Build No. 2 - Reduction Drive - now w/ NEW video

[mauimart]

I am planning a second build based on a Turnigy 80-100 outrunner running a reduction drive using a generic pocket bike transmission/gearbox (ebay). I have put together a mock-up of how I'd like to see things come together. The outrunner and gearbox would be mounted to 1/4 inch thick plate aluminum and that assembly would be mounted to the underside of the downtube. The motor would drive the transmission using either a belt or chain with something like a 3:1 or 4:1 reduction. The gearbox has a ratio of 5:1. The output of the gearbox would then drive a freewheel crank with another reduction. The overall reduction could be anywhere from 15:1 to 30:1 and would hinge on what motor wind I use and what type of performance I ultimately decide I want. (I have a stock delta 130kv and a rewound wye 70kv and most of my riding would involve an average grade of about 5-7%.) I would like to maintain the function of the rear cluster/derailleur under electric drive as I have found this works a treat on my existing e-bike and the hilly terrain.

Before I proceed beyond the mock-up stage and order more parts, I have to resolve a few issues and make certain there are no show stoppers to this design.

1. The input shaft to the gearbox is somewhat thin and a bit short (that's what she said anyway). It originally had a centrifugal clutch bell attached to it and after having removed that I can see that it's roughly 8mm in diameter and threaded (see pic). Assuming it is strong enough for this application in the first place, I somehow need to extend it a bit and make it compatible with the mounting of a belt or sprocket. Any ideas here?

2. Once item 1. is resolved I think getting the output of the transmission to drive the crank with a chain should be doable. The outupt shaft is about 9mm and splined and the gearbox unit came with a T8F, 22T sprocket. My hope here is that I could somehow adapt and fit a freewheel, bicycle size sprocket on the gearbox output shaft.

3. I don't have any experience with a freewheeling crank so any help here would be appreciated. Are there any good threads that outline the process and describe a good source for these parts.

The pictures show the reduction mock-up mounted to an old Costco full suspension mountain bike. The pulley's and belt were just laying around and are used simply to get an idea as to how the bits will fit together. Also, I have not decided on the actual bike yet but I figure this thing will make for an adequate test subject. I will probably end up using a bike with a bigger triangle for more battery and controller space. Rear suspension is undecided at this point but I figure would be easy to implement with a crank driven setup. All feedback and questions are welcomed.

Martin

[spinningmagnets]

You might consider the option of using the 80-85 in this application. If you drove the rear axle directly, that would draw more amps than this project. By driving the BB, you will be giving the motor some gears to use. 80-100's barely get warm in the rear axle drives (except perhaps when racing), and in this application there would be even less motor heat, so the extra mass of the 100mm length may not be neccessary.

The 80-85 is shorter by 15mm and lighter, however, if you do not want to re-wind it to a custom kV and terminated in Wye (Wye reduces controller heat even more), the stock delta winding is for a 170kV. The stock 80-100's are available in a very usefully low 130-kV and also 180-kV. The 170-kV may be too high to get an easily fitted reduction onto the BB.
http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=5141

Heres the non-hub resources freewheeling-crank (thats what she said) links viewtopic.php?f=28&t=7192#p165215

[Whiplash]

I have done this already and I can tell you to keep the amps down LOW, I sheared the teeth off the pinion gear with only 1200 watts..... Not very strong gearboxes, I think its the cheaply made gears...

[TopCat]

Damn that guy's fast

I was reading this post and was just about to let mauimart know about Whiplash's attempts at this kinda setup but...

Regards
Tom

[Whiplash]

Haha funny! I get up for work pretty early and that is the only time I get to do this in the morning.. yes to the original poster if you can power low you should be okay. And don't accidentally take off in tall gears...

[spinningmagnets]

In a motor that size (80-100), I would keep the volts and amps down, and slowly creep up on your desired watts. Seriously consider buying an 80-85. The 80-100 is so popular now, you would have no problems selling it for full price (since HK is frequently sold out of these recently).

