AEM in BRZ

[Mitch P]

@Dustin@Dynosty, No offense taken.

My post was more in response to the post after your's where the same question you asked was posted again, with little additional information nor an extension of the original question.

Additionally, when it comes to ecu brand, the box itself can do a lot more than most people think it can. It all depends on how the end user wires the box up and sets up the software. I have done some out of the ordinary stuff with AEM's, Motec's, Pectels, etc... and it all depends on how creative you get with the software and hardware. Given, you might need to add an external power stage or signal conditioning circuits here or there, but one can usually find a way to make it work.

Same thing with this project. When it comes to the AEM EMS, I have pretty good idea of what it's capable of and I know what's needed to have a properly running DI system. I have an idea on how I'm going to do it, but until I actually do it and see if it works or not, one can't say, "yes it works" or "no it doesn't." Many people throughout history have said "You can't do that" but many other people have proved them wrong.

In the end, maybe I can't get it to work, but a least I'm going to try. That's the whole point of R&D, start with an idea, build it, test it. If it works, great, if not, find another way.

And as far piggybacking ECU's, again, it's all in how the end user sets it up and tunes it.

[Dimman]

Originally Posted by Mitch P

@Dustin@Dynosty, No offense taken.

My post was more in response to the post after your's where the same question you asked was posted again, with little additional information nor an extension of the original question.

Additionally, when it comes to ecu brand, the box itself can do a lot more than most people think it can. It all depends on how the end user wires the box up and sets up the software. I have done some out of the ordinary stuff with AEM's, Motec's, Pectels, etc... and it all depends on how creative you get with the software and hardware. Given, you might need to add an external power stage or signal conditioning circuits here or there, but one can usually find a way to make it work.

Same thing with this project. When it comes to the AEM EMS, I have pretty good idea of what it's capable of and I know what's needed to have a properly running DI system. I have an idea on how I'm going to do it, but until I actually do it and see if it works or not, one can't say, "yes it works" or "no it doesn't." Many people throughout history have said "You can't do that" but many other people have proved them wrong.

In the end, maybe I can't get it to work, but a least I'm going to try. That's the whole point of R&D, start with an idea, build it, test it. If it works, great, if not, find another way.

And as far piggybacking ECU's, again, it's all in how the end user sets it up and tunes it.

Could you use 8 sequential injectors and run 4 to port and 4 to direct? Or do they need different signals? What exactly does the DI box do?

[Mitch P]

Again, if one knows how it works and understands the software settings, you can get it to do what you want, but if you don't know how it works or understand the software settings, then you're probably going to have problems with it.

My personal experience with the DBW idle control on the AEM EMS's for the STi is that it works pretty well. If anyone is having problems with it, I can give some pointers on how to set it up it to idle like you want. I know the STi DBW throttle control system particularly well, as I helped to design it.

[Mitch P]

For DI control, you must at least send the DI box 4 separate injector signals plus another signal for the spill control valve in order to control the amount of rail pressure. You can use injector, coil or other crank angle timed outputs properly setup in software,with the correct hardware interface, to control these types components.

The amount of fuel injected is a combination of the injector pulse time and the fuel pressure, just like a regular port injection system. With direct injection you only have a limited crank angle window to inject fuel (you don't typically want to inject fuel with the exhaust valve open as it just goes out the pipe, but sometimes you might) so the crank angle through which the direct injector fires is important. To get more range in available fuel flow for given crank angle, you also need to adjust the fuel pressure.

With a port fuel injection system, if you raise your fuel pressure from 36psi to 81psi, you get approximately 50% more fuel. Physics tell us that a fuel pressure change will effect fuel delivery by the ratio of the square roots of the pressure change. In our example: sqrt(81)/sqrt(36) = 1.5, or in other words, a 50% increase. Now, a DI system runs at a much higher pressure range. According to the FR-S/BRZ manual, this system runs from 2.4 MPa to 20MPa (~350 to ~2900 psi) so from lowest to highest pressure, you get an approximate fuel flow change of sqrt(2900)/sqrt(350) = 2.88 or a 188% increase in base fuel delivery. (min to max rail pressure) Now, in reality, this might not hold exactly true due to a number of factors (injector non-linearity, cylinder pressure, etc...) but the idea is generally true.

One additional facet of fuel pressure control is that the spill control valve output must also be timed specifically to a certain crank angle window for proper operation of the high pressure fuel pump. The pump is mechanically run off a triangular-ish lobe on the camshaft. (Kinda looks like a 13B rotor) The ECU control signal must be timed to each of the three lobes for proper pump operation.

So in the end, you needed send the DI box 5 signals with correct pulse widths and engine position timing to get it to work properly. Calculating or mapping those value is a matter of software setup and some math. The DI box also may need additional error checking signals to ensure that it operates as though the stock ECU is still controlling it. We have not tested that yet as we need to get the DBW throttle going to characterize what the DI box is looking to see without the stock ECU closing the throttle on us.

[Mitch P]

The great thing about the FR-S/BRZ's DI box is that it is fairly simple so integration will be easier than with other factory DI boxes. Once we get this worked out, this DI box could be used on other DI engines where the stock DI box is a bit more complicated.

Additionally, the fact that the FA20 is both port and direct injected means that it's bit easier to do DI testing as you can keep the engine running with the port injectors while you do testing with the direct injectors. With a DI only setup, you have to get pretty far into the R&D to even get the engine to start and idle.

[SkullWorks]

Originally Posted by Mitch P

Additionally, when it comes to ecu brand, the box itself can do a lot more than most people think it can. It all depends on how the end user wires the box up and sets up the software.

...

And as far piggybacking ECU's, again, it's all in how the end user sets it up and tunes it.

This is exactly why it is so bothersome to me when people will freely express what an ECU can or cannot do.

In internet land it seems people do not understand that most boxes available today are limited by the tuner and not the hardware/software. when I hear people say things CANNOT be done so definitively, that I have seen Mitch do, it just lets you know who gives up and who perseveres against trials.

Watching Mitch play with Ecu's is like watching a kid play with legos


Did you know mitch thinks math is fun, he gives me lectures (which I enjoy) on physics and engine dynamics, electronic control theories etc.

He has a DBW throttle body in the middle of his desk that he just plays with...frankly I think some of his screws are a tad bit loose, I think some might be ball screws too....

[No Limit Motorsport]

Mitch got AEM Series 2 working in one of our customers 2007 STI's when no one else could. Thank Mitch!

I can't wait to see what you do with the new Infinity box.

[Visconti]

Bump

[SkullWorks]

yeah...sorry Mitch had to help set a new record at Pikes peak, help judge Monterey Historics, then help win Formula D Las Vegas, we will be back in the lab shortly finishing this up.

[Jesse@JDLAutodesign]

looking forward to this!

[SkullWorks]

We are making great progress on the AEM Series 2 front.

Mitch Re-RE-tuned my car this weekend, quite happy with our results.