Electric bicycle building
 

Hi,

I thought I build an elecric bicycle before I jump head on into converting a motorcycle just to see how it works... I decided to have a "blog" here.

I'll try to keep it cheap.

Things I'd like to achieve:
- to go to work. :)

I live in a city... (well, a few cities built together in a huge industrial area) and I live some 14kms from work. getting to work by bus or train is some 90 minutes, with the bike it could be a little less. I could charge (at least partially) the batteries at work, so they will have to be able to give me some 20-25km range.

So far the idea is:

- bike: a simple Dutch roadster, slim tyres
- motor: 12V power steering motor (because it's cheap and I trust it more than the chinese motors one can find on internet.) the idea is not mine, I stole it from Julian Edgar, I hope he does not mind. ( Browser Warning)

- batteries: 2 x Greensaver 12V 20Ah/2Hr, I found them on internet. They were new/old stock sitting in a warehouse for three years where they were recharged every 6 month. (so the seller claims) they arrived in original packaging with 12.80 v so it does look promising. I hope they are in good shape.

- drivetrain: I was considering friction drive, but If I can manage I'd like something better. The huge ratio between the rpms of the motor and wheel makes a simple chain connection hardly possible. I found somwhere that the gearbox from an angle grinder was used. I checked mine, the ratio is 1:3.08, which could work with a 1:4 - 1:5 chain transmission. I do not know if the angle grinder gears will be able to handle the torque, and also I could not find exact data on steering motor RPM, so I'll have some shooting in the dark here.

- controlling elecrics: copy of Julian Edgar's... I plan to pedal up to a certain speed where I can engage the motor on 12V and we'll see how fast it will be able to go. I think for shorter periods 24V (batteries connected serial) will do no harm. So the bike will be definitely hybrid :Start and acceleration with human power, keeping the speed with electric. I plan to put two small buttons on the handle and a pair of relays by the batteries. Wireing will be simple. I think I will get an ampermeter and voltmeter to check and test the operation at least at the beggining.


Calculations: I used Bicycle Speed (Velocity) And Power Calculator to check what power I need. Assuming, that the steering motor will be able to give me 200-300W, it should give me some 25km/h, which is quite ok. Will see, how it works.

Any Ideas, Recommendations?
(If this works, I might step forward to buy an old palett lift and convert my GS450 which has a half dead engine anyway... ;) )

Ok, here some pictures of the basic parts I've already got:

the battery: http://p1.bikepics.com/2012%5C02%5C0...44814-full.jpg

the motor (12V power steering motor (Koyo electrics))
http://p1.bikepics.com/2012%5C02%5C0...344813-800.jpg

http://p1.bikepics.com/2012%5C02%5C0...344812-800.jpg


http://p1.bikepics.com/2012%5C02%5C0...344816-800.jpg

next step is to get an old angle grinder from the fleahmarket and to figure out how to connect the motor, the gearbox and the sprocket on the output shaft.

wow! what a great idea on the angle grinder!

and, you certainly seem to have a great project going.

consider using mosfets and your thumb to create a cheapy motor controller - push the button for a few seconds to get a boost. the longer you hold the button down, the faster you go.

also, you could use a capacitor/resistor delay setup like a wiper delay. use an adjustable reostat for variable power.

[QUOTE=drmiller100;285551]wow! what a great idea on the angle grinder!
QUOTE]

An angle grinder is very high rpm, I don’t think the gears would hold up long at all. You might want to try something like a Hole-Hog. In the electrical industry we actually use them (with additional gear reduction) as cable pullers. They are incredibly robust and would probably get you closer to the rpm you would need.

After looking at a hole hog more carfully, this might be really difficult to adapt....

One more thing, I would bet money this is a series wound motor. This is a good thing for you as the torque goes up as the rpm goes down. The only bad thing is they really have no speed limitations, if you don’t put load on them and run the motor, it can over speed some motors.

