Notices
ECU Flash

Speed Density 2.0 (3D VE Tables, Baro)

Thread Tools
 
Search this Thread
 
Old Feb 14, 2013 | 01:02 PM
  #571  
10isace's Avatar
Evolved Member
15 Year Member
Liked
Loved
Community Favorite
iTrader: (12)
 
Joined: Nov 2008
Posts: 1,058
Likes: 16
From: East of the Rockies
Originally Posted by cdrinkh2o
10isace - Thanks for the feedback.
So you used the same source files that I did..hmmmm...I did not have 94170715 working before SD...I'm coming from 96530706 and copied fuel, injector scaling & latencies, ignition, MAF Comp, inserted my 3D VE table built from MAF logging.... I'm using a JDMMAP sensor and GMIAT in the IM wired into fuel ADC....so there has to be some difference between your XMLs and mine.

Can you email me your XMLs to cdrinkh2o@gmail.com so I can filediff them?
No problem.

What happens if you open a the 94170715 MAF version?

Just sent it. Do you need the Bin file too?

Last edited by 10isace; Feb 14, 2013 at 01:06 PM.
Reply
Old Feb 14, 2013 | 02:47 PM
  #572  
Dynotech Tuning's Avatar
Former Sponsor
iTrader: (12)
 
Joined: Jan 2011
Posts: 1,312
Likes: 9
From: Seekonk, MA
You need to download the MAF and SD versions since it references both XML files.

-Jamie
Reply
Old Feb 14, 2013 | 04:03 PM
  #573  
10isace's Avatar
Evolved Member
15 Year Member
Liked
Loved
Community Favorite
iTrader: (12)
 
Joined: Nov 2008
Posts: 1,058
Likes: 16
From: East of the Rockies
I lied. It's not working. I'm having the same issue, starts up, goes lean if trying to throttle, misfires and dies. I had the MAF plugged in and tuned. When I pulled the MAF plug

I've inherited all 3 roms. I had this trouble with the 2D Rom. So I'm sure it's a problem with the 2D Rom.

Just loaded 96531706 2d SD. No MAF, works great. DAMN!!!
Reply
Old Feb 15, 2013 | 05:28 AM
  #574  
Biggy VIII's Avatar
Evolving Member
 
Joined: Nov 2008
Posts: 424
Likes: 3
From: Russia, Moscow
Originally Posted by cdrinkh2o
^ BIGGY - I had a similar experience with 96534706 on a USDM 2005 Evo 8. That is why I am now trying 94174715.


OH! I feel much better now, as I'm not the only one with that problem!!!

I'm definately going to try 9417 SD 2.0 rom. So I have several questions, hope somebody can answer me:

- If I want to use Fuel Temp pin as MAT input, I put 0x03 in "SD Convert IAT to MAT (0x03 -> 0x0E)" ? Should I make any other changes?

- Do I need to make any changes in "MAP sensor" section of the rom? I have EDM Evo 8 that doesnt have MAP sensor in stock...

Last edited by Biggy VIII; Feb 15, 2013 at 10:06 AM.
Reply
Old Feb 15, 2013 | 09:57 AM
  #575  
99EclipseGSX's Avatar
Evolving Member
 
Joined: May 2005
Posts: 100
Likes: 0
From: Santa Monica, CA
Originally Posted by Dynotech Tuning
These are correct in the 3D roms I have used.

A good way to see if its scaled correctly is to set the entire VE table to 100% and adjust your injector scaling temporarily so the car is driveable. Under normal acceleration, the KPA and LOAD should log the SAME values.

-Jamie
Using your scaling values seemed to help with one issue, but this patch still is giving me lots of load problems. It's reporting way too low of a load at all times, even with high VE values. As a test I put 100% VE in the table at anything above idle, and I hit 14 psi and load still only reported a max of 115, where of course since kPa to Load is set 1:1, and so is Map Sensor kPa, so I should be seeing closer to 190 load at that boost level (approx 190 kPa boost).

