast accel between 25 and 38mph. It is a narrow range but it
suits my needs very well. It does require pedalling along with a little effort and
gentle throttle use all the way up to 20mph. Or, you could accel REALLY
slowly - no pedalling required. I can pedal up to around 33mph with sensible
cadence so I will get the fat burning activity I require!

58mph was with a 88 tooth gear and a 10 tooth pinion.

Currently running the 91 tooth along with the 10. Top speed is 38mph according
to GPS - measured over a 1 mile distance.

I am not happy with the performance from start to 20mph - thats a fact. Yesterday
I had the opportunity to use a fast current logger and it showed spike peaks on
the order of 280A. As suspected - this is the reason for the bad low speed
performance. Until the motor gets to 3000rpm its basically 'stalled' and draws huge
current.

What I would really like to do for a general use version would be to gear it
down - say something like 25:1 then bump up the voltage until the current spikes
are just below the Jazz's cut-off point (220A it seems from measurements). Geared
that way I think it might be possible to have a top speed of 40mph or so with
faultless accel.

Thats what I want next - it is very much like my old axi powered scheme which
worked fine but drove the pedals.
Scott

Better performance with about half the power!

http://www.endless-sphere.com/forums/viewtopic.php?f=6&t=3904&p=99074#p98906

One thing I learned with the Mariner is that even though it tops
out at about 2700W, with the extra torque benefit that the gearing provides, it
actually has better performance than the 5303/72V setup it replaced, and that
one hit 5000W peaks.

Optibike is widely considered one of the best ebikes runs through the gears (Derailleurs).

The pi bike climbs 25% grades good performance with a 500 watt motor and the system works very nicely.

http://endless-sphere.com/forums/viewtopic.php?f=6&t=5297#p79741

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...

Anyway, he's never had a single problem with any of the Nexus hubs.

So there are several ways (Derailleurs,Nexus 8 speed and 3 speed) to get excellent performance with a through the gears system. Depending on how its done this can actually be simpler to set up than a left side drive.

A nice setup might be a 3 speed nexus in the rear with a 3 speed Derailleur system on the front crank (like Deecanio's front crank set up).

Given the substantial advantages and the fact that its not very much harder or no harder to set up I don't understand why there is virtually no interest in this option. People are actually discussing adding a whole separate transmission to the left side rather than using the gears that are already in place or that can be easily added by installing a hub gear system.

The only issue I see is that this requires careful planning (Motor rpm/gearing and sprocket sizes) for both pedaling and motor to work well together.

 

[safe]

Gears Rule!

For those of us that actually use multiple gears the choice is a no brainer... gears are simply better. I could go over all the physics (again) about it, but the bottom line is that a motor with multiple gears performs like one that is double the power. It's like a power doubling device for your motor. You get better climbing, higher top speeds, less heat and better range.

The only negative is that they are mostly custom built bikes right now (except for the Optibike) so they simply are not easily available for regular people. Combine an RC motor with your derailler (seemlessly) and you will get an excellent result.

Hopefully the lack of gearing will change in the future.

 

[perbear]

Yes, the link that I posted was to Michael Kutter's patent. I've lost interest in it a bit because the pedal gearing changes so much depending on whether the motor isn't being used, or not.

There's also a similar epicyclic differential drive on the Prius. These things have been around for quite a while...

I like the idea of doing something similar with 2 RC motors

I have been working on such a design for some time. I use it independently of the pedal gear, using a disc brake rear hub and replacing the disc with a syncro pulley (and use rim brakes). The target is to make it fit on most cycles without modding the frame. This design can do very efficent regen into High Current Charge batteries (A123 cells) so the rear wheel can brake quite well without disc

This two-motor/epicyclic gear design may seem rather complex at first, but after some thinking and calculation it seems quite possible to produce. Two motors with integrated epicyclic gear should be achievable at a reasonable cost by clever integration.

My design use a coreless dual-PM disc motor in the center surrounded by an LRK outrunner with 110mm stator. The disc motor is directly connected to the sun gear, while the LRK motor drives the ring gear. Power output goes via the planetary carrier to a final belt gear with suitable ratio to the wheel.

I chose a coreless motor since it has no loss while freewheeling which is important when coasting without electric power.
The coreless motor has four functions;
1) free-wheel for starting since sensorless ESC require some RPMs to sync with motor,
2) free-wheel for coasting,
3) as a CVT to match the cycle speed to the high efficiency power band of the LRK motor and finally
4) to prevent LRK from exceeding maximum RPM during fast downhill coasting.

