MS3 pre-1.6 alpha 1 firmware (superceded)

[jsmcortina]

To announce the release of “MS3 pre-1.6 alpha 1” this adds a number of changes over “MS3 1.5.2 release”.

Who should use this code?
Users wanting to test the new features or as directed by tech support to workaround known issues in ‘release’ version. It is required that alpha/beta testers assist developers by reporting any issues found and providing tunes and datalogs to demonstrate the issue.

Who should not use this code?
Unless directed by tech support, new users should use the current release code instead – see the Downloads page.

Main changes since 1.5.2 firmware
Bug fixes
Don’t trip overboost when not-running or cranking.
Allow dome test mode to run regardless of MAP.
Fix stepper idle problem limiting homing steps in error.
Fix stepper idle test that wasn’t holding at homing position.
Don’t apply idle RPM correction when idle target RPM is zero.
Batch fire crank resync fix – reset glob_sequential after crank if it was changed.

New/changed features
Closed-loop boost, single-solenoid. Add bias range option.
Add TPS-based MAP prediction.
Add support for Kia IMU over CAN for accelerometer inputs, X, Y, Z=yaw.
Add feature to lookup launch limit based on MAP to help with turbo spooling.
Add 11th,12th spark outputs for MS3PRO hardware.
Add cam decoder for Barra. (Set 6 cam teeth.)
Add “Linear 0.5-4.5V” Generic Sensor transformation.
Add on/off bitfield datalog channels. Tweaked version of data supplied by EFI Analytics.
Change layout of I/O testmodes to show usage too. (Needs checking on all hardware.)
PWM fan control.
Launch – option to ignore RPM requirement.
Add VSS1 vs. VSS2 timing retard table as additional Traction method.
Changed injector deadtime logic to use lookups only and can include pressure.
Extend boost vs. speed/time etc. to 12 points.
For boost vs. time, hold at zero time during launch.
Apply rewritten VQ35 decoder from 1.5.2.1

Minor/ internal
Remove PW4X.
Add recommended braces to A/C idle/fan condition.
Change MAP calibration displayed units to kPa.
Fix typo on lotus dash setting.
Do set DDRS for MS3, MS3-Pro gen1, MS3-Goldbox.
‘gear’ changed to signed in ini.
Reduce no. knock channels from 16 to 12 to free up outpc space.
E46 CAN broadcast tooltip note.
Shift light bulb test.
VVT delay.
Allow knock_conf_debug on MS3Pro if num_cyl is small enough.
Return an error if incorrect data length sent to calibrate AFR table.
Attempt workaround for injOpen1 in ini to stop TS complaining.

The firmware zip is available from: https://www.msextra.com/downloads/dev/m ... -6/alpha1/
** Injector deadtime method has changed. You MUST check and reset your deadtime. **

Be sure to read the README.txt and RELEASE-NOTES.txt that are supplied in the zipfile.

Some of the ini changes will result in harmless warnings when loading MSQs from 1.4.x Users upgrading from 1.2.x or earlier codes need to pay more attention.
TunerStudio 3.1.xx or later is recommended for correct serial validation and SDcard format operation.

EDIT: Bugs that have come to light since release:
Knock is broken. Fixed for next release.

[nathaninwa]

Thanks James! I will get this loaded and I have an event next weekend.

[mapdaddy]

So how will this change my deadtime exactly?

Can we add to this a scramble boost feature?

And thank you so much as always!

[jsmcortina]

Per request from DIYAutoTune, the injector deadtime uses the curve to lookup the value itself, instead of being a percentage modifier.

inj deadtime new.png (42.63 KiB) Viewed 3831 times

I've been using this firmware (more or less) for the last three weekend's racing and about 500 miles. The features I use appear to work as expected.

Do remember that this is an "alpha" code though! Be sure to have copies of your previous firmware and as always "File > Save Tune As" to keep a copy of your old tune, in case you run into a problem and need to revert.

James

[Six_Shooter]

Per request from DIYAutoTune, the injector deadtime uses the curve to lookup the value itself, instead of being a percentage modifier.

inj deadtime new.png

I've been using this firmware (more or less) for the last three weekend's racing and about 500 miles. The features I use appear to work as expected.

Do remember that this is an "alpha" code though! Be sure to have copies of your previous firmware and as always "File > Save Tune As" to keep a copy of your old tune, in case you run into a problem and need to revert.

James

This will make it so much easier than finding the calculator that someone made to convert from what is usually provided with injectors to what MS needed.

