DrifterDrive Ver 7.1

[drifter]

I have an idea of using a bike freewheel with a friction roller drive system. This uses a back wheel hub, axle and bearings and a screw on freewheel sprocket
with 18 teeth. I found an 8 tooth 1/2" pitch sprocket which I drilled out the bore to fit the 10mm motor shaft and got a friend to turn the teeth down to 1/8"
on his lathe. I will need to fit a piece of metal tubing,maybe muffler pipe or what ever I can find, over the spoke flanges to give me a roller size of approx. 2.5" to 3" to get approx. 20 to 25 KPH max speed, I can also go up a tooth or two on the motor sprocket if needed. I plan to fit this to the front wheel so I can keep the short leads on my watt meter and also can better see how it is working.

[Bluefang]

It will work very well, thats almost exactly what i used for my friction build viewtopic.php?f=28&t=22108 Tho mine was not quite as compact as yours.

I ended up using some PVC pipe as a sleeve to go over the lot, drilled 3 holes around it and filled the inside with a bit of epoxy to make sure it didn't flex to much and then wrapped it in some old sand paper, grips the tire really really well

[spinningmagnets]

Very clever! I like it. Whatever issues arise, I am certain you will find a doable solution. Several builders have raised the option of using a 3-speed hub in discussions, and you are the cutting edge of solving any concerns that stand in the way.

[drifter]

Hi Bluefang, I like your roller engagement setup. I think I will have a look at fitting hall sensors too because I am having sync problems with my Turnigy ESC.
Also like the sandpaper (actually no such thing as sandpaper but that word is in general use)

[GGoodrum]

Interesting, but one thing you might try is eliminate the extra tube you want to add, and just put the non-slip tape directly on the hub. This will allow you to use a larger motor sprocket. I can tell you, the small 8t sprocket is going to make a ton of racket, due to the chordal effect. It is hard for a chain to get around a sprocket that has fewer than 11 or 12 teeth. It ends up sounding like bucket of bolts in a washing machine.

-- Gary

[drifter]

Thanks Gary, It would be a lot easier to change the sprocket than to add the tube. Now where is my calculator.

[Bluefang]

Be carefull mate, i have been riding around for a few days now and since the novelty factor of the new Ebike has worn off i am noticing that my drive is making way to much noise from the chain and free wheel setup, yours might be better as i assume your going to have it completely off the tire when not in motion, but chains running at high rpm are very very noisy.

By the way if your interested in selling your other drive with the belt on it send me a pm with cost, its pretty close to what i will build so i should be able to easily mod it myself and make my own roller. Tho i am going over to the US for 6weeks in a weeks time so i wont be able to do anything till after that lol.

[drifter]

BF, I hooked up the power today to see how noisy it is and it doesn't seem too bad. The other drive doesn't exist any more as i used bits of it for the new one sorry. I would like to use a belt instead of the chain but I haven't yet examined how to fit a belt pulley onto the freewheel.

[John in CR]

4 bolts are warranted instead of 2. All the torque is passed through them at a significant distance of leverage.

[Solcar]

I wonder if sprockets could be made of rubber. As rubber belts are used with metal pulleys, I'd like to see a chain used with rubber sprockets.

[spinningmagnets]

I recall a friends motorcycle that had nylon sprockets and a metal o-ring chain on a Honda 750 (novel to us at the time in 1974). Ran quieter and cleaner, as each link has some grease sealed inside, no spray-lube needed.

Don't know about longevity of wear, but it has been done. Perhaps some plastic sprockets could be found for #219 chain? SDP-SI, McMaster-Carr?

edit: Found $12 (11T) nylon sprockets for #25 chain, 9T up to 45T available. Bore is 1/4", but theres plenty of room to bore it out bigger, easy to thin if you wanted to use bicycle chain...
http://www.mcmaster.com/#nylon-sprockets/=9ke2g2

[drifter]

I did a test with and without the chain fitted and was surprised to find that without ,the meter shows 51W and with,it shows 124W. That is quite a bit of power loss, might be because of the small sprocket, might go back to timing pulleys.
John in CR, yes I am going to change to 4 bolts when I get everything sorted out. BTW what is CR.

