MS3 pre-1.6 alpha 2 firmware (superceded)

[Motterdude]

I do appreciate the help, thanks.

Given that, there is still a question I aksed pertaining alpha firmware that remained open.

2) What is the added "VVT Delay" feature in version pre-1.6 alpha 1 supposed to do? Is it just a delay in the VVT setpoint application (like VVT Maximum Slew) or is it something else?

Thanks James

These questions aren't related to the alpha firmware, they are just general support.
The loop 'outputs' are on the "Programmable on/off outputs 2" page.

James

[jsmcortina]

2) What is the added "VVT Delay" feature in version pre-1.6 alpha 1 supposed to do? Is it just a delay in the VVT setpoint application (like VVT Maximum Slew) or is it something else?

The "After start delay" ? Please check the Tooltip.

James

[Motterdude]

You are very much right.

Appreciate it.

2) What is the added "VVT Delay" feature in version pre-1.6 alpha 1 supposed to do? Is it just a delay in the VVT setpoint application (like VVT Maximum Slew) or is it something else?

The "After start delay" ? Please check the Tooltip.

James

[Djovani_jr]

Is it possible to add this feature?

viewtopic.php?t=71348

[AbatelliCristian]

In my first test, CANIN not work well in alpha2. I have test canin in ALS activation and burnout activation but no work. It is always active
Now reload 1.5.2 and retray..

[jsmcortina]

In my first test, CANIN not work well in alpha2. I have test canin in ALS activation and burnout activation but no work. It is always active
Now reload 1.5.2 and retray..

If the problem is confirmed, please attach tunes and datalogs from 1.5.2 and pre-1.6 alpha 2.

James

[AbatelliCristian]

I have reload firmware alpha2 and now work. I don't know what happened in my previous update.
But I have other question. ALS with DBW enabled, with dbw enabled it doesn't follow tables and As soon as I touch the gas pedal, the strategy switches off automatically.
Is normal with DBW?

I attach video and log
https://youtu.be/zFcYTB4ITqo

[MalcolmV8]

Is it possible to get MAT based timing retard table expanded to take into account engine load in 1.6? That way it only pulls timing when in boost or high loads and not at idle and part throttle driving.
More detailed post here
viewtopic.php?t=77324

Thank you

[Blondienys]

In my first test, CANIN not work well in alpha2. I have test canin in ALS activation and burnout activation but no work. It is always active
Now reload 1.5.2 and retray..

If the problem is confirmed, please attach tunes and datalogs from 1.5.2 and pre-1.6 alpha 2.

James

James would you be able to send me a private message please

[Mustang5L5]

I'm curious about the PWM fan control. I don't have 1.6 so i can't see the menus myself.

Is there any setting for A/C on in terms of commanding a min duty cycle? In other words, i want to control fan duty cycle based on coolant temp, and turn the fan off above a certain vehicle speed, but if i put my A/C on, i want my fan speed to default to MAX speed, or perhaps a min duty cycle setting (80%) and allow coolant temp to command it higher if my table is created to call out 100% at higher temps.

I was doing this with a blend of building a PWM output and a few on/off relays connected to triggers, but would be nice to go back to a single pwm wire and free up those on/offs and have more control.

[prof315]

I would still LOVE to see rev limit by gear for endurance racing.

[PhoB]

I know its not this exact firmware version related, but VVT was mentioned here.
Is it good idea to make some return rate parameter on Hold duty? I mean if i use hold duty value, no matter of PID, ecu jump to hold duty value immediately when target is met.
Single value seems not exact for every rpm, oil pressures/temperatures , so it tend to make like sawtooth like duty result and also actual cam angle drifts up and under target constantly. Assuming slowing return rate to hold value, can make it more stable?

[Reverant]

VVT afterstart delay seems to be a "power on" delay and not a true afterstart delay. Ie if you set it to 10 seconds, and just wait 10 second without starting, VVT will start working immediately. Can we get a true afterstart behavior?

[jsmcortina]

VVT afterstart delay seems to be a "power on" delay and not a true afterstart delay. Ie if you set it to 10 seconds, and just wait 10 second without starting, VVT will start working immediately. Can we get a true afterstart behavior?

That is not desired behaviour for sure. Do you have a datalog to demonstrate it?

James

[Reverant]

No, I tried it on a customer's car a few hours ago. The car has a single tooth on the cam with a VR sensor, and the MAX9926 doesn't like single teeth cam setups at cranking. After it starts, its fine, but until then it reads crazy VVT angles and maxes out the duty. We updated to alpha 2 to solve this issue but I noticed that it still happened some times. I narrowed it down after noticing that if you crank immediately after power on, there is no problem. If you wait for the configured delay (in our case, 10 seconds) then start, the problem is there. If you wait 5 seconds then crank, VVT pid kicks in after another 5 seconds.

