MS2/Extra 2.0 bug list (OLD, for REFERENCE ONLY)

[Maglin]

This is probably just a ini change required. In Knock Sensor Settings you have Recovery Advance (deg) greyed out in safe mode and available in Aggressive mode. Well It's only used in safe mode according to the MSII doc's. Also it's a little hard to initially find the knock setting. You currently have to go into the threshold table and then into settings. Maybe it's always been that way but it took me a few minutes to find it.

[Keithg]

I noticed during a warm restart (and bump start) that if you are on the throttle upon engine start up that o2 regulation will start before waiting for the sensor to warm up. The result is very lean running until the sensor warms up.

Also with o2 regulation, would it be a good idea to have a delay to reengaging after a fuel cut? I have set my transport delay according to my log and it still wants to overly enrich after a fuel cut. Maybe it is better to have a too long transport delay than a short one. With a NB sensor, would this make any difference?

I am running an LC-1

KeithG

[wuffwaff]

Hi,

at the moment it is not possible to use two wide band sensors (one for each cylinderbank) together with EAE.
It is also not possible to use only one lookup (VE) table for both sensors.

William

[Keithg]

is there a problem with regular AE? When I use regular AE, GEGO acts very odd. It will snap to max % and then back. This is with PID and with transport delay. I have not yet tried regular GEGO.

EDIT... It appears that I needed more MAP lag as AE was being triggered inadvertantly. disregard...

KeithG

[muythaibxr]

The next beta will have fixes for all listed bugs except:

1) delay after fuel cut before re-enabling O2 correction
2) Fixes to allow EAE to work with dual table, and fixes to allow dual O2 to work wit ha single table.

This (hopefully) will be done later this week.

Ken

[muythaibxr]

For those who have reported that entering a number greater than 25.5 on the tpsdot threshold in the accel wizard causes it to roll over, you're right.

However, it appears that this is a megatune bug in the standard accel wizard dialog as far as I can tell, so there's not really anything we can do about it.

Ken

[UnaClocker]

Megatune is open source.. ;) Someone should fix it. ;) Right?

[muythaibxr]

Someone could, but not me!

Ken

[Keithg]

ken,

Will the idle code be changed at all from the last S19 you posted? I drove on it this weekend quite a bit. I can still hgang the code pretty often at a DC and rpm above the threshold and it will not drop into PID control unless I pull engine rpms down with the clutch. This is very apparent during WUE and after WUE ends, though it can also happen on a warm engine.

KeithG

[muythaibxr]

Set the rpm threshold a bit higher then, or the dashpot a bit lower.

Otherwise, I will have to just disable the rpm threshold altogether, and make it purely time-based.

That's what I had done before, but at least one person didn't like that either.

I'm sure if you tinker a bit with the rpm threshold, you can get it working.

Ken

[Keithg]

I'll see what I can wrestle with. Went out for lunch and had no stalls.

I think I may have found a bug... IN the regular Accel Wizard, It seems to ignore the MAPdot threshold. I have it set at 60 kpa/s and it triggers accel on the steps set in the blocks even if they are below the threshold.

KeithG

[muythaibxr]

OK, I'll look at it.

You gave up on EAE?

Sounded to me when you last mentioned it you were going to figure out what your opening time was, and set up the lag compensation.

Ken

[Keithg]

Ken,

I have not given up on EAE, merely reduced the number of variables I
am wrestling with currently. I got pretty overwhelmed with the details
of installing MS2/E. So much more to think about and I am trying to
get my VE table dialed in first. My plan is to get idle control dialed
in next, then deal with EAE. I think I understand what is going on
with EAE, but the 6 graphs which all interact, but can only be
addressed separately is a bit much to throw at me all at once with
trying to keep the car from stalling.... Is there any way to put more
than one of these on the same graph? I know that is an MT issue and
not MS2/E, but it would help to see at least 2 at once. Also the
ability to see more than the one gage at once on the graph would help
tuning it.

Another thing that I did not consider was all the jitter and other A/D
and D/A issues that are needed to keep MS2/E smooth. MS1/E was slow
enough that these had very little effect, but they are kind of
critical here because they all interact. Map jitter triggers EAE, EGO
correction messes with EAE if RPM or MAP varies while it is
correcting... O2 lag affects PID o2 control which then fights EAE and
PWM idle and it all just messes you up!

