** Advanced Tuners, please read and CORRECTwhat ever I typed in this post, please ! **
While trying to understand my past fuel issue, I learned a lot on how the computer was using the injector characteristic, external inputs and the VE table.
And I think I understand better now why some people were saying that tuning the LTFT had to be done, but at one point you would have to adjust the VE table. I think that I am at this point now in my “learning process”.
If I start from a stock tune and add bigger injectors, the first suggested approach was to multiply the injector flow by a ratio and then fine tuning them using the LTFT as a reference.
But at what point do we have to stop playing with the injectors and start adjusting the VE table? I have read that around plus or minus 5 it was considered more than fine. I guess that with this margin the external environment would be the major responsible of the variation and this would indicate that the VE table would need to be adjusted to reflect the actual efficiency.
But how exactly can I tackle this table? I think that by planning it before starting doing it can only help me (and everybody that is in the same situation)
Now let’s assume that I would be able to use a diablosport to datalog instead of my current aeroforce.
In closed loop, which are more than often low to mid rpm and low manifold pressures, I thought of monitoring the LTFT, RPM, MAP and TIP (or Pressure Ratio if available) , and Pulse Width.
First, if I see that at the same pulse width, at different RPM and different Page Ranking, I get a spike in the LTFT, that would mean that some adjustments would need to be done on the injector size. Once I could not do a relation between the LTFT and the PW anymore, I could switch to the VE table.
With the Pressure Ratio and RPM, for any point on my log I would be able to locate in which cell I have to adjust the VE, and with the LTFT know by how much to adjust it : Negative LTFT = PCM having to remove fuel = Raising the VE value ; Positive LTFT = PCM having to add fuel = Lowering the VE value. The goal would be to collect as much data as possible in the same conditions for the same cell, do averages, correct the table, repeat until satisfied, and try to sweep the lower part of the Page Ranking and as much RPM as possible.
In open loop, I would have to monitor the RPM, MAP/TIP and AFR since the computer doesn’t see shit at these level.
Again, with the Page Ranking and RPM I would locate the cell and than adjust the VE value with the same rule as above but using the AFR as feedback. Repeat the process at the same PR until my AFR matches the PEAS values.
But how would I know that it is not the injector flow rate that is responsible of the AFR variation and not the VE table? Do I absolutely need to perform an injector flow test?????
And something I have to understand before starting any of this is WHY! Will I notice any difference? Going back to my first injector scale made me realize how much smoother the car was running in daily driving when the LTFT were around zero (now back to -32 at idle). I didn’t had to wait for the car to be hot to have a nice AFR, like if the PCM was waiting for the engine to be at operating temperature before loading up the LTFT… and it still moves a lot more than when they were adjusted !
Should I be expecting similar results at WOT with an adjusted VE Table or is it just because we notice all the little bugs when driving slowly?
Don’t be afraid to give your inputs, I might be wrong from the beginning but I do need to know more.
I read above , and it seems right to me (by no means an expert) .
One potential snag . Im afraid there are so many things that modify your final AFR that you will not be able to correlate legitimate data from incidental data . For example , knock , extrapolated exhaust temp , battery voltage , max injector pulsewidth vs. rpm table , ramp value when going richer/leaner , and more will affect the AFRs . This could make it very difficult to decipher the cause of fluctuations .
I too suspect that without a good handle on actual injector flow at multiple PWM points , you will end up building on a foundation of best guesses . I find it impossible to believe that the characteristics of flow between two different injectors , of two different flow capabilities , can be expressed in a single , linear rescale calculation . But you have heard me gripe about this before , nobody publishes comprhensive flow data . So if you are on non-Mopar injectors , like me , your kinda shooting blind .
It seems to me you have a good plan of attack . Dispite my sorta-concerns , I hope you get the results your after . I will be interested in your progress , as I plan going to a non-stock turbo setup next year . which will , no doubt , screw up my VE tables .
Higher VE value = more fuel. Lower VE value = less fuel.
yeah you're right, that's what I meant but typed it all wrong.
By biggest concern is : Does it worth it ?
I took me like 5 reflashes just to get the LTFT in check. But I do have a better drivability out of it. Will it do only a minor difference with the right VE or close to nothing ?
If I have exactly the right AFR (commanded compared to actual) at a certain manifold pressure. Can I say that I will have the same VE at another boost pressure ?
ex :
I ask for 11.4 AFR everywhere in my PE table.
I get exactly 11.4 AFR @ 17 psi
I raise the boost to 20 psi
I still ask for 11.4, but get a oscillating AFR between 11.2 and 11.8.
Can I copy/paste de VE values of 17 psi and paste them at the 20 psi line ?
You do know it is all in Pressure Ratios, and not boost, right? This is actually a very important point. You could use that '17 psi' line as a start to a higher level, but I would add some into that, so you do not go lean.
You do know it is all in Pressure Ratios, and not boost, right? This is actually a very important point. You could use that '17 psi' line as a start to a higher level, but I would add some into that, so you do not go lean.
Yes I know the difference between pressure ratio and boost.
I will just type everything I have in mind, so if you see something that is wrong tell me.
**From now on I will be using a constant atmospheric pressure to simplify my ideas**
The VE table is nothing more than the EFFIENCY of the engine for specific RPMs and PRs. That means that the computer does all the calculations to figure out how much fuel he has to dump in the engine and then multiply the result by the corresponding VE value.
If the calculations are good for 17 psi, that means that the amount of fuel calculated AND the VE values are accurate (or that both are wrong and that the errors on each side are canceling each others, but I don't want to go this way for now).
The "first calculations" of the PCM are 100% accurate because they are based on physic laws. All the other variables which could affect the model (the "unknowns") are put together inside these VE values. One multiplied by the other equals "realistic" amount of fuel.
I changed the manifold, turbo, exhaust, coldpipe, and soon intake manifold. I modified the caracteristics of the engine. I didn't changed the physic laws. But I did modified a good part of all the "unknows".
By saying that the VE values are exactly the same for different boost level, I'm assuming that the calculations are ALWAYS wrong by the same amount (%). Which is wrong. Cool, I just answered myself !
Last edited by SRT-What? : 08-25-2009 at 04:49 PM.
By saying that the VE values are exactly the same for different boost level, I'm assuming that the calculations are ALWAYS wrong by the same amount (%). Which is wrong. Cool, I just answered myself !
I would keep the same "rate of change"(gradient) that currently exits in these rows above 17psi(~2.2-3.0) and just increase the whole lot by an equal %. At least for the immediate row im adjusting. For example if your starting at 2.2 ajudt everything above it also. If the next row is 2.5 adjust it and everything above it at the same time, and so on until you're at the last row.
The turbo mass flow rate can stagnate and higher Pratio depending on its design(slope of compressor map), so the difference between rows may not be very significant. The differences between colums should be more consitant since the only independant variable is RPM(more predictable change to ve) . So I'd adjust entire rows before zeroing in on the RPM.
The VE table is more easily thought of as just a "air mass flow rate correction factor".
__________________
"Just because someone goes fast doesn't mean you should do whatever you think they did.
This path assumes they knew what they were doing, which isn't necessarily the case"..........RB Racing
Last edited by duster360 : 08-26-2009 at 08:21 AM.
The AutoGuide.com network consists of the largest network of enthusiast-owned enthusiast-operated automotive communities.
AutoGuide.com provides the latest car reviews, auto show coverage, new car prices, and automotive news. The AutoGuide network operates more than 100 automotive forums where our users consult peers for shopping information and advice, and share opinions as a community.
VE Table fun
_____ 
Linear Mode
