1992 Kawasaki KLR250 Build Thread

[KLRbee]

Hello All,

I'm doing a conversion of a 1992 Kawasaki KLR250 to electric. This was chosen as I found one with an iffy engine for fairly cheap, titled and road legal. I'm going to convert it to a road legal dual sport. Current plan:

- Sevcon Gen4 controller
- Motenergy motor - ME0913 probably
- '60v' nominal battery (18s Chevy Volt cells)
- 1k onboard charging (probably MeanWell power supplies like a 2013 Zero)
- BMS to be determined
- Pot box throttle using a hall effect TPS
- Really cheap made in China '250 watt' DC-DC converter, which will be load tested for stability
- Other decisions yet to be made

Current to do items:

- Get the existing engine running, take video, pull it, and sell it - funds the project, plus gets it out of the way. It is currently hard to start due to bad valve adjustment.
- Model the bike and place components. It's going to be tight, this is part of the reason for a 3kWh battery size instead of bigger. It's a small bike.
- Figure out gearing. Reading suggests that anything under a 12t front sprocket will be very hard on chains. It looks like I need to hit about a 5.5:1 final gear ratio which should be doable with a 13t front and a 72t custom rear. If I can use a 10t front the rear can be a much more reasonable size. If this worked well I suspect Zero would have done it, and they haven't.

 

[KLRbee]

Saving this message for later if needed.

Here is the Google Drive folder which I'll be putting images and files related to this in: https://drive.google.com/folderview?id=0B_4YMWOqAX8AWFNwOGl5X25VbDg&usp=sharing

 

[KLRbee]

Here is the bike as originally purchased.

 

[KLRbee]

View attachment 1



Here is the original wiring diagram and my current modified wiring diagram. It does not yet include the Hall effect throttle I'm planning to use. Right now it is set up for a pot box.

 

[DanGT86]

Looks like a cool project. You seem very well organized with your wiring diagrams. Good to go into stuff like this with a clear plan an and realistic budget.

I am not an authority or an experienced e-moto builder, just a guy who reads a lot and shops for parts I cant afford. Here are a few thoughts that jump out at me when reading your plan.

-sevcon controller. I have heard lots of horror stories about programming sevcons for specific motors. I think you want to buy that as a kit with the motor from a supplier that has programmed them to get along. You may just want to check on that before ordering.

-leaf cells. great bang for the buck, but they are pretty big. Being that they come in 60ah groupings, you end up stuck with a large metal tin. They might be hard to fit in the tight confines of a moto frame. With current 18650 cell technology you could make smaller packs that fit better in a few places. You could even buy multiple ebike packs if they are on sale and series parallel them together to get your 3kwh. Also look at chevy volt cells. With the water cooling ears cut off the bottom of the plastic cell separators they end up being pretty compact. That width seems like the biggest issue in a motorcycle frame. With leaf cells at 60ah you are looking at almost 6kwh by the time you get 96v. With volt cells you would be around 4kwh at 96v. If you are going to take them apart then you can make smaller packs. The volt cells are already configured to make an easy 24s3p pack of about 86-88v and 45ah or 4kwh.

-Steel frame. Love it. That is a big help when custom building. Way easier to weld and cut without inducing structural failure like aluminum. I say that not knowing your welding skill level. You could be an aluminum welding master.

-Common bike with lots of parts availability. Also a big plus. Probably tons of options and experience out there for suspension and brake upgrades.

Sorry if you weren't asking for opinions. I figured since you were posting on a forum you were open to them. Should be a cool build. I look forward to watching it since its very similar to a build that's been stuck in my head for a few years.

 

[mistercrash]

Leaf modules is a definite yes. Get the latest generation modules that are not closed like a sardine can. If needed you can take the aluminum covers off and build a custom box in which the cells will be protected and suspended in the box by the four threaded rods. I think agniusm has a few pics of the way he did it. You can go see the elmoto forums, lots of great info on motorcycle conversions. http://www.elmoto.net/forum.php

 

[KLRbee]

-sevcon controller. I have heard lots of horror stories about programming sevcons for specific motors. I think you want to buy that as a kit with the motor from a supplier that has programmed them to get along. You may just want to check on that before ordering.

At this point I'm set up with an IXXAT and DVTC and I have a sample DCF from John over at Motenergy. I'm getting ready to build out the motor/controller/battery setup on the bench to program and test. This is going to be a headache. It also seems pretty doable.

-leaf cells. great bang for the buck, but they are pretty big. Being that they come in 60ah groupings, you end up stuck with a large metal tin. They might be hard to fit in the tight confines of a moto frame. With current 18650 cell technology you could make smaller packs that fit better in a few places. You could even buy multiple ebike packs if they are on sale and series parallel them together to get your 3kwh. Also look at chevy volt cells. With the water cooling ears cut off the bottom of the plastic cell separators they end up being pretty compact. That width seems like the biggest issue in a motorcycle frame. With leaf cells at 60ah you are looking at almost 6kwh by the time you get 96v. With volt cells you would be around 4kwh at 96v. If you are going to take them apart then you can make smaller packs. The volt cells are already configured to make an easy 24s3p pack of about 86-88v and 45ah or 4kwh.

