progress on the banshee...

[ericjon262]

I drove the car to work, on the way there, idle got progressively worse, to the point it was surging all over, and not wanting to maintain running, I was looking at the log, and for some reason, when it was idling ok, it was in the 40 KPA cells, when it was idling like trash, it was in the 35 KPA cells, I looked at the spark table and noticed there was a step in the table between 35 and 40 KPA. with about 3 degrees less timing right there. it was the only thing I could see that was different in that area, I'll implement the change before I leave work and see how much it helps.

[ericjon262]

I've been driving the car and trying to get something resembling a tune started, but I think I may have one small issue that I need to rectify. The valvetrain has been super noisy, and I did some math the other day, and I think the pushrods are to blame. the pushrods I have in the engine are about .025" longer than stock, I installed .030" shims under the rocker pedestals, so I suspect the lifters have minimal preload, if any at all. in the past, I had seen recommendations of .020" longer pushrods with the cam I am running. I emailed Ben at WOT-Tech, and we both agreed that a set of pushrods .050" longer than stock would probably help quiet things down a bit, they should be here early this next week.

idle tuning continues to be a bear, I think I may be on the ragged edge of what the deka 60's can do, I'm going to keep working with it and see where I can get, but it seems like it either idles on the rich side, or super lean.

[pmbrunelle]

idle tuning continues to be a bear, I think I may be on the ragged edge of what the deka 60's can do, I'm going to keep working with it and see where I can get, but it seems like it either idles on the rich side, or super lean.

Hmmm, some thoughts/ideas...

If the valves aren't opening/closing exactly like they're supposed to, maybe that's causing your issues? Maybe solve this before trying to tune with a known problem present?

Have you attempted to do a characterization of your injector dead time (function of battery voltage), and have you programmed this in the ECU?

Understand that compensation is never perfect, it is best if dead time is as short as possible. Are you running a manifold-referenced fuel pressure regulator, or atmosphere-referenced? With manifold reference, there is less differential fuel pressure across the injector at idle. The less fuel pressure there is, the less the fuel pressure is keeping the injector forced shut. With less fuel pressure, the injector can snap open quicker.

You have a constant voltage from your alternator? If voltage is low, then the injector is slower to open.

If you retard the idle timing down to the 5-10° BTDC range, then the thermal efficiency of the engine will decrease. To stay alive, the engine will need to idle at a greater MAP kPa. More air -> greater injector pulse width.

What is the idle pulse width? How does it run during warmup, when the mixture is richer, and pulse width is longer?

What AFR are you targeting, and what is the measured AFR when it swings lean or rich?

[ericjon262]

idle tuning continues to be a bear, I think I may be on the ragged edge of what the deka 60's can do, I'm going to keep working with it and see where I can get, but it seems like it either idles on the rich side, or super lean.

Hmmm, some thoughts/ideas...

If the valves aren't opening/closing exactly like they're supposed to, maybe that's causing your issues? Maybe solve this before trying to tune with a known problem present?

Have you attempted to do a characterization of your injector dead time (function of battery voltage), and have you programmed this in the ECU?

Understand that compensation is never perfect, it is best if dead time is as short as possible. Are you running a manifold-referenced fuel pressure regulator, or atmosphere-referenced? With manifold reference, there is less differential fuel pressure across the injector at idle. The less fuel pressure there is, the less the fuel pressure is keeping the injector forced shut. With less fuel pressure, the injector can snap open quicker.

You have a constant voltage from your alternator? If voltage is low, then the injector is slower to open.

If you retard the idle timing down to the 5-10° BTDC range, then the thermal efficiency of the engine will decrease. To stay alive, the engine will need to idle at a greater MAP kPa. More air -> greater injector pulse width.

What is the idle pulse width? How does it run during warmup, when the mixture is richer, and pulse width is longer?

What AFR are you targeting, and what is the measured AFR when it swings lean or rich?

when I got home from work, I pulled a bunch of timing, and started attacking idle and trying my best to get it better, I did have a couple things going against me, I forgot to turn off idle advance, and I also had DBW idle control enabled, I think the combination of these two weren't helping, I got it to about 14:1, PW is around 1.1-1.2 or so, pretty short still. tomorrow, before I leave work I'll try disabling DBW idle control again, and see how I can get it. hopefully I can further improve.