Since similar builds which drive the BB (like whiplash) have enjoyed good success with the smaller 63mm diameter motors (easily capable of 2100W/3-HP),...an 80mm motor at 48V (or even 36V) would likely be breaking things very soon. Best of luck with whatever you decide...

[mauimart]

Thanks for the input guys.

I have done this already and I can tell you to keep the amps down LOW, I sheared the teeth off the pinion gear with only 1200 watts..... Not very strong gearboxes, I think its the cheaply made gears...

I was hoping to do about 3kW peak with this setup yet it seems that could be out of the question. Don't those pocket bikes put out more power than that? I guess I'm not too surprised at the mediocre power handling capability -- you can't ask for a whole lot at a cost of $20 + shipping. Do you know of any other gearboxes with a similar compact profile that could handle the power and torque of the large RC motors? It just seems like an elegant drive train solution that lends itself well to easily configurable gearing. Back to the drawing board then...

You might consider the option of using the 80-85 in this application. If you drove the rear axle directly, that would draw more amps than this project. By driving the BB, you will be giving the motor some gears to use. 80-100's barely get warm in the rear axle drives (except perhaps when racing), and in this application there would be even less motor heat, so the extra mass of the 100mm length may not be necessary.

Thanks for the tip on the 80-85. Do you know if the mounting hole locations are the same as the 80-100 as the motors have the same diameter? Almost a drop-in replacement if the 80-85 also had a 12mm shaft instead of 10mm.

[spinningmagnets]

The shaft on the 80-85 is also 12mm (same as the 80-100, but 15mm shorter). The two endcaps are identical, so they would drop right in. The biggest question is would the 170kV work for your goals?

[Whiplash]

I am actually in development of something like that right now but the cost will be much higher than that little cheats 20 dollar box, although my power output rating should be in excess of 3000 watts possibly closer to 5000, and it will actually be even more compact than your setup. I am using a 3 stage internal gearbox, oil filled, and the production unit will be milled from a solid block of billet aluminum. Check out my build thread for other details. If you want I will keep you posted.

[Bluefang]

Well that sized gearbox should be able to handle quite abit if the gears are good quality, a 50CC scooter has a 7-1 gearbox at the end of the CVT that is about the same size and handles up to 20hp in a tuned scooter, not for long tho. But back to my point it should be fairly easy to find a pinion gear thats alot stronger then the current one to replace it with? That would also solve your problem with the input shaft been too small. Even replacing the main gear should not be that hard, with the sheer number of these boxes around you would think they are all the same basic design. Or you can go out and buy the high end version which will certainly handle the power.

So who can explain how to measure a gear like the one we have?

[Whiplash]

It could have something to do with the fact that the gas engine has the slipper clutch to take the shock loads out. That could be used to make it last but will kill efficiency due to the slippage in my opinion and it would add complexity..

[Miles]

So who can explain how to measure a gear like the one we have?

http://www.gizmology.net/gears.htm

Identifying Unknown Gears

Count the teeth;
Add two;
Divide by the Outer Diameter.
This gives me the Diametral Pitch. It rarely gives me an exact pitch, due to errors in measuring, but it is usually very close to one of the common pitches of 32, 24, 16, etc.

If this gives me a goofy answer like 22.1, I suspect it is a metric gear, and divide 25.4 by the pitch (or divide the pitch by 25.4 and hit the 1/x button on the calculator) and see if it comes reasonably close to one of the standard metric moduli of 0.5, 0.8, 1.0, 1.25, 1.5, 2.5, etc.

If this still gives me a goofy answer like 1.14, then I figure it might be a Stub Gear, and try this:

Count the teeth (it doesn't usually change from the last time);
-Add 1.6;
-Divide by the Outer Diameter (which also doesn't usually change).
-It should be close to one of the common pitches, but, alas, it often isn't. If I have a gear with which it meshes, I try this:

Add up the teeth on BOTH gears;
-Divide by two;
-Divide by the Center Distance.
-If this doesn't work, it must be a profile-shifted gear, so I try swearing, which also doesn't work.

