[noisymime]

Ive been working on the same kind of thing, but what im finding the injectors will still fire even if the fuel maps are Zero'ed for the map/rev range you don't want the injectors to work on. I don't fully understand why but its been a problem for me so far, the next thing i was going to try was use the VVT output and map, but i haven't had a chance to test it yet.
I was going to ask Josh about this problem when i got a chance.

I have a suspicion I know what this is. If you're playing around with it again, try setting the injector open time to 0. If that fixes it, I'll write up a proper solution, but I suspect that even with the table set to 0, it's still trying to open for the 1ms or so that the injector take to open.

[MadCow]

First let me thank everyone for all the great responses. Sometimes people ignore the newbs. OG I am not sure about boost. I am more in the planning stages. Weighing options and so on. But honestly the manifold pressure isn't that important to me if the speeduino doesn't care. I only want it to know one thing, maintain Afr with input from the wideband O2. I basically want it to work like a more sophisticated Hobbs switch. People have setup a Hobbs to turn on an injector that sprays full duty cycle when a preset manifold pressure is reached.

NickZ its hard for me to say on certain things cause I have so little knowledge. Just my reading and videos I have watched. I don't fully understand what makes SD happy in its function and how much is needed for what I want. Perhaps SD isn't what I need and maybe just an arduino that can read an O2 and control an injector. Seems simple. After reading I thought SD would put me ahead but maybe it is more complex than necessary for what I want to do. I haven't committed to anything so if you guys know of a more simple way let me know. thanks

[PSIG]

Enjoy the ride, and stick to your goals, as you are far from the first to envision something like this. The simplest way to achieve a good running turbo engine would generally be to properly mod and tune a correctly-sized carb for the correct AFR on its own, just like without a turbo. However, if the correct carb or mods are not to be used (limiting the fueling capacity with boost), then your scheme has obvious merit.

The primary challenge to your simple scheme is the controller's ability to sense AFR in a timely manner, and then for the ECM to correct the fuel flow accurately, based on that sensor info. Just putting a variable fuel nozzle with an oxygen sensor tends to result in wild swings in AFR as the system attempts to compensate, first lean, then rich, then "oh crap it's too rich, so make it leaner, uh oh it's too lean..." "Smart" damping systems are complex and can only help so much. One simpler solution to this scenario going out-of-control is a better 'tune' for the added fueling, so the system does not have so much correction to do. This is why the seemingly more complicated SD EFI basis is suggested, to increase initial fueling accuracy during boost, minimizing necessary corrections, increasing system stability, safety and performance. Better is ... better.

Any mode that accomplishes this stability can be used, though MAF is more involved and undeveloped, and Alpha-N is less accurate as fuel requirements change under boost, leaving SD as the best available option with Speeduino at this time. Fortunately, SD is relatively simple, and with good setup and tuning should address your needs quite well. That's up to you. The choices and results are yours; so keep learning (as we all are) to help yourself make good choices.

David

[noisymime]

Ive been working on the same kind of thing, but what im finding the injectors will still fire even if the fuel maps are Zero'ed for the map/rev range you don't want the injectors to work on. I don't fully understand why but its been a problem for me so far, the next thing i was going to try was use the VVT output and map, but i haven't had a chance to test it yet.
I was going to ask Josh about this problem when i got a chance.

This is fixed now on github, or should be anyway.

First let me thank everyone for all the great responses. Sometimes people ignore the newbs. OG I am not sure about boost. I am more in the planning stages. Weighing options and so on. But honestly the manifold pressure isn't that important to me if the speeduino doesn't care. I only want it to know one thing, maintain Afr with input from the wideband O2.

I honestly think SD is the way you want to be going here. Trying to tune based on AFR alone is a fools errand in my experience. If it was all you had available to you then you could make something work, but you will get a far, far better result by adding the MAP sensor and doing a proper table for the boosted area you're interested in.

It comes down a what you want to achieve. Using AFR alone is better than the Hobbs switch you describe, but not by a lot. It's still taking a huge guess about how much fuel it should be injecting. Doing it with a proper load based map will be far more flexible and accurate, but for your application you may not need that accuracy.