Alternator is pretty consistent unless RPM gets really low, then it starts to dip a little, at normal speeds it's fine though. as far as characterizing dead times, I used some tables or data I found somewhere on the internet as a base and haven't made any further adjustments yet.

FPR is manifold referenced, I'd like to keep it that way for now.

Overall, I believe it may have been premature to say that I think I need smaller injectors, but I definitely need to keep working on it.

[ericjon262]

I drove it home from work last night without DBW idle control, but I also set the throttle to 10% minimum for the drive. once I was home I played around with lowering the limit down some, and seemed to get an OK idle. This morning it was ok on the drive to work, but hunted quite a bit. I gave it some thought, and made some adjustments to the break points in the RPM scales of the VE and ignition advance tables. I also realized that the first three columns of the VE and Advance table were being used pretty much exclusively for idle or starting the car, I had been working on tuning idle in a 3 column by 4 row area, but I really don't need to limit it to 4 rows, as during driving conditions, I won't be there anyways. I think this was also part of the cause of the hunting, because the engine would barely hit one of the outside cells, and throw itself off equilibrium. I'm going to implement these changes before I leave work tonight and see what happens.

in other news, the heater isn't working in the car. The heater core has flow, so I suspect there's an issue with the blend doors, but I haven't investigated much yet. if anyone has any ideas, I'm all ears.

[ericjon262]

idle at the beginning of the log fluctuates between 733 and 981 RPM, which seems reasonable to me, at the end, it's from 547-1349 RPM, and totally out of control, the only difference is the coolant temperature and runtime, closed loop fuel correction is turned off, and warmup enrichment is 100% at both points. I'm not really sure what's going on. it doesn't appear to have a misfire, and runs fine outside of idle. my new pushrods were supposed to be here today, but I suspect it will be tomorrow or wednesday before I have them in my hands. when they get here, I plan to swap them in, along with new spark plugs, and check torque on all of the exhaust, the head studs, and the intakes.

[Shaun41178(2)]

I know you said you have closed loop ego correction off but looking at your a/f plot when at 180 degrees, that does not appear to be the case

I am not using an iac on my engine. My idle used to hunt badly once closed loop kicked on. My af ratio looked like yours. So I adjusted the ego correction above 1500 rpm only as well as locked the ve table in the idle cells so the computer will never try to adjust them.

Maybe that's what you need to do.

Also it looks like you are trying to control your idle by lowering your timing to only 11 degrees. On my 3900 I think I am around 19 degrees at idle. Trying to control idle with timing can work but it's not the correct way.

[ericjon262]

I know you said you have closed loop ego correction off but looking at your a/f plot when at 180 degrees, that does not appear to be the case

I am not using an iac on my engine. My idle used to hunt badly once closed loop kicked on. My af ratio looked like yours. So I adjusted the ego correction above 1500 rpm only as well as locked the ve table in the idle cells so the computer will never try to adjust them.

Maybe that's what you need to do.

Also it looks like you are trying to control your idle by lowering your timing to only 11 degrees. On my 3900 I think I am around 19 degrees at idle. Trying to control idle with timing can work but it's not the correct way.

right now, my main goal is to get it to stay stable, an IAC or DBW throttle can't compensate nearly fast enough to stop that kind of rolling/hunting, in the bad part of the log, RPM is fluctuating 350+ RPM in less than 1/2 a second, in the early parts of the log, it's fluctuating 165 RPM over 1.3 seconds.

I lowered the idle advance that low to see how the engine reacted, and if it would help with my overall issue, so far it seemed to like the low timing, except when the cooling fan kicks on, which to me, means it needs more, so I will be advancing it back, in the process, I should also be able to walk the throttle position back down to maintain a similar idle speed.

