1992 Suzuki GS500e
Hey, I'm converting a '92 GS500e. I bought the bike non-running with a clean title off Craigslist for $300. It was in great shape except for the engine bit. Nice, upgraded brakes. The perimeter frame of the GS appealed to me because you can put pretty much whatever you want into it. For me, that consists of:
24s2P Headway LiFePo4 pack. 76.8V 30AHr. 300A continuous discharge
Kelly KLS8080IPS sine controller- 96V 500A
The "motor" is a modified heavy equipment alternator. Rip out the rectifier and drive the field winding separately and you've got a 3 phase synchronous AC motor. I've done this before- some coworkers and I build and race little alternator-powered go karts.
Misc- Albright contactor, Kelly DC/DC (12V, 35A), Domino throttle
Finish bottom-balancing the pack. I'm sure there are plenty of opinions on this here, but that's what I'm doing, at least for now. Once the cells are all discharged, I'll assemble and charge the pack.
Weld the last bits to the frame. Yesterday I welded in the new motor mounts. I need to put in the controller mounts next.
Mount everything up, make some harnesses, and test ride! Then do everything else.
Here's a pic of the rolling chassis with the battery enclosure mocked up in it:
Post Thanks / Like - 1 Likes
Cool! I am interested in the process of converting the alternator and how the performance is. Do you have more information on that somewhere? What's the efficiency like?
Also, I top-balance my Leaf cells and use no BMS. Never had a problem.
The conversion is fairly easy- rip out the rectifier and regulator and bring the three windings out of the case. The rotor is an electromagnet instead of a permanent magnet, so that has to be driven with some sort of constant current (current = field strength). I designed a little bang-bang constant current buck converter to do the job. From there, you can use a BLDC controller to drive it, no problem.
Originally Posted by nedfunnell
As for efficiency, it's hard to say. You have the static losses of driving the field coil. In my case that's about a constant 75W. Non-trivial, but not awful in a 20kW system. Without some sort of dyno it's hard to say what the efficiency really is. The alternators perform really well in the little cars. The nice thing is they're designed to run in hot environments (engine bays) so they're thermally overbuilt.
Glad to hear your bottom-balanced Leaf rig is fairing well.
Here's my discharging rig:
Three Chinese constant current dischargers (with LV shutoff). 3 sets of 16 cells in parallel, connected with copper strap for grounding water pipes. The dischargers can only do 10A, so it's slow going. I'm 17 hours in and only at 3.22V!
Once I hit my LV shutoff of 2.75V, I'll put the whole pack in parallel and let it settle before I build it up and charge it. Soon...
Post Thanks / Like - 1 Likes
Batteries are discharged. All batteries sat in parallel for a few days. Motor and controller mounts are fabricated and welded into the frame. Next, I need to make a few more bus bars, tin all the bus bars, finish building the pack and then charge it. In the meantime I'll start making cables and harnesses and I may very well be doing a little test ride in the alley soon... pics (video?) forthcoming.
The only thing I haven't completely figured out is how to mount the main fuse and shunt such that they're protected and safely isolated. I'm thinking just a nice plastic enclosure. I also need to fab some sort of tray for all the lesser electronics- fuses, relays, etc. It will likely be a plastic or aluminum sheet.
I made lots of progress this past week. Battery pack is built. Fuse/shunt box is built. All cabling was made. DC/DC is wired up. Contactor relay setup is wired up. Basically, everything's ready to test. I tried to give it a first spin last night but I discovered that there's some sort of issue with my field drive (a constant current supply I designed to energize the field coil on the motor). I tested it at 30V but I had not yet tested it at Vbatt. I brought it to work today to troubleshoot, hopefully I can figure it out. On the plus side, everything else work and I was able to talk to the controller with the Kelly software. Progress!
Worst case, I'll hook up a small battery to the field coil just for testing and figure out the field drive. Hopefully I don't need to redesign it...
Will post more pics soon!
Ok, I'm a bit frustrated.
Everything is wired up an ready but I'm having trouble with the Kelly "auto identify" function. I started a thread on it in the "controller" section, but basically the auto identify routine does its thing but then never seems to complete? It's supposed to throw a "reset error" when it finishes but I get no response good or bad. Anyone know anything about this? I'm in contact with Fany at Kelly but it's slow going what with the time difference.
Hi - I am also using the Kelly KLS8080i - assuming you changed the Identification Angle to 170? After it does it's thing and you power off / on the controller, does the Identification Angle now show 85? If it does, it should be good to go.
Does the motor move back and forth a few times whilst it's Identifying?
Yes, I write 170, reset the controller, it does its little routine (appears to be stepping through a full 6 commutations)...and then nothing. ID angle doesn't go to 85, there is no information on the monitor screen. I've let it sit and waited 5, 10, 15 minutes and nothing. Strange!
Originally Posted by mcress
According to Fany, all my motor parameters are correct so without more under-the-hood info I'm at a loss as to what to do next.
OK - does it show 85 in the identification angle once you have powered off / on the controller? Does the motor not run at all?
One other thing - are you using Android to program the controller or PC? I was (and still am) unable to do certain things in android (they just don't work!) whereas the same things I can do in the PC version just fine. I need to speak to Fany about that also....
When I power on/off, it comes back showing 170 and starts trying to identify all over again. It will do this ad infinitum unless I manually write 85 before I reboot.
I'm using the PC version.