Why Your FSAE Team Should Use FreeEMS!
Posted: Wed Jul 10, 2013 10:49 am
Reasons to use it:
1) Cost report, replace 1000+euro with 100euro, without any loss in performance of your engine
2) Customisations, add custom logic to control aspects of vehicle within the same box saving weight, space and keeping complexity low.
3) Control, having the source code and superior data available for diagnosing issues during dev or at the track = gold
4) Right now, you'll be ahead of the curve and at an advantage due to the above, sooner or later, everyone will be doing it
Reasons not to use it:
1) Learning curve steep and high :-(
2) All tools are still alpha-quality :-(
3) Sweet sponsorship deal signed :-p
4) Fear of the unknown and/or dark :-p
EDIT 2016:
Still, the 2013 Elith Racing Team, electronics/engine control lead by Jonas Bromo, managed to pull it all together, and put a competitive car on stage in England as a complete package.
The ECU wasn't the weakness, in the end. Weaknesses were:
1) Brake bias too rear because team member who didn't care and eventually left designed that. Same calipers, pads, and disks out back as out front, balance bar fully stacked to the front was not enough to make it neutral or front biased. Took too long to get through brake check, missed out on drag race which would have been a winner.
2) Sprocket not locktighted into place on gearbox, fell off 3 laps into circuit session, took car out of running.
3) Wheels insufficiently strong, broke in Finnish event, no major harm done.
4) Apparently the chassis was pretty flexible.
5) The pneumatic shifting ran out of air too quickly
ECU wise, the biggest challenge was for a green set of students to correctly wire, install, and tune the system. That's why I flew over to give them some tips and pointers as they went. They learned a lot in a short space of time and carried it through to success.
There were minor niggles about the ECU and tune:
1) Below 2k not properly tuned, high and unstable idle
2) No tip in enrich meant response wasn't perfect at low RPM
3) Key-on priming pulses on a car that gets turned on and off a lot without starting = blown up inlet manifold, and some hilarious moments
4) 6 channels meant staged injection wasn't an option, so power was falling in the upper end of the rev range (though they had not planned to include 8 anyway, and it could have been run as 2, 2, 2.
5) Fan control that required the engine running (charging) to operate would have been good
6) Firmware updates without losing tune etc were a bit painful/difficult/scary
7) IAT was brass, and on the roll cage in the sun = heat soak, leanness, etc
Of those:
1, 7 was insufficient setup/tune
4 was engineering understanding, mostly, code secondarily, top 10% of engines, power wise, despite this.
2, 3, 5, were code
2, 5 got solved, 3 got worked around.
6 system maturity/insufficient time to do things the slow/hard/right way given the system
Despite all of that, the car would have done really well if not for the brakes, sprocket, and later, the wheels. The starting and off idle shittiness didn't really pose a problem, it was just a bit annoying for the driver/pit crew.
Jonas implemented:
- 2 step launch
- pneumatic paddle shift box control
and a few other things I've forgotten, in about 200 lines of Fred-reviewed code.
If you planned this a bit earlier and did things by the Fred-book from the start, some of those niggles wouldn't have been an issue.
Again, the ECU issues were the least of their worries, the device and code performed flawlessly within the boundaries of what was available, and never let them down.
First two reasons lists written at the time of the original post. All latter content written tonight at the time of this edit.
1) Cost report, replace 1000+euro with 100euro, without any loss in performance of your engine
2) Customisations, add custom logic to control aspects of vehicle within the same box saving weight, space and keeping complexity low.
3) Control, having the source code and superior data available for diagnosing issues during dev or at the track = gold
4) Right now, you'll be ahead of the curve and at an advantage due to the above, sooner or later, everyone will be doing it
Reasons not to use it:
1) Learning curve steep and high :-(
2) All tools are still alpha-quality :-(
3) Sweet sponsorship deal signed :-p
4) Fear of the unknown and/or dark :-p
EDIT 2016:
Still, the 2013 Elith Racing Team, electronics/engine control lead by Jonas Bromo, managed to pull it all together, and put a competitive car on stage in England as a complete package.
The ECU wasn't the weakness, in the end. Weaknesses were:
1) Brake bias too rear because team member who didn't care and eventually left designed that. Same calipers, pads, and disks out back as out front, balance bar fully stacked to the front was not enough to make it neutral or front biased. Took too long to get through brake check, missed out on drag race which would have been a winner.
2) Sprocket not locktighted into place on gearbox, fell off 3 laps into circuit session, took car out of running.
3) Wheels insufficiently strong, broke in Finnish event, no major harm done.
4) Apparently the chassis was pretty flexible.
5) The pneumatic shifting ran out of air too quickly
ECU wise, the biggest challenge was for a green set of students to correctly wire, install, and tune the system. That's why I flew over to give them some tips and pointers as they went. They learned a lot in a short space of time and carried it through to success.
There were minor niggles about the ECU and tune:
1) Below 2k not properly tuned, high and unstable idle
2) No tip in enrich meant response wasn't perfect at low RPM
3) Key-on priming pulses on a car that gets turned on and off a lot without starting = blown up inlet manifold, and some hilarious moments
4) 6 channels meant staged injection wasn't an option, so power was falling in the upper end of the rev range (though they had not planned to include 8 anyway, and it could have been run as 2, 2, 2.
5) Fan control that required the engine running (charging) to operate would have been good
6) Firmware updates without losing tune etc were a bit painful/difficult/scary
7) IAT was brass, and on the roll cage in the sun = heat soak, leanness, etc
Of those:
1, 7 was insufficient setup/tune
4 was engineering understanding, mostly, code secondarily, top 10% of engines, power wise, despite this.
2, 3, 5, were code
2, 5 got solved, 3 got worked around.
6 system maturity/insufficient time to do things the slow/hard/right way given the system
Despite all of that, the car would have done really well if not for the brakes, sprocket, and later, the wheels. The starting and off idle shittiness didn't really pose a problem, it was just a bit annoying for the driver/pit crew.
Jonas implemented:
- 2 step launch
- pneumatic paddle shift box control
and a few other things I've forgotten, in about 200 lines of Fred-reviewed code.
If you planned this a bit earlier and did things by the Fred-book from the start, some of those niggles wouldn't have been an issue.
Again, the ECU issues were the least of their worries, the device and code performed flawlessly within the boundaries of what was available, and never let them down.
First two reasons lists written at the time of the original post. All latter content written tonight at the time of this edit.