Double checked to make sure EGO correction is off, and it still is, so it definitely shouldn't be that. I also doublechecked the log file, and EGO correction is unchanged from beginning to end at 100%

it was pointed out on the MSextra forum

One thing I noticed in your log is that "good early" the VSS shows 0 (zero), while the "bad late" shows initially that the VSS is zero, but then jumps up to a few MPH, and also indicates that the car is in first (based on calculations and the fact the ECU is seeing vehicle speed. Was teh car actually rolling a t a slow speed or sitting still and the VSS signal needs some attention?

as noted earlier in this thread, my VSS has a noise issue, I have the new sensor that I still need to install, engine state settings had "Use VSS" enabled. I don't think it would be causing the problem, as all closed loop idle control is disabled, but I went ahead and disabled that setting until I can replace the sensor with the cam sensor and adapter, hopefully later this week. I did review the log, and the idle state was stable in the earlier, good idle, and fluctuating alot in the later, bad idle. I don't think this would cause the problem, but I would be quite relieved if it fixed it.


I appreciate you and Patrick's input, both of you have more experience tuning MS than me.

[The Dark Side of Will]

I am not using an iac on my engine.

Trying to control idle with timing can work but it's not the correct way.

I get a kick out of this sequence

The correct way to control idle is controlling both airflow and timing.
The steady-state idle should use slightly higher airflow with slightly lower timing than what would be required to simply keep the engine turning at that speed. When a transient load like A/C comp, power steering pump if equipped or an alternator load transient like turning the cooling fan, HVAC blower or headlights on or off. Pop-up headlights are particularly fun, as they represent a double transient because both the lights and the motors cause transients sometimes together and sometimes not.

When the ECM detects a change in RPM at idle, it can immediately increase timing to add torque to prevent the RPM from sagging, then more slowly retard timing as the airflow mechanism catches up by increasing airflow. Once it reaches the new steady state, it's ready for the next transient.

My idle used to hunt badly once closed loop kicked on. My af ratio looked like yours.

This is a strong point... your AFR varies wildly with RPM... the change in torque with AFR is what's destabilizing your idle. The AFR slope or rate of change looks extremely steep, as though you're traversing a discontinuity in your VE table.

Is there a "rolling idle" that would change settings based on VSS? GM computers have that to kick the idle up a little bit while the vehicle is moving in a parking lot, in order to better absorb power steering loads.

Also, how does the ECM know you're at idle? Zero on the pedal sensor?

Also, what kind of temperature dependency do you have in your AFR table?

[pmbrunelle]

This is a strong point... your AFR varies wildly with RPM... the change in torque with AFR is what's destabilizing your idle. The AFR slope or rate of change looks extremely steep, as though you're traversing a discontinuity in your VE table.

So near stoichiometry, the addition or subtraction of fuel causes a direct change in engine torque.

As the mixture is enriched, torque will increase up to a maximum, then torque will decrease if the mixture is further enriched.

Around the maximum (which occurs around 12.5 AFR), small fluctuations in AFR do not cause fluctuations in engine torque. Engine torque remains stable despite changes in AFR.

Eric mentioned running near 14 AFR, so engine torque will depend significantly on AFR here.

Of course, besides reducing the engine's sensitivity to AFR fluctuations, we should also try to reduce the AFR fluctuations themselves.

AFR fluctuations can exist for a number of reasons:
Sensor noise
VE table issues
Numerical noise in the ECU's calculations
Dead time compensation is not close enough, combined with short pulse width
Wall-wetting -> fuel delivered (evaporated from the puddle) into the combustion chamber now does not reflect the present airflow; it reflects the airflow measured at the time of injection into the port.

The wall-wetting time constant is very much a function of RPM. The faster the air rushes through the port, the shorter the time constant.

On my own Fiero, the time constant may be ~1 second at 700 RPM. Up at 1200 RPM, the time constant may drop to ~0.5 seconds.

By increasing idle RPM, the wall-wetting time constant (and therefore whatever AFR fluctuation it may produce) is reduced.

right now, my main goal is to get it to stay stable, an IAC or DBW throttle can't compensate nearly fast enough to stop that kind of rolling/hunting

Without adding any of the closed-loop idle or EGO functions (which just adds more complexity into the mix), you need to get a stable non-oscillating idle with a constant throttle opening.

