MS2/Extra Alpha 3.0.3 20090507 (was 20090426)

[racingmini_mtl]

That's just a warning and there may be more of the same. There must be some other error message. Some functions have been moved to other memory banks so it is possible that if you were jumping to one you will now need to call them (call instead of jmp).

Send me the complete output text you get from make and your code if you can't figure it out.

Jean

[the_joker82]

Hi, finally we can try to start our vmax!
We have all the components to finish to assemble the ms (i.e. addictional drivers for the injectors),
but we are not sure how they are linked to the ms.
Were we can find the schematic to add new circuits?

Thank you very much

Luca

[racingmini_mtl]

This shows you where to connect the new injector drivers on a V3.0 board. More information about the needed modification is available on the same page.

Jean

[spyro]

im trying to use semi sequential on a 60-2 wheel but with no success it always runs rich like you are using 950cc injectors (im using 470cc on a 1600cc engine)
is there any manual on how to setup semi sequential
i have tried different angles but with no success

i have new 950cc injectors but with normal ms2extra code i cant control them, low pulswitdh thats why im trying semi sequential

[racingmini_mtl]

Semi-sequential may not be the best with huge injectors. The problem is that you will have 2 squirts per cycle which means very short pulse widths. And if your opening time is not correct it will be even worse.

If you have a working VE table on the non-sequential code, you should be able to use the same one with the sequential code as a starting point. However, you have to make sure that it is for the same number of squirts or make the needed adjustment to the req fuel value. If you had 2 squirts alternating before then with sequential it will be the same but with semi-sequential it will be equivalent of 2 squirts simultaneous.

The table here tells you how many squirts per cycle you engine configuration will have with the sequential code. You have to make sure the req fuel value reflects that.

Jean

[kritip]

Just to confirm I have run the latest beta for 10-20 miles and have no issues to report so far.

Sometimes when i start though, it runs lovely and smooth, other times it seems a bit lumpy, as if it's not synced properly. Is there anyway to confirm that the cam sensor is being picked up and it's sync'ing correctly?

One I go COP it will be more obious as if the sparks in the wrong place, then it not going to run at all lol.

Cheers,

Kristian

[racingmini_mtl]

What you can do is have a look at the composite logger. If there is an issue with the cam trigger you should see it there and it would cause an issue with syncing.

Jean

[kritip]

On the bench today, the composite logger was giving loads of sync errors whilst running. Turning it off cleared them up. I have tried the composite logger on the car, but it doesn't always show the 60-2 wheel, and when it does, i get random bursts of data on cam sensor. I gave up trying with it

[racingmini_mtl]

Yeah, it seems to not be behaving well with 60-2 on the bench for me too. I hadn't tried it yet with 60-2. I'll have to check with Phil and James to see if the problem is on the MS side or the TS side.

Jean

[Dennis930]

racingmini_mtl wrote

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.

Jean,

I removed and reloaded TS and reloaded the firmware on MS, but I still have the selection for type of Idle Control greyed out. However, if I load the MS2 Extra example MSQ provided with TS, the Idle Control dialog box is fine. Is "base.ini" the correct ini file for the alpha code?

-Dennis

[racingmini_mtl]

Do not use base.ini. Use "megasquirt-ii.ini.ms2extra" and rename it "megasquirt-ii.ini". It is located at the same place as the "download-MS2-firmware.bat" and "ms2_extra.s19" files. This assumes you have an MS2 and not microsquirt.

The base.ini file is not a real ini file but is used by the make file to generate the ini files for MS2 and microsquirt.

Jean

[Dennis930]

Thanks Jean, that fixed it!

-Dennis

[KaPower]

How does cranking mode work in semi seq? Do the injectors squirt twice every 720 degrees, instead of squirting 4 times every 720 (once per cylinder) like the old cranking system?

What do you think is the best injector wiring for semi-sequential? 1-4 for Bank1, 2-3 for bank2 seems like the best pairing to me, since at least 1 of the squirts can be on a closed valve right before it opens during the intake stroke.

It's too early for me to comment on any possible improvement that semi-seq has made for me, but I can say that so far everything seems to be functioning.

[KaPower]

Semi-sequential may not be the best with huge injectors. The problem is that you will have 2 squirts per cycle which means very short pulse widths. And if your opening time is not correct it will be even worse.