I finally managed to test the motor and "measure" the RPM.
I pushed the shaft against the rear wheel of the bike and checked the revolutions the rear wheel took under a minute. Seems like with this load (bike off the ground, practically minimal load) the motor spins with 1950 RPM.

Electictracer, I have similar doublts about the angle grinders. Checked RPM/Torque data and found out, that they produce like a 5th of the torque I might use as continous...

I happen to have some leftover parts from a motorbike gearbox, so in theory I can can put together a reduction gear which can do the trick...

the data I have so far:
Motor RPM (minimal load): 1950
Bike wheel RPM @ 20km/h : 160
reduction gear I can make with the gears available: 19/12: 1:1.58

big question: what RPM would the motor have, provding enough torque for a continous motion with the speed of about 20km/h? If no load RPM is 1950, with what RPM would I have to count for the gears?

With this site
Bicycle Speed (Velocity) And Power Calculator
I calculate 100W-s for my humble carcass on the frame, 0% incline.
1%: 180W
2%: 260W
0% - 260W ---> 28km/h

Of course I do not know the accuracy of the site but it gives a good idea. I'll be able to play with the final gear ratios as it will be with stock rear sprockets, so I'll figure it out.

Any ideas about the RPM under load for such a motor?

GA7021 7 in. Angle Grinder-GA7021 at The Home Depot

Move up to a 7 inch angle grinder. They are TOUGH.

15 amps at 120 volts is 1800 watts, which is over 2 horsepower.

if you are making over 2 horsepower continuously, you don't need electric assist!!!!!!

Yes, I was considering the same idea. I have an auction on an eurotec 3150W angle grinder. That is a beast. sold as broken as some rubber part is missing. I can't wait to lay eyes upon it! :cool:

The big grinder would work great. Calculating the rpm under load is tough. You might have to do some sprocket changes. Good luck, I love these electric bike projects.

seems to be hard to get a broken down beast of grinders... I'll try my luck on the fleahmarket this weekend. I saw ppl selling old/broken power tools...

I might just end up with a friction drive. At least with that I'd get a picture of the RPM under load. Planning the drivetrain would be straightforward after that.

About the electrics.
I'm planning a precise charger for the batteries. I found a charge controller, UC3906, which seems to be up top the task pretty well.

My problem is that i did not find a schematic which would be able to charge bigger batteries, and I would want something more than 1-1.5A maximum current. (battery pack will be 2x 24Ah @ 20h 12 batteries in paralel)

I got the datasheet for the batteries and with the datasheet of the UC3906 I'm able to plan the thing, but I'm a bit lost at the power switch part. :D It was too long ago I last dealt with transistors... :D So If somebody could tell me taking a look at the datasheets, weather the chip will be able to control 10A with a BDW94C darlington, I'd really appreciate that.
I guess it can handle it, but better not fry the chip, right?

Finally I managed to get a half big grinder for chips on the local fleamarket.

I have no idea what machine it could be but probably some 1.5 - 2.3 kW model.

I made a little "plan" in Paint about mating it together with the motor.

http://p1.bikepics.com/2012%5C03%5C0...61249-full.jpg

My concerns:
The small cog was originally put on the shaft of the grinder only by the friction force what a well tightened nut gave to it. Withouth machining tools I'm a bit reluctant to try to create an insert for the motor shaft to use the same solution.
The idea: I cut a little, 4mm wide, 4mm deep slot on the motor's shaft which is 18.8mm in diameter. Also I cut out a fitting piece from the bottom of the cog, so they will sit together and this joint will transfer the torque.

the motor output shaft has a hole drilled into it, some 11mm in diameter, with tiny teeth inside. The cog has a bore inside of 10mm diameter, so what I thought is that I have a 10mm steel bolt (I happen to have one already cut) which I could fit into the hole on the motor shaft using some soft metal ring and loctite. This would center the cog perfectly and the thread on the end would keep the cog escaping. Hopefully.

