Been stuck on this for months... what in the world is this?

[Artorias]

Hi y'all. This is pissing me off

Been stuck on this for quite some time. Log attached. Gen 2 Coyote, return style 1:1 55psi base, twin 61mm journals, ported 18 manifold with IMRCs, FIC1000s

I have a lot of experience with other brands but have been stuck on figuring out the cold start behavior on a Coyote. There seems to be no way to multiply airflow based on ECT. The log below will show the best demonstration of what the issue is. Been fighting slightly rough cold starts, and the issue is if I get fueling dead on when the car is at operating temp, it will be WAY too lean upon cold starts. I'm talking 20-30% too lean when closed loop kicks in. The only luck I've had is to add to the MAF curve, but if I do that then all of it will just get pulled when the car is at operating temp.

I cannot seem to get fueling to stay consistent through the temperature range, I can either have perfect fueling at operating temp but awful cold fueling, or I can have great cold starts but extremely rich operating temp fueling of which all gets pulled by LTFT and then applied to the next cold start which puts me right back to square one.

The log below has two idle periods, one at 120-130 degrees and another closer to 180 degrees. Look closely at how NOTHING changes except for pulsewidth and flow rate. Airflow is the same, period is the same, IAT the same, MAP the same, load the same, VCT the same, transport the same, AC pressure (fuel pressure) the same... am I not logging something I should be? I do know my torque model isn't perfect yet, and I am aware the car targets Torque Control/Idle Speed Limit every single time it's idling. I'm also aware the car is underspeed upon cold starts, as in it's commanded to target 1250 yet it always idles about 70 RPM under.

Fuel rail temp modifiers do nothing, as it appears no matter what is populated in the entire table the car only listens to whatever is multiplied against specifically 60 degrees. Assuming since these engines are inferred rail temp and default to 60 degrees, maybe it always assumes 60 degree rail temp. So haven't had much luck with this

There has to be something here to fix cold fueling when closed loop kicks in. What in the world am I missing here?

[Pistol_91]

Post up your tune file

[Artorias]

Post up your tune file

Thought i did, my mistake. Here is the tune

Car hasn't hit boost yet so the torque and spark tables aren't scaled at all currently. Been working my way up through drivability first. Car drives great except for return to idle and the cold fueling issue right now

Twin Turbo FIC1000 IMRC - V139.hpt

[murfie]

Your fueling looks much worst than it should every where except at idle between 140*-160*.

I think the problems you are having is bad injector data. Have you tried just using the HPT data tab from FIC, setting the fuel regulator rail pressure to hold 55PSI, and disconnecting the 1:1 vac reference?

[Artorias]

Your fueling looks much worst than it should every where except at idle between 140*-160*.

I think the problems you are having is bad injector data. Have you tried just using the HPT data tab from FIC, setting the fuel regulator rail pressure to hold 55PSI, and disconnecting the 1:1 vac reference?

I have tried that but I suppose I can give that another go. The behavior did change a little but the overall shape was the same, as in still sensitive to changes in temperature.

Granted with the log and latest revision above, my MAF curve is excellent but the values are still higher than they should be in low airmass areas so that could be why fueling still looks off. During acceleration it's dead on as far as I can tell but decel is quite rich of course. Just very odd to see trims pulling 10-15% only when ECT goes from cold to about 170 and above.

When it's at operating temp things are much more stable, but like you said when cold it's much different. Just can't see why ECT is affecting trims so heavily.

[Artorias]

bump. no conclusive direction, still fighting this

[Eric@HPTuners]

This is from not having correct injector data. The only way you can fudge this would be screwing around with the lost fuel tables.

[Pistol_91]

This is why I chose ID's over FIC's. What exact data did you input?

[murfie]

I think FIC's data is fine. He used the return style data for the FIC 1000H at 55PSI base pressure. Not the short body 1000 data they also provide. This removes all the multipliers for pressure differences as it is suppose to maintain a 1:1 relationship between MAP and rail pressure. Looking at the AC pressure voltage in his log it appears to be doing that fairly decently but theres still a few points where its either higher or lower. I dont know what 5v pressure sensor he is using for fuel rail pressure so I can quite figure out his base fuel pressure and if its actually close to the 55 PSI base.

