Page 3 of 4

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Tue Oct 13, 2009 4:14 am
by SimonKho79
Hi Jean,

Need some help here. I am thinking to try staged injection with your code.

My setup is a 3 cylinder engine, with 3 injectors running sequential code now. From your staged injection table from your site. I noticed that in order to use the staged injection. #Inj seems to be 1 and #Inj for staged is 2. What does this means? Does it mean I have to inject the 3 primary injectors as simultaneous? and another 1 or 2 extra injectors at the throttle body inlet as staged?

If that were the case. Can I just connect ALL my 3 injectors to ONLY 1 driver connected to Channel 1, and the extra injector to the driver connected to Channel 3? Leaving Channel 2 unused. Will one driver enough to cater the needs to trigger my 3 primary injectors at the same time?

Thanks Jean.

Cheers,
Simon

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Tue Oct 13, 2009 8:35 am
by racingmini_mtl
To run staged injection with sequential on a 3-cylinder, you'd need 6 drivers so that's not possible. The only other alternative is to use 2 drivers and have the 3 primaries on the first driver and the 3 secondaries on the second driver. And when using two drivers, you have to use channels 1 and 2. And that will mean 3 squirts per cycle.

As for putting 3 injectors on the same driver, it depends which drivers you have but I assume you have my 4-channel board so I would not recommend using 3 injectors on this.

I must say that it's not the best setup to run staged injection on a 3-cylinder. You probably won't get much benefit from timed injection with 3 squirts. You may be better off with the regular untimed injection where you can set the number of squirts per cycle or use bigger injectors that don't require staging. With sequential injection, you may be able to get away with using very large injectors.

Jean

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Tue Oct 13, 2009 9:05 am
by automat
that's why I switched to sequential on my 3 cylinder. I needed bigger injectors than what untimed injection could comfortably make work at idle. I could have used staged injection but with sequential it runs very well at idle with big injectors. it saves me the hassle to have an intake manifold with 3 extra bung welded.

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Tue Oct 13, 2009 10:17 am
by SimonKho79
racingmini_mtl wrote:With sequential injection, you may be able to get away with using very large injectors.
So does the statement above means that I can use a slightly bigger injectors to replace my my natural aspirated stock injectors for my boosted application?

Thanks for the reply Jean.

Cheers,
Simon

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Tue Oct 13, 2009 10:24 am
by racingmini_mtl
Yes. If you go to a boosted application, replace the stock injectors with the needed big injectors. And make sure they are big enough to leave some margin. Unless you go to huge injectors, you should be able to get a very good idle and low load behaviour.

With the MS2 precision, sequential injection and Ken's MAP sampling in the latest code (20090912) you have all the tools to make that work without having to go with the complexity of staged injection.

Jean

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Sat Oct 17, 2009 3:40 pm
by SQLGUY
Hi all,

Just installed 3.0.3h (20090912) on my Microsquirt, and reapplied my 3.0.3d tune, which had been running pretty well. Everything still looks correct in TunerStudio, but I am now seeing 800% or more fuel load just turning on the key (not yet started). And, yes, it really is trying to inject that much fuel. Flooded to death.

The application is a 1982 Yamaha SECA 750 - air cooled four cylinder eight valve motorcycle. I'm using a 36-1 wheel for CPS, and a single throttle body with four pipes going to the four cylinders. Port injection. Tune attached.

Please let me know if you have any thoughts about what's happening here. I'll keep poking around to see what I find.

-----------------Edit------

OK, found the problem. Pre-upgrade, my realtime baro sensor was configured for "SpareADC", but, after upgrade, this is not an option, and MicroSquirt is no-longer reading the 2nd MPX4250 at all. I can run currently by switching to "Initial MAP reading" for baro comp, but how can I re-enable the 2nd sensor?

-------------Edit 2-----------

OK, got that figured out... apparently SPAREADC is now MS2 AD6/JS5. Baro sensor is working again.

However, I still have two other problems from the new code:

1. My narrow band sensor is now only reading .18 or .38 volts, regardless of anything else. I did recalibrate AFR, but that didn't help.

2. Something definitely changed in close-loop idle. I now have to set the idle adder to 3000 RPM (it was 300) to ensure that the valve doesn't stay pegged at about 75%. With the high value of adder, CL works OK. Without it, the idle stays up at 3200 RPM or so.

------------Edit 3-------------

Seems the CL PWM problem is that the IAC now goes to cranking % when above target+adder RPM. My setup requires a decent amount of extra air during cranking, in order not the have to crack the throttle when starting. With the current code set, though, this is unworkable, because the cranking % value for IAC results in way too high of an idle. I've reduced the cranking percentages and put the adder back at 300, and that's OK except that now I have apply throttle during cranking. Please let me know why the change was made and if there's some other way I could be setting this up now to not have to compromise.