The LRK motor has been selected since it has quite high torque and low RPM. The LRK motor supplies more than 70% of the total motor torque and handles most of the power. LRK RPM is kept within its high efficiency band as much as possible, while the coreless motor speed is adjusted to match. That way the total system efficiency can be quite high, even with a low efficiency from the small coreless motor since most of the power (and energy consumption) goes into the LRK motor working at max efficiency.

Planetary gear needs strong and sufficient cogs and to get hold of these is probably the most challenging part of the project... Prototypes will likely be made by laminating CNC stamped sheet metal parts, sun wheel by steel milling, and planet wheels (four) in tough fibre reinforced plastic to reduce noise.
I have found some suppliers of planetary gears in India that seems capable of making good and low priced gears in quantities not so large. If the prototypes works well I hope many would be interested so I could order some nice helical cut hardened steel gears as a group by or something.

Electronics are not very special or complex. It consits basically of two ESCs with governor functionality controlled by a small MCU. This MCU use a pre-loaded look-up table to find RPMs for the motors based on wheel speed, throttle and load. In addition a small BMS/regen module is needed for safety, but that can be added later.

If anyone is interested I can post some pictures of my 3D model plus my Excel power calculations. As for the real thing so far I have the LRK stator laminates and it's magnets. And a brand new bike with hydraulic brakes in the garage waiting for the motor

Cheers,
Per

 

[Miles]

Yes, please post the model pics & spreadsheet, Per

I've just been playing with the new epicyclic module in the latest version of MITcalc...

 

[voicecoils]

If anyone is interested I can post some pictures of my 3D model plus my Excel power calculations. As for the real thing so far I have the LRK stator laminates and it's magnets. And a brand new bike with hydraulic brakes in the garage waiting for the motor

Cheers,
Per

Cool! More info and photos would be sweet 8)

 

[Miles]

Welcome to the RC Groups forum, too, Per :wink:

I like the way you solved the regen or freewheel problem......

 

[docnjoj]

Wow Per! Way cool! Welcome! Yes please post (and patent!)
otherDoc

 

[Jason]

Hey guys I'm also trying out the rc outrunner/esc combo on my giant mountain bike. I have a 19/1 reduction through belt drive with a turnigy sk 50-65 380kv bldc outrunner. I also ordered up the Turnigy sentilon 100Ahv ESC with a turnigy aerodrive wattmeter/servo driver. Unfortunately I'm a little disappointed with the quality of the wattmeter and speed controller since both were dead on arrival. I've since fixed the wattmeter which had a bad solder joint and then found that the speed controller seems to be missing a diode. I was hoping to find someone on here that has had this speed controller apart and could snap a highly detailed picture for me so I might be able to see the style of diode that goes in the slot. While I was here I also wanted to throw out this link for a two speed transmission. It's for a large scale buggy and seems to be pretty simple/efficient in design. Hopefully it'll come to market soon.http://www.teamgonads.com/Merchant2/merchant.mvc?Screen=PROD&Product_Code=4500&Category_Code=TeamGoNads&Product_Count=24

 

[johnrobholmes]

Helluva first post Perbear! Welcome, and we would all love to see what you have come up with!

 

[perbear]

Yes, please post the model pics & spreadsheet, Per
quote]

Here is a picture of my model, I still miss the belt pully - have to decide what standard (i.e. cost level) to go for.

Per

 

[docnjoj]

How 'bout them beans? A Sanyo tranny and RC motor combo!

http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=2744

We have truck-loads of these hubs here, many of them just abandoned in the streets, Woo woo! I found an abandoned one just this week.

I never seriously considered using this hub because I didn't want to deal the built-in electronics I knew nothing about.

Hey DJ
Those are just regular brushed motors inside. The design and belts is great! If you ever want to sell one, send me a PM.
otherDoc

 

[Miles]

Here is a picture of my model, I still miss the belt pully .......

Thanks.

Nice

I think this deserves its own thread

 

[perbear]

The 3D model of the "Nohassel" epicyclic motor^2 is not finished (missing locking rings, screws, nuts, washers and spacers and more) but it at least shows the principle. All bearings are 608 - low cost. I will follow up with machining drawings. Weight target is less than 5kg inclusing everything (motor, battery, ESCs, housing, gear, pullys etc). Peak power is several kW but only for regen. Continous output is 250W in Pedelec mode, and 2-4 times more in "scooter" mode.

A new thread would be nice. I would like to post some files (Excel and drawings) but I have not found a file area - do I have to post some more before file area is revealed?

 

[Miles]

I would like to post some files (Excel and drawings) but I have not found a file area - do I have to post some more before file area is revealed?

There's no specific area for files, you simply upload them to a Post using the Upload Attachment section.