[drifter]

I have decided to ditch this idea too. Whatever type of transmission reduction is used, chain, belt, or gears introduces some efficiency loss into the system. When using limited power to begin with, any type of power loss is not welcome especially when using a smaller power supply (battery) to reduce weight. So this takes me back to the start of my experiments again. I like the motor used as roller system by Kepler but I want a smaller diameter roller for better hill climbing so I have put together another mock-up to try. Using parts from the junk box of previous builds I have used a Turnigy 63-74 motor, a one way roller bearing, a pulley from an old RC starter motor and various bits of shelving angle iron and aluminium.
Works well in the workshop (loungeroom) .When it stops raining I will road test.

[recumpence]

Hey Drifter,

That friction drive looks cool! If it slips, all you need to do is add grip tape to the roller. But, it looks really good.

I love simple "Think outside the box" designs.

Matt

[spinningmagnets]

A year and a half ago, Grinhill posted his first RC build (March 2009). He later went to a bigger motor (63mm) which I have copied (Thanks, G!). Although...I have always liked the way he used the brake mount bosses as a attachment point for the motor on his first one.

http://endless-sphere.com/forums/viewtopic.php?f=28&t=9172#p141788

By adding longer bolts and an intermediate friction-nut as a divider, you can still have brakes on those points, with the mount slid under them. The problem it addresses is that bike frames have a wide variety of shapes. This causes a problem when someone wants to build a copy of a successful drive, but the bike he has (or most wants) is a shape that restricts certain drive configurations.

As luck would have it, I already have a hardtail DH-frame with a low top-bar. After seeing Todds build, I would have bought one if I didn't already have one. Regardless of frame geometry, the relationship between the common brake types and the tire are very consistent, with only a tiny amount of adjustment needed for any drive thats attached there.

Looks Good Drifter!

edit: you have prompted me to finish my "volt/kV/roller diameter/speed" chart:
http://endless-sphere.com/forums/viewtopic.php?f=28&t=21365&p=326424#p326424

[drifter]

What is the problem with these bloody Turnigy ESC's. I am just about to spit the dummy and give up on using RC motors. I have a Cyclone kit that I can put back on the bike till either someone figures out how to modify the Turnigy ESC to fix the sync issues or I fit hall devices and use a sensored controller, I have one I got with a second hand hub motor, or buy a Castle Controller, I don't want to spend so much.

[adrian_sm]

What sync issues are you talking about. Mine seems fine (Turnigy 85Amp ESC + Turnigy 5065) , but I only engage the drive once I am moving. Maybe if you give a few more details someone might be able to help out.

I appologise if you have already had a rant somewhere about it, but I couldn't find it with a quick search.

- Adrian

[Kepler]

I thought you had the the same setup Adrian.

Drifter, either your expectations are too high or you have a faulty component. If you want me to test your gear in one of my drives, let me know. For the sake of the cost of postage, might answer things once and for all.

[drifter]

Kepler, thanks for the offer. Please PM your address and I will send my controller for testing.
Robbo

[Hillhater]

Drifter brother,....
you have my full sympathy....I feel your pain and frustration !

A fellow sync sufferer (waiting for alternative ESC's to try ! )

HH

[bzhwindtalker]

to try or to fry?

[Kepler]

Kepler, thanks for the offer. Please PM your address and I will send my controller for testing.
Robbo

PM sent.

I will be testing drifters ESC on one of my drives and test it both with a simple servo tester interface and a prototype smart interface. Will also test it with a 63-74 and the 50-65 motor.

Drifter, I presume you have experimented with ESC programming settings such as timing etc.

Should be an interesting little exersize

[drifter]

Kepler very kindly did some testing of my ESC with 2 of his motors and found it had sync problems with the 50-65 motor and performed OK with the 63-74. He also reported that there were no problems when using a Castle Controller. This tells me that the ESC (Turnigy Kforce 100) is usable if everything else is setup ok. I am now wondering why there is a difference in the results using different motors, is it the different windings, different number of magnets or something else. I will keep trying as I am still coming up with ideas I want to try.