I'm also having to use WAY too low of an injector scaling value to get the car to about where it should be fuel trims wise at light load. Typical Actual engine VE will typically be below 60% at idle, and there's no way I'd be able to get the car to run down there. It runs perfect with the older 31706 2D SD stuff.

This is why I usually stick to standalones.. It's fun playing with patches though but I really don't like skewing/bandaiding data to get the car to run properly. I should be able to input an accurate injector scaling value, accurate MAP sensor data, and realistic VE values and have the load match actual load (kPa) and have actual AFR match target AFR.

Beau
Reply
Old Feb 15, 2013 | 10:05 AM
  #576  
Biggy VIII's Avatar
Evolving Member
 
Joined: Nov 2008
Posts: 424
Likes: 3
From: Russia, Moscow
Originally Posted by 99EclipseGSX
Using your scaling values seemed to help with one issue, but this patch still is giving me lots of load problems. It's reporting way too low of a load at all times, even with high VE values. As a test I put 100% VE in the table at anything above idle, and I hit 14 psi and load still only reported a max of 115, where of course since kPa to Load is set 1:1, and so is Map Sensor kPa, so I should be seeing closer to 190 load at that boost level (approx 190 kPa boost).

I'm also having to use WAY too low of an injector scaling value to get the car to about where it should be fuel trims wise at light load. Typical Actual engine VE will typically be below 60% at idle, and there's no way I'd be able to get the car to run down there. It runs perfect with the older 31706 2D SD stuff.

This is why I usually stick to standalones.. It's fun playing with patches though but I really don't like skewing/bandaiding data to get the car to run properly. I should be able to input an accurate injector scaling value, accurate MAP sensor data, and realistic VE values and have the load match actual load (kPa) and have actual AFR match target AFR.

Beau
Sounds similar to my problem with 96534706. But my load won't go any higher than 75ish...
Reply
Old Feb 15, 2013 | 10:11 AM
  #577  
Dynotech Tuning's Avatar
Former Sponsor
iTrader: (12)
 
Joined: Jan 2011
Posts: 1,312
Likes: 9
From: Seekonk, MA
I'm really surprised you guys are having these issues. I never have any of these issues myself *shrug*.

-Jamie
Reply
Old Feb 15, 2013 | 10:40 AM
  #578  
Architechnik's Avatar
Newbie
 
Joined: Dec 2012
Posts: 18
Likes: 0
From: Portland, OR
It sounds like you've missed some of the scaling options. You'll need to go to each place that references the MAP sensor and go into the settings by editing the parameter. I think there were three of them when I first went in to edit mine from the 4-bar Kavlico or whatever the initial was to the GM 3-bar.

There, you will select the type of MAP sensor your vehicle has and save it. It will change one of the columns of numbers. You'll copy those numbers over into the other column and then you'll have all the settings correct.

These ROMs all use multiplier tables to get the scaling correct and if you have it set to anything other than what your MAP sensor output is providing, then your kPa will be off. There are some other custom versions you may need, but you have to be sure to copy it into each available parameter group.

As a checklist, try going to the original tephra7 install instructions: https://www.evolutionm.net/forums/ec...ched-roms.html
This should give you a good baseline with the 3D setup, since that was the starting point for the original setup. The only thing that really changes are the inputs for VE table at various points... you just get more resolution here, nothing else.

If problems still persist, there's a good chance you have a mechanical failure that was not necessarily noted in the previous ROMs you were using. Similarly, as noted a couple posts back, be sure you're using the right ROM for your type of vehicle, EDM, JDM, or USDM.

You should also build out a 3D VE table using an excel spreadsheet if you had a previous SD setup. Consider the two tables as multipliers to get the correct value at a particular point. The provided one is only between 80 and 100, while mine is 58 to 135.

