I work at a college and I just went to a software fair where our students show off what they have been working on this year. Well, a couple guys have been working on a user-friendly graphical interface for the Megasquirt unit. It will also work as the controller/interface for the stand-alone ignition system as well. They will be releasing a beta model of their software very soon.

Well, it is one thing to say here is how it works and quite another to put it on a car and make it work. They had a 70 Chevele sitting there running. It looked like a very nicely done; yet inexpensive system that worked well.

They have told me that their software made it much easier for a non-programmer person to setup the Megasquirt. One guy is also talking about creating a setup wizard that will walk you through the process. What else is cool about this? Well, the code is open source software and anyone can download it (here is their url - http://heartofgold.sourceforge.net/index.php, but there is no code there yet). All the code is in C and Java and should work with Macs, Pcs, and on Linix systems.

So, what am I thinking of doing? Well the Megasquirt "likes" GM sensors and I was thinking of going to pick-a-part and picking up what I need. I would plan on using my Lakes intake or maybe one for an IDA and making an adapter plate to use a GM throttle body injection. I saw one out of a 90 Caprice with a small block that was a two-barrel about the same size as my Dellorto. It has a fuel pressure regulator bult in and GM plugs. I also would pull all the sensors I need from GM car and build a system from there. I know there are other ways to do this, but I like the idea of learning on a system where I start with cheap used parts. Is it perfect? No. Is it a cheap way to learn about injection system? Yes, that it is.

After the racing season, I think that I will have a new project.

 

That sounds interesting to me. I've always liked the megasquirt but I wanted control over spark also--they say that is coming. For me spark is a requirement as turbos need to be able to adjust timing. Some way to integrate a knock sensor would also be a big plus.

I've got lots of mechanical, electrical, and programming experience so when I get my turbo FB done, I'll be jumping on this--sounds like fun. Being open source will work good for me, I already work with alot of unix open source right now. Right now too busy getting the car finished up--but next year looks good.

I want to be able to plug in a laptop and do on the fly adjustments while I watch the sensor readings. If we can do the base unit with software for less than $400 I think it will be a hit. Henrik over in the second gen forum has deassembled the S4 ECU so maybe we could get some base info from him?

Scott