MS2/Extra Alpha 3.0.3 20090507 (was 20090426)

[racingmini_mtl]

This is the new release of the alpha 3.0.3 code (sequential code):

removed link

This release includes all the latest updates from the beta 2.1.0 20090413. It also includes all the features and options that were left out of the previous release. This means that this release should be the same as the beta 2.1.0 release with the same features and behaviour and have the sequential injection option in addition. I haven't tested all features to make sure that I didn't break any of them but the tests I did do went fine. But do remember that this is still and alpha version.

Jean

EDIT: There is now a new version which brings the code inline with the RC release from 20090505. It's basically the same as the last alpha release with the correction of the afr target bug. Here is the code:

http://www.msextra.com/ms2extra/files/a ... 090507.zip

I updated this instead of starting a new thread due to the small change in the code and to keep the information in the same thread.

[tristanlee85]

I drive 3 miles to work everyday so if I break down it won't be a big deal. I'm downloading the code now and I'll let you know how it works on my car. I'm not doing sequential so I won't be much help there, but I'll be able to test and see if some of the 2.1.0 features are broke.

[racingmini_mtl]

Thank you for testing it.

While you're at it, you could try the semi-sequential mode. It doesn't require any hardware modification and allows you to play with the injection timing.

Jean

[tristanlee85]

Maybe. After I burned this code, I was trying to figure out why my boost control wasn't working, and when trying to test for some voltages, and I bridged the 12V and middle pin of U5 and it sparked, and now when I connect my computer to the MS, I get "Invalid Data Received" so I don't know if this is in any way related.

[racingmini_mtl]

This code is based on beta 2.1.0o and not the 2.1.0n shown in your signature so there might be differences that would have affected the boost control. The middle pin of U5 is ground so it's no wonder there was a spark. You should try reloading the 2.1.0 code and see if that works.

Jean

[tristanlee85]

Well it wasn't working on the 2.1.0n code either. It was, then it stopped after I was messing with the IAC.

I also tried swapping in my other MS2 that has the 2.1.0n code on it and I can't connect to it through the laptop either so I think I killed some IC on the board somewhere.

[racingmini_mtl]

I had to replace U6 at one point and I think that was due to something like what happened to you (but I'm not sure since it was something like 2 years ago). So I would check U6 for sure.

Jean

[UnaClocker]

Does this version have the support for Sequential on the Neons without COP? I know it was a private version that would do that on the last alpha.. Hoping it got rolled in this time around.

[racingmini_mtl]

Does this version have the support for Sequential on the Neons without COP? I know it was a private version that would do that on the last alpha.. Hoping it got rolled in this time around.

Yes it does. Actually, the last version also had it. It was a pre-alpha version that did not have it.

This page shows all the mode that are available for the different versions of the MS2/Extra code and for the JimStim: http://www.msextra.com/ms2extra/MS2-Extra_wheels.html.

Jean

[tristanlee85]

Once I get the serial issue figured out I'll mess with semi-sequential... though I have no idea what it does. You said it doesn't require any hardware mods, so Im assuming it will work fine with my Ford EDIS setup and no cam position sensor? Once I get done eating I will read up on the documentation. I just figured I would throw the question out there anyway.

[racingmini_mtl]

EDIS and dizzy setups are the 2 setups that won't work with sequential/semi-sequential because the CPU doesn't know what the engine position is. The CPU doesn't know which cylinder it's firing: the EDIS or the dizzy takes care of it. Which means you can't do sequential injection because the CPU wouldn't which injector channel to fire.

So that was a good thing to ask.

Jean

[tristanlee85]

I knew you couldn't do that with sequential, but I wasn't sure of the difference between that and semi-sequential. The reason for upgrading to the new code wasn't for the sequential support. I just like trying out new code and give feedback whenever possible.

[racingmini_mtl]

I thought that was you intent but I also thought you had some sort of toothed wheel setup which is why I mentioned it.

In any case, any feedback on testing is appreciated because that's the only practical way to actually go through the code and get some confidence that it's doing what it should.

But with the EDIS setup, you could just add one ignition driver, connect the VR sensor to the MS, and control the coil pack directly. Then you could do semi-sequential injection. Just kidding (you could but I'm not really suggesting you do that)

Jean

[kritip]

I will load this code before i return home on friday on a 160mile run

Why isn't this folded into the 2.1.0 release, as it seems to be integrated and working so well?

[racingmini_mtl]

There hasn't been a lot of test coverage yet and the potential interactions (bad ones) between the sequential code and the rest of the code have not been fully covered. So this is still considered experimental code but it seems to be behaving correctly and the advantages of sequential injection on engine performance seem to be significant for all those who have reported on it.

The 3.0.3 branch will become the main development branch at some point which should not be too far in the future but James and Ken will have to confirm when. The 2.1.0 code has had much more test coverage and is stable enough to go to a release version which would not be the case with the addition of the sequential code.

Jean

[Dennis930]

Jean,

I have been testing the semi-sequencial code (Alpha 3.0.3 20090313 and 20090426) on the Jim Stim for a six cylinder engine. All the injector control mods I made appear to be working porperly. I have one problem with Tuner Studio but I do not know if it is code related or Tuner Studio related, or if I am missing something. When I try to set idle control, the algorithm is greyed out. It works with the sample 2.0.1 MSQ that came with Tuner Studio. I tried it with MT and the algorithm is functional.

Another observation concerning injection strategy for semi-sequencial; If you want to avoid valve overlap one must be mindful of the cycle position of the "wasted" cylinder. The primary cylinder can be injected at closed valve, but the "wasted" cylinder is going to inject with opening of the intake valve. To avoid valve overlap I am planning to inject at approximately -30* start of pulse for my engine. One reservation I have is what happens when the engine is WOT and full boost, will the "wasted" cylinder pulse continue on past the close of the intake valve, and if it does, is really an issue since this engine originally ran CIS.

-Dennis

[racingmini_mtl]

If you have something working in MT and not in TS then the issue is either in the ini or in the TS code. It won't be in the firmware.

If you want to inject on a closed valve, you could try to use end-of-pulse and set the timing to end as the valve starts to open. If your duty cycle stays low enough then you'll have both pulses on a closed valve.

Jean

[ashford]

im considering this code for a 2.5 turbo dodge where im using the stock distributor which has 2 hall sensors feeding ms while running one coil. the wheel decoder in the beta for this dizzy is capable of coil on plug, so can it be used as sequential since the ms has all the info for cop or does it need to be in 4 spark out mode for the code to work in sequential.

btw the engine synched wrong when the 2nd trigger wasnt working right

[racingmini_mtl]

As long as you have the triggers that will allow COPs and the wheel decoder can do it, you'll be able to have sequential injection even if you configure the ignition for a single coil. I actually just tried it on the JimStim and it does work with the Chrysler 2.2/2.5 mode (which I assume is the mode you use).

Jean

[Keithg]

it runs om my install. 36-1 with cam sensor. The only issue I noted was a MSQ problem on my side. I had 60 in the 'Trigger Angle/Offset' box. In previous codes this did nothing on my install. With this code, it retarded my ignition. Once I set this properly and verified timing, all was well.

Also, I tried to add my VSS code to this and had a compile error and do not get an S19

: reference to a banked address [3a:9a43] in the normal address space at d383
make: *** [ms2_extra.elf] Error 1

KeithG