I suppose that I would suggest trying to make the idle work at 12.5 AFR, and 1200 RPM.

Once you get a stable baseline, then you can work on reducing the idle speed and leaning it out more to your liking.

[ericjon262]

My idle used to hunt badly once closed loop kicked on. My af ratio looked like yours.

This is a strong point... your AFR varies wildly with RPM... the change in torque with AFR is what's destabilizing your idle. The AFR slope or rate of change looks extremely steep, as though you're traversing a discontinuity in your VE table.

Is there a "rolling idle" that would change settings based on VSS? GM computers have that to kick the idle up a little bit while the vehicle is moving in a parking lot, in order to better absorb power steering loads.

Also, how does the ECM know you're at idle? Zero on the pedal sensor?

Also, what kind of temperature dependency do you have in your AFR table?

I don't think there is an idle modifier based on VSS, the MS3 determines "Engine state" via a few parameters, throttle position, VSS, and RPM change. I turned off the VSS input to the engine state calculation with the intention of turning it back on once I handle the noise problem, but I also haven't driven the car since doing so. yesterday, I got the new VSS mounted, and the wire routed, I still need to terminate the MS3 side, and tie back the wiring. I'll test the new VSS after I install my new pushrods that showed up yesterday.

This is a strong point... your AFR varies wildly with RPM... the change in torque with AFR is what's destabilizing your idle. The AFR slope or rate of change looks extremely steep, as though you're traversing a discontinuity in your VE table.

So near stoichiometry, the addition or subtraction of fuel causes a direct change in engine torque.

As the mixture is enriched, torque will increase up to a maximum, then torque will decrease if the mixture is further enriched.

Around the maximum (which occurs around 12.5 AFR), small fluctuations in AFR do not cause fluctuations in engine torque. Engine torque remains stable despite changes in AFR.

Eric mentioned running near 14 AFR, so engine torque will depend significantly on AFR here.

Of course, besides reducing the engine's sensitivity to AFR fluctuations, we should also try to reduce the AFR fluctuations themselves.

AFR fluctuations can exist for a number of reasons:
Sensor noise
VE table issues
Numerical noise in the ECU's calculations
Dead time compensation is not close enough, combined with short pulse width
Wall-wetting -> fuel delivered (evaporated from the puddle) into the combustion chamber now does not reflect the present airflow; it reflects the airflow measured at the time of injection into the port.

The wall-wetting time constant is very much a function of RPM. The faster the air rushes through the port, the shorter the time constant.

On my own Fiero, the time constant may be ~1 second at 700 RPM. Up at 1200 RPM, the time constant may drop to ~0.5 seconds.

By increasing idle RPM, the wall-wetting time constant (and therefore whatever AFR fluctuation it may produce) is reduced.

right now, my main goal is to get it to stay stable, an IAC or DBW throttle can't compensate nearly fast enough to stop that kind of rolling/hunting

Without adding any of the closed-loop idle or EGO functions (which just adds more complexity into the mix), you need to get a stable non-oscillating idle with a constant throttle opening.

I suppose that I would suggest trying to make the idle work at 12.5 AFR, and 1200 RPM.

Once you get a stable baseline, then you can work on reducing the idle speed and leaning it out more to your liking.

the largest point of confusion for me, is why would the idle be good at the start of the log, and totally out of control at the end? the only variable changed is the coolant temp and run time. I think idle was more than reasonable at the start of the above pictured log, the engine was out of WUE, and the VE table is flat in the area in question. I'm hoping the VSS replacement does fix some or all of my problem, but I'm not going to hold my breath. if it doesn't, I think my next course of action will be to start the car, and let it idle until it starts getting bad, and see what changes with no other inputs, as the above log illustrates, the car can idle happily at 13.8-14.3. I also plan on increasing my ignition timing some, it needs more idle torque to minimize the effects of transients on RPM. if it continues to destabilize hot, I'll bump the RPM up, and get it stable at some engine speed, and then try walking it down to a stable idle RPM of around 800-900 RPM,

[The Dark Side of Will]

the only variable changed is the coolant temp and run time.

