Hello Electrified Miata Fans! We've been testing the Electrified Miata software in the go-kart. Bruce zipping around the cul-de-sac in the video. There is more work to do, but testing the go-kart is much more fun than other tasks like documenting the software or soldering batteries. Any python software developers interested in producing documentation in exchange for go-kart rides? Contact me. Join us in this journey to explore how to go fast using rotating magnetic fields.
Last weeks goals were:
1. Test go-kart
2. Clean up shop.
3. Move Miata back to my garage.
4. Start on battery balancing code when charging the go-kart.
To test the go-kart, we still had to hook up the main power. After discovery that heater hose is a poor insulator , I went to Home Depot and picked up some 'Seal-Tite' which is a good insulator and made for the purposes of protecting wires. After getting the main power connected, Bruce came over and we tested the go-kart. As you can see from the video, the go-kart scoots along in a sprightly fashion bringing that grin when a driver is presented with plenty of very controllable power. Remember the "more of that please" line from Ford vs. Ferrari? Oh Yeah!. The go-kart worked well but the BMS software process died. Time to work on the software, so we can learn more about these Miata battery packs.
There are endless tasks on the software side of this project. This week, we added battery temps to the display, sped up the battery data rate and added some 'debugging' logs.
Technical mode on:
I mentioned that the BMS process died, but didn't say how it died. Initially, I didn't know. I needed to capture this information. Python informs us with a 'stack back trace', which tells us the condition the code was in when an exception happened. I had to save the stack back trace in a file since there is no 'screen'. Saving the information in /tmp directory is fine, but requires that I modify how often the /tmp directory is cleaned up by Linux. It's that Linux system arcana that you vaguely remember exists, but the actual details take time to research and test.
Technical mode off:
In the process of testing the software, the kart just stopped. Was this a software issue? The web interface seemed unresponsive, but then came back to life.. Sigh, push the kart back up the driveway into the garage. After a reboot everything seemed OK. I made some software changes and did more testing. Again, the kart died. Push it back in the garage again. This time, the web interface remained dead. I took the top off the Pi box and none of the Pi's had power! The meter showed the regulator wasn't putting out any power. The regulator wasn't hot, but when I touched it, it came to life. It turns out the toriodal coil shown in the below picture had vibrated loose (marginal solder joint).
I ordered another regulator, re-soldered the coil and used some RTV to physically stabilize the coil. After installation, I was able to do some more testing. Oh Yeah! With all that testing, I'd drained the battery. It needed re-charging. After adding some charging connectors and charging, the kart was ready for car night.
In preparation for car night, Bruce and I towed the Miata back to my house. It is so light, that the two of us, with a running start, were able to push the Miata up into it's parking spot. When car night arrived, we unloaded the Leaf/Miata transmission and engine hoist and pushed those into the garage. Thanks to Mark Rogalski for letting us borrow that. With the hard work out of the way, it was time to start testing the kart! Joe hopped in first and started zipping around. His comment was something like "It goes plenty fast". Brian hopped in next. He managed to do a 270, topping
Joe's 180 and on his way back shouted something about 'There is a problem, shut it off'. Brian's sense of humor can be dry at times so I was unsure if he was joking or serious since he appeared to be having fun. Bruce spotted the fact that he was on the brake trying to slow the kart down and it wasn't. We shut the power off and the brakes were literally smoking hot. Turns out the connection between the throttle pedal and the throttle controller (a potentiometer) had come loose after twenty years leaving the throttle partially engaged, causing Brian some momentary consternation.
After re-alignment and reinforcing the connection with tie-wraps, we got back to testing. :)
The kart is relatively quiet but you can hear the rear wheels slipping on the pavement as the solid rear axle fights the kart when turning. You can also hear the tires make the sizzling noise as nearly everyone was doing 180's or more. It was hard to get any heat in the tires with the pavement being cold. My neighbor, Andy Bruner came out to see what all the noise was about. I introduced him to the group and he was given the nickle tour of making an Electrified Miata.
After the guys went home, my wife remarked I had that smile again. You know
the ear to ear, toothy grin with sparkly eyes? There was a giddiness from being able to spread joy to others and a sense of accomplishment. I really enjoyed seeing my friends, especially when they were able to have fun just playing around. Thanks guys for the smiles and fun memories.
Next weeks goals are:
1. Upgrade firmware on BMS boards.
2. Get battery balancing code working.
3. Install hardware switch to cut power
4. Start on logging and integrating a 120GA Solid State Drive (SSD) on the cart
As always, thanks for reading!
Bill likes cars that understand the 'go fast now' pedal