I have spent the last few evenings building up the ECU kit for my multi port throttle body project, and thought you might like to see some pics before I do the full write up on my site.
So far I have uploaded some code to it and had it running on the bench with a simulator board that replicates all the normal sensor inputs like MAP, throttle position, RPM, IAT etc. (first time my desk has done 15,000 RPM!) Then I spent this evening fabing up the mounting bracketry and starting the wiring looms. Surprisingly its all going according to plan ...so far. Still LOADS more to do though. Here's the pics:

The kit:


Relay board mounted. This connects to all the sensors and injectors and provides power to the various sub-circuits.


ECU mounted. That loom was a real pain to make up!


PC port and status lamps.

 

Ok, now I'm getting a woody, Woody ! LOL

 

Holy crap. Um, what Derek said!!.1!

Any chance you'll make some up for those of us dying for power for track/autocross?

 

your about 1 million times smarter than me. anything beyond ikea instructions manuals confuse me, you could be on your way to making loads of money buddy.

 

geeez... lol... Hddp... lol... said it best!!!

...talk about CROSSFIREINNOVATIONS (.com)!!!

 

Is that the Mega Squirt system? I've seen a few people run it and it seems to have a following.
I used the Electromotive TEC II on my hot rod. It's an older system (they have new ones) but it works great and I like the distributorless ignition.

 

Yes Feets, its a Megasuirt. I chose it because it is all 'open source' and lends itself to being modified to integrate with the other systems on the car. There a lot of issues still to sort, like support for the quad 02 sensors and 'drive by wire' etc but anything else is just way out of my budget, even to buy, let alone develop!

Derek: LOL, perhaps I should rename it the viagrR.

 

megasquirts are known to lead to lots of engine management problems, at least in my old world of rx7's. but best of luck to you, this is yet another thread i will be keeping a close eye one...you figure this out i may buy my crossfire after the lease...

 

We looked at the megasquirt system also, and it has a lot going for it. The open architecture and astoundingly low price being two biggies. At the end of the day, decided it was too limited - at the time it could not handle ignition, only fuel, but a newer, enhanced version was due out.

Does yours do both fuel and ignition?

Some of the guys I race with have a megasquirt & love it.

Henry

 

Yes, mine is the latest MSII with the faster processor and ignition support and CAN. There are definitely limitations but for controlling extra enrichment and altering ignition timing I think it will be fine. I will be leaving the stock ECU to take care of the rest of the car so the MS will only need to perform a limited role. I am looking at it as part of an integrated solution, not as a complete replacement for the stock ECU but it will need modifying to do the other things I am looking at. For the price there is nothing to touch it. The best part is that all the development suites are free; with a lot of the other systms you buy the ECU then end up paying a fortune to set it all up.

 

good choice on using the megasquirt... my '98 turbo RT ran like a factory tuned car with a MS i built to control the extra 4 injectors...

about the control for the "quad o2 sensors" .... the downstream o2 sensors are just used to monitor the catalyst efficiency ... the two upstream o2 sensors are what you have to worry about as they are what monitors actual air to fuel ratio ... you really do not need to monitor both, one side will be fine cause the MS is a batch fire system not sequential, therefore the air fuel ratio will be even on both sides ...

what i would do, is remove all the o2 sensors, place one wideband o2 sensor in one of the upstream positions, and plug the remaining 3 with readily available o2 plugs (summitracing or jegs) ...

now the bigger worry is making it work with the crank sensor signal, getting a proper coolant temperature reading, and making it start in all sorts of temperatures without stalling out ... using a good fluke voltmeter you can reverse engineer the cold start curve by monitoring the startup pulsewidth over the seasons with the stock ECU ...

Woody, i am very happy to see this kind of enthusiasm from a fellow XF owner. This is exactly what this car needs, true pioneers in XF modifications...

 

Thanks for that Inteseblu.
I've not decided exactly what to do with the 02's yet, but I am altering the headers and exhaust anyway as part of my 'TrackSport' conversion so ultimately won't be using the stock setup. Reverse engineering all the curves is not really a problem as I have all the kit for that (Fluke digital storage 'scope etc) being that designing electronic temperature monitoring systems is my business!
My biggest concern is making sure the rest of the car behaves itself (or even works!) after messing with the ECU. Like you say there is an issue with running the MS from the crank sensor rather than the ignition but figuring out this sort of stuff is half the fun.
Thanks for the support.

 

Glad you think it's fun...

 

Well, more fun than the day job anyway. I've not had the enthusiasm knocked out of me yet.

 

.I'm sitting here setting-up a new edit system that I just purchased, that's the easy part... then I have to figure out how to use the software... that's the not so fun part...

I'm watching CNN and the middle east going up in flames... knuckleheads

 

I guess you're going to run this as a piggyback system instead of a replacement?
It will be interesting to see what you get out of it. We all know that they use a conservative tune so the vehicle runs well at any temperature or climate. Being able to mess with it will open new doors.

 

woody - considering what your line of work involves, i could not think of a better person to tune the MS cold start function. haha

btw, from what feets said, are you running it as a piggyback as a controller for extra injectors or as a full standalone with the stock ecu yanked out or a combination of the two leaving the stock ecu to do ignition and the ms injection ?

 

Well, the whole project will cover anything from just running a 7th injector in a TB plate I have designed, to running multi-port throttle bodies using the stock injectors. It will be a process of evolution. Initially I will try and use the MS to take over fuelling duties leaving the stock ECU to run the ignition (accepting that there may be all sorts of issues with incorrect ignition maps coming in to play). I don't think it a viable option on a road car to completely pull the stock ECU as it runs so many other systems in the car. You can follow everything on my site as it happens. That way it saves me writing everything up here in adhoc posts AND on the site - documenting everything and answering emails/post is now a job in itself, and I still have to a day job to do too. Only had about 6 hours sleep in the last 3 days. Still its shows there is some real interest in the project which has to be a good thing.

 

I can certainly understand your position. I ended up in a similar situation when I started building the hot rod. Not too many people had dropped turbos on an older Mopar. When all was said and done, it was lots of fun, lots of work, and a real interesting car.
Luckily, that was several years ago and I've been able to improve upon it since then.
Installing the EFI was fun too but my system doesn't have the open support that you can get with the MS. It turns out that my biggest problem was a fuel pressure regulator that would fail at high engine speeds. Most annoying. The symptoms all pointed to a programming error.

It will be interesting to see how well the aftermarket stuff integrates with a highly evolved and involved OEM system.

 

ooh i got one tid bit of info for you for when you go with the 7th injector part of the project...

make sure to use the dual table code to run the 7th injector... the standard code injects fuel even if the VE map is set to 0% across the board cause it still takes cold accel enrichment into account amongst other values... the dual table code was designed to run progressive nitrous control, so 0% means 0 pulsewidth no matter what...