Speed Density 2.0 (3D VE Tables, Baro)
Dec 24, 2012, 12:23 PM
#526
I'm transitioning from a fully developed MAF tune to 3D SD. This means that my AFR, boost, timing, injector size and latency, MAF Size and Comp are all perfect for my current build. Most of the past two months in fact was spent tuning MAF Comp from idle to cruise, to get STFT under control. Let me know if I can help you here.
I have good MAF load vs kPa data from evoscan so I can set the starting point for 3D VE. There are a few of important questions that I have not found answers to:
1. Should kPa2Load be left at 1:1 if using a 3D VE map built from my MAF tune? (I would have thought that this could keep AFRs too lean and timing too advanced at full boost, since my actual load (310) exceeds kPa (280). )
2. Should MAF Scaling and MAF Compensation tables be copied over from to 3D SD tune?
3. Should kPa2Load be set based on data from MAF tune (see below for my setup - big cams)?
I have good MAF load vs kPa data from evoscan so I can set the starting point for 3D VE. There are a few of important questions that I have not found answers to:
1. Should kPa2Load be left at 1:1 if using a 3D VE map built from my MAF tune? (I would have thought that this could keep AFRs too lean and timing too advanced at full boost, since my actual load (310) exceeds kPa (280). )
2. Should MAF Scaling and MAF Compensation tables be copied over from to 3D SD tune?
3. Should kPa2Load be set based on data from MAF tune (see below for my setup - big cams)?
I created a mockup table just to show how your settings actually created a VE curve. Then I placed this setting in a 3D table so you can see how it would be affected by varying RPM ranges. This assumed the 2D VE table was all set to 100%. IF you have less than 100% in the 2D VE table, you are making things even leaner.
Remember, VE= LOAD/KPA*100
The problem is you have only created a VE curve for the LOAD axis and not the RPM axis. You can effectively use the standard 2D VE table to help simulate a 3D VE Table but it is kind of pointless when they have already created a 3D VE patch these days.
-Jamie
Last edited by Dynotech Tuning; Dec 24, 2012 at 12:26 PM.
Dec 24, 2012, 12:49 PM
#527
Good way if explaining it . 3d surely makes more intuition out of it.
Could you reasonably substitute your map ve for rpm ve in your 3d table and make the same point? I think yes but I'm looking at my gutted maf here thinking it would be nice to give it a shot anyways.
Could you reasonably substitute your map ve for rpm ve in your 3d table and make the same point? I think yes but I'm looking at my gutted maf here thinking it would be nice to give it a shot anyways.
Dec 24, 2012, 01:05 PM
#528
Yes, you could represent the 2D VE table in the 3D VE Table in the same manner. Simply fill in the entire rpm column with whatever value you have in your 2D VE table. The kpa2load and VE table are effectively both multipliers of VE. So if you created a VE of 57.8%@61kpa and in your VE table you had a VE of 81%@2000rpm, your result VE at 61kpa & 2000rpm would be:
Kpa2loadVE*2DVE=total VE
57.8%*81%= 46.8% total VE at 61kpa & 2000rpm
In a nutshell, it's so complicated to truly see how you are simulating a 3D VE table by using the two tables in the 2D VE patch that it makes so much more sense to just use the 3D Ve patch and save yourself a world of mathematical headaches.
-Jamie
Kpa2loadVE*2DVE=total VE
57.8%*81%= 46.8% total VE at 61kpa & 2000rpm
In a nutshell, it's so complicated to truly see how you are simulating a 3D VE table by using the two tables in the 2D VE patch that it makes so much more sense to just use the 3D Ve patch and save yourself a world of mathematical headaches.
-Jamie
Last edited by Dynotech Tuning; Dec 24, 2012 at 01:15 PM.
Dec 25, 2012, 11:15 PM
#529
What you effectively are doing is just another way to create a VE curve.
I created a mockup table just to show how your settings actually created a VE curve. Then I placed this setting in a 3D table so you can see how it would be affected by varying RPM ranges. This assumed the 2D VE table was all set to 100%. IF you have less than 100% in the 2D VE table, you are making things even leaner.
Remember, VE= LOAD/KPA*100
The problem is you have only created a VE curve for the LOAD axis and not the RPM axis. You can effectively use the standard 2D VE table to help simulate a 3D VE Table but it is kind of pointless when they have already created a 3D VE patch these days.
-Jamie
I created a mockup table just to show how your settings actually created a VE curve. Then I placed this setting in a 3D table so you can see how it would be affected by varying RPM ranges. This assumed the 2D VE table was all set to 100%. IF you have less than 100% in the 2D VE table, you are making things even leaner.