Ideally, you wouldn't set the multipliers to 1 and the inffered rail pressure to the same 55PSI. You would leave the multpliers as they are, then fill out the inffered rail pressure with the correct values based on the fuel flow. You have a sensor that you could plot this out with. You could also use the inffered MAP and offset it doing it this way works with what the ECU thinks should be happening.

The easiest thing tho would be to remove the 1:1 reference, and just set the regulator to a PSI near the factory 58psi. it makes filling out the inffered rail pressure tables a bit easier, if even needed, and makes it work very similar to the factory returnless FDPM pump DC control and pressure relief.

There really isn't much cylinder head temperature does to affect what the MAF is saying is going into the cylinder. If your transfer is accurate,fuel mass will be accurate. The variable in all this is your injector control, which relys on MAP pressure and rail pressure to determine the injectors flow rate, so it knows how long of a pulse it should do. A returnless system with a 1:1 therotically makes the injectors only flow one rate all the time, its just knowing and entering that specific value into the tune trips people up. Then 1:1 systems are not perfect and capturing the error they have in the ECU never ment to have a return style fuel system takes some creativeness.

[Artorias]

This is why I chose ID's over FIC's. What exact data did you input?

I think FIC's data is fine. He used the return style data for the FIC 1000H at 55PSI base pressure. Not the short body 1000 data they also provide. This removes all the multipliers for pressure differences as it is suppose to maintain a 1:1 relationship between MAP and rail pressure. Looking at the AC pressure voltage in his log it appears to be doing that fairly decently but theres still a few points where its either higher or lower. I dont know what 5v pressure sensor he is using for fuel rail pressure so I can quite figure out his base fuel pressure and if its actually close to the 55 PSI base.

Ideally, you wouldn't set the multipliers to 1 and the inffered rail pressure to the same 55PSI. You would leave the multpliers as they are, then fill out the inffered rail pressure with the correct values based on the fuel flow. You have a sensor that you could plot this out with. You could also use the inffered MAP and offset it doing it this way works with what the ECU thinks should be happening.

The easiest thing tho would be to remove the 1:1 reference, and just set the regulator to a PSI near the factory 58psi. it makes filling out the inffered rail pressure tables a bit easier, if even needed, and makes it work very similar to the factory returnless FDPM pump DC control and pressure relief.

There really isn't much cylinder head temperature does to affect what the MAF is saying is going into the cylinder. If your transfer is accurate,fuel mass will be accurate. The variable in all this is your injector control, which relys on MAP pressure and rail pressure to determine the injectors flow rate, so it knows how long of a pulse it should do. A returnless system with a 1:1 therotically makes the injectors only flow one rate all the time, its just knowing and entering that specific value into the tune trips people up. Then 1:1 systems are not perfect and capturing the error they have in the ECU never ment to have a return style fuel system takes some creativeness.

Thanks for the responses, I'll be focusing on injector data to solve this. I suppose half the reason I went return style was to simplify the tuning process, and also should I specify I'm using an A1000 Gen 2 Aeromotive regulator should that matter. Fuel sensor being used is 0-75psi 0.5-4.5v

Regarding FIC's data, I specifically have IS403-1000H injectors. You are correct that I didn't use the short body data. These are the two data sheets I have for these injectors, and I emailed FIC months ago about these. They specified the "1000H" sheet was "older" data and the other sheet without the H was newer, so I've used the newer sheet. Unsure if this older sheet is useful in any way

FICLINIC_1000H_FORDDATA_SCT_WEBSITE_RAILTEMP.xlsx
FICLINIC_1000_FORDDATA_SCT_WEBSITE_RAILTEMP (2).xlsx

[Pistol_91]

I used the data from the 1000 sheet same as you and it didnt match what you had from the tune you posted in post #3
Try this
Twin Turbo FIC1000 IMRC - V139 (New data).hpt

[murfie]

Using that voltage and pressure range of the sensor, Im getting a base of about 40-41PSI. This is equal to setting the fuel pressure to 55PSI when there is ~14.7 PSI (STP atmosphere) of air pressure in the manifold. A true 55PSI base would be closer to 70 PSI when there is that much pressure in the manifold. Some manufactures account for it, other treat it as the base.