What I have problem with with this two joints is: I have no idea how strong they will be...

anyone with some mechanics - machining knowledge could give me a hend here? :D

The shear strength on the 4mm x 18mm area will depend on the material of the cog. Would it be possible to put a steel sleeve over the outside of both the cog and motor shaft then use four 6mm set screws to hold it in place?

Any improvements in aerodynamics will reduce power requirements.

Hi, thanks for the idea.

The cog has a 10mm diameter bore in it. It was for the original motor shaft, and the cog was held on it only by a simple nut and the friction force what this tightened nut gave to it. So the actual face of the metal is not 4x18mm... It's smaller due to the hole in the middle of the cog. The shaft of the motor I use also has a hole in it with tiny grooves. I figure I could drill that up to 12mm or so, get a little shaft welded into the cog and drill and tap a few small screws into the motor shaft to hold the shaft of the cog...? Would that work? Would that be any stronger then the 4x 10-ish slot?

let me see if I get this. The hole in the cog is 10mm. The hole in the motor shaft can be drilled up to 12mm. Could you press a shaft into the cog? Could you press that whole thing into the motor shaft? maybe some combination of pressing and welding? Small screws would probably break, its hard to get any strength out of them.

Yes, I could make a shaft, which is something like this:

http://p1.bikepics.com/2012%5C03%5C1...64321-full.jpg

instead of the nut I can actually weld the cog to the 10mm shaft.... this way the only thing to do is to ensure that it will stay aligned with the motor and that it will be capable of transmitting the power...

I like the looks of that much better. Pressing the peices together, done properly, should transmite the torque in the amounts your dealing with.

No really update here... I found a mechanic workshop who could do the trick of machining the pieces together, however they are exactly opened during those hours when I work...
probably next week I can pop in there though.

Let me recommend a great resource for electric bike projects:
Endless-sphere.com • Index page

A good press fit will be needed to secure the parts. Using a hydraulic press on a motor shaft can be tricky as you don't want to bent the shaft at all. Either thread the adapter to 12mm and use a jam nut or grind two flats on the adapter 100 grads apart then secure with 6mm 9.8 grade set screws. You may need to separate the gear from the motor at some point. The people at the mechanic shop may have some good ideas too.

Finally I got it done. A guy from the neighborhood knew a turner and he coupled the thing nicely. As usual I forgot to take a picture, but the thread is alive and there is progress.
Right now I'm looking for a metal plate to bolt the motor housing together with the gerabox housing.
Once that is done, only the task to find appropriate chain / sprocets and to fix 'em on the bike.

We are getting there. :D

I've got some plywood and two metal plates to build a mounting plate for the gearbox and motor. In case my knowledge of the word is not correct: I mean under plywood that kind which is glued together of wooden sheets with their directions being perpendicular. I'll have 20mm of that. As far as I know that is the strongest wood-pane you can get.)
this 20mm of wood is between 2 steel plates being 2mm each.
I drilled the mounting holes and managed to align the motor with the gearbox... well... more or less at least... seems I'm not that precise with a standing drill as I'd like to be.

either way it's waiting to be put together for a test.
after that I'll glue the pieces of this sandwich together, and paint it.

I have no idea how strong it will be, but probably it will be able to hold it together well enough.

I'll post pictures soon.

Well, If done correctly this type of construction can be very strong. I image the plate being used to hold the motor and resist the torque it is producing.

Just make sure you use a "glue" that is approved for the materials you are bonding. Usually this means using an epoxy.

That is stronger than the floor pans we build for rail cars! Epoxy would work and let it cure under pressure; park the car on it over night. Seal the edges with epoxy too.

Hehe, I'll look rather strange gluing something on the street with the car jacked and lowered on it. :D
But I might just do that... ;)

At worst, you could end up with the stuff glued to your tire :-P

But that would not improve my swift's economy, would it?
:D

nevermind, bad pun.

Not to worry. A piece of wax paper will keep the goo off your tire and the pavement. Remember More's Law: If enough is good too much must be better. Actually a thin coat works fine. Most shops use a 1 ton hydraulic press to lay up sheets of laminate spaced with fiberboard.