Remember, VE= LOAD/KPA*100
The problem is you have only created a VE curve for the LOAD axis and not the RPM axis. You can effectively use the standard 2D VE table to help simulate a 3D VE Table but it is kind of pointless when they have already created a 3D VE patch these days.
-Jamie
So far I've set up my 3D VE table by logging ([load1B]/[JDMMAPkPa])*100 by RPM in Evoscan. Fuel tables (High/Low Octane, Injector Scailng and Latency, MAF Scaling and MAF Comp) have all been copied over from 96530706 to 96534706. Ignition (High/Low) have also been copied over.
Resulting AFRs at open loop idle and cruise are where I want them around 14.7. However tip in and transition to boost at anything over 80 kPa goes 20-30% too lean. The question is should I increase values 20-30% in the 3D VE table around 60-80 kPa and 1500-2500 RPM to richen the transition, or should I bump loads 20-30% from 60-80 kPa in the kPa to Load table?
Thanks in advance for any advice you can give.
Dec 26, 2012, 05:54 AM
#530
With 3D SD, keep the kpa2load at 1:1 at all times and just use the 3D map. If the 60-80kpa areas are lean, add VE in that area of the VE table. Just be sure not to confuse load and KPA. Generally, you will see your VE rise from about 80% at 60-70kpa to about 90-93% around 100-120kpa.
-Jamie
-Jamie
Last edited by Dynotech Tuning; Dec 26, 2012 at 05:20 PM.
Dec 26, 2012, 04:06 PM
#531
With 3D SD, keep the kpa2load at 1:1 at all times Nd just use the 3D map. If the 60-80kpa areas are lean, add VE in that area of the VE table. Just be sure not to confuse load and KPA. Generally, you will see your VE rise from about 80% at 60-70kpa to about 90-93% around 100-120kpa.
-Jamie
-Jamie
Jan 31, 2013, 10:59 AM
#532
Evolving Member
Hey mates!
I am having troubles with 3d SD..
I have tried a lot of kpa2load combinations, played a lot with 3d VE table, but my car behaves the same way. It starts perfect, idles well, but when i touch the throttle and try to rev it it leans out misfires etc.
Even if I floor the pedal (gradually or rapidly), the map sensor (JDM Map) reads 95-105 KPA but the load goes to 70 but never higher than that, AFR leans out to 18-20 as soon as I push the throttle. If I hold WOT it keeps super-lean AFRs and missfires, generaly regardless of any 3d VE or KPA2LOAD changes.
It seems to me that my ROM may be corrupted, or I have downloaded something nonworking. I have tried downloading that rom from the 1st and 2nd posts of the "Speed Density 2.0" topic. I am using 96534706 as I own EDM 2004 Evo8. Evo9 JDM Map and GM IAT both mounted in the inlet manifold.
I have also tried the old 2d SD rom. Set it up in a couple of minutes, transfered my fuel and ignition maps, injector parameters from my finished MAF tune and it worked fine - fired right up, idled, revved. It holded some kind of reasonable AFR at idle revved fine without any harsh enleanment and generally felt smooth. I have even tried some all-around driving in 1-3 gears, it has pretty rich AFRs for cruise, but I haven't tuned it at all.
What else can i do or check? Could you give me a link to the known working version of 96534706 3dSD ROM?
I am having troubles with 3d SD..
I have tried a lot of kpa2load combinations, played a lot with 3d VE table, but my car behaves the same way. It starts perfect, idles well, but when i touch the throttle and try to rev it it leans out misfires etc.
Even if I floor the pedal (gradually or rapidly), the map sensor (JDM Map) reads 95-105 KPA but the load goes to 70 but never higher than that, AFR leans out to 18-20 as soon as I push the throttle. If I hold WOT it keeps super-lean AFRs and missfires, generaly regardless of any 3d VE or KPA2LOAD changes.
It seems to me that my ROM may be corrupted, or I have downloaded something nonworking. I have tried downloading that rom from the 1st and 2nd posts of the "Speed Density 2.0" topic. I am using 96534706 as I own EDM 2004 Evo8. Evo9 JDM Map and GM IAT both mounted in the inlet manifold.
I have also tried the old 2d SD rom. Set it up in a couple of minutes, transfered my fuel and ignition maps, injector parameters from my finished MAF tune and it worked fine - fired right up, idled, revved. It holded some kind of reasonable AFR at idle revved fine without any harsh enleanment and generally felt smooth. I have even tried some all-around driving in 1-3 gears, it has pretty rich AFRs for cruise, but I haven't tuned it at all.
What else can i do or check? Could you give me a link to the known working version of 96534706 3dSD ROM?
