Electric Peugeot JetForce conversion

[circuit]

Just to showoff myself...
Converted from this:



to this:



... and loads of work and photos later... to this (showing off at national scooter championship):





---
Hilights to "tell the story":















Thats it, I think.

More info on my blog: http://e-motion.lt/ and forum thread on http://e-motion.lt/bbpress/topic/12

[fechter]

Awesome!
The battery looks like a major project all by itself.

How's the performance?

[circuit]

Performance is not as good as I'd like it to be. Well, front wheel does lift off at launch and accelerates to 20 km/h very quickly (it's hard to hold on), but then acceleration slows down (when belt starts to climb up). Accelerates to around 60 km/h on flat. Uphill only ~40-50 km/h or so (the CA is still not configured for correct wheel size).
There are still many problems to solve... Such as:
1. Complete the programming of BMS (had no time before the championship);
2. Install battery cover and charging port;
3. Modify industrial telecom UPS charger for voltage of this pack (22 pcs. 20Ah A123 cells);
4. Change weights in variator. Now belt starts to climb at 5500 RPM, that needs to be around 7000-7500 RPM. I get less torque because of this.
5. Change the belt to Kevlar racing one... This one is slipping (with red (hardest) spring on rear pulley);
6. Ramp up current limit to 130 Amps. Now I got only ~36 Amps. I use Lyen 18 mosfet controller with default settings. Mine looks to be faulty and does not accept any configuration (my friend has identical one and it works ok). I shorted the shunt with a small wire on the field, during national championship... It increased the current a couple of times, but still not enough. Will have to calculate correct shunt and do it properly.
7. build and install 12V buck converter for lights and horn.
8. Change all lights to LED (xenon maybe for headlights) and horn to better (more efficient) one.

[maydaverave]

Awesome build. If you have six weights you can take every other one out and see how it performs at half weight. Not a permanent solution but will give you some idea of what weights you should get.

[circuit]

Awesome build. If you have six weights you can take every other one out and see how it performs at half weight. Not a permanent solution but will give you some idea of what weights you should get.

Thanks for suggestion. Tried that. The weights are 8 grams each, I took half of them out, so that leaves us 4 grams. That's a bit too light, belt hardly tries to cling at max RPMs. I guess it needs ~5-6 gram weights. Anyway, still lots of work to do.
Remembered one more thing to do: buck converter for 12V lights... Added that to previous list as #7.

[gwhy!]

Awesome build. If you have six weights you can take every other one out and see how it performs at half weight. Not a permanent solution but will give you some idea of what weights you should get.

Thanks for suggestion. Tried that. The weights are 8 grams each, I took half of them out, so that leaves us 4 grams. That's a bit too light, belt hardly tries to cling at max RPMs. I guess it needs ~5-6 gram weights. Anyway, still lots of work to do.
Remembered one more thing to do: buck converter for 12V lights... Added that to previous list as #7.

This looks really cool, I had to up the weight of my rollers to around 18g ( needs to be slightly heavier) and changed the clutch launch rpm to 1k ( just removed the springs in the end ) this was on a stocked geared 50cc gilera, I used ( what looks like the same motor ) as you have. Current limit on my 12fet controller is set to only 70A so only pulls around 3500W ( so loads left to play with ) but the scoot pulls very well with a top speed of around 40-45Mph, My project has be put on hold for the time being . Keep at it ( go heavier with the weights )

Edit: My build thread http://endless-sphere.com/forums/viewtopic.php?f=12&t=17576

[circuit]

Power limit at ~40%.

[gwhy!]

Hi Circuit,
How are you getting on with this, please give us a update.

[circuit]

Hi. I have upped the power a little bit, now it is limited mechanically: variator does not lift the belt all the way up and I get a top speed of 55 km/h on flat, 60 km/h downhill and 50 km/h uphill. It should hit 100 km/h theoretically. Acceleration is similar to original 50cc engine with limiters removed. CA shows 5,2kW peak (60% of what I expect).
Now I have machined solid rollers from brass and they are 12,3 grams each. I guess I will have to make them a lot heavier. Maybe will have to do some lead molding... How did you squeeze 18 grams in there?