Screenshot 2024-10-30 185950.png

Try entering what the FIC 1000 sheet, returnstyle tab, spits out with a 41 PSI base fuel pressure. Or use the 39PSI data from the FIC 1000H sheet.

[Artorias]

I used the data from the 1000 sheet same as you and it didnt match what you had from the tune you posted in post #3
Try this
Twin Turbo FIC1000 IMRC - V139 (New data).hpt

I guess what confuses me a little with the attached file is seeing the flow rate multipliers occupied and not set to 1. The return style tab in both excel sheets from FIC show setting all multipliers to 1 since pressure is now regulated via a mechanical regulator with vacuum/boost reference, so the calibration no longer has to compensate for changes in manifold pressure by altering injector flow. The regulator now takes care of it in sync with manifold pressure. Maybe I'm not understanding something about return style 1:1 systems?

Using that voltage and pressure range of the sensor, Im getting a base of about 40-41PSI. This is equal to setting the fuel pressure to 55PSI when there is ~14.7 PSI (STP atmosphere) of air pressure in the manifold. A true 55PSI base would be closer to 70 PSI when there is that much pressure in the manifold. Some manufactures account for it, other treat it as the base.

Screenshot 2024-10-30 185950.png

Try entering what the FIC 1000 sheet, returnstyle tab, spits out with a 41 PSI base fuel pressure. Or use the 39PSI data from the FIC 1000H sheet.

What you wrote makes sense and was, to my understanding, correct. Maybe I'm misinformed on how a return style 1:1 works. I was under the impression "base" pressure was the set regulator pressure with engine off and vacuum off (so I set my regulator by temporarily wiring the pumps to run with ignition so when in accessory mode, the pumps run and I can set the pressure with 0 vacuum since the engine isn't running). So 55 psi would be at 0 vacuum. Then when in vacuum such as idle, fuel pressure drops in sync. And the other way around, fuel pressure rises when losing vacuum/entering boost.

So 55psi base and 41-42psi at idle with vacuum reference sounds correct to me but from what you're saying, I believe you're specifying that isn't correct.

So I guess at this point now I'm really confused. What's the technical/proper way to set a 1:1 regulator or did I just interpret the definition of "base" incorrectly?

[Pistol_91]

Sorry I missed that in your post. I read it one day and then looked at everything a few days later. Slipped my mind. Disregard then.

[murfie]

That is how you set base pressure.

There is two different zeros, gauge and absolute. One being zero vaccum and the other being zero pressure. They give you two different rail pressures as your "base". The sheets provided never specify if its the pressure delta across the injector or if its the starting value you set the rail pressure at. I believe FIC its the delta, and ID is the starting rail pressure. Both accurate flow values, just the end users can interpret thier meaning of "base" wrong and get the wrong values.

[Artorias]

That is how you set base pressure.

There is two different zeros, gauge and absolute. One being zero vaccum and the other being zero pressure. They give you two different rail pressures as your "base". The sheets provided never specify if its the pressure delta across the injector or if its the starting value you set the rail pressure at. I believe FIC its the delta, and ID is the starting rail pressure. Both accurate flow values, just the end users can interpret thier meaning of "base" wrong and get the wrong values.

Murfie, thank you for the good responses. I believe I understand what you're doing, and if I understand properly we are taking the base regulator pressure (55psi) and subtracting my atmospheric pressure (14.7psi, 29.9inhg) and that's how we're getting to 40-41 psi like you're seeing in my log at idle. Engine off, vacuum is zero but there's still pressure from the atmosphere inside the manifold, which is 14.7psi. I'm only about 150 feet above sea level in my area. So this makes sense thinking about it when setting the regulator to it's "base" pressure. Even though the engine is off there's still pressure in the manifold. I never considered this until now, but I assume this variable only really matters because I'm using a 1:1 reference?