What parameters cause AFR to change with coolant temp?
Pat's idea of wall wetting is a good candidate, as that's going to change with metal temp.
What cold start temperature compensations do you have operating?

Idle is the hardest engine state to tune.

the car can idle happily at 13.8-14.3.

Maybe? The AFR varying significantly across idle speeds indicates that it's not staying in that range.

I also plan on increasing my ignition timing some, it needs more idle torque to minimize the effects of transients on RPM.

Idle ignition timing should ramp smoothly through the idle table. Idle ignition timing should be higher at RPM below the idle set point and lower at RPM above the set point. That way when RPM sags from a transient, ignition timing comes up for increased torque. When the opposite transient happens, ignition timing comes down to lower torque and slow the engine down.
AFR should be flat through the idle table.

[ericjon262]

I had 0 intention of running the car until the new pushrods were installed, boy do I lie to myself alot. I finished up the VSS install, unfortunately, I'm not getting a signal. I'm going to try removing the O-ring I installed and see if getting the sensor slightly closer to the wheel fixes that, I'll use some kind of sealant to prevent leaks.

I started the car, and let it warm up and go until the cooling fan had cycled several times. the engine idled like the early part of the datalog the entire time(20+ minutes) I took the car for a drive, and paced it out a few times, made a trip about the same distance as it would be to go to work and back, got home, idled fine. let the cooling fan cycle a few times again, still, peachy. I'm beginning to think the VSS was causing more problems than I thought, the only changes I made between the above posted screenshots and today, were the different speed sensor, and the parameter adjustments to use it.

[pmbrunelle]

the VE table is flat in the area in question.

This statement struck me as bizarre, so I downloaded your *.msq tune file from the other forum to understand better.

Looks like you have exactly the same VE values in the idle area:

Eric VE 2D.jpg (215.71 KiB) Viewed 911 times

Same table viewed in 3D:

Eric 3D VE.jpg (112.46 KiB) Viewed 911 times

Commanded AFR, with idle area in red rectangle:

Eric 2D AFR.jpg (127.2 KiB) Viewed 911 times

AFR should be flat through the idle table.

At the moment, AFR in the idle area cannot be constant due to two reasons:
1. Engine VE varies with RPM and load, but the VE table (in the idle area) is constant. Therefore, the VE table (as it is today) cannot accurately model the actual airflow.
2. Commanded AFR is not constant in the idle area.

Speaking of AFR, you're not running includeAFRtarget?

AFRtarget.jpg (14.37 KiB) Viewed 911 times

I would recommend it switching it on, so that the VE table is only modeling the volumetric efficiency, without having the target AFR baked into it. When includeAFRtarget is activated, with a properly tuned VE table, you can simply change the AFR by changing the commanded AFR, without having to revisit the VE table.

********************************************************************************

Your VE table is currently a horror show. Discussion of dead times, ignition timing retard, idle RPM, those are details that don't matter with the current state of your VE table.

VE is generally a smooth function, so the VE table should reflect that. In something such as a VTEC Honda with lobe-switching, I suppose you might find a crease in the VE table at the switchover point.

Here's my Fiero's table:

Pat 3D VE.jpg (126.72 KiB) Viewed 911 times

The red cursor is at the idle point. There is nothing particular about the idle point. The VE table attempts to model VE accurately for all conditions of MAP/RPM.

For idle stability, it is not sufficient for the VE value at the idle point to be correct. The gradient of the VE table in the idle area must also reflect the gradient of the engine's VE.

To transform the VE table into a good representation of the engine's VE, you should use the VE Analyzer Live in TunerStudio as you drive around. Try to hit as many cells as possible in near "steady-state" conditions. I generally pick an RPM (keeping constant RPM by left-foot braking), and then slowly open the throttle with my right foot to move through the cells.

The VE table as computed by the automation won't be perfect. It will be spiky, and some unreachable areas won't be tuned (for lack of data). So, manually adjust the table so that it looks smooth, while trying to imagine what the true VE function may look like beneath the noise. Do some logical extrapolations to areas of the table you weren't able to reach.