Also I have put clutch back in, as it was very hard to push it (same as car in first gear) and it was very sensitive on start, front wheel was lifting off the ground very easily. Removed the springs completely, but I still have a problem - it is slipping badly. I got a new clutch today, will try to fit it in tomorrow or so.

By the way, I got into local media news (sorry, lithuanian) - it's easy in such small country.
http://www.lzinios.lt/lt/2010-10-13/tra ... entus.html

[gwhy!]

Hi. I have upped the power a little bit, now it is limited mechanically: variator does not lift the belt all the way up and I get a top speed of 55 km/h on flat, 60 km/h downhill and 50 km/h uphill. It should hit 100 km/h theoretically. Acceleration is similar to original 50cc engine with limiters removed. CA shows 5,2kW peak (60% of what I expect).
Now I have machined solid rollers from brass and they are 12,3 grams each. I guess I will have to make them a lot heavier. Maybe will have to do some lead molding... How did you squeeze 18 grams in there?

By the way, I got into local media news (sorry, lithuanian).
http://www.lzinios.lt/lt/2010-10-13/tra ... entus.html

Sounds good and all going in the right direction , I bought some roller shells ( there are a little thinner than stock ) and cut pieces off of a 16mm bar into the correct length to fit inside the shells, also your rollers may be slightly smaller in diameter than mine .

Cool got in the news, I will run it though a web page translater and have a read, Thanks for the update.

[gwhy!]

I read the news article and well done, Keep plugging away at it and good luck , and please keep us upto date with future developments.

Cheers.

[gwhy!]

Hi Circuit,
I have just been having a look and yes my rollers are a bigger diameter than yours ( mine 19.5mm yours i think is 16mm ) so you will have a hard job upping the weight to 18g, but I have also noticed that you are using a strong torque spring this will also prevent the the belt ramping to the top of the variator ( try going back to stock ) but as you have already said this may cause a problem with belt slip. I had no problems with my clutch slipping once the rpm got upto around 3k ( starts engagement at 1k rpm ) a good clutch should allow around 1 to 2k rpm (motor) of slippage.
Cheers.

[circuit]

Thanks for advices, I think the same way. fitted new clutch hour ago and it is performing better. Done a few burnouts. Will try to solve roller problem somehow.. Unfortunately that red torque spring was there when I purchased it, so will have some additional troubles finding stock one.

edit:
yes, my rollers are 16x13mm and inside of the shell is 13x11 mm. If I made a mold from lead, it would weight 21.5 grams. But I'm afraid it will not hold a pressure and will deform.. So I need some kind of strong shell or ...
Or I could simply machine new 16x13mm rollers from solid brass/bronze and use them without inserts. Of cause they should be lubricated well. And I doubt that aluminum variator would last long... That would be 22 grams.

[gwhy!]

Thanks for advices, I think the same way. fitted new clutch hour ago and it is performing better. Done a few burnouts. Will try to solve roller problem somehow.. Unfortunately that red torque spring was there when I purchased it, so will have some additional troubles finding stock one.

edit:
yes, my rollers are 16x13mm and inside of the shell is 13x11 mm. If I made a mold from lead, it would weight 21.5 grams. But I'm afraid it will not hold a pressure and will deform.. So I need some kind of strong shell or ...
Or I could simply machine new 16x13mm rollers from solid brass/bronze and use them without inserts. Of cause they should be lubricated well. And I doubt that aluminum variator would last long... That would be 22 grams.

Personally I think you will be better sticking with your 12'ish g weights or go as heavy as you can but keep the shells ( or it will become very noisy and will ware the ramps in the variator double quick time ) try to get hold of a weaker torque spring this way you can get the max speed from the scoot, once you have that you can start upping the power by increasing the roller weights.

