Design an ECU project?

True, I hadn't thought of the latency. The OEM traction control I'm familiar with has not been a performance feature, just for safety.

It was a hard limit I believe. It would randomly drop spark when the limit had been reached. It's similar to what you're planning in that it dropped the cylinder completely, just with spark instead of fuel.

Right, hence the low-speed-only use on OEMs.

MSD has a spark box all the 10.5W tire drag racers use that has a form of traction control in it. To set it up the racer logs a few good test runs and the box records the engine rpm curve. To affect traction control on a run the computer compares the current engine speed with the recorded curve and retards timing based on the amount of overspeed. There is a certain percentage of slip allowed before any timing is pulled to allow for improvement. These are 2000hp cars on 11.25 inch wide tires going 6.40s @ 230+mph. Obviously this is not anything you could use but it does demonstrate the effectiveness of retarding timing.

How do high performance OEM traction controls work? Lots of cars have selectable amounts of tire slip now. Seems like whatever method they use would be best unless it's too complex.

 

My Toyota Camry has a throttle body with two solenoids. I know, a Toyota Camry is not high performance, but that doesn't matter. The response time is not coupled to number of cylinders or rpm, but is simply dependent on the mechanical bandwidth of the solenoid setup. If a solenoid can throttle back air in the time it takes for a solenoid to shift gears in an F1 transmission, isn't that fast enough for traction control?

The traction control in the Camry is very smooth. When roads are snow-packed and I dump it it just feels like there is no power. It is certainly fast enough and I don't sense that the engine is missing at all so I suspect it is all done with throttling back air and fuel.

I suspect the same is true for the 360 and F430. Rifledriver would know?

That said, your engine doesn't have solenoid actuated TBs and the scheme you describe sounds reasonable to me. I am an ardent fan of the KISS philosophy if nothing is to be gained by doing something more complicated. Fun to think about though.

 

Can't be. The Competition modes on Corvettes and Ferraris work at speeds that would cook the brakes if they were the primary control device. Emission tests are not run under conditions that would induce tire spin.

 

Yes, you are correct. They would have to use it to establish a preset limit, like 50% off or all off.

I may have been wrong about the two solenoids. That is what my mechanic told me over the phone, but that doesn't explain how they do drive-by-wire, which would require a motor of some sort. I don't think it has to be a voice coil though. My first guess is that it might be a microstepped step motor, which would have sufficient bandwidth to give good throttle response as well as do traction control. I'll take a look myself this weekend.

 

The system most likely uses the ABS sensors, but not the brakes themselves, to sense tire slip, just like they sense wheel lock up for the brakes. Tie the ABS sensor input into the throttle and do with the throttle what ABS does for brakes, modulate to prevent wheel slip. Not too difficult to do. Their description states the system can modulate the throttle up to 100 times per second, much like an ABS cycles the brakes. Digital throttles allow a lot of tricks like this. The 2010 Corvettes also have a launch control mode, which uses the same system.

Taz
Terry Phillips

 

I just found this on the Internet:

On late model Camrys, engine torque is reduced by cutting fuel to individual
cylinders and by controlling transmission shift timing to supplement wheel
braking control.

These systems are very effective at reducing wheel spin and maintaining
traction on the road. It is obvious that extensive use of the traction control
may result in accelerated brake pad wear over the life of the car. However, the
cost of a new set of brakes is far cheaper than the cost of an accident.
Jeff Ness, Toyota Master Diagnostic Technician

The post was made 11 years ago so I'm guessing what was used then is not necessarily used now. Also techniques used for traction control when the intent is to prevent occasional slippage on snow/ice may not necessarily be the best techniques to use when 500 ft. lbs. of torque is being applied to the wheels on dry pavement. It sounds like manufacturers have used different techniques over the years - combinations of braking, air/fuel throttling, and fuel cut off to individual cylinders.

I wonder what the 360 and F430 use. Rifledriver?

 

Overkill. I found a used throttle body from a drive-by-wire Toyota Yaris on the Internet and it has two simple servo motors.

 

I was wondering the same thing. Why reinvent the wheel when it already exists. The M880 will handle anything you can dish out.

 

Hey Mark

Just came across this while researching another project.
Dont know if you looked into this or not.

http://wolfems.cart.net.au/details/635751.html

Richard

 

I have developed and designed two automotive ECU's from scratch... well, one was a piggyback controller that adjusted spark and injector curves, and the other was a standalone ECU. I also built, from scratch, the transmission and traction control computer for my racecar using a different transmission.

Here's what I can add...

If you can buy something off the shelf, even if it costs thousands of dollars, it is an *infinitely* better idea than doing it yourself. If someone offered me $50,000 to design them an ECU, I'd turn it down. I'd turn down 5 times that. Just because the sheer amount of work involved is gigantic.

I know it sounds somewhat simple on paper, but so does the space shuttle It's easier when you are dealing with a single setup (one engine on one car) vs. trying to make something more widely applicable, but there are always thousands of hours spend chasing and fixing gremlins that are difficult to catch and resolve. I would go with Motec first, even if it required jury-rigging something up. If that was not gonna work, I would go with something like the AEM universal system and jury rig it however necessary to work.

If you take a company like Motec or AEM, they will sell thousands of units at multiple thousands of $$'s each. When you do a one-off, you are really doing pretty much the same amount of work as they do when they develop a kit, but you don't have the wealth of knowledge these guys do to get you started at 75% completed instead of 0% completed.

If you decide to go it alone, I'd be happy to answer any questions you have.

 

Mark, get ahold of Neel at http://www.apexspeedtech.com/ Lookie, there's a second hand M800 available. http://www.apexspeedtech.com/phpbb/viewforum.php?f=2&sid=ec8266f27ed1be49f9ff41ea1afe4765

 

Those are some nice options. Do you need two of these ecus? How much do they run?