It's an iterative process; you won't get the table like mine on the first go. With each iteration, increase the "difficulty" from Easy, to Normal, then to Hard in VE Analyzer.

In the final iterations of VE table refinement, you should datalog (always trying to hit each cell in near-steady-state) over a bunch of different conditions. Hot/cold, humid, dry. The final VE table correction should be based on this average data over different conditions. MegaLogViewer can compute a corrected VE table based on multiple datalog files. Manually review and smooth if necessary.

Generic values of injector dead time will be OK for your initial roughing-in of the VE table, but before getting to the final stages of VE table refinement, you should characterize the dead time for your specific car.

********************************************************************************

I thought you had a MAF sensor? Why aren't you using it? It saves the fuss of getting the VE table to model the engine accurately.

********************************************************************************

Eric Settings.jpg (118.71 KiB) Viewed 911 times

What is Semi-sequential?

How many times does each injector fire per 4-stroke cycle?

[Shaun41178(2)]

Maybe we need a refresher on what the setup is. Cam, injector size, blah blah

I have 80 lb injectors and have my injector dead time at .845. I have high z injectors

Af ratio on 93 octane should be at 14.1. Remember there is 10% ethanol which means you gotta set the target to 14.1
At idle and low load

[ericjon262]

the VE table is flat in the area in question.

This statement struck me as bizarre, so I downloaded your *.msq tune file from the other forum to understand better.

Looks like you have exactly the same VE values in the idle area:
Eric VE 2D.jpg

Same table viewed in 3D:
Eric 3D VE.jpg

Commanded AFR, with idle area in red rectangle:
Eric 2D AFR.jpg

The VE table is flat in the idle region, and above/below, with closed loop fuel correction turned off, the AFR table shouldn't be making changes to fueling. I did that so I would have one less variable changing things while I tried to get the car to idle consistently.

AFR should be flat through the idle table.

At the moment, AFR in the idle area cannot be constant due to two reasons:
1. Engine VE varies with RPM and load, but the VE table (in the idle area) VE is constant. Therefore, the VE table (as it is today) cannot accurately model the actual airflow.
2. Commanded AFR is not constant in the idle area.

Speaking of AFR, you're not running includeAFRtarget?
AFRtarget.jpg
I would recommend it switching it on, so that the VE table is only modeling the volumetric efficiency, without having the target AFR baked into it. When includeAFRtarget is activated, with a properly tuned VE table, you can simply change the AFR by changing the commanded AFR, without having to revisit the VE table.

most EFI applications I have used in the past (granted, use was limited) performed calculations similar to what was described by not including target AFR, having the VE table control fuel, so I stuck with that, or at least I think that was my mindset at the time. include target AFR sounds simpler by the way you describe it.

Your VE table is currently a horror show. Discussion of dead times, ignition timing retard, idle RPM, those are details that don't matter with the current state of your VE table.

VE is generally a smooth function, so the VE table should reflect that. In something such as a VTEC Honda with lobe-switching, I suppose you might find a crease in the VE table at the switchover point.

Here's my Fiero's table:
Pat 3D VE.jpg

The red cursor is at the idle point. There is nothing particular about the idle point. The VE table attempts to model VE accurately for all conditions of MAP/RPM.

For idle stability, it is not sufficient for the VE value at the idle point to be correct. The gradient of the VE table in the idle area must also reflect the gradient of the engine's VE.

To transform the VE table into a good representation of the engine's VE, you should use the VE Analyzer Live in TunerStudio as you drive around. Try to hit as many cells as possible in near "steady-state" conditions. I generally pick an RPM (keeping constant RPM by left-foot braking), and then slowly open the throttle with my right foot to move through the cells.

The VE table as computed by the automation won't be perfect. It will be spiky, and some unreachable areas won't be tuned (for lack of data). So, manually adjust the table so that it looks smooth, while trying to imagine what the true VE function may look like beneath the noise. Do some logical extrapolations to areas of the table you weren't able to reach.

It's an iterative process; you won't get the table like mine on the first go. With each iteration, increase the "difficulty" from Easy, to Normal, then to Hard in VE Analyzer.