I have FIC 1000s scaled at ~948 with higher latency than recommended for Evos, but I'm finding I have way more control and I'm not dumping fuel when it's not needed. It's probably part of why my scaling is much more broad than the one provided.
Reply
Old Feb 15, 2013 | 10:57 AM
  #579  
10isace's Avatar
Evolved Member
15 Year Member
Liked
Loved
Community Favorite
iTrader: (12)
 
Joined: Nov 2008
Posts: 1,058
Likes: 16
From: East of the Rockies
Originally Posted by Architechnik
It sounds like you've missed some of the scaling options. You'll need to go to each place that references the MAP sensor and go into the settings by editing the parameter. I think there were three of them when I first went in to edit mine from the 4-bar Kavlico or whatever the initial was to the GM 3-bar.
I've done this, but what about the KPA to Load? That reads Map sensor KPA.
Reply
Old Feb 15, 2013 | 11:12 AM
  #580  
99EclipseGSX's Avatar
Evolving Member
 
Joined: May 2005
Posts: 100
Likes: 0
From: Santa Monica, CA
Originally Posted by Dynotech Tuning
I'm really surprised you guys are having these issues. I never have any of these issues myself *shrug*.

-Jamie
Originally Posted by Architechnik
It sounds like you've missed some of the scaling options. You'll need to go to each place that references the MAP sensor and go into the settings by editing the parameter. I think there were three of them when I first went in to edit mine from the 4-bar Kavlico or whatever the initial was to the GM 3-bar.

There, you will select the type of MAP sensor your vehicle has and save it. It will change one of the columns of numbers. You'll copy those numbers over into the other column and then you'll have all the settings correct.

These ROMs all use multiplier tables to get the scaling correct and if you have it set to anything other than what your MAP sensor output is providing, then your kPa will be off. There are some other custom versions you may need, but you have to be sure to copy it into each available parameter group.

As a checklist, try going to the original tephra7 install instructions: https://www.evolutionm.net/forums/ec...ched-roms.html
This should give you a good baseline with the 3D setup, since that was the starting point for the original setup. The only thing that really changes are the inputs for VE table at various points... you just get more resolution here, nothing else.

If problems still persist, there's a good chance you have a mechanical failure that was not necessarily noted in the previous ROMs you were using. Similarly, as noted a couple posts back, be sure you're using the right ROM for your type of vehicle, EDM, JDM, or USDM.

You should also build out a 3D VE table using an excel spreadsheet if you had a previous SD setup. Consider the two tables as multipliers to get the correct value at a particular point. The provided one is only between 80 and 100, while mine is 58 to 135.

I have FIC 1000s scaled at ~948 with higher latency than recommended for Evos, but I'm finding I have way more control and I'm not dumping fuel when it's not needed. It's probably part of why my scaling is much more broad than the one provided.
As 10isace said, there only appears to be one spot where scaling needs to be corrected based on the MAP sensor, which is "MAP Sensor kPa". That is confirmed to have correct scaling, the MAP sensor itself reads perfectly and I have the values set 1:1 across.

"kPa to Load" has scaling that doesn't appear to be based on a MAP sensor but generic for any of them. Again, it's set 1:1. I've even tried setting all the auto-baro values to 1.010 as I don't really trust that yet check yet.

Again, if I do a compare ROM to my 2D SD ROM (31706), there are no significant differences other than of course the SD tables, and that runs awesome with somewhat realistic VE values and injector scaling values. Also, it reports accurate load that very closely matches boost in kPa. I'd stay with it although we all know VE can only be accurately represented in a 3D table, and there are some areas where I could benefit from true 3D control, hence my switch to 3D.

Beau
Reply
Old Feb 15, 2013 | 01:31 PM
  #581  
Architechnik's Avatar
Newbie
 
Joined: Dec 2012
Posts: 18
Likes: 0
From: Portland, OR
Oh, I apologize; I think I was messing with the boost level items for the other MAP conversion inputs.

Have you copied over the cells from the 2d version of the tables into a multiplied out 3d version?
If you're not very savvy with spreadsheets, try downloading this and filling out the highlighted rows. If you don't have a value for a cell, get the average of the cells next to it.
https://docs.google.com/spreadsheet/...=sharing#gid=0

P.S. Don't mind the values in there currently, they were only test values.