[etard]

Awesome project, these electric scooters through the variator are turning out to be a great platform toward a light motorcycle build. I am wondering what updates you might be able to share on your scoot Circuit? Any videos?

Also, I was wondering if you had thought about a lower KV motor such as the Collossus or even an Astro 3220. My Collossus motor has a KV of 75 so maybe more friendly to your rollers?

[richerson]

Great project, looks very professional.

[Teh Stork]

Uhm, why run the electric power through the variator? This is just losses (like 20-30%) when it comes to this application. Torque isn't a function of rpm on a electric engine, so using a variator as a torque converter is not needed Hall sensors will give you high bottom torque with the correct controller...

Looking to complete such a project myself. But I will have the engine direct drive to the rear wheel, regenerative braking and I'll keep the ICE intact to be able to charge the pack as I go. Also thinking of including a small engine at the front wheel. 3kW ish.

[Karolis]

^ He already done that. Just don't have time to write it.

[circuit]

^ He already done that. Just don't have time to write it.

Yes, I did.

Uhm, why run the electric power through the variator? This is just losses (like 20-30%) when it comes to this application.

Somewhat true.

Torque isn't a function of rpm on a electric engine, so using a variator as a torque converter is not needed

False. Power = torque*rpm. While maintaining the same torque on input shaft, you can run it on much higher rpm and higher power. That transforms to higher torque on output shaft.
My front wheel was lifting off the ground with variator and ~4kW of power. After variator was removed, it no longer does so.
I had my own (other) reasons to remove it.

Hall sensors will give you high bottom torque with the correct controller...

Sensorless motor in EV should be out of question.

Oh, and my scooter does have a regen.

[gwhy!]

Uhm, why run the electric power through the variator? This is just losses (like 20-30%) when it comes to this application.

Swings and roundabouts its, its no where near 20-30% losses through a variator ( by my testing ) but it is lossy, on the plus side you dont need such a big controller or motor so you dont get the massive current hit on hard acceleration. I think it balances out because a variator system is so efficient at what it does but not so efficent on how it does it. I would rather run a variartor system anyday over a chain drive or hub system.

[circuit]

Yeah, but one small issue: variator does not work with electric motor due to constant tension. The belt just does not climb up under load, even with tripled weights. This was the main reason to ditch it.
The issue also applies to NuVinci CVT hub. Both tested and dealt with by me.

[gwhy!]

Yeah, but one small issue: variator does not work with electric motor due to constant tension. The belt just does not climb up under load, even with tripled weights. This was the main reason to ditch it.
The issue also applies to NuVinci CVT hub. Both tested and dealt with by me.

Interesting I didnt run into this problem, the belt would climb right to the top of the front pulley. Did you try a different ramp profile or a weaker torque spring?

[circuit]

did you try it under load? I.e. while accelerating, not with lifted wheel spinning freely in the air.

[gwhy!]

did you try it under load? I.e. while accelerating, not with lifted wheel spinning freely in the air.

Yes I was riding it around on some "private" roads it cruised at round 35-40mph So i know the rpm from the motor was the limiting factor ( 8000 rpm) for the max speed, I did start looking into changing the gearing in the back wheel (for a higher top speed ) and also upping the voltage to get more rpm but because of the trouble I was having getting legal on the road the project sort of stopped ( but not forgotten ).

[circuit]

Yes, but how was the acceleration? Mine did take the speed, but acceleration was very poor and motor's power draw was only 1-2kW, as it was running on max rpm all the time. My power limit was at 6kW, but still it drew only 1-2kW due to maxed-out rpms. Especially this was seen in uphills: maxed-out rpms, 1-2kW draw and no speed.
After ditching the variator, acceleration is way better (except first 0-15 km/h). Goes ~70 km/h.

Anyway my project is on hold now as I am highly disappointed with chinese motor controllers (LYEN included). Few months ago I started my own company JSC Elektromotus, where we are designing EV-related stuff, such as advanced BMS systems and motor controllers. After our motor controller is built, I will register this scooter to be street legal. Hope to be first in my country.