EV Automatic Gearbox Controller Project
Thought I'd start a thread on a new project I have just begun. Plan is to develop an open source arduino based controller for automatics. The ZF 5HP24 is my first victim as I need it for a new conversion project but also plan on looking at the 5HP18 and the CAN bus controlled 6HP26.
First installment is a general intro and look at the flexplate coupler design : http://www.youtube.com/watch?v=7tl9aOdjSoQ |
Part 2 : Looking at the range switch and working out basic software to control the valves.
http://www.youtube.com/watch?v=Pr8GBK8dR0w |
subscribed
|
|
very cool, do you forsee doing away with the torque converter before it is all sorted out?
The efficiency curve is my main reason for asking... http://jmclutch.com/site/sites/defau...e_OutInRPM.png |
1 Attachment(s)
First look at the controller hardware
|
More work on the adapter plate. Now finished and ready to be mated to the gearbox. Just hope the damn thing fits ....
http://www.youtube.com/watch?v=oiZZWdsaZGU |
|
wow! that trans is massive next to the motor. Do you already have some idea what the electrical pins do for controlling it?
|
Why not just use direct drive and eliminate losses?
|
It was stated early on in the first video that part of the diligence here is in verifying claims of inefficiency in an automatic-electric conversion. So props for that (though they are pretty well understood). A little amusing that spinning this transmission unloaded at that speed would burn out my only EV motor in short order :) (yes it is a bike)
|
A free spinning electric motor uses very little power. I don't see how it would burn your motor out other than normal wear and tear.
|
with the transmission in neutral his motor was drawing 64 amps @ 24 volts. Since he isn't dictating, you kinda need to watch the videos to play along here :)
|
sorry, I'm stuck at a silent computer.
|
$5 earbuds from walgreens?
Quote:
|
Just a note since other people are doing the same sort of thing as the position switch, to minimize the possibility of fudgy readings, using the midpoint between readings is useful:
readings sorted by value and midpoints: Code:
n 643 Code:
//map the position switch resistor reading to an actual position |
Thanks for the comments guys. Working on the code now. Lot of this is gonna be trial and error as the finer points of how the pressure regulator valves are applied is not well documented. Nor is even if the valves run at 5v or 12v. So I'm starting at 5v. More updates soon.
|
Sorry if this is obvious, but if you know of an existing vehicle with this trans, then maybe you could borrow it and scope the waveforms? With a ready made adapter harness no less.
|
Hey damien, maybe this might help.
Its a link to the 2003 range rover service manual, which uses this same transmission. More detailed on the transmission chapter, than the regular 5hp24 manual. :rolleyes: cant post a link with under 5 posts to my name. Google search " 2002 range rover workshop manual" |
That chart P-hack posted gave me an idea: Just having the torque converter between the motor and differential. (One that has lock-up preferredly)
|
First run of the 5hp24 with arduino control :
https://www.youtube.com/watch?v=MLm0TP9oWfk |
Wow, you don't wast any time :)
2 observations: 1. Again sorry if this is obvious, but since this is trial and error, can you use the various speed sensors to ensure the clutch packs engage fully? i.e. clutch pack input rpm/output rpm = a known gear ratio. 2. Is line pressure rpm dependent, and does that affect pwm duty cycle? 3. when you disabled the valves in reverse, the output shaft doesn't look like it changed direction. Is that "normal"? This is interesting also because a lot of people would like to make their ICE automatics behave more like a manual if they could do so short of a transmission swap. |
@P-hack
1. Yes that is the plan. Once I have the LM2907 circuits built and interfaced to the arduino I will have rpm readouts from the turbine and output shaft. Then its simply a matter of checking the ratio to ensure the intended gear is engaged. 2.Line pressure is maintained constant mechanically by the pump volume control valve between 500 and 6500rpm. PWM control of the EDS1 valve allows it to be varied electronically. As the arduino uno has only 8 bit pwm this gives me 255 levels. Good enough I reckon. 3.yes. All that should happen in reverse is that the line pressure defaults to max when power is lost and the pump loads up slightly more. So this afternoon I wrote some code and added two shift buttons to the controller for up/down shift. Fired up in neutral and selected drive. Went into first no probs. Shifted to second and the box locked up:eek: into third , fourth and fifth no probs so looks like i have some more programming to do!! :) |
|
|
All times are GMT -4. The time now is 08:15 AM. |
Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
Content Relevant URLs by vBSEO 3.5.2
All content copyright EcoModder.com