In the final iterations of VE table refinement, you should datalog (always trying to hit each cell in near-steady-state) over a bunch of different conditions. Hot/cold, humid, dry. The final VE table correction should be based on this average data over different conditions. MegaLogViewer can compute a corrected VE table based on multiple datalog files. Manually review and smooth if necessary.

Generic values of injector dead time will be OK for your initial roughing-in of the VE table, but before getting to the final stages of VE table refinement, you should characterize the dead time for your specific car.

I've actually been using the VE analyze feature, in MLV, but not the autotune as much. I also have not hit all parts of the map, and having the throttle capped limits the ability to hit the entire table. I ran the analysis on the current table with the data from the log I took yesterday after work, and the 100 KPA table smoothed out significantly. I'm hoping now that I have isolated a part of my issues to the VSS input, that things will tune smoother and faster. honestly, the car felt like it ran smoother with no VSS than it did with the noisy VSS.

I haven't seen a VE table that looks like your's before, but most were using different algorithms for fuel control, so they would look differently. I'd imagine enabling incorporate AFRTarget makes the majority of that difference.

I thought you had a MAF sensor? Why aren't you using it? It saves the fuss of getting the VE table to model the engine accurately.

I had a MAF before I installed the intercooler, once I installed the intercooler, I made an entirely new charge pipe, and the MAF was left out.

Eric Settings.jpg

What is Semi-sequential?

How many times does each injector fire per 4-stroke cycle?

That's the way I interpreted it, currently MS3 doesn't support my crank and cam position sensor arrangement for actual sequential injection.

Maybe we need a refresher on what the setup is. Cam, injector size, blah blah

I have 80 lb injectors and have my injector dead time at .845. I have high z injectors

Af ratio on 93 octane should be at 14.1. Remember there is 10% ethanol which means you gotta set the target to 14.1
At idle and low load

that's a good point I had not considered, I have a Flex sensor, but I have it disabled to remove added variables. I'll have to run a few ethanol tests to compare the ethanol content to the content observed by the sensor.

MS3 Pro Ultimate
LX9 3.5 V6 from a 2005 Malibu
DBWx2 with 76mm LS4 throttle
60 pound deka injectors (.700 dead time currently)
14point7 spartan 2 wideband
3 bar C6 ZR1 MAP sensor
"LSx" truck ignition coils, round style
standard GM IAT sensor

WOT-Tech stage 2 turbo cam, installed straight up.
265@.050 .506" lift intake
273@.050 .506" lift exhaust
113 ICL, 114 LSA

Ported heads and intake
Precision 6267 gen 1 turbo, journal bearing
aluminum flywheel
5 speed manual
Currently non-intercooled.

[Shaun41178(2)]

According to MS literature high z injector dead times are closer to .900. I am guessing your injectors are a.chinese variant? I am guessing you got no literature with them as far as dead times?

[ericjon262]

According to MS literature high z injector dead times are closer to .900. I am guessing your injectors are a.chinese variant? I am guessing you got no literature with them as far as dead times?

The injectors are legit Siemens Dekas, I bought them almost 10 years ago, and don't have any literature for them, but I did dig all over to find people running them with dead times around where I have them set at .700.

[The Dark Side of Will]

What OE application were they used in? You may be able to find OE injector table data for them... that you know is backed up by a lot of test time.

For example, Injector Dynamics sends injector data in various OE formats (GM, Ford, Etc), so that tuners can copy/paste the right data into the tune when using those injectors. The closest schmoes like us can get is copying out of a factory tune that uses those parts. That's what I'm doing with the LC3 injectors in my engine.

[ericjon262]

What OE application were they used in? You may be able to find OE injector table data for them... that you know is backed up by a lot of test time.

For example, Injector Dynamics sends injector data in various OE formats (GM, Ford, Etc), so that tuners can copy/paste the right data into the tune when using those injectors. The closest schmoes like us can get is copying out of a factory tune that uses those parts. That's what I'm doing with the LC3 injectors in my engine.

I'm not aware of them being used in an OE application, I have thought about getting another set and having them flowed and getting as much data as I can from a fuel injector service, I'm not sure how necessary that is though.