I made something!

"roflcopter" (roflroflroflcopter) 04/21/2014 at 11:33 • Filed to: None

3

10

What you see here is an ECU, that I built, for our Lemons car. Preliminary tests show everything working the way I want it to, gonna take it with me to school this week and throw it through it's paces on the bench, then it'll be time to test it on the car. I am excite.

mazdaspeed2 > roflcopter 04/21/2014 at 11:38

0

That's awesome! I am working towards an electrical engineering degree right now. What went in to building it?

roflcopter > mazdaspeed2 04/21/2014 at 11:42

1

Well the board I'm using is an ARM based Arduino clone, all the ease of programming with some real power under the hood. Most of what went into getting this thing together was a lot of back of the napkin math to make sure I wouldn't be overtaxing the hardware(it's amazing how slow engines are in computer terms it turns out) and then making a circuit to be used to actually drive the injectors. They draw about 2A max current, which there is no way that logic level outputs from the board would be able to drive that. The code itself is actually really basic, I'm using Alpha-N fuel calculations to begin with, it'll be speed-density by the time I'm done though. All of the scheduling is done through hardware timers and interrupt channels(why I wanted to go with an ARM chip) and the fuel calculations are just done in the main loop.

This thing is REALLY rough, but I'm just trying to get it working, then I'll improve from there.

Demon-Xanth knows how to operate a street. > roflcopter 04/21/2014 at 11:55

0

Make sure your circuitry can handle the noise from the ignition and injectors without going all whack. It seems trivial, but always seems to trip everyone up.

roflcopter > Demon-Xanth knows how to operate a street. 04/21/2014 at 12:05

0

The injector driver circuit is fully clamped and I'm running a more than robust enough voltage regulator to get a solid 5V to all of that. The board I'm using is actually made for industrial/automotive applications and can handle 9V-30V unregulated input. The injector driver circuit is two layers away from the IO pin on the board as well. Hopefully it'll be fine, I plan on test it on the bench with function generators and stuff and simulate some noise and all until things start to break down, then I'll know what my upper limits are.

BJ > roflcopter 04/21/2014 at 12:47

1

Very cool! Are you planning on sharing your schematics and code, or is it an industry secret? ;-)

Demon-Xanth knows how to operate a street. > roflcopter 04/21/2014 at 12:48

0

Good plan. It's easier to design with it in mind than it is to have an afterthought duct tape fix.

mazdaspeed2 > roflcopter 04/21/2014 at 12:57

0

That's really cool. Some of that is a bit over my head right now, haha, but it is really cool to see a practical application in something I'm interested in. Best of luck to you!

roflcopter > BJ 04/21/2014 at 13:05

0

I'll share all of it once I have something worth sharing, it's definitely not going to be a plug n play solution though, understanding of code and circuits will be needed.

roflcopter > Demon-Xanth knows how to operate a street. 04/21/2014 at 13:07

0

One of the reasons this has been so slow going is due to the planning that went into it. I'm sure I've overlooked things but hopefully nothing big.

BJ > roflcopter 04/21/2014 at 13:25

0

From an educational point of view, I think it'd be wonderful if you could share. Out of curiosity, why did you have to build this? Was the original part missing?