[spinningmagnets]

I was hoping to do about 3kW peak with this setup yet it seems that could be out of the question. Don't those pocket bikes put out more power than that?

I'm sure this would have occurred to you soon on your own, but...The peak power of a gasoline engine is at the higher RPMs, when the gears and chains are already spinning very fast, and the load is more spread out.

With an electric motor, the full torque can be applied at 1-RPM.

[mauimart]

Some more thoughts on my reduction drive...

Well that sized gearbox should be able to handle quite abit if the gears are good quality, a 50CC scooter has a 7-1 gearbox at the end of the CVT that is about the same size and handles up to 20hp in a tuned scooter, not for long tho. But back to my point it should be fairly easy to find a pinion gear thats alot stronger then the current one to replace it with? That would also solve your problem with the input shaft been too small. Even replacing the main gear should not be that hard, with the sheer number of these boxes around you would think they are all the same basic design. Or you can go out and buy the high end version which will certainly handle the power.

I could try to source higher quality helical gears (if they exist) for the cheapo gearbox. I would still be limited to an input shaft diameter, using existing bearings, of 9mm. This seems to be the weakest link to this setup. I did some searching for a similar gearbox with higher performance (for the larger scooters) but have found nothing yet. Do you have any links to this 7-1 gearbox?

I did however stumble upon one of these:

http://www.anaheimautomation.com/manual ... 0Sheet.pdf

It's a planetary gearbox with reductions ranging from 3:1 to 10:1 with single stage, capable of max 6000rpm input, 95% efficient, max torque of 95 N-m, rated torque of 47 N-m. (Anyone have come ballpark torque numbers for the large outrunners? ) The major drawback is price, ~$250. Outside of that seems like a possible solution if it can handle the torque.

Martin

[mauimart]

I think I may have found a gear that will work...
It's currently being used at the local sugar mill. I will inquire if they have a spare.

[mauimart]

IT"S ALIVE!
I scrapped the 5:1 pocket bike transmission and put the planetary reduction on hold and this is what I came up with.
Standard dual reduction using 5/8" jack shaft.
From motor to jack shaft: 9T:54T, no. 35 chain.
From jack shaft to freewheeling bottom bracket: 14T:48T (Overall reduction from motor to crank: 20.6:1)
From crank to wheel: 24T:32T (1st gear), 24T:11T (8th gear). Overall reduction from motor to wheel: 27.4:1 (1st gear) and 9.4:1 (8th gear).
The motor is a rewound turnigy 80-100, 9-turns, 14AWG, wye, internal halls at 120 deg.
Controller is a Kelly KEB72800.

Initial test rides have been very encouraging and fun. I love the full suspension coming from my first build which has none. I also like the ability to be able to pedal the bike. I am currently running a 18s1p 5Ah pack of Lipo's. Full-up weight is 57 lbs. Controller is set to 50% rated current. This thing has got some giddyup.
The plan is to do some more testing to make sure the mechanical drive train will survive what the motor will dish out and to see if the motor can handle my commute (7 miles, 1800 ft elevation gain) without overheating or blowing up. Once satisfied with the performance I will begin cleaning-up the build - using nuts an bolts instead of zip-ties, trimming aluminum, some lightening holes in the 54T sprocket, finding a permanent location for the controller, and adding a bigger batter pack (hopefully ~1kWh in the triangle). I'll try to get some video...
Martin

[adrian_sm]

Very nice. Great to see you get a dual suspension/RC build up and running. Oh, and I like the digital dash.

Be careful with the clearence to the front wheel, wouldn't want the motor mount bracket to hit the wheel at speed.

What speeds can you actually pedal assist to. Looks like a pretty small chainring.

- Adrian

[mauimart]

Very nice. Great to see you get a dual suspension/RC build up and running. Oh, and I like the digital dash.

Be careful with the clearence to the front wheel, wouldn't want the motor mount bracket to hit the wheel at speed.

