Auto VE Blues

[Texas Terminator]

what am i missing here. i am pegging the WBO2 at anything above roughly 80% throttle. :( everything below that is fine.

[deezel]

What do you mean by pegging the wideband?
Is it 14.7AFR below 80% throttle and then something like 11.9-12.7AFR above 80%?
This may be showing that your PE mode is set to kick in at 80%tps.
If this is not what you mean, please elaborate.

[Texas Terminator]

opposite direction. any attempt to hit higher MAP's, it go's WAY lean. like scary cutting out knock retard lean. and upon looking at my scan, it looks more like anything above 50% throttle or so. PE is disabled per the instructions, so, it has no effect. i have followed the auto ve instructions to a "T" with the only exception of unplugging the MAF, i need it for the IAT but the MAF is zeroed.

[TAQuickness]

did you scale your entire VE table up by 15% before starting?

[Texas Terminator]

did you scale your entire VE table up by 15% before starting?

yes, everything by the book. all i can guess to do would be to bump my VE's in the upper MAP's another 15% or so and try again. i am starting from scratch with my stock 4.8 tune as i was unhappy with the results i achieved with (COUGH!) EDIT. my motor is very solid, but, i am still not interested in testing it's integrity with the lean conditions i am encountering. like i said, lower MAP's are fine.

what if i scaled the VE table way high? will the "auto tune" compensate backwards too? my motor LOVES fuel. FWIW, it's a very low compression too, only 8.3:1, i was originally going to go FI but gave up on the blowcharger and don't have the bank for what i want to do, so, i am going to try a two stage shot after i get a good base tune on it.

[deezel]

Ok, you are starting with a stock 4.8L tune and you have bored it out to a 408ci (or 6.7L) engine, correct?

I think the VE table would be way off between a 4.8L and a 6.7L engine, not considering the cam being used. Your idea of scaling the VE table way high and working your way back down is probably a good choice for this situation.

[Texas Terminator]

Ok, you are starting with a stock 4.8L tune and you have bored it out to a 408ci (or 6.7L) engine, correct?

I think the VE table would be way off between a 4.8L and a 6.7L engine, not considering the cam being used. Your idea of scaling the VE table way high and working your way back down is probably a good choice for this situation.

you are correct sir! i did change the displacement in the ECM. i am not entirely sure what if any effects it has though.

so, it WILL calculate both backwards and forwards? such as, it will decrease the VE table if it sees a rich cell?

[caver]

Yep it works both ways

[Texas Terminator]

so, if it will compensate in either direction, why even scale the VE tables at all to begin with?

[caver]

opposite direction. any attempt to hit higher MAP's, it go's WAY lean. like scary cutting out knock retard lean. and upon looking at my scan, it looks more like anything above 50% throttle or so. PE is disabled per the instructions, so, it has no effect. i have followed the auto ve instructions to a "T" with the only exception of unplugging the MAF, i need it for the IAT but the MAF is zeroed.

Thats why.
You need a starting point so the car is driveable without melting pistons

[Texas Terminator]

opposite direction. any attempt to hit higher MAP's, it go's WAY lean. like scary cutting out knock retard lean. and upon looking at my scan, it looks more like anything above 50% throttle or so. PE is disabled per the instructions, so, it has no effect. i have followed the auto ve instructions to a "T" with the only exception of unplugging the MAF, i need it for the IAT but the MAF is zeroed.

Thats why.
You need a starting point so the car is driveable without melting pistons

LOL! ok, but in my case, it made things worse. i can't be the only one that has had this problem?

[Texas Terminator]

damnit! this time i scaled my entire VE table an ADDITIONAL 20% and it STILL goes way lean at upper TPS's and MAP's!! :x anyone care to take a peek at my tune? this is getting quite frustrating!

[dfe1]

When you installed the larger displacement engine, did you change injectors? You may be running out of injector if you're trying to fuel a 6.7 liter engine with injectors designed for a 4.8. Another possibility is if you did change injectors, the actual flow rate is not as advertised. I'd be happy to look at your file and give you an opinion. Just PM me.

[Texas Terminator]

When you installed the larger displacement engine, did you change injectors? You may be running out of injector if you're trying to fuel a 6.7 liter engine with injectors designed for a 4.8. Another possibility is if you did change injectors, the actual flow rate is not as advertised. I'd be happy to look at your file and give you an opinion. Just PM me.

oh no, more than enough injector, 57lb to be exact. and they are set correctly at 8.29 across the board (i am not running the regulator) and no issues with pressure or flow. all of this started when i did the AutoVE tune mods.