EDIT:
It will give you the 3D values you should start with that would be the same as the 2D ones. You can then copy/paste values into the excel document so you're just getting the values, then copy those into the 3D VE table.

Last edited by Architechnik; Feb 15, 2013 at 01:35 PM.
Reply
Old Feb 15, 2013 | 02:05 PM
  #582  
99EclipseGSX's Avatar
Evolving Member
 
Joined: May 2005
Posts: 100
Likes: 0
From: Santa Monica, CA
Originally Posted by Architechnik
Oh, I apologize; I think I was messing with the boost level items for the other MAP conversion inputs.

Have you copied over the cells from the 2d version of the tables into a multiplied out 3d version?
If you're not very savvy with spreadsheets, try downloading this and filling out the highlighted rows. If you don't have a value for a cell, get the average of the cells next to it.
https://docs.google.com/spreadsheet/...=sharing#gid=0

P.S. Don't mind the values in there currently, they were only test values.

EDIT:
It will give you the 3D values you should start with that would be the same as the 2D ones. You can then copy/paste values into the excel document so you're just getting the values, then copy those into the 3D VE table.
I did this at first actually, but later as a test I set the VE abnormally high (100% at anything above idle/light cruise) just as a test, and it's still way, way far off. The load calcuation is miles off. There's no way I should be calculating 115 load while I'm at 15 psi (MAP and boost gauge read this), 4000 RPM and 100% in the VE table, with all scalings selected correctly and set to 1:1. What else goes into the load calculation? All MAF settings are set to the factory values as they should be.

Beau
Reply
Old Feb 15, 2013 | 02:27 PM
  #583  
Architechnik's Avatar
Newbie
 
Joined: Dec 2012
Posts: 18
Likes: 0
From: Portland, OR
You said the sensor conversion is low in an earlier post, but the conversion I see is like this:
Code:
-43.1
-22.456
 -1.812
 18.831
 39.474
 60.539
163.758
370.616
While I'm running a 3-bar and the conversion shows this:
Code:
  0
 14.781
 29.531
 44.312
 59.062
 74.156
148
295.968
I'm guessing the end point at 370 is about right, while I'm not sure why there are negative numbers at the start. The 5 bar Kavlico shows the same sort of lower end negative conversion numbers.

The only thing I can think is that the Kavlico sensors only read from 2v - 5v or something along those lines. Does that sound about right?
Reply
Old Feb 15, 2013 | 02:52 PM
  #584  
99EclipseGSX's Avatar
Evolving Member
 
Joined: May 2005
Posts: 100
Likes: 0
From: Santa Monica, CA
Originally Posted by Architechnik
You said the sensor conversion is low in an earlier post, but the conversion I see is like this:
Code:
-43.1
-22.456
 -1.812
 18.831
 39.474
 60.539
163.758
370.616
While I'm running a 3-bar and the conversion shows this:
Code:
  0
 14.781
 29.531
 44.312
 59.062
 74.156
148
295.968
I'm guessing the end point at 370 is about right, while I'm not sure why there are negative numbers at the start. The 5 bar Kavlico shows the same sort of lower end negative conversion numbers.

The only thing I can think is that the Kavlico sensors only read from 2v - 5v or something along those lines. Does that sound about right?
The AEM/Kavlico/SSI are all the same and read from 0.5-4.5v.

I noticed similar numbers initially but I changed them all from my own values from 0 to 500, but as long as they're 1:1, this shouldn't be an issue..
Reply
Old Feb 15, 2013 | 03:15 PM
  #585  
10isace's Avatar
Evolved Member
15 Year Member
Liked
Loved
Community Favorite
iTrader: (12)
 
Joined: Nov 2008
Posts: 1,058
Likes: 16
From: East of the Rockies
So does the SD Map sensor ve and calibration look right? If I try to change it to a 1:1 then Map Sensor KPA changes.

Name:  SD20Mapsesorkpa_zps927d0117.png
Views: 0
Size:  157.7 KB

Last edited by 10isace; Feb 15, 2013 at 03:25 PM.
Reply



All times are GMT -7. The time now is 06:29 AM.