NoGo

iTrader: (2)

I have been working on the VE table quite a bit lately and I think I found the solution.

Thanks to gameover for some key hints.


The VE table looks as though it is in meters cubed, it is just not used like a conventional VE table. The VE values are such that the PCM can directly backcalculate to g/cyl, the primary means to determine fueling and timing.
This is what makes it so confusing. You can't solve for air mass, you have to solve for g/cyl.

Anyhoo, here is the equation

VE = ((massflow * IAT / (MAP * RPM * Displacement))

Massflow: grams/sec
IAT: Degrees Kelvin
MAP: Bar
RPM: Duh!
Displacement: Cubic Meters

To solve for the massflow in g/sec simply re-arrange the equation.

Massflow = (VE * MAP * RPM * Displacement) / IAT


My old method of "Divide by 30" works okay because we are inadvertantly solving for a volume ratio. The molar mass of air is 28.96 g/mol.


I have compared the above equation to every bit of data that I have ever collected for a stock car, and the data matches up perfectly.

Any corrections, comments, or blinding errors please let me know.

Thanks,
Kevin

HumpinSS

iTrader: (1)

How much more accurate is this formula compared to the old on

Looks like i need to get cracking on an access DB and some queries to help tabulate this data for use in EDIT

NoGo

iTrader: (2)

The old one should be off by about 4% given the difference between the actual molar mass of 28.96 and the value of 30 that we were using.

I should probably re-iterate that your MAF has to be accurate in order for the above calc to work. I just got done re-checking some more and was about ready to shoot myself when I realized I was comparing against a vehicle with a ported MAF. It makes a big difference. The MAF error directly translates to the equation error. So if your MAF is off by 5%, then your VE calcs will be off by 5%.

HumpinSS

iTrader: (1)

Oh ok that may explain why I had to adjust my ifr TABLE BY 4% after doing some calculations based on Chris B's ve calculator (assuming he used the same formula). My maf is stock and the tables are stock so I should be ok, I do plan on chaging the maf but it will be to the 85mm Z06 chumpy which i have tables for via www.ls1howto.com

Bink

Congrats!!!
So, for now, I need to rescale my VE table by 96%.
You are the MAN.
joel

HumpinSS

iTrader: (1)

That will work also. Damn I didnt even think about that. I wouldve just recalculated the whole table amd went from there. We need ChrisB to verify that he is using the old formula

CHRISPY

awesome!!

Brains

iTrader: (7)

Whoohoo!! Excellent work, now I can do some math rather than my tedious trial-and-error method

Bink

I think ChrisB's formula is the same ...except we changed 25 to 30 as the max scaled value.

28.96/30.00 = 97% FWIW.
joel

gameover

Try THIS

As always, open to corrections

Bink

gameover - Thanks

From your HP Tuners site:
I have these ???


g/cyl * 8192 = MAP * 51.2 * VE / (T * 32)

8192 - where is this from?

T * 32 - Does the PCM blend the IAT & ECT values?? Where does "32" come from?




VE = 8192 * 28.96 * 32 * V / (8.3145 * 51.2) - Where did T go??


Can this be -> VE% = VE value / (178.33 * CylVol) if so , then is VE Table Value = VE% * 178.33 * 0.708 ??




joel

NoGo

iTrader: (2)

Hello,

I came at your formula from several different angles and I couldn't get it to check out. Correct me if I am wrong.

From your determinations

VE = 17833 * V

solving for VE in terms of VE:

V = VE / 17833

Substituting the determined value of VE into the ideal gas law.

PV = nRT
n = PV / RT


Running just a simple check, I have tried some WOT data that I have for a bone stock 2000 Z28.

P = 100kPa -----converted to 100000 pA for the equation
V = 2349 / 17833 -----determined from VE table at 100kPa & 5600 rpm
R = 8.314 -----just to iterate this value of R locks the units to m3, Pa, K
T = 304 K ----- dyno done on a 88 deg F day.