[joecar]

they are set correctly at 8.29 across the board (i am not running the regulator)

You mean your regulator is not MAP/vacuum referenced and your IFR table is flat (no slope) at 8.29 g/s...?

If that is so, then that may be your problem.

If the regulator is MAP referenced, the IFR table has to be flat (no slope).
If the regulator is unreferenced, the IFR table has to be sloped to allow the PCM to compensate for variations in flow rate due to variations in MAP.

When you run in closed loop, the LTFT's will compensate for your flow rate variations due to flat IFR table.

[Texas Terminator]

they are set correctly at 8.29 across the board (i am not running the regulator)

You mean your regulator is not MAP/vacuum referenced and your IFR table is flat (no slope) at 8.29 g/s...?

If that is so, then that may be your problem.

If the regulator is MAP referenced, the IFR table has to be flat (no slope).
If the regulator is unreferenced, the IFR table has to be sloped to allow the PCM to compensate for variations in flow rate due to variations in MAP.

i think that is backwards. with a decrease in vacuum, the regulator increases the rail pressure respectively, changing the effective flow rate of the injectors.

the regulator is disconnected from manifold vacuum, thus, a steady pressure at the fuel rail, that is why i have the IFR table the same all the way across. at any rate, even if it was scaled, it would not be enough of a difference to make the drastic change in A/F like i am seeing. it is going to 22:1 and even a little higher! eeekkk!!

if i change the tune back, it is fine, even with the MAF disconnected.

[dfe1]

If it all started when you did the Auto VE mods, then that's the logical place to start. I did a bunch of playing after switching to speed density and found that the VE table had to be changed considerably from one that I worked over when running with the MAF. My fuel trims were slightly negative when running the MAF, but went way lean when I switched to speed density. Try looking at pulse width and see if it looks reasonable when compared to the amount of air the engine is using. You may be getting into areas of the VE table that you're not expecting. If pulse width looks reasonable, then it's most likely a delivery issue. Either that, or your sensor is whacked. I've seen cases where a sensor is reasonably accurate under some conditions, but not under others.

[Texas Terminator]

If it all started when you did the Auto VE mods, then that's the logical place to start. I did a bunch of playing after switching to speed density and found that the VE table had to be changed considerably from one that I worked over when running with the MAF. My fuel trims were slightly negative when running the MAF, but went way lean when I switched to speed density. Try looking at pulse width and see if it looks reasonable when compared to the amount of air the engine is using. You may be getting into areas of the VE table that you're not expecting. If pulse width looks reasonable, then it's most likely a delivery issue. Either that, or your sensor is whacked. I've seen cases where a sensor is reasonably accurate under some conditions, but not under others.

makes sense, but, i have done everything that the tutorial says to do exactly. so, why would i be having this problem and nobody else is? weird i tell you! as for my injector pulse width, i have never messed with it nor had to, why now? and why would it be so drastic of a change? it is perfect until i hit higher (above 50% TPS and MAP) numbers then it spikes lean. also, nobody has even been able to tell me what a "correct" pulse width should be with my injectors (Racetronix 57lb) should be. in fact, i e-mailed Racetronix directly and they said it should not be changed from stock. as for the sensor being whacked, i am guessing you are talking about my WBO2? not the case, it is brand new and freshly calibrated. beyond that, it is QUITE obvious when the motor starts cutting out and detonating severely.

did i mentinon this sucks?

[Texas Terminator]

so are there any other ways to get the VE table straight?

[Blacky]

Getting back to the IFR table.
If your FRP is NOT referenced to manifold vacuum, then it MUST be a sloped line.
If the FRP IS referenced to manifold vacuum then it SHOULD be a flat line.

Can you log the PID {SAE.FUELSYS}, it shows when the PCM changes from closed to open or open to closed loop. It will show you if the fueling is going lean as soon as the PCM enters open loop.

Regards
Paul

[caver]

Have you checked your fuel pressure under load? Might be a issue there. What is your inj duty cycle running at when it gets lean? Max duty cycle at 6000rpm should be around 85% if its way lower than that then go beserk on the VE table if its around there with those injectors than you definitely have a supply problem.

[Texas Terminator]

Have you checked your fuel pressure under load? Might be a issue there. What is your inj duty cycle running at when it gets lean? Max duty cycle at 6000rpm should be around 85% if its way lower than that then go beserk on the VE table if its around there with those injectors than you definitely have a supply problem.

fuel pressure is fine under load, steady at about 57 PSI.