Jan 31, 2013, 02:04 PM
#533
What VE settings do you have in your VE Table between 90-120kpa where it goes lean? There is no need to mess with the kpa-to-load table with 3D SD, keep it at 1:1. Thats the whole point of having a 3D VE table. Just make sure you selected the proper KPA scaling in the KPA-toLOAD table for the MAP sensor you are using.
-Jamie
-Jamie
Jan 31, 2013, 02:49 PM
#534
Evolving Member
What VE settings do you have in your VE Table between 90-120kpa where it goes lean? There is no need to mess with the kpa-to-load table with 3D SD, keep it at 1:1. Thats the whole point of having a 3D VE table. Just make sure you selected the proper KPA scaling in the KPA-toLOAD table for the MAP sensor you are using.
-Jamie
-Jamie
I am using JDM Map from JDM Evo9. The formulas I am using in "kPa JDM3barMAP" scaling are "x/3" "x*3".
Idle cells are set to 67-70% VE to make AFR closer to 14.7, than a smooth transition to 90%+ after 3000rpm, and a smooth transition to 100 at 90kpa.
The whole columns of 90-120 load are set 100-107% VE respectively.
When I hit 98-105 kpa (logged via JDM MAP in evoscan and confirmed visually by a defi boost gauge) my 1byte Load hitting only 70 and not any higher.
Changing the kpa-2-load was just my desperate attempt to make it substantialy richer and hit higher load cells any way I could.
Feb 1, 2013, 05:38 AM
#536
Evolving Member
Yes i can select any map scaling if that help to sort things out.
Yes I can try 9417 for example. But is 9417 compatible with 9653? Should I make any specific changes because of map-sensor tables in 9417?
Feb 1, 2013, 08:13 AM
#538
https://www.evolutionm.net/forums/sh...079&styleid=27
Should be direct swap
Find someone with an e8 and your 9653 rom for a starter rom. Something with comparable cams would be best
If you still suffer the same issues then idk..
we'll finger it out
Should be direct swap
Find someone with an e8 and your 9653 rom for a starter rom. Something with comparable cams would be best
If you still suffer the same issues then idk..
we'll finger it out
Last edited by 211Ratsbud; Feb 1, 2013 at 08:16 AM.
Feb 8, 2013, 11:38 PM
#539
I understand the whole point of having a 3D VE table, thats why I am struggling so stuborn to make it work, and refusing to stick with 2d SD.
I am using JDM Map from JDM Evo9. The formulas I am using in "kPa JDM3barMAP" scaling are "x/3" "x*3".
Idle cells are set to 67-70% VE to make AFR closer to 14.7, than a smooth transition to 90%+ after 3000rpm, and a smooth transition to 100 at 90kpa.
The whole columns of 90-120 load are set 100-107% VE respectively.
When I hit 98-105 kpa (logged via JDM MAP in evoscan and confirmed visually by a defi boost gauge) my 1byte Load hitting only 70 and not any higher.
Changing the kpa-2-load was just my desperate attempt to make it substantialy richer and hit higher load cells any way I could.
I am using JDM Map from JDM Evo9. The formulas I am using in "kPa JDM3barMAP" scaling are "x/3" "x*3".
Idle cells are set to 67-70% VE to make AFR closer to 14.7, than a smooth transition to 90%+ after 3000rpm, and a smooth transition to 100 at 90kpa.
The whole columns of 90-120 load are set 100-107% VE respectively.
When I hit 98-105 kpa (logged via JDM MAP in evoscan and confirmed visually by a defi boost gauge) my 1byte Load hitting only 70 and not any higher.
Changing the kpa-2-load was just my desperate attempt to make it substantialy richer and hit higher load cells any way I could.
Hey BIGGY your scaling is correct for the JDM MAP sensor, but in your Map Sensor kPa table, the Map Volts (kPa) values are wrong.
They should look like this:
I made the same mistake.
The values change in the left hand column of this table when you first change the scaling. You have to make a copy of the values , then change the scaling, before finally pasting the original values back in.This should increase the kPa values read by the ECU and therefore calculated as Load during transition from idle to boost. Let us know how you go.
Last edited by CDrinkH2O; Feb 8, 2013 at 11:41 PM. Reason: typo
Feb 9, 2013, 12:28 AM
#540
MAF Mapped 3DVE - How's It Look?
Guys - Here's the 3DVE table I created as a starting point for 96534706 by mapping Load1B, kPa and RPM then calculating VE (Jamie's method), using my MAF tune on 96530706. (Mods in sig below).
My question is do the VE values look OK as a starting point for SD 2.0?