Yep, you and I have come to the same conclusion. I'm going to actually go for a '72V nominal' setup with Volt cells. It looks like I will be able to sneak them in the frame as-is. It's fewer Ah than the Leaf cells, and smaller, but more vulnerable. Probably I will have to set up some guards and skid plates. This will be a good spot to use expanded steel and aluminum plate as they are easy to work with. If it doesn't work I got the Volt pack for cheap enough, I'll scrap the idea and build 20p4s modules out of Samsung 25R 18650 cells for twice the cost and a little bit less capacity.

-Steel frame. Love it. That is a big help when custom building. Way easier to weld and cut without inducing structural failure like aluminum. I say that not knowing your welding skill level. You could be an aluminum welding master.

-Common bike with lots of parts availability. Also a big plus. Probably tons of options and experience out there for suspension and brake upgrades.

I'm not a great welder, and I did want to pick a common bike. Both of these were totally thoughts I had in mind when choosing the bike. I'm into the bike for very little money, so if I end up changing my mind I can flip it and likely lose nothing in this detour. Meanwhile I am building up and testing my drivetrain. Since I might be at lower voltage and ultimate power is not a concern, I'm using the ME1012.

 

[maydaverave]

Nice I have been thinking of going a similar route with a 1979 yamaha xs400. Lots of details on motor and controller set up and installation please. I'm thinking either the motor and controller your using or a max-e adapto with a revolt 160.

 

[maydaverave]

What I'm wondering about a motor mount is why can't I just use the transmission case? It's firmly attached to the bike. It's got a hole right where my sprocket needs to go. Hell I could get the liquid cooled mars and use the trans case as a radiator. Just wondered if you thought about it and why its a bad idea?

 

[DanGT86]

Transmission case takes up much needed space in most cases. Some have used 2stroke transmissions fkr electric conversions because the oil in the trans is separate from the motor crankcase as its ow. Sealed unit. You can use it if its easier.

 

[maydaverave]

Transmission case takes up much needed space in most cases. Some have used 2stroke transmissions fkr electric conversions because the oil in the trans is separate from the motor crankcase as its ow. Sealed unit. You can use it if its easier.

Interesting, we need a lot less space than we used to. Everything keeps shrinking. I guess if needed one could cut all the excess off and just keep enough for a structurally sound mount.

 

[KLRbee]

If the transmission were not a stressed member I could see machining it away to use it as a motor mount. In this case since the engine is in decent shape I am planning to pull it and sell it to fund part of the conversion cost. The KLR250 is a pretty good engine but as with all dirt bikes they need top ends occasionally.

 

[KLRbee]

Updates: I have the Sevcon running the motor on the bench. The harness is mostly built. I forgot to put a circuit in for brake regen so I'll add that to the wiring diagram and perhaps put one more wire on a digital input for reversing which I can cap off until I decide to add a reverse switch. Since I'm planning to run the forward switch to the kill switch I can add a switch for reverse that will only work when the kill switch is off. The seat switch is going to be hooked to the side stand.

Mistakes I've made so far:

-tried to use a 12V Gigavac GX11 contactor. No dice. The coil resistance is too low so even if set up for a 12V contactor the Sevcon issues an error that the output is shorted. Switched to a White Rodgers 586 style 48V contactor which works great. Still have to experiment with what level to economize the coil to as it's probably lower than I think it is.

-needed to use different firmware on the Sevcon as it came with AC motor firmware.

-didn't set up a load for the motor. This means setting gains is hard because even just a little throttle application and it tries to toss the bench over. Should be easy to solve once I have it on the bike and have the inertia of the drive train available.

-bought 6 pin connectors to interface with the bike but not smaller ones for building breakout points in my harness. Other ones on the way, but since they all come from China they take weeks to arrive.

-figured I could use a Meanwell 12v power supply brick as a DC-DC, got one for cheap. No dice so now I have a wasted 12v brick sitting here.

-bought a battery voltmeter that needs 12v to show voltage. Not useful as I want to put it right after the conctactor and have it not depend on the DC-DC to work.

So here's the good stuff. First a picture of the current setup:

https://drive.google.com/file/d/0B_4YMWOqAX8AUmhLV1FrRDdZTUE/view?usp=sharing

Second a video of a drive test:

https://drive.google.com/open?id=1dRFs1F-884HcnLAKMPzqADVIeKt6iBy3Cw

 

[KLRbee]

No big updates but wanted to keep track of the thread. Using my free wrenching time lately to clean out the garage and get some of the other motorcycle and car projects out of the way, so the drivetrain has just been sitting on my bench ready to go. I still need to solve DC-DC and figure out what is going to mount where as I know it's going to be tight, and will need to figure out how much protection to put around the battery.