Megasquirt Traction Control
Turbo Mini Forums

cheers rod pm on its way,

some form of traction control is something i would really like to get working, unfortunately i would not know where to start learning even the basics of the code so thats a long way off for me atleast

im planning of building/testing a 3rd circuit using the knock detection menus to control spark retard coupled with interrupting the boost control solenoid and possible switching maps completely

it may be a fudge but it might just work

i would like to use the knock detection for detecting knock aswell but im a bit sceptical on just how accurate that would be with an ohv pushrod engine and an integrated gearbox so its not somthing i would miss (knock sensor) but would be nice to have

i think you would prefer a more software related approach where as what i have in mind is a more hardware based method, but my findings either way will be updated onto here

Yes i moved to the darkside

Instagram @jdm_brett

Brett,

I'm all for hardware solutions with minimum use of the software/firmware if it can be done, and usually rely on others to write the software.
Electronics is just a hobby for me, my core qualifications are all in mechanical and heavy electrical engineering and making a PCB is just CAD now thanks to the Far East.

I don't think it will be too difficult to get the software to work with later versions of the MS-Extra code, I guess it's just low on the priority list.....

If you are thinking of building your own circuit for knock control, I'd recommend KiCad (for PCB design) and iTead to get some very cheap prototypes made up.

Both those recommendations came to me from Jean but I haven't looked back.

Schrödinger's cat - so which one am I ???

Hey guys, I just thought I'd chime in here. I hope my interloping is not unwelcome, but I think you've got a really great technically focused community here. I actually got a copy of 3.0.3r with the diff file merged in from Patrick, I believe. Much appreciated.

At any rate, looking at the data sheet for the IC from the German paper, a tweak from their setup should allow the use of ABS sensors (square wave output) for the speed sensing at each wheel. By tapping in to those, I'm hoping to construct a system for my car (build here: http://www.r3vlimited.com/board/showthread.php?t=205514 )

I'm not sure that the ABS sensor solution will help you guys at all, but the schematic modification is fairly simple and I can post it for comments if there is interest. I haven't hooked it up yet, since my engine is in pieces currently.

The PCB is really, really awesome, btw. If I wasn't using a tweaked circuit in my own application, I'd be all over that. Great stuff.

Hi lurker27, just had a quick glance at your build thread, looks like the traction control will have alot of work to do

any chance i can have a copy of 3.0.3r i have never actually tracked one down

of back to read the build thread

Yes i moved to the darkside

Instagram @jdm_brett

Hi lurker27.

Certainly looks an interesting project, RWD is so much more fun.

Sounds like you had a bit of bad luck with the cam drive ???

I've got a Sierra Cosworth (Ford's performance car for the UK market late 80' early 90's), toothed belt drive to the cams and a few years back the belt lost a couple of teeth, slipped the timing and bent all sixteen valves.....

Re the sensors, the ones you see in Paul's pictures are Hall switches, my setup (not yet finished) is Hall switches and when I tested the circuit(s) last year I built a simple square wave generator (just on breadboard) to test it all and it was fine.

For this application, I think Hall switches are better as their on/off voltages are allways the same so you only measure frequency (speed) whereas with a VR sensor its voltage, as well as frequency, goes up with speed.

I've also gone for ABS ring gears as the trigger - there is a very common VW/Audi setup over here in Europe where the toothed "wheel" is a push fit over the CV joints or rear hubs and is almost the perfect size for the Mini configurations - just need to machine 1mm off the back of the CV joints and make a small spacer tube for the rear hubs.

Brett, I'll email you the modified 3.0.3r I thought you already had it :)

Schrödinger's cat - so which one am I ???

Yes, I had a timing chain failure - the head is toast, but I have a new one I hope to swap on this weekend with a bigger cam + race rockers.

I believe that, looking at the data sheet, the variant voltage is irrelevant - part of the circuit essentially establishes a voltage baseline, and then the circuit counts the number of "crossings" past that trigger voltage, and then finally converts that into a final voltage output. This may not be the case with how the original FSAE circuit was conditioned, however.

I will host a diagram illustrating what I mean tonight. In my case, the sensors are already in place (Gotta love a 1989 with 4 wheel disc ABS stock)

OK, I wasn't exactly right - If you want to use a VR sensor (swinging voltage both ways, you must use the 8 pin packages, which DON'T have an externally set voltage threshold - they just cross ground and count that as frequency. Everything else is roughly the same. Sorry for any confusion i may have caused there.

Good point, although looking at the datasheet again I don't see why the 14 pin package wouldn't work just by tying pin 11 to ground instead of a reference.

But, TBH, our use of the 14 pin packages was more based on that's what Patrick used, so we had a circuit and component values which we knew would work.

All I did was add a bit of paranoia with the voltage clamp to protect the CPU against any other faults sending more than 5V its way :)

Schrödinger's cat - so which one am I ???