My 2400cc engine with 740cc high Z injectors, actually has a smoother idle, and much, much, much! better throttle response with 2 simultaneous or 4 squirts alternating, 2 alternating would run ruff, and have a major hesitation off idle during tip in. The injector pw was quite low at idle, 1.3ms with 4 alt.

IMO the biggest problem with low pulsewidths is non linear fuel flow, MS assumes fuel flow is linear so corrections made at low pulswidths (GammaE, mat, clt, ego, ve, map...) are not accurate...

[racingmini_mtl]

How does cranking mode work in semi seq? Do the injectors squirt twice every 720 degrees, instead of squirting 4 times every 720 (once per cylinder) like the old cranking system?

Yes, it's one squirt per rev. The sequential code doesn't change the injection mode when cranking so semi-sequential mode will have 1 squirt per rev and sequential mode will have 1 squirt per cycle (720 degrees).

What do you think is the best injector wiring for semi-sequential? 1-4 for Bank1, 2-3 for bank2 seems like the best pairing to me, since at least 1 of the squirts can be on a closed valve right before it opens during the intake stroke.

The injectors should be paired the same way as the ignition coils which is what you have.

It's too early for me to comment on any possible improvement that semi-seq has made for me, but I can say that so far everything seems to be functioning.

Good to know that it seems fine. Let me know if you find anything strange and also what improvements, if any, you have once you've played with the settings.

Jean

[racingmini_mtl]

Semi-sequential may not be the best with huge injectors. The problem is that you will have 2 squirts per cycle which means very short pulse widths. And if your opening time is not correct it will be even worse.

My 2400cc engine with 740cc high Z injectors, actually has a smoother idle, and much, much, much! better throttle response with 2 simultaneous or 4 squirts alternating, 2 alternating would run ruff, and have a major hesitation off idle during tip in. The injector pw was quite low at idle, 1.3ms with 4 alt.

IMO the biggest problem with low pulsewidths is non linear fuel flow, MS assumes fuel flow is linear so corrections made at low pulswidths (GammaE, mat, clt, ego, ve, map...) are not accurate...

With sequential injection you should be able to use a single squirt per cycle (which is what you'll get anyway with the sequential code) and still get a smooth idle and good throttle response because the squirt for each cylinder will be done at the same time in the cycle so you have the same behaviour for all cylinders.

With batch injection you will instead have one cylinder with raw fuel shot in the cylinder, another one with the full cycle to vaporize the fuel in the intake manifold before it goes in the cylinder, and 2 others with something in the middle. And since all the injection pulses are separated by 180 degrees, you get a more up to date pulse width during transitions which will help throttle response.

While I can see how the non linearity will be bad for short pulse widths, if you can double the (effective) pulse width just because you're using half the squirts you will get a much smaller effect from these non-linearities.

Jean

[KaPower]

While I can see how the non linearity will be bad for short pulse widths, if you can double the (effective) pulse width just because you're using half the squirts you will get a much smaller effect from these non-linearities.

Agreed, but the improvement in idle quality and tip in throttle response out weighs the improvement of a slightly longer pulse width. (1.6ms 2squirts alt, 1.3ms 4 squirts alt...)

-JM

[racingmini_mtl]

While I can see how the non linearity will be bad for short pulse widths, if you can double the (effective) pulse width just because you're using half the squirts you will get a much smaller effect from these non-linearities.

Agreed, but the improvement in idle quality and tip in throttle response out weighs the improvement of a slightly longer pulse width. (1.6ms 2squirts alt, 1.3ms 4 squirts alt...)

-JM

But like I said in the part of the message you cut off, the idle quality and throttle response are both improved by running fully sequential injection even if it only has a single squirt per cycle. So you would have both advantages. You should try it if you can.

Jean

[KaPower]

But like I said in the part of the message you cut off, the idle quality and throttle response are both improved by running fully sequential injection even if it only has a single squirt per cycle. So you would have both advantages. You should try it if you can.

Jean

Ok you talked me into it! I was going to wait for MS3 this fall, but I just ordered your sequential PCB. Looks like I have to modify my wiring harness agian

JM

[KaPower]

I don't see a parts list, or any pic's that answer my questions about the sequential injection pcb, so I'll ask you here. What resistors do I need? are both DB15 connectors included?

-JM