So when you say to keep the regulator at 55psi but input fuel data from FIC based on ~41psi, now it's starting to make sense to me. This entire time I've been assuming fuel injector data accommodates the already existing atmospheric pressure but I guess that isn't always the case. I'm going to give this a try later and see how the car acts.

That brings me to another question however. I want to keep the 1:1 reference, so if I subtract atmospheric pressure and pop in new data then how does this affect the car if I, for example, drive all the way to the rocky mountains. Would fueling still be correct or would you in theory have to redo the regulator "base" pressure/fuel injector data any time the car sees major changes in atmospheric pressure? I hope I worded that all correctly

[skythunder83]

I have seen GEN2 going much happy with 55+ PSI like going with 58 or 60, I agree that injector data could be the issue here also in relation to boost, but base pressure is also something could help

[murfie]

That brings me to another question however. I want to keep the 1:1 reference, so if I subtract atmospheric pressure and pop in new data then how does this affect the car if I, for example, drive all the way to the rocky mountains. Would fueling still be correct or would you in theory have to redo the regulator "base" pressure/fuel injector data any time the car sees major changes in atmospheric pressure? I hope I worded that all correctly

The 1:1 reference eliminates atomospheric pressure as an issue. I connects the rail pressure and manifold pressure so the injectors flow one rate no matter what. If you dont use this, you need a barometric pressure sensor or to make sure your calculated baro is reasonably correct.

When you set the base pressure with the engine off near sea level, the data on some manufactures sheets will be accurate. If you did this at the top of a 10,000ft mountain, not sure who would and not know to watch out for this, it would be close just not as accurate.
If the manufacture gives you the delta pressure flow rates, which is what the base pressure is actually setting, then it wouldn't matter because you would have set the rail pressure and subtracted the barometric pressure to find it.

The regulator you are using has a max of 75psi, so if you try to set a 55PSI injector delta pressure, you dont have much room for boost before you max this regulator out. Stay with the 39-41, that most people know as a 55 psi base pressure, is normal for a returnless system.

[Artorias]

The 1:1 reference eliminates atomospheric pressure as an issue. I connects the rail pressure and manifold pressure so the injectors flow one rate no matter what. If you dont use this, you need a barometric pressure sensor or to make sure your calculated baro is reasonably correct.

When you set the base pressure with the engine off near sea level, the data on some manufactures sheets will be accurate. If you did this at the top of a 10,000ft mountain, not sure who would and not know to watch out for this, it would be close just not as accurate.
If the manufacture gives you the delta pressure flow rates, which is what the base pressure is actually setting, then it wouldn't matter because you would have set the rail pressure and subtracted the barometric pressure to find it.

The regulator you are using has a max of 75psi, so if you try to set a 55PSI injector delta pressure, you dont have much room for boost before you max this regulator out. Stay with the 39-41, that most people know as a 55 psi base pressure, is normal for a returnless system.

Here's a new log, this is with regulator set at 55psi engine off and with 40.3psi injector data from FIC's newer data sheet like what you suggested to try. Car feels slightly smoother but the behavior didn't quite seem to change much. Also saw a huge spike where trims went almost max positive so that was odd.

This car won't see more than 7-8psi of boost, so figured the regulator has plenty of room to accommodate that at 55psi base. Still a little lost, how does this log look?
Twin Turbski FIC1000 IMRC - V140 log 1 cold start cold drive.hpt

[murfie]

Not sure why I had to change it to a HPL.
Thats much better than your first log. Goes lean when you let off and it should go to DFCO. Still needs some work. I will have too look into it more later.

Im not seeing where it went max lean, and its not going max rich anymore.

Screenshot 2024-11-06 060527.png