MS3 pre-1.4 beta 13 released

[jsmcortina]

To announce the release of “MS3 Pre 1.4 beta 13″
ETA on this 1.4.x code becoming release is 2 months.

Who should use this code?
Experienced users wanting to test the new features.

Who should not use this code?
New or casual users – please use the current release code instead – see the Downloads page linked above.

Main changes since pre-1.4 beta 12
Bug fixes:
Fuel temperature (primary) degC fix.
Fuel temperature (secondary) degC tweak – applies to all sensor transforms, not just CLT,MAT
VVT change – set vvt_calc on each cam tooth always. (PF issue.)
Dual Fuel accel-pump AE was applying to both channels in error.
Fix a bug where AC/fan idleup can fight with PID when PID control interval is lower than the AC/fan idleup delay.
Fix a small ini error in boost_vss. Tool-tip said the feature worked with OL boost. It doesn’t.
Fix pinport usage code for VSS3,4

New / changed features:
Add some manual links to ini file.
Default to not dividing by outpc.baro in fuel calc. Add compatability option for it.
Remove TPS bypass limit, use hard limit.

Minor / internal:
ini tweak testint
Rearrange revlim calc.
Baro defaults to off.
Re-enable sensors logging for Coyote VVT3 debugging. Disable VVT1 debugging.

Known bugs
–

The firmware zip is available from:
http://www.msextra.com/downloads/dev/ms3/1-4/beta-13/

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

Upgrading users are advised to enable the “old baro calc” for backwards compatability.

Some of the ini changes will result in harmless warnings when loading MSQs from 1.3.x
Users upgrading from 1.2.x or earlier codes need to pay more attention.
TunerStudio 2.6.x is REQUIRED.

James

[prof315]

I'm going to load this up on my car today. Got a question though..... Will there ever be a 4G63 decoder with sequential phasing for 1.6L Miata engines? I have a customer who wants to switch to COP and sequential fueling, but not until there is a decoder option for it so we can avoid rewiring the injectors

[jsmcortina]

I'm going to load this up on my car today. Got a question though..... Will there ever be a 4G63 decoder with sequential phasing for 1.6L Miata engines? I have a customer who wants to switch to COP and sequential fueling, but not until there is a decoder option for it so we can avoid rewiring the injectors

How about if you added 360 deg to the tooth#1 angle and the code re-arranged the outputs to suit? That enables maximum code re-use and would be in the docs. Or is it better to add a specific mode name?

James

[ashford]

will give this a try to-nite

also the link has a "-" in it the website does not so it 404's
http://www.msextra.com/downloads/dev/ms3/1-4/beta13/#

[jsmcortina]

Thanks for pointing that out. The original link should work now.

James

[prof315]

I'm going to load this up on my car today. Got a question though..... Will there ever be a 4G63 decoder with sequential phasing for 1.6L Miata engines? I have a customer who wants to switch to COP and sequential fueling, but not until there is a decoder option for it so we can avoid rewiring the injectors

How about if you added 360 deg to the tooth#1 angle and the code re-arranged the outputs to suit? That enables maximum code re-use and would be in the docs. Or is it better to add a specific mode name?

James

That works fine. I wasn't aware that I COULD add 360 deg to tooth#1 with that decoder. I know it has negative degree limits and I thought it had positive limits as well.

[jsmcortina]

I'm going to load this up on my car today. Got a question though..... Will there ever be a 4G63 decoder with sequential phasing for 1.6L Miata engines? I have a customer who wants to switch to COP and sequential fueling, but not until there is a decoder option for it so we can avoid rewiring the injectors

How about if you added 360 deg to the tooth#1 angle and the code re-arranged the outputs to suit? That enables maximum code re-use and would be in the docs. Or is it better to add a specific mode name?

James

That works fine. I wasn't aware that I COULD add 360 deg to tooth#1 with that decoder

You can't! (yet)

I was mainly posing the question of what is most logical and works best for the user going forwards.

James

[prof315]

Ok got it. I'm all for keeping the code tight. And yeah a line in the 4G63 documentation noting that for 1.6L Miatas you add 360 deg to your trigger angle for correct phasing should work.

The idle up "bug" seems fixed. My car used to exhibit the described behavior and my settings meet the parameters for it to happen but my fans don't pull rpms down hard enough to make a fuss over it.

[shaodome]

Hey James,

Does the below also address the issue with VSS2? In doing some testing with Ben from DIY it was found that VSS2 was not responding to digital in 2.

Code: Select all

Fix pinport usage code for VSS3,4

[jsmcortina]

Hey James,

Does the below also address the issue with VSS2? In doing some testing with Ben from DIY it was found that VSS2 was not responding to digital in 2.