What speeds can you actually pedal assist to. Looks like a pretty small chainring.

- Adrian

Thanks Adrain.
The picture is a bit deceiving -- the clearance is ok. I have fully compressed the front shock and there is no interference. I will eventually trim/round that front corner of the bracket to give me even more clearance.

The chainring is indeed small (24T) but in the highest gear I can pedal up to around 16 mph which for 5-7% grades is more than enough. I reckon I will pedal only if there is a problem with the electric drive. I save all of my pedaling energy for my road bike.
Martin

[spinningmagnets]

Too bad there is currently no affordable outrunner motor selection halfway between the 63mm motors and the 80mm. I doubt you will ever have to worry about the motor heat. Looks good, best of luck!

With 18S LiPo and giving the motor some gears, I am certain you will be happy with its climbing ability.

[Kurt]

The bike looks great. I am keen to see what data you can collect. top speed, amps, temps and so on . A video or two would be great .

Too bad there is currently no affordable outrunner motor selection halfway between the 63mm motors and the 80mm

The 80mm motors are not that big or heavy really when you think about it. Not having everything running on the limit is a good thing in my book.

Interesting how all the medium and large turnigy out runners from HK have been on back order for a very long time. Will be interesting to see if HK come up with some new low KV large RC motors to meet the demand for them.

The great thing about using a cheap RC motor is its such a simple job to change motors compared to a hub motor having to rebuild a wheel every time

Kurt

[spinningmagnets]

Not having everything running on the limit is a good thing in my book

I couldn't agree more. I just bought a house in Kansas, so I am fresh out of time and money, but if I suddenly found myself with a lot of both, I'd say the 80-85 is the hot ticket for this application. Even with a complete reworking of the motor immediately upon arrival, The shaft/shell/baseplate/and magnets are worth the HK price.

Disassemble, add the proper epoxy between the magnets to lock them in shoulder-to-shoulder, swap the bearings to the mid-grade NSK for a big improvement thats still very affordable, rewind the copper per Thud to a the lowest practical kV, and terminate the windings in Wye to ease the current and heat @ the controller (of course add a cheap temp sensor to the windings). A low amp-draw system also means you can get high performance from a non-LiPo chemistry (C-rate limited).

Add Berties optical sensors and a Lyen 6-FET run at 44V-48V, and the motor/controller would never be the weak link in this configuration. You would need some way to prevent power from being applied when at a stop in a high gear, or things would break.

I would even look for some way to snub the throttle to prevent shock-loads due to a sudden application of full-throttle. Between Justins RC-CA, Keplers interface, and Adrians brain box, I'm not sure yet which would best provide that with a system such as I've described. I just checked the Cyclone-USA website and the closest kit to this would be the 48V 1200W single-cog kit for $969 (+ T&S), and thats even with the need to immediately upgrade some of the components...

[mauimart]

video finally...

[knoxie]

very nice looks and sounds like a good setup there, be interested to see it all tidied up and how it fares on longer rides, you figuring to up the battery capacity then?

[AussieJester]

That is one of the best edited e-bike videos I have seen what program do you use? Nice job on the setup will look slick when finished !

KiM

[mauimart]

Thanks for the compliments.

very nice looks and sounds like a good setup there, be interested to see it all tidied up and how it fares on longer rides, you figuring to up the battery capacity then?

The plan is to clean it up and add 10 to 20 Ahr, 18s pack. I have tried a number of times to tear it down for rework but I can't seem to bring myself to doing so as I am having too much fun riding this thing. I like to call it extra qualification and validation testing... I plan to get the controller out of the triangle and reserve that space just for batteries. I'm really enjoying the off-road experience this bike has to offer and in the process I'm noticing how crappy the suspension is. It has me wondering whether I should invest in something a bit more modern and capable? In any event, lots to do to make it better - after I ride it a few more times.

That is one of the best edited e-bike videos I have seen what program do you use? Nice job on the setup will look slick when finished !

KiM

I used iMovie '09 to edit the video. It's really the only thing I use my mac for.