[Texas Terminator]

Getting back to the IFR table.
If your FRP is NOT referenced to manifold vacuum, then it MUST be a sloped line.
If the FRP IS referenced to manifold vacuum then it SHOULD be a flat line.

Can you log the PID {SAE.FUELSYS}, it shows when the PCM changes from closed to open or open to closed loop. It will show you if the fueling is going lean as soon as the PCM enters open loop.

Regards
Paul

hey Paul, i am doing the AutoVE right now, so, it doesn't even enter closed loop, at least it shouldn't be. on that note, it is very lean for about thirty seconds or so upon initial startup, barely idles etc., then it levels out, almost like it is going into closed loop.

please explain the whole idea of why the IFR table should be sloped if the FPR is not referenced, it just doesn't make sense to have the PCM think the IFR is changing if the rail pressure is staying constant. i don't get it.

[Texas Terminator]

FWIW, i have bumped up my VE tables by the recommended 15% across the board, then another 25% throughout the middle and 50% in the upper MAP ranges and all seems to be well. it is still SLIGHTLY lean, but it improving considerably with the AutoVE adjustments.

[Blacky]

Getting back to the IFR table.
If your FRP is NOT referenced to manifold vacuum, then it MUST be a sloped line.
If the FRP IS referenced to manifold vacuum then it SHOULD be a flat line.

Can you log the PID {SAE.FUELSYS}, it shows when the PCM changes from closed to open or open to closed loop. It will show you if the fueling is going lean as soon as the PCM enters open loop.

Regards
Paul

hey Paul, i am doing the AutoVE right now, so, it doesn't even enter closed loop, at least it shouldn't be. on that note, it is very lean for about thirty seconds or so upon initial startup, barely idles etc., then it levels out, almost like it is going into closed loop.

please explain the whole idea of why the IFR table should be sloped if the FPR is not referenced, it just doesn't make sense to have the PCM think the IFR is changing if the rail pressure is staying constant. i don't get it.

Assume the FRP is not referenced to MAP and stays constant, for example assume it is 50psi.
When the manifold pressure is 5psi (closed throttle), the pressure differential across the injector is 50-5=45psi.
When the manifold pressure is 14psi (WOT), the pressure differential across the injector is 50-14=36psi.
In this case, the injectors will flow different amounts of fuel at those different pressures. In other words, as the MAP increases it becomes more difficult for the injectors to inject fuel into the higher pressure air. The IFR table calibrates for that condition with a sloped line.

If the FRP is referenced to manifold vacuum, then as the manifold pressure increases (i.e. vacuum decreases) the FRP increases accordingly so:
When the manifold pressure is 5psi (closed throttle), the pressure differential across the injector is (50+5)-5=50psi.
When the manifold pressure is 14psi, the pressure differential across the injector is (50+14)-14=50psi.
In this case the IFR table will be flat since the pressure drop across the injectors is constant no matter what the MAP is.

(UPDATE: the above figures do not take into account that MAP is absolute pressure and FRP is gauge pressure. See joecar's more accurate figures two posts below...)

Regards
Paul

[xtreme]

I've seen this before.

Just a reminder: The actual calculation for injection time is done behind the scenes and is a function of the cylinder volume, VE, target A/F, and injector flow rate. So, if you said the injectors were bigger or smaller, the pulse width would adjust accordingly to get you back to the same target A/F with the same VE.....etc

If it is still lean, then the VE table needs to be larger values in order to reach the targets.

I'm still curious. What are some of the VE values around about 4000-5000 rpm 100%

[joecar]

Non-referenced regulator:


delta = pressure difference across injector
= FP + BARO - MAP
= FP + VAC

IFR = Rr.SQRT(delta/Pr)
= Rr.SQRT((FP + VAC)/Pr)

where
Rr = rated flow rate
Pr = rated pressure (relative to BARO)
= pressure difference across injector during rating

IFR varies as VAC (manifold vacuum) varies.
IFR table must reflect this variation to allow PCM to compensate for this MAP-induced flow rate variation. Edit: i.e. IFR table must be sloped.

MAP-referenced regulator:


delta = pressure difference across injector
= FP + BARO + MAP - MAP
= FP + BARO

IFR = Rr.SQRT(delta/Pr)
= Rr.SQRT((FP + BARO)/Pr)

where
Rr = rated flow rate
Pr = rated pressure (relative to BARO)
= pressure difference across injector during rating

IFR stays constant (as long as BARO is constant).
IFR table must be flat (no slope) since flow rate is constant regardless of MAP variation.