 

[swbluto]

Does anybody have a rough idea of what kind of power I could get in a hi-torque outrunner with dimensions equal to or less than 3"x5"(3 inches being the diameter of the "circle")? I'm looking forward to upgrading a really small electric scooter with one of these miraculous motors but I'd like to know what kind of "Power range" is currently available.

 

[recumpence]

Well, an AXI 5345 is good for 4,000 watts continuous and that motor is far smaller than your space requirement.

If you want to spend a fortune, you could look at a big Plettenberg. My Pletti pulls 7,500 watts.

Matt

 

[MitchJi]

Hi,

Hacker has an A80 series that isn't on the U.S. Web Site (bigger than swbluto on is looking for).

I'm not planning on using one but I'm curious, will this drive configuration (Standard 6-bolt prop flange) work for an ebike without jumping through too many hoops?:

Hacker A80 Max 5000 watts - $669.00
RPM/Volt (Kv) 218 or 180
1450g / 3,2pounds Weight
shaft diameter: 8mm
motor diameter: 89mm - 3.5"
motor length: 107mm - 4.2"

More info here (English and German):
http://www.hacker-motor.com/images/Catalog08/Hacker-2008-WEB 21.pdf

 

[fechter]

You should look for one that has the shaft coming out of the mount side. If you use a belt or chain, there will be a lot of radial load and you don't want the sprocket to be too far away from the bearings.

 

[MitchJi]

You should look for one that has the shaft coming out of the mount side. If you use a belt or chain, there will be a lot of radial load and you don't want the sprocket to be too far away from the bearings.

Thanks!

 

[Jeremy]

Hi

I am an e-bike beginner and have some questions.

On thread page 2 this little motor was shown:

http://www.rctoys.com/rc-toys-and-parts/HI-5018-530/RC-PARTS-BRUSHLESS-MOTORS-HIMAX.html

How is it possible, that such a small motor can move more than 200 pounds (bike and driver)? With a gear to get the torque???

Would it be possible to combine such a motor with a hub like the Rohloff Speedhub:

http://www.rohloff.de/en/products/speedhub/

Motor + Speedhub + gear, everything together in a big bottom bracket.

This old Flyer-F http://www.flyer.ch/topic9885.html has a big motor. The empty case of that motor should be big enough for everything.

Thanks for the answers (to the hopefully not too stupid questions)

 

[Miles]

Motor + Speedhub + gear, everything together in a big bottom bracket.

And pedals, too............?

 

[swbluto]

Hey everyone. I'm currently in the process of developing a speed-prediction program(available online, eventually. Perhaps within a week.) so you can predict how fast you'll go given your environmental details, voltage, motor, gearing, etc. Thought that may be of interest to some of you interested in doing these kinds of conversions(I know I am! I want to know how fast my scooter will go up hills!) and also other conversion like hub motors. I think the accuracy will be pretty high(my quick calculation showed an accuracy of ~97.5% for my bike and I didn't actually tune all the details!) for the vast majority of calculations, but it won't be perfect. There's going to be "should be negligible" unaccounted for errors like motor overheating and magnetic field saturation from too much current. :wink:

Anyways, during the midst of my "research", I've discovered something that may explain why doubling the amount of motors does not necessarily double the amount of power and definitely not double the amount of torque(unless the application is a "constant speed" load which our bikes aren't.). I'm sorry if this has been already answered and/or this topic is too old(I don't remember which page this discussion happened on, haha.), but the main issue is that the power curve is a downward facing parabolic-curve over the RPM range and the torque is a negatively sloped curve over the RPM range(With linear segments in the non-controller limited region). As you increase the "total amount of power" from doubling the motors, they run at a higher RPM which decreases their individual contribution from the amount of power and torque one motor was operating at before(this will always be true if the motor was previously operating in the right half of the power curve; If it was previously operating in the left half of the power range, then that might not be true, but most of our motors aren't operating in the inefficient left-half of the curve at their top speed!). Since their individual contribution decreases, this results in a not-quite doubling of the power and torque that would be true if they maintained the same amount of torque and power output at higher RPMs. Just thought this might clarify a few things. I think a graph would make it clearer but my access to graphing tools are kind of... limited.

Straight-forward logic chain: Higher RPM reduces power and torque from a given motor. Doubling the amount of motors increases the RPM. Therefore, twice the amount of motors means less than twice the amount of power and torque. I just thought the above paragraph might have contained too much not-so-relevant information.

 

[Miles]

There's going to be "should be negligible" unaccounted for errors like motor overheating and magnetic field saturation from too much current. :wink:

Armature temperature is certainly something you need to account for.....