The EGO issue is still there, though, and it's really annoying. I was hoping to use the narrow band for some of my tuning today.

Thanks,
Paul

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Sat Oct 17, 2009 7:29 pm
by muythaibxr
I believe that I have a fix for the CL idle problem, I found the same issue on my rx7 in ms3.

I'll be merging a bunch of code and fixes over the next few days back to ms2/extra, and that will be one of the fixes I merge.

Ken

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Sun Oct 18, 2009 10:35 am
by SQLGUY
OK. Sounds good.

Any ideas on the EGO reading? Is there something I should be configuring differently, or does narrow band not work in 3.0.3h?

Or, is it possible that TunerStudio is not correctly setting the EGO cal information?

I will try calibrating it from MegaTune instead and see if there's any differece there.

-------------Edit-------------------------

Yup, seems like TunerStudio isn't setting up the AFR table correctly for narrow band. After recalibrating with MegaTune, my narrow band is reading correctly again, in both MT and TS.

Do you think this is a problem with TS or with the base.ini?

Thanks,
Paul

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Tue Oct 20, 2009 7:06 am
by tpsretard2
I am not 100% sure where to post this but as this will become the main line of code i thought maybe i should post it here..

I am trying to get the Subaru 6/7 code working. There was a post a little while ago that cop mode had been added for this wheel mode.
I have tryed and it dos not work, i checked the code yesterday and it only has 2 triggers.. is this going to be added?

Unfortunately the post saying it was added seems to of been lost in the form crash or i just can't find it using the forum search.

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Tue Oct 20, 2009 7:46 am
by jsmcortina
Subaru 6/7 hasn't been added yet.

James

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Tue Oct 20, 2009 8:13 am
by tpsretard2
:( any time line on this..
i know the wheel decoder is there for it.

it dos dos not have cop support..

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Sat Oct 24, 2009 11:56 am
by SQLGUY
muythaibxr wrote:I believe that I have a fix for the CL idle problem, I found the same issue on my rx7 in ms3.

I'll be merging a bunch of code and fixes over the next few days back to ms2/extra, and that will be one of the fixes I merge.

Ken
Hi Ken,

Any update on the next build?

Can you tell me in what build the current CL code was introduced.I'd like to get back to the older Alpha code until you've got the fixes integrated.

Thanks,
Paul

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Sat Oct 24, 2009 2:07 pm
by muythaibxr
The code that didn't have the issue you're having is 2.1.0. I think I introduced the fix in the first 2.1.1. I do not know which 3.0.x to tell you to go back to to get that change out.

I believe I already have the fix worked in to 3.0.x, so the next 3.0.x release should have it fixed.

I'm working on merging a bunch of other stuff though, so no idea when the next 3.0.x will be released.

Ken

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Sun Nov 01, 2009 3:13 pm
by warp21
I just finished my build and loaded ms2extra_alpha20090912. When I try and start the car the injector signal seems inverted, staying on most of the time with blips of off time. I'm set up for full sequential 4 cyl and this is how my circuit has bee built. I've also included my msq.

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Sun Nov 01, 2009 3:23 pm
by jsmcortina
What about the flyback?

James

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Sun Nov 01, 2009 3:25 pm
by warp21
I was under the impression that flyback was only needed for lowz injectors so I left it out.

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Sun Nov 01, 2009 3:31 pm
by racingmini_mtl
With the drivers you used, the signal will indeed be inverted. You used an inverting FET driver which is not what you want. You want a non-inverting one such as the IXDN404PI. And that also means, you want to have pull down resistors instead of pullups (the four 10K resistors should be connected to ground instead of VCC).

Reading what is on my site, I realise that it's not very clear. While I mention the use of 4 identical drivers, I don't specifically mention that they should not be inverting. I'll correct that.

And as for the flyback, since the IRFIZ34N is fully avalanche rated, it should be able to handle the flyback from a single high Z injector.

Jean

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Mon Nov 02, 2009 10:26 am
by warp21
Since I have the circuit built I was wondering if I could just add a 2n2222a transistor to each of the injector inputs ahead of the ixdi404. This should invert the processor signal correcting my problem.

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Mon Nov 02, 2009 10:57 am
by racingmini_mtl
Yes you could do that but remember to also put a current limiting resistor on the base of the transistor. A 1K resistor should be fine.

Jean

Re: MS2/Extra Alpha 3.0.3 20090627 (was 20090623)

Posted: Mon Nov 02, 2009 10:59 am
by racingmini_mtl
I should add that you want the transistors before the pull up resistors. You probably knew that but I just wanted to make sure.

Jean