[Texas Terminator]

Getting back to the IFR table.
If your FRP is NOT referenced to manifold vacuum, then it MUST be a sloped line.
If the FRP IS referenced to manifold vacuum then it SHOULD be a flat line.

Can you log the PID {SAE.FUELSYS}, it shows when the PCM changes from closed to open or open to closed loop. It will show you if the fueling is going lean as soon as the PCM enters open loop.

Regards
Paul

hey Paul, i am doing the AutoVE right now, so, it doesn't even enter closed loop, at least it shouldn't be. on that note, it is very lean for about thirty seconds or so upon initial startup, barely idles etc., then it levels out, almost like it is going into closed loop.

please explain the whole idea of why the IFR table should be sloped if the FPR is not referenced, it just doesn't make sense to have the PCM think the IFR is changing if the rail pressure is staying constant. i don't get it.

Assume the FRP is not referenced to MAP and stays constant, for example assume it is 50psi.
When the manifold pressure is 5psi (closed throttle), the pressure differential across the injector is 50-5=45psi.
When the manifold pressure is 14psi (WOT), the pressure differential across the injector is 50-14=36psi.
In this case, the injectors will flow different amounts of fuel at those different pressures. In other words, as the MAP increases it becomes more difficult for the injectors to inject fuel into the higher pressure air. The IFR table calibrates for that condition with a sloped line.

If the FRP is referenced to manifold vacuum, then as the manifold pressure increases (i.e. vacuum decreases) the FRP increases accordingly so:
When the manifold pressure is 5psi (closed throttle), the pressure differential across the injector is (50+5)-5=50psi.
When the manifold pressure is 14psi, the pressure differential across the injector is (50+14)-14=50psi.
In this case the IFR table will be flat since the pressure drop across the injectors is constant no matter what the MAP is.

Regards
Paul

you have me thrown a bit when you say "PSI". this is not a FI setup.

so, in laymans terms, it is basically boosting pressure to the injector because it is harder (less vacuum) to pull the air/fuel charge into the motor?

[Texas Terminator]

I've seen this before.

Just a reminder: The actual calculation for injection time is done behind the scenes and is a function of the cylinder volume, VE, target A/F, and injector flow rate. So, if you said the injectors were bigger or smaller, the pulse width would adjust accordingly to get you back to the same target A/F with the same VE.....etc

If it is still lean, then the VE table needs to be larger values in order to reach the targets.

I'm still curious. What are some of the VE values around about 4000-5000 rpm 100%

this is where things get a bit weird and maybe someone can clarify it for me. my VE tables (2000 OS) can not be set any higher than "10". i have seen other peoples VE tables and they go considerably higher. what gives?

as far as the actual numbers are concerned, i am showing anywhere from a low of 1.5008 at 4k RPM at 20kPa to a high of 3.5232 at 4800 RPM at 95kPa.

[Texas Terminator]

Non-referenced regulator:


delta = pressure difference across injector
= FP + BARO - MAP
= FP + VAC

IFR = Rr.SQRT(delta/Pr)
= Rr.SQRT((FP + VAC)/Pr)

where
Rr = rated flow rate
Pr = rated pressure (relative to BARO)
= pressure difference across injector during rating

IFR varies as VAC (manifold vacuum) varies.
IFR table must reflect this variation to allow PCM to compensate for this MAP-induced flow rate variation. Edit: i.e. IFR table must be sloped.

MAP-referenced regulator:


delta = pressure difference across injector
= FP + BARO + MAP - MAP
= FP + BARO

IFR = Rr.SQRT(delta/Pr)
= Rr.SQRT((FP + BARO)/Pr)

where
Rr = rated flow rate
Pr = rated pressure (relative to BARO)
= pressure difference across injector during rating

IFR stays constant (as long as BARO is constant).
IFR table must be flat (no slope) since flow rate is constant regardless of MAP variation.

WHOA!! huh? where is the "flew over my head" smilie? :)

so again, the increased IFR is simply there to help "push" more fuel in at higher vacuum levels?

[Blacky]

so again, the increased IFR is simply there to help "push" more fuel in at higher vacuum levels?

Yes. But higher pressure, lower vacuum - but I guess it depends on your point of view :)

[Blacky]

this is where things get a bit weird and maybe someone can clarify it for me. my VE tables (2000 OS) can not be set any higher than "10". i have seen other peoples VE tables and they go considerably higher. what gives?

as far as the actual numbers are concerned, i am showing anywhere from a low of 1.5008 at 4k RPM at 20kPa to a high of 3.5232 at 4800 RPM at 95kPa.

