I just got done trying different stuff for the throttle, since it has a bit of a thunk when taking off. I believe the thunk has to do with my coupler not being what it could be. Maybe my spider is getting worn out. or maybe there is some play somewhere else in the mechanical parts of the car. I can't wait for the others to tell me how the various throttles feel to them. I sort of regret the bucking bronco stuff that I did below now:
Hey Jay! Kd is 0, Kp is 9/16 and Ki is 149/32000. hehe. Sort of little, huh. I could make Ki larger but it was converging so nicely, and I just quit testing. I ran out of time. Also, the car started rocking back and forth pretty violently, so I thought I was going to ruin my coupler (the e-brake was on). I was testing a bunch of Ki's. I'd make a throttle step from 0 to 50, and then from 50 to 0, increment Ki a bit, watch on the laptop what was happening, and then close my eyes and pretend that I was on a bucking bronco!
Hey Nikki! I haven't heard of anyone doing a UK parts list. It was brought to my attention that at least 1 component wasn't RoHS compliant (had Pb in it), so we should change that. There are also a few minor changes needed on the control board, but I still need to start from scratch I think (I don't have the software that made it in the first place) and do the layout in expressPCB or another free one. Or anyone else want to change the layout? hehe. I've been sort of busy with school and getting the throttle perfect on my crappy $40 car, which may have mechanical problems that I'm trying to fix with stupid code! aaahh!
dcb: I DEFINITELY want to add an RPM sensor. It would make trivial a lot of difficuties I have had. I don't think the clutchless thing will be as much of a problem anymore, though, because I added some throttle ramping code. The PI loop is still super tight (current chases throttle better than a cougar chases it prey!), but now the throttle changes more gently. That seemed to take care of the shifting difficulties.
MCUDogs: It looks really beautiful! Comments: I think we can now get away with 4 holes for each of the 3 bus bars instead of 5 for 2 of them. I just used 5 before because the free bus bars I had already had holes at 1.5" increments, so I wanted to use those. 5 definitely doesn't hurt anything, though! It gives VERY good contact between the heat spreader and the heat sink. The holes between the capacitors should be perfectly in the center so as to be as far away as possible from the edges of the capacitors since the holes will be vias, and there needs to be room for soldering 10 gauge wire (which gets pushed into the holes).
Are those the actual dimensions for the TO-247 parts? It would be really nice not to have to bend the leads to fit them into those holes for the mosfets. It looks like you took care of that problem!
We could make the gate lead through-hole also. If we did that, then we would have to get rid of the tiny hole on the gate pad (which was for the gate resistor) and just solder the gate resistor to the other side. We would have to make the pad on the back side larger, and would have to move it so that the gate lead of the mosfet would go right through it. I think that would be an improvement (but a minor improvement). Thank you for all your hard work!
EDIT: MCUDOGS, I think those white borders around the 5/16" holes on the bottom side shouldn't be there.
EDIT AGAIN!: MCUDOGS, and if you really want to go crazy, on the bottom side if you move up the gate resistor pads, you could move up the B- 5/16" holes too, since B- bus bar gets attached at the bottom! That's one of the things I wanted to do different next time.
Last edited by MPaulHolmes; 07-20-2009 at 12:04 AM..
|