[nathaninwa]

I like the test feature that tells what pin assignment is.

[rsabol77]

Is it just me or is the stop engine latching in or something? I used it last night and I had to power cycle ecu to get spark cut to clear after using the engine stop button.

[PhoB]

Thanks, developing is not death i stared to be sceptical

Quick look, on Knock setup, there is missing input "internal" its probably bug due adding new "analog" option?
Is there any explanation how analogue knock sample and window input works?

Boost closed loop min/max range is still not enought, my opinion and experience, it need to be set positive and negative separately. And PID delay timer to avoid too soon control also(same like already exist in idle control). Otherwise its usable maybe for dragrace, not on streetcars.

EDIT: sorry, yes analogue knock , sample and window is already in prevousion version, i just used internal option every time so now, when it was not possible to choose, saw these parameters.

[jsmcortina]

Thanks, developing is not death i stared to be sceptical

Quick look, on Knock setup, there is missing input "internal" its probably bug due adding new "analog" option?
Is there any explanation how analogue knock sample and window input works?

Internal is supposed to be there and has gone AWOL due to a typo. Analog isn't new, it originated in MS2.

There's a description of the settings in the manuals.

James

[PhoB]

I tried to rewrite row 707 of ini, INVALID to Internal and option is sucesfully back. Will it work as temporary fix ?

And if developers worked around knock, why is still there "ignore knock above MAP" ? is it difficult to change this to "under MAP" which some of us called for here, as it make more sense of use or ?

[jsmcortina]

I tried to rewrite row 707 of ini, INVALID to Internal and option is sucesfully back. Will it work as temporary fix ?

The typo was only in the ini file, so your change is correct and will work.

There were no firmware changes around knock.

James

[wes kiser]

The injector deadtime changes will make life easier for alot of people.

Thank you for the PWM fan. One suggestion is their needs to be a "duty adder" to the minimum with AC on. Also, regarding the start behavior, does it go straight to minimum duty or does it ramp up over a time period?

[jsmcortina]

Thank you for the PWM fan. One suggestion is their needs to be a "duty adder" to the minimum with AC on. Also, regarding the start behavior, does it go straight to minimum duty or does it ramp up over a time period?

My thinking when adding this feature was that due to the soft-starting, the closed-loop idle control should cope with the increased load on its own.
The fan will start at the minimum duty and then ramp from there.

I'm using the Dorman 902-310 SSR on my fan and the fan won't move at very low duties. My car could probably do with a larger radiator, but with the 14" single fan and PWM control it has been behaving better than ever before.

James

[wes kiser]

I mean there should be a duty adder for the fan for AC on. For instance, if minimum duty is 20%, then if there was a 20% adder for AC on the minimum duty with AC on would be 40%, then ramp up with engine temp toward max duty. This is more for AC operation than engine cooling.

[Dennis_Zx7r]

Changed injector deadtime logic to use lookups only and can include pressure.

This is great, thanks! However when I first selected this to have a look at it it messed up my DT table with -214748364.8 everywhere and even restore tune wouldn't fix it when switching back to "classic" DT compensation and junk all the DT tables. Filled the tables and will try this asap, but may take some time as the season's basically over.

[jsmcortina]

I wouldn't expect "restore tune" to help. This deadtime vs. voltage curve and deadtime table are new data, so you won't have anything to restore.
However, when connected online, TunerStudio should populate the default values that are in the firmware instead of -infinity.

James

[5.0Thunder]

James, how is that 902-310 relay working for you? I had it on a pair of 1700CFM spal fans and after using it for a couple days in the texas heat, it started acting up. As the engine bay temps would get very high, the fan speed would slow down and oscillate. I think the relay was overheating.

Also on the topic of this firmware, if i had flex fuel set up previously, will the new baseline percentage calculation effect my tune?

[jsmcortina]

My SSR has worked for me so far, but UK late summer is very different from Texas heat !

Were you previously using the Flex baseline setting?

James

[5.0Thunder]

Were you previously using the Flex baseline setting?

James

Yes, Since my blend was 75% ethanol at the time I used 75% as the setting.

[andreiB]

I tried to rewrite row 707 of ini, INVALID to Internal and option is sucesfully back. Will it work as temporary fix ?

The typo was only in the ini file, so your change is correct and will work.

There were no firmware changes around knock.

James

hello!I changed the ini file as written in line 707,knock appeared but the sensor settings menu knock parameters is not active

2022-10-11_20-58-39.png (20.43 KiB) Viewed 3249 times