That's the first I've heard of that. Please start a new topic with MSQ/datalog.

James

[ashford]

i did a quick test on the coyote, it don't appear to have problems above 1000 rpm. currently i am using one of jeans duar vr boards which gave much worse behaviour than when i used the opto, on a scope the first tooth of the "three group" triggers looked longer than it should like it was triggering early, this may be the current issue. i can go back to the opto setup if you want me to use that for testing.

in this log i took i had the min and max angles zeroed at the beginning of the log, then switched to old settings about 1/3 of the way in. i noticed when it was zeroed the vvt3 error was not flashing, but when settings inputted it started to.

was there any changes in the code to improve this as it seems to act better now?

[jsmcortina]

i did a quick test on the coyote

It appears that you are triggering on the end of each tooth, have you tried flipping the capture edge on VVT3 ?

What I _think_ may be happening is that some teeth overlap the missing tooth region when the cam tooth count gets reset to zero and it is then confused about which tooth is which.

OR... it is getting a false trigger on one cam edge which then puts the other teeth out of line. I've got an idea for adaptive filtering for 1.5, but don't want to make that change at this development stage in 1.4.

James
EDIT: if you are able to test with the capture edge on cam 3 flipped, please record a datalog, composite log and engine log if the angle is still jumping around.

[ashford]

if memory serves me swapping capture edge resulted in about 15* retard with rpm. ill give it a try though, but it does seem to act better than beta 12.
if im not layed up after work ill, try and get some scopes of it along with logs etc

[jsmcortina]

if memory serves me swapping capture edge resulted in about 15* retard with rpm. ill give it a try though, but it does seem to act better than beta 12.
if im not layed up after work ill, try and get some scopes of it along with logs etc

As you have VVT1,2,4 behaving correctly, I'm confident that VVT3 can be made to work too. I'd like to get this completed before 1.4 becomes "release" in the very near future.

James

[ashford]

i took all requestid logs and 2 scope traces one raw from the sensor with 660 ohm pullup and on after the vr conditioner. to look at them download the software from the usbscope i have http://www.hobbylab.us/discodownload . it appears that the rising edge of the signal floats about a bit(collector opens)at lower rpm.

if the code relies on both edges this will cause a big problem. both intake and exhaust cams use the same trigger wheel it just that the exhaust use a vr sensor and the intake uses a geartooth sensor. it is described as a vr sensor with internal conditioner. the other thing that bugs me is that when the collector is closed voltage is @2v.

i am going to switch back to the opto since that gave me the best results so far, then see how it works.

[jsmcortina]

i took all requestid logs and 2 scope traces one raw from the sensor with 660 ohm pullup and on after the vr conditioner. to look at them download the software from the usbscope i have http://www.hobbylab.us/discodownload . it appears that the rising edge of the signal floats about a bit(collector opens)at lower rpm.

You are correct - floaty edges won't help.

if the code relies on both edges this will cause a big problem.

Just a single edge.

. the other thing that bugs me is that when the collector is closed voltage is @2v.

i am going to switch back to the opto since that gave me the best results so far, then see how it works.

That sounds like a good approach.

James

[ashford]

did some testing with vvt3 back on the opto. engine logger looks much better. about the only pattern i have seen is when a rapid increase in cam angle at lower mid rpm(changed commanded) through 10-15 degrees is when most errors occur. generally it would wildly oscillate after this sometimes without stopping, i reduced the pid numbers a bit and the oscillating duration decreased to a few engine cycles after the error, seems to have no effect on vvt1 though. i am out of ideas to try, about the only thing i can think of is to swap vvt1 and 3 sensors around, but this requires the removal of the intake. if you think it is worthwile i will do it.

[aidandj]

I'm going to load this up on my car today. Got a question though..... Will there ever be a 4G63 decoder with sequential phasing for 1.6L Miata engines? I have a customer who wants to switch to COP and sequential fueling, but not until there is a decoder option for it so we can avoid rewiring the injectors

I and quite a few others are running sequential fuel and spark on 1.6 miatas. What is the issue that is stopping you from doing it?

[prof315]

I'm going to load this up on my car today. Got a question though..... Will there ever be a 4G63 decoder with sequential phasing for 1.6L Miata engines? I have a customer who wants to switch to COP and sequential fueling, but not until there is a decoder option for it so we can avoid rewiring the injectors

I and quite a few others are running sequential fuel and spark on 1.6 miatas. What is the issue that is stopping you from doing it?

Are you doing it with the stock 4G63 decoder and an unmodified CAS? The phasing is set up for a 1.8L which has the CAS driven by the exhaust cam it's 360deg off when used on a 1.6 (CAS driven by the intake cam)

[prof315]

I'm issue free with beta 13 and over 500 miles on the car.