Printable View
Awesome work
Someone got the parameters listing to add
2312 ?
2127.240 ?
Im looking for a datasheet to know parameters listing
Thanks
The full math parameter: GMVE STFT =
(([16.71]*(1+(.01*[6.156]+.01*[8.156])/2)-[12.56]/60*4*[11.92]*[2312]/[2127.240])/([12.56]/60*4*[11.92]*[2312]/[2127.240]))*100
Is there any reason why when I plug the math parameter in there are two sensors that are in red?
Attachment 104239
You have to record them in the channels list for them to be available to the parameter.Quote:
Originally Posted by sgod1100
i knew it was something stupid I was missing...thanksQuote:
Originally Posted by smokeshow
thanks for laying this out smoke, i came across it a month or two ago while trying to dial in new (larger cu in) engine. having it as a check against the error% helped a bunch and finally got it close to move onto full SD tune
Been a while since I got brain fog from math, so bear with me.Quote:
Originally Posted by jsllc
Since fuel trims are an error, why is WB Error Ratio ((WB_EQ - EQ Commanded)/EQ Commanded)*100 not used in this math?
It's there. Your statement is equivalent:Quote:
Originally Posted by JtblckZ
((WB_EQ - EQ Commanded)/EQ Commanded) = ((WB1/EQ Commanded)-1) = WB_avg
Can you replace the GMVE_af calculation by logging the "Volumetric Efficiency Airflow" parameter from the E67 ECU and let the ECU do that math? I've tuned my MAF with a WB and I compare the logged "Mass Airflow (SAE)" value against the "Volumetric Efficiency Airflow" value to help tweak my VVE table.
I want to update/clarify part of this process. There are a number of things in my bag of calibration tricks that I don't necessarily keep to myself but just neglect to heavily detail due to complexity and/or implementation time. One of these things is the first order lag filter that the software uses to model the actual charge temperature. You will know it as the 'complex cylinder temp model' with the bias and filter tables. When I come across data that suggests the fueling isn't lining out due to temperature model issues, I can run the IAT/ECT data through my own scripting and correlate the data with a temp calibration problem that way...but this isn't possible unless you're familiar with MATLAB or Octave. However, as HPT continues to improve their definition files, this becomes unnecessary in some applications.
Any vehicle where HPT has the CAN variable 'Manifold Air Temperature' or 'MAT' defined, you should absolutely use it in place of IAT in the original math parameter. This is the charge temp estimate that is actually input to the GMVE speed density equation after the complex cylinder temp bias/filter has performed it's calculation. While IAT is pretty close, it is insufficient to properly model cylinder airmass at low engine airflow. This of course begs the question, is my bias/lag filter calibrated properly...that depends on your engine hardware mods, and I haven't detailed that here or made a writeup on it. But MAT will net you better results during your VVE calibration than IAT will, if you have it available. So use it if it shows up in your parameter search in the HPT scanner.
Hey smokeshow, thanks so much for sharing this tip. It has helped me out heaps and I use it all the time. I had to modify your parameter slightly to suit my setup i.e. LTFT + STFT and WB EQ ratio error.
I am using manifold air temp sensor.
I have had the calculator out a number of times for sanity checks comparing the math result with the the expected result. For testing I tried the following in your formula:
- VVE air flow PID (g/s) -not ideal as it maxes out at 512g/s
- VVE PID (mg*K/MPa)
- + all of the above with LTFT + STFT and WB EQ Ratio Error
All methods came out near smack on.
Well done, and thanks so much once again :cheers:
This brings up another point for those of you who have added blowers and hopefully relocated your IAT. There are a couple of biasing charts for MAT based on ECT, time, vehicle speed that should be zeroed out since your relocated IAT is now reading directly from the manifold. Some of the canned tunes that come with a blower kit do not do this, even if they provide a IAT relocation in the kit.Quote:
Originally Posted by smokeshow
No problem. The math parameter is meant to be adjusted anyway, depending on your specific vehicle and whether you're working in OL or CL. This procedure assumes the user is comfortable with calibration enough to edit the math parameter to their liking.Quote:
Originally Posted by hjtrbo
Will this work on a gen v using e92 ecms? I ask because I assume it will not but maybe with slightly different math? Or no because of the vve and not a true ve?
It does. I'm using it on my E92A ecm now, after some help from Smoke. Just make sure the math is set up for your specific application.Quote:
Originally Posted by tjd1991
It’s hard, if there is a video tutorial, it’s OK
Easiest method is calculating GMVE from the Dynamic Airflow when in MAF only mode.
Dynamic Airflow = the filtered airflow used by the ECM to fuel the engine. In MAF only mode this will = Filtered MAF airflow.
Once you have made wideband and/or fuel trim corrections to Dynamic Airflow you can then use that value as "Corrected Airflow" to then use the GMVE model to derive a GMVE value. You can then compare that Generated GMVE or what I call "CalcGMVE" to the GMVE delivered from the VE/VVE which is logged as GMVE. A simple error calculation can then be used to give you a GMVE Error to plot against your VE histogram to make corrections to while you are also using normal error calculations to dial the MAF in.
Quote:
Originally Posted by LSxpwrdZ
I think you're swapping some definitions around there. Dynamic airflow should not be used to calculate the VVE. It is a modeled representation of the nonlinear relationship between the two (technically three) methods for estimating cylinder air mass and the actual cylinder air mass that gets trapped in the cylinder. With the distance between the MAF and the intake valve, the MAP sensor and the intake valve as well as the tuning effects of the manifold and cam phasing, in an airflow transition both the MAF and VVE are inaccurate. Dynamic air exists to correct that inaccuracy during a transition - but it requires the MAF and VVE to be steady-state accurate already. MAF and VVE air estimation are the starting point in the calculation which dynamic then modifies during transients...so you can't calculate GMVE values from dynamic airflow. It only works the other way around.
Do you have a good video of that process? I was getting just a little lost on each step. :)Quote:
Originally Posted by LSxpwrdZ
Definitely not mixing definitions around. You only see all those models when you have it in blended mode... Obviously filtering transients which should be done anytime you are trying to dial in a fuel model anyhow. Hence why I also said using it in MAF only mode. Dyno tuning everyday with an eddy brake dyno proves this functional 100%. One WOT pull both MAF and VE/VVE are within 1-2% error.Quote:
Originally Posted by smokeshow
No I don't currently. I may make one on it one of these days.Quote:
Originally Posted by TheMechanic
This same system works on Gen5 stuff also on certain years (at least the years GMVE PID is logable).