:D :D
So I guess my car does not qalify. It's 750kg altogether... :D

But I'll fix it, no worries.

finally i got the pictures.

You can see the grinder gearbox coupled with the electric motor with the used of that metal-plywood sandwich. Also you can see the way we solved the connection of the grinder's cog and the motor shaft.

Ok, ok, the sandwich is not the best engineering product I've ever seen, but I guess it will do...

http://p1.bikepics.com/2012/05/17/bi...402330-800.jpg

http://p1.bikepics.com/2012/05/18/bi...402734-800.jpg

http://p1.bikepics.com/2012/05/18/bi...402735-800.jpg

http://p1.bikepics.com/2012/05/18/bi...402736-800.jpg

http://p1.bikepics.com/2012/05/18/bi...402737-800.jpg

http://p1.bikepics.com/2012/05/18/bi...402738-800.jpg

http://p1.bikepics.com/2012/05/18/bi...402739-800.jpg

http://p1.bikepics.com/2012/05/18/bi...402740-800.jpg

http://p1.bikepics.com/2012/05/18/bi...402741-800.jpg

Ok, we arrived to the step of transferring the power to the wheel...
the question is how...?

The structure of the motor and the grinder is that it turns only to one direction. (I think, even if it would be possible to reverse the motors polarity, the nut which holds the smaller cog would loosen up...) This direction forces me to drive the rear wheel from the left side.

So, how do we put a sprocket on the left side? (the 6 gear gearbox is on the other one)

My idea is that I get a big sprocket (some 40 teeth) with a small hole in the middle and put that one on the left side of the hub, fastening it with M6 or maybe M8 bolts going through the little triangles formed by the spokes by the hub... I could carve out some rubber pieces which would ensure that the sprocket is nicely mounted to the spokes / on the hub...

The good candidate for such a sprocket is a BMX front sprocket, being a full disc so I can drill wherever I want. It's also a bit wider than usual bike sprockets. size is: 1/2 x 1/8. chain is 410

to match it, I found that the front sprocket from an old german moped (it was really popular here, parts are abundant) Simson could fit nicely on the grinder shaft. That sprocket needs chain 420, which is only wider, the pitch and the roller diameter is the same. Get 420 chain, and we are done...

Better idea?

I was also considering putting a rearwheel hub to the front and have the bike front driven. I'm a bit reluctant to do so... probably because I never saw a front driven bike. :D

There are some e-bikes that have front wheel drive; using a hub motor. The advantage is it leaves room for the rear gear cluster, and the possibility of better regenerative braking.

Yes, I saw quite a number of those fromt wheel driven, hub motor powered bikes. (last week I saw a hub motor in a wheel for chips on the fleahmarket... GRRRRR :D ) I'm reluctant to put this huge motor/gerbox assembly to the front though. It is really bulky.

I got the idea:
I'll get a whel hub which can have a disk brake and on the disk brake mounts I fasten the big 44 teeth BMX sprocket.

soon it will be done.

wow it was logn ago when I wrote last time...
Still, the project is alive... interferred with another project named: refurbishing the house :D
So, The idea of the power transfer is:
Motor - gearbox - small freewheel (like 16 teeth) - chain 1/8 - big sprocket (BMX pedal sprocket) fixed on a wheelhub which has thread on both sides for freewheel / fixed gear. (stunt bikes and delivery-boys use them... calling the bikes fixie)

so the main "invention" is the new wheel hub which has the thread on both ends. It will allow me to put a normal 3/32 freewheel on the right side to be driven by the pedal, while on the left side there will be a fix threaded sprocket locked with a locknut. (all can be ordered from bike shops for chips) The sprocket I choose has some holes on it for weight reduction. They gonna be perfect to bolt the BMX sprocket on. I could get a wheel hub which takes a disc brake too, but let's face it: this is much cheaper and I can place the sprocket much closer to the wheel... I gonna need it as the frame we not designed top have anything on the left side.

So here we are at the moment, haggling with the internet shop.