The 0..10 limit is when the VE table is displayed in g*K/kPa. (internal PCM units)
When the VE table is displayed as % you will see values from 0 to 100 or even slightly higher. And much higher for boosted applications (200, 300% for 2 and 3 bar)

Change the VE unit display in Properties->Display->Volumetric Efficiency values. You must re-load the *.tun file for the change to take effect.

Regards
Paul

[Texas Terminator]

so again, the increased IFR is simply there to help "push" more fuel in at higher vacuum levels?

Yes. But higher pressure, lower vacuum - but I guess it depends on your point of view :)

so, pretty much EVERYONE has their stock regulator disconnected? i say this because of the IFR calculator that everyone uses. would this be a correct statement? now that i look back at my stock tune, it is flat (regulator connected to manifold vacuum). and i just looked at some tunes that are in f-bods and corvettes and they are sloped. do they not have a manifold vacuum referenced regulator?

[joecar]

so again, the increased IFR is simply there to help "push" more fuel in at higher vacuum levels?
Sloped IFR is there to help PCM calculate the correct injector pulse width.

With unreferenced regulator, injector flow rate increases as vacuum increases
(due to increased pressure difference across injector caused by vacuum);

at idle (vac = 65 kPa) flow rate is more than at WOT (vac = 0 kPa) due to vacuum;

IFR table slopes up as VAC increases; the kPa scale is VAC (which is BARO - MAP);

PCM figures out current vacuum (BARO - MAP);
PCM figures out mass of air in cylinder(s);
PCM figures out what air:fuel ratio to use;
PCM calculates the mass M of fuel required for air:fuel ratio;
PCM looks up IFR table to know how much mass the injectors flow at current vacuum,
and uses this to calculate injector pulse width to spray M.

All F-body and most Y-body models came from factory with unreferenced regulator (and sloped IFR table).

If you want to run turbo/SC, then a MAP-referenced regulator makes life easier (since IFR table scale goes 0 to 80 kPa vacuum only)
since this will maintain a constant pressure difference across the injectors regardless of MAP (vacuum or boost) so they will have a constant flow rate.

[joecar]

Texas,

With wrong IFR values or wrong IFR slope,
PCM commands wrong injector pulse widths, causing AFR to go rich or lean
(or some combination in different cells), causing LTFT's to adjust;
and engine appears to run just fine.

When you go into SD tuning mode, LTFT adjustments are disabled.

What were your LTFT's before going into SD tuning mode...?

[Texas Terminator]

Texas,

With wrong IFR values or wrong IFR slope,
PCM commands wrong injector pulse widths, causing AFR to go rich or lean
(or some combination in different cells), causing LTFT's to adjust;
and engine appears to run just fine.

When you go into SD tuning mode, LTFT adjustments are disabled.

What were your LTFT's before going into SD tuning mode...?

first of all, mucho thx for all the input!!

as for the Ltrims, i really can't answer that. when i got EFI flashscan, i started from scratch (modified essentials from stock tune). i had EDIT before and never really got a good tune on it. if i were to guess, i would say that they were definately positive.

perhaps i should have mentioned this is a truck from the beginning? oops!

[dfe1]

To help put things in perspective, just think of the IFR table as the electronic equivalent of a vacuum line connected to the fuel presure regulator. The regulators in trucks are vacuum referenced, those in F- and Y-bodies are not. That being the case, the IFR table for a truck cal is flat while those in F- and Y-body files are sloped. There's no magic to a sloped IFR table, it's just an electronic means of compensating for lack of an actual pressure regulator vacuum reference.

[marcink]

IFR table slopes up as VAC increases; the kPa scale is VAC (which is BARO - MAP)

Does it mean that when you have SC or turbo you MUST have your FPR vacuum/boost referenced? How will the IFR table work when MAP>BARO? What happens when index is out of range? What value will be taken from the table?

I am interested in that as I will be installing 2 bar MAP, custom OS an a TT soon. Please explain that if you know.

Thanks,
MarcinK

[Blacky]

IFR table slopes up as VAC increases; the kPa scale is VAC (which is BARO - MAP)

Does it mean that when you have SC or turbo you MUST have your FPR vacuum/boost referenced? How will the IFR table work when MAP>BARO? What happens when index is out of range? What value will be taken from the table?

I am interested in that as I will be installing 2 bar MAP, custom OS an a TT soon. Please explain that if you know.

Thanks,
MarcinK

Anytime a table index runs of the end of a table the PCM uses the value at the edge of the table. See image.

Regards
Paul