[5.0Thunder]

Posting an example of new deadtime settings for Ford Racing LU80 Injectors (Siemens Deka 80lb) for confirmation. Please look over and let me know if this is the method we should follow.

Their data is found when testing at 39.15psi so used that exact data in the 39.2psi row of my table, then used their multiplier "OFFCOMP" table to apply the multiplier for the other pressure rows.

[BlueScreen]

Is it possible to get MAT based timing retard table expanded to take into account engine load in 1.6? That way it only pulls timing when in boost or high loads and not at idle and part throttle driving.
More detailed post here
viewtopic.php?t=77324

Thank you

This! It's also a big issue for me. When I slowly roll to a stop, or stopped at a red light with the engine fully warmed up, enough timing is pulled to stall the engine. I can work around by setting too much timing at idle, or not removing timing base on MAT and risking burning pistons when MAT is high enough. Both aren't great.
You can see in the screenshot from the attached datalog, at the end, the engine was struggling to run with ~5* of timing, would almost die, pick up some advance, then restart the loop and ultimately stalled. This was happening because 6* was remove because of MAT, but it wasn't necessary since I was stopped at a red light, 0% TPS, no timing should have been removed.

Screenshot.png (386.6 KiB) Viewed 3191 times

It's a feature that has been asked for years => viewtopic.php?t=69613. It could be very easy to do in my case, I simply don't want MAT retard when I'm not in boost. From the code available of version 1.4, it could be simple as that (untested):

ms3_ign.c

Before:

Code: Select all

outpc.mat_retard = intrp_1dctable(outpc.airtemp, NO_MAT_TEMPS, (int *) ram_window.pg8.MatTemps, 0, (unsigned char *) ram_window.pg8.MatSpkRtd, 8);    // %

After:

Code: Select all

if(outpc.ignload > 100)
{
    outpc.mat_retard = intrp_1dctable(outpc.airtemp, NO_MAT_TEMPS, (int *) ram_window.pg8.MatTemps, 0, (unsigned char *) ram_window.pg8.MatSpkRtd, 8);    // %
}
else
{
    outpc.mat_retard = 0;
}

It could of course be expanded by making the minimum load configurable, or being a 2d table with MAT and LOAD for axis, but this simple code would go a long way for me and surely many others. If source code for 1.5.2 or more recent is made available, I'm willing to code it and test it myself is necessary.

PS: I love your YouTube channel Malcolm. You don't know, but you helped me a lot with my MS3Pro PnP on my 97 Mustang GT this Winter.

[simon k]

If source code for 1.5.2 or more recent is made available, I'm willing to code it and test it myself is necessary.

Yes please!

[BlueScreen]

Posting an example of new deadtime settings for Ford Racing LU80 Injectors (Siemens Deka 80lb) for confirmation. Please look over and let me know if this is the method we should follow.

Their data is found when testing at 39.15psi so used that exact data in the 39.2psi row of my table, then used their multiplier "OFFCOMP" table to apply the multiplier for the other pressure rows.

5.0Thunder, I suspect there is an error in the calibration data. Look at the FNPW_OFFCOMP at 50psi, I quite convinced the correct value should be 1.0990. It would not make any sense that it's lower than 45psi and 55psi. That's why your 50psi row in TunerStudio look wrong compared to the row above and under.

I would also like to know if it's the correct method to calculate deadtime because I have those exact same injectors.

[KWIK427]

Posting an example of new deadtime settings for Ford Racing LU80 Injectors (Siemens Deka 80lb) for confirmation. Please look over and let me know if this is the method we should follow.

Their data is found when testing at 39.15psi so used that exact data in the 39.2psi row of my table, then used their multiplier "OFFCOMP" table to apply the multiplier for the other pressure rows.

5.0Thunder, I suspect there is an error in the calibration data. Look at the FNPW_OFFCOMP at 50psi, I quite convinced the correct value should be 1.0990. It would not make any sense that it's lower than 45psi and 55psi. That's why your 50psi row in TunerStudio look wrong compared to the row above and under.

I would also like to know if it's the correct method to calculate deadtime because I have those exact same injectors.

IDK. Look at the data from Ford Racing for the LU47's. There are numbers that don't follow the trend in two sets here....
Those number could be correct.

47lb Injector Data.PNG (110.5 KiB) Viewed 2843 times