As for opening time, I have figured it to be about 0.745 ms on mine. I
did this by idling and adjusting the OT and MAP until there was little
difference between 1 squirt and 2. It is funny that with MS1/E HiRes,
I determined it to be 0.72 and somewhere along the way, I was
convinced it was closer to 1 and now I am back where I was... How do
you verify OT? I tried to load it and wait for EGO to bring it to 14.7
with 1 and 2 squirts. With EAE, this is nigh impossible as the 1% of
so that EAE swings around can have a significant effect on OT unless
you can put it on a dyno and hold it constant. I have to do this while
driving and logging. I was not able to get 2 runs at the same load/rpm
(I have to do this solo, which makes it harder) Which is why I turned
it off for now. Once I am happy with OT, PWM idle and VE, I'll go back
to try to get EAE right.

KeithG

[muythaibxr]

Ken,

I think I understand what is going on with EAE, but the 6 graphs which all interact, but can only be
addressed separately is a bit much to throw at me all at once with
trying to keep the car from stalling.... Is there any way to put more
than one of these on the same graph? I know that is an MT issue and
not MS2/E, but it would help to see at least 2 at once. Also the
ability to see more than the one gage at once on the graph would help
tuning it.

If you follow my suggestions for tuning, it becomes a lot easier to tune... You basically tune the main curves at cruise rpms, then tune the rpm curves to make idle and high rpms better, then tune CLT until the car becomes drivable while warming up.

Another thing that I did not consider was all the jitter and other A/D
and D/A issues that are needed to keep MS2/E smooth. MS1/E was slow
enough that these had very little effect, but they are kind of
critical here because they all interact. Map jitter triggers EAE, EGO
correction messes with EAE if RPM or MAP varies while it is
correcting... O2 lag affects PID o2 control which then fights EAE and
PWM idle and it all just messes you up!

I have added Al's sampling rate change from one of the later ms2-standard codes to ms2/extra. I did not copy his code exactly, as his code is a lot different from ours, but I made it work the same. Basically the MAP sensor is only ever sampled immediately after a "tach" pulse. This change should keep throttle response alright, but smooth out the MAP signal to the point where you can possibly even increase the MAP lag factor.

Once I am happy with OT, PWM idle and VE, I'll go back
to try to get EAE right.

KeithG

This is probably the best way of doing things anyway. Once your VE is tuned right, EAE (or any AE for that matter) becomes a lot easier to tune.

Ken

[muythaibxr]

Bugs outstanding (even in today's beta, which I'm about to release):

1) delay after fuel cut before re-enabling O2 correction
2) Fixes to allow EAE to work with dual table, and fixes to allow dual O2 to work with a single table.
3)IN the regular Accel Wizard, It seems to ignore the MAPdot threshold. I have it set at 60 kpa/s and it triggers accel on the steps set in the blocks even if they are below the threshold. (Not yet reproduced).

Ken

[Ballistic]

Has next-cylinder mode ignition been addressed?

[Gokart]

beta 2;
Dwell gauge (when activate in megatune) is not displaying the value we input into the 'more ignition option' table. The cranking and running values are way too high as displayed in the gauge. I have to decrease almost half the value to get to a target dwell. Which is right? the gauge or the entered value?

[UnaClocker]

I've always felt the dwell control was WAY off.. I've been told it calculates it differently, but your findings kind of confirm my feelings. I run a dwell of 1.4 in MS2, and 3.6 in MS1.. Kind of an extreme difference..

[UnaClocker]

Ok.. BUG REPORT...
In the latest beta (2nd beta), PA0 is no longer working. Not sure if it was working in the 1st beta, I didn't run the engine up to temp to trigger my radiator fan.
Warmup LED is now working, wasn't back when PA0 worked for me.

[muythaibxr]

I've always felt the dwell control was WAY off.. I've been told it calculates it differently, but your findings kind of confirm my feelings. I run a dwell of 1.4 in MS2, and 3.6 in MS1.. Kind of an extreme difference..

The dwell in ms2/extra is voltage corrected. Most likely this is what is causing your problem.

This was done on purpose.

I have to check, but I believe the entered number is at 14 volts, not at 12 like it was with ms1/extra... which would explain dwell being a lot higher at idle and crank than you expect.

Ken