Solving......

n = [100000 * (2349 / 17833)] / [8.314 * 304]
n = 13172.20 / 2527.45
n = 5.211 moles

Using the molar equation: mass = # moles * molar weight
mass = 5.211 moles * 28.96 g/mol
mass = 150.92 grams

This doesn't work out. The actual air flow at WOT is closer to 250 grams/sec. I tried your formula several other ways and against real world data for several other operating points and it still doesn't work out.

Also, how did you come to this equation? Hex editoring through the code??

Thanks

CAT3

iTrader: (9)

Ok, Kevin, is there going to be a chart (spreadsheet) or something to plug our data into or am I a bigger math flunky than I thought? I would love to solve my VE tables, more so after the new cam, and my friend as well.

Basically I need to log the following for your calculations:

MAF
MAP
IAT
RPM
and I might be able to make a SS, if thats right?

Thanks.
Charlie

T/A Medic

iTrader: (1)

so how would u figure it out if u didnt have a MAF to register the incoming air? ie spd density

Bink

You need a sensor to measure air.

gameover

A few things:

1. your VE table values you are using are scaled values not the real ones in the PCM (try multiplying by 5.12)
2. the calculation determines airmass per cylinder (ie. g/cyl) not flow in grams per second as the MAF would measure. You need to convert from g/cyl to g/sec using the normal calc mult by RPM/15 etc.
3. also the V from my equation is in litres not m3 as is required by your calc you need to convert (/1000)

i make it as about 0.77 g/cyl using your numbers which is ~280 g/sec.

NoGo

iTrader: (2)

1. I thought 5.12 was the scaling factor that the PCM used for the RPM due to our unique RPM signal. I find it odd that it would apply the same scaler to the VE table???

2. I already converted solving for g/cyl. You can just replace the massflow in my equation with the equvilant g/cyl (ie: g/cyl = (15 * Massflow) / RPM

3. Okay, I redid your calcs with the above suggestions, and I still can't get it to work out with the recorded data. Scaling the VE by 5.12 and dividing by 1000 yields a result less than 1.


I just pulled up some WOT data on 3 different cars. The massflow at 5600 RPM was 270.19, 262.63, 258.47 (grams). Using my calculation you can determine 258 g/sec. Pretty close if you ask me.


Somewhere, we're not lining up. I think we are using different VE values for solving.



If somebody could post up or link the LS1 edit values, that would be very helpful. It's not letting me upload my pics for some reason.

gameover

The VE table value of 2349 you are using is not correct it is out by a factor of 5.12 (ie. the actual value stored is more like 12026). They are out by a factor of 5.12, it has nothing to do with the RPM, the PCM does not scale them...

Again, the formula of V = VE / 17833 gives V in litres so when you put it into PV=nRT you need to convert to m3 by dividing by 1000.

so you'd get:

n = 100 * (12026 / 17833) / (8.3145 * 304) = 0.02668 mol

g/cyl = 28.96 * 0.02668 = 0.773

g/sec = 0.773 * RPM / 15 = ~288

NoGo

iTrader: (2)

Okay, that makes a little more sense.

ITS THE SAME EQUATION!! One of ours is just missing a factor

I just did a comparison between your formula and mine for the entire VE table and all solution values differ by an identical 1.1754

Sooooo, it is the same equation just one of ours is missing a factor. We just need to find out who's.

Example. Your solution at WOT compared to mine.

For identical VE values:

VE at 5600 RPM and 100 KPA and 31.1 deg C

You solution yields 288.28 g/sec
My solution yields 245.25 g/sec

The difference between the two is a factor of 1.175

You can do this for any solution point on the table and the difference is the same. Soo, it is the same solution, just one of us is missing a variable or constant.

HumpinSS

iTrader: (1)

Damn my head hurts. Anyways I flashed last night with 96% across the whole table and now i seem to be getting some KR very easy. My ltrims are also off by another 2-3% leaving me at 91% for the IFR table from stock. The car feels a little more responseive and smoother. I have to correct my IFR table to see if that gets rid of my KR, if not i will be force to multiply the whole table starting at 2% to get the KR squared away.

Thanks for all the help and infomration guys it is appreciated. Now all we need is a fuggin spread sheet or something of the sort to get all this data back to edit.