[MadCow]

Sorry for the misunderstanding. I actually thought you guys were abbreviating speeduino as SD. I didn't realize you meant speed density. I was meaning speeduino when I put SD. In that perhaps there is a more simple solution than speeduino for my purpose. But if nicks has it figured out then I will be all over it

[NickZ]

hopefully testing tomorrow with the new code.
All i have done is made a micro squirt , ive just used 1 injector driver Driving a 440cc Injector into the air stream before the throttlebody, map sensor and tacho input, I did add a TPS input but not currently using it.
Made a VE Table starting at 100kpa to 180 kpa and RPM from 2000 to 6500 RPM
first row 100kpa 2000-6500 RPM is 0 and if new code is working ok it shouldn't fire the injector, I then have a steady slope in the VE table up the boost and rpm range adding more fuel.

We added a Supercharger to a NA ford v8 302w in a 2000 model AU falcon,Running around 9psi boost, Not wanting to spend the $1200 on a new flash tune we are trying this method first.
Problem really is the Australian AU falcon V8 doesn't have a factory knock sensor (stupid piece of crap) so we are going to drown the top end in fuel.
without the micro squirt 4500 RPM up its in the 14:1 A/F ratio.
I want it to be 11:5 to 12:5 A/F Ratio under boost.

[noisymime]

Made a VE Table starting at 100kpa to 180 kpa and RPM from 2000 to 6500 RPM
first row 100kpa 2000-6500 RPM is 0 and if new code is working ok it shouldn't fire the injector, I then have a steady slope in the VE table up the boost and rpm range adding more fuel.

Keen to hear how you go with this, if theres any issues let me know as this is a use case I want to make sure can work OK.

[PSIG]

Problem really is the Australian AU falcon V8 doesn't have a factory knock sensor (stupid piece of crap) so we are going to drown the top end in fuel. without the micro squirt 4500 RPM up its in the 14:1 A/F ratio. I want it to be 11:5 to 12:5 A/F Ratio under boost.

I don't know if you have a specific tuning routine or plan you use or not, but if not (and ideas for others trying various tuning methods) these are just suggestions; but even if the engine had a knock sensor I would not suggest tuning in a manner that needed it, as they are not consistent enough in sensitivity to trust an engine to. There are plenty of scrap engines that used knock sensors as primary tuning tools. Old school plug reading for heat, fuel, and ignition timing is one example of tuning methods that can be much safer, while ensuring max power. I would also suggest tuning for what the engine wants for indicated AFR, and not pre-assign any particular numbers to it. One of the reasons for this, is that real-world AFR readings must be regarded as relative, rather than absolute. I would prefer to target what the engine likes and needs, and establish the reported best AFRs from there for my AFR table (if used).

Your approach of starting with a wedge table is useful of course, but no need to drown it in fuel if your timing is good. Or, are you making it sloppy rich to retard the burn, due to no timing control? I would encourage being patient and sneaking up on it, developing the table in areas before you even get there. There is no requirement to jump straight to the goal with crossed-fingers, and the tables should be built progressively anyway so it has a complete tune, and not just off-boost and WOT along the edge of the table. These are only suggestions that may help avoid issues and have more fun with it.

David

Just for example; if tuning is not progressive, not only is det or damage a greater gamble, but the orange transition area in the example table below would have unknown results for both fuel and spark advance/retard if there were a later throttle-up from midrange rpm:

[MadCow]

hopefully testing tomorrow with the new code.
All i have done is made a micro squirt , ive just used 1 injector driver Driving a 440cc Injector into the air stream before the throttlebody, map sensor and tacho input, I did add a TPS input but not currently using it.
Made a VE Table starting at 100kpa to 180 kpa and RPM from 2000 to 6500 RPM
first row 100kpa 2000-6500 RPM is 0 and if new code is working ok it shouldn't fire the injector, I then have a steady slope in the VE table up the boost and rpm range adding more fuel.

We added a Supercharger to a NA ford v8 302w in a 2000 model AU falcon,Running around 9psi boost, Not wanting to spend the $1200 on a new flash tune we are trying this method first.
Problem really is the Australian AU falcon V8 doesn't have a factory knock sensor (stupid piece of crap) so we are going to drown the top end in fuel.
without the micro squirt 4500 RPM up its in the 14:1 A/F ratio.
I want it to be 11:5 to 12:5 A/F Ratio under boost.

Great. Sounds about like what I want to do. Post up how it goes tomorrow please. Thanks

[NickZ]

Still haven't tested new code for this, but if all goes well (i'm sure it will) I have another board ready to go off for production for this purpose.
Has mpx4250 map sensor, 1 Injector output, optical input for tach signal (will work from 5v to 12v, But is able to work for higher with resistor change) and O2 sensor input so can monitor A/F ratios.

microsquirt.jpg (199.76 KiB) Viewed 5742 times