Andante

Okay, here's something that has been floating around in my head for a while now.

Given the fact that we have a lot of highly skilled, intelligent and dedicated people frequenting this board, what would it take to unite and come up with a custom developed ECU for our single cylinder Indian bikes. There are already similar projects that are openly available online, but they all cater to multi cylinder engines (4, 6 or 8). Perhaps it could draw on the information from the MegaSquirt project or something.

I am aware that this would probably be a somewhat expensive undertaking. More so if conversion from carb to FI would be done. But how about bikes that are already Fuel Injected? Or have fuel injected siblings. I'm pretty sure an ECU that can accept a custom map for fueling and ignition would be a valuable tool to some and perhaps a thrill for others.

Also I'm am completely aware of the fact that it's quite an undertaking to tune the fueling maps and ignition timing for an engine. The fact that it requires a good amount of time on a dyno as well as requiring a broadband lamba probe. However, I'm interested in discussing possibilties.

So the questions are:
It is a reasonable thing to do?
Which disciplines are needed? (Things like software, electronics, what else?)
Could a project like this actually succeed?
Could it be a community style project? (Volunteer basis, perhaps companies sponsoring development?)
What applications could it have? (Tuning, conversion of existing bikes?)

I'm putting this forward as a discussion point. Let me know what you think.

Aryan

General Biking Discussion Approved.

__________________
...in search of that perfect world - My Travel Blog :)

svjhonda

Sounds excellent to me.

But problem is we're spread all over the country. Even developing the logistics will need frequent and efficient communication between whoever is working on the project. Also consider the tools and hardware needs. Ideal place would be a place with a dyno. Rest hardware can be shifter/bought/borrowed.

But, this is too large and unreal for my imagination. Primarily because it'll need a hell lot of dedication, effort and monetary help.

Andante

Well, the fact that it's large project doesn't mean it's impossible. Look at projects such as the linux kernel and many other open source projects. If that's possible, why wouldn't this be possible? Granted building a dedicated community would take some effort.

As for things like tools and hardware needs, there's something which I agree with to just some extend. The basic hardware you need isn't anything really expensive, there's lots of embedded chips that do not cost the world and have more than enough processing power to do the job. Of course, things like hardware design and testing requires some specific tools, but I'm sure there's people with access to thing like logic analyzers and scopes.

The real squeeze will be developing the software to do the dirty work of calculating the right amount of fuel to inject, but I'm sure that it can be worked out. The basic principles are clear, and the way most FI works is with look up tables, so the algorithms are not even in the higher math regions. Some complexity could be found in the transient response algorithms (responding to things like throttle opening changing). But even those could be implemented as look ups, I suppose.

As for communication, we've got internet. A dedicated forum or website, would go a long way to make communication good enough. As for testing if you've got a few people with the right equipment, they could take turns at testing the changes and giving approval. As I see it, it can be done.

How about everyone else? I'd love to hear your ideas/comments.

svjhonda

I'd love to be a part of something like this, but I feel I have insufficient knowledge to be actually of help. Also I'd have to shell some cash out, which is a bit beyond me. Still 19 and a student.

I'm really waiting for Joel's commnets on this.....

Andante

As I said, I'm putting this up for discussion to see what comes out of it. As for being able to help or not it doesn't matter the fact that you're discussing is good enough. More so since there is no project, just an idea.

And yea, I am also curious what some of the long time members have got to say on this one.

Andante

I suppose there's a conclusion to be drawn here. There isn't much or any interest in a project like this one. Anyway, that's fine too.

Everyone, please move along. Nothing to see here!

Joel

This is not an impossible project and its not easy either.
We all know the working principles of a lot of things around us, but materialising them is a different ball game.
A megasquirt is for 4,6,8,etc cylinder engines. Yes, you can probably find one for a single cylinder engine. Next what you need is a supporting software. Now thats a task.
Also to build a good ignition system, you really need to understand a lot more about engines and then work towards building the right base.
All said, this is a good project more from an academic point of view. Building a system for real world conditions and then for racing is a different ball game.

FYI, I'm already involved in my Fuel injection project for racing and I know what a task it is. I'm in the process of testing and getting data. Unfortunatly I wont be able to disclose anything at this point.

Joel

__________________

joel@raceconcepts.in

Andante

I totally agree with what you're saying. This was an idea that has been floating around in my head and I needed to get it out. I was hoping to get some people into the discussion and reach a conclusion somewhere.

Now, I am a software engineer. I'm pretty much aware of what goes into developing software that can actually get this right. The hardware is probably a challenge too.

So anyway, by the time you are in a position to talk about what you've been working on, I'd love to hear about it. I find such things highly interesting, more so since the future of all IC engines is going to be FI.

The idea of this thread was to have a discussion on the subject, but the response has been very less, probably because no one wants to touch the subject. Anyway, that's fine by me.

Joel

The future of all IC engines is indeed electronic fuelling. Having said that, every maker is trying hard to develop complex systems which can sqeeze the best and the last drop from a given power plant.
However what you need to realise is that, to build an electronic system for an Internal combustion engine, you need to know a lot more than merely software development. People who know the best of engine tuning and electronics, strike the best equation. When you build a system, you need to know what the engine demands and where. Why and how. Understand how performance enhancements effect the fuelling and ignition. What are the mechanical limits and what are the electronic limits of a given engine.
When you take the route of forced induction, the electronics will need to work more efficiently else its a 100% blow-up. Tuning an engine, mechanically and electronically to get the desired results is as close as doing a black magic! Its a mad puzzle to solve and its done best by only a hand few in the world.

__________________

joel@raceconcepts.in