For your LTFT low....make sure your idle Hz is lower than the minimum crossover point in the "Closed Loop LT Trim Control" table. Big cams, turbo upgrades, intakes, and even filters all affect the idle Hz.

My settings....
Low -> Mid = 62.50
Mid -> Low = 56.25

My car idles between 38 and 50 Hz depending on the air temperature.

As for the boost error....I cannot find any variances between the stock and v7 xml. The issue could be with EvoScan itself.

 

Looking at the tables posted above it appears that in stock form the last value in the "RPM/Speed to Gear" table that corresponds to the 0 gear = 2900 should match the "Boost control High/low Gear Range crossover" = 2900.

You changed that boost crossover to 75. This might....I repeat, might, cause a small problem.

 

As far as I know, Boost control high/low gear range crossover is "kind of" gear dependent boost control switch (low and high gear) from factory.

That determines which desired load, base WGDC, and WGDC correction interval is to be used.

But since we are using Tephra's gear based boost control, I guess that value now only determines which WGDC correction interval is to be used. If the value remains stock (2900 rpm/MPH), we always use the high gear range value.

 

So what determines the low or the high gear since we have 5 and 6 speed cars?

 

Factory RPM/SPEED value is more like a switching point for low/high gear, so stock value 2900 RPM/MPH means all gears are high gears regardless of 5 or 6 speed.

For example, for 03 EVO VIII with 5 speed gears, if you set it to 102 RPM/MPH which is about the half of 88 (for 3rd gear) and 117 (for 2nd gear) RPM/MPH, then 1,2 are low gears, and 3,4,5 are high gears.

 

Yes that is the way I understand it and the way (I think.....LOL) I have been using it.

I wonder if the gear based boost not working has anything to do with my blown head gasket? Not complaining as I realize that playing with modded roms has its dangers...sure has been a pita though.

 

The crossover table is the switch for where you wish the change to happen. 75 represents that I want it between 3rd and 4th. The 2D table being reffered here shows the rpm/mph per gear. That 2900 that is in the 2D table looks odd to me too? I remember 2900 being the disable/enable.

This is the way I understood it as well. Hence why I utilized a 75 there. I want 1-3 gears as low, and 4,5 to be high gears.

But before I can test my gears, etc, I need boost control. With a negative number always in the boost error, the WGDC Correction table always trys to increase the WG, hence increasing boost. This doesn't allow for any accurate reading of boost, AFR, etc, to fine tune. At that point gear based boost is mute...

I am idling at 31hz fluctuating up to 37.
My settings I believe are stock:
Low->Mid = 56.25
Mid ->low = 43.75

Do you think I should scale my Maf here?

As to Evoscan, I verified that the lookups are looking in the right location. My Boost error reading in evoscan is -3.087, never changing. I beleive that the pointer it looks at in the ecu is not being changed within the ECU. Very weird if you ask me.

EDIT:One other thing Paul, is your load error working? Also where does the ECU get its information for load/boost error? Not the memory address, but the actual reading that the ECU bases it's calculation off of?

 

no you won't need to do any scaling. My HKS suction pipe causes all kinds of MAF issues.

I haven't logged boost error. Sorry I can't help.


Have you tried an older version of EvoScan???

 

I have tried both V.2.5 and 2.7 on the EVO Scan. I can switch back to my V5 map with either evoscan's and both the boost error/load error tables read right and work in evoscan. It is something in the 94170715.v7 rom.

I would challenge anyone to look at there logs, and check the boost error/load error during full boost or spool up. You should find that boost/load error never changes. Your turbo is only being controlled by the Max Upward WGDC Vs. TPS table. If you change this table to reflect your max boost error correction or load error correction, you will see that you will over boost, due to the tables always thinking there is that much boost error. With my TPS VS WGDC table only allowing for max 2 upward, I am able to keep the WGDC from increasing to far and blowing well past the 21 psi I have loaded in my boost tables. This is vital to controlling boost. If not, you will spool and spool and spool, and the wastegate will keep compensating upward...

EDIT:My only conclusion after looking and verifying the .xml's is that the memory address reffered to in the MUT8A location is not being passed data from where ever the ecu gets its source for boost error. Since load error utilizes this same location with a different memory address, it too is not being passed the info to recoginze load error. I then think where these two addresses get there source... if it is from the boost tables themselves, etc? If so, then current creation for the new boost tables and wgdc tables might not be linking or reporting the boost/load error over to the MUT8A location.

 

Has anyone had any luck contacting mrfred?

 

Good observation. If I have disassembly tool (and skills), I would examine if boost control loop gets the right target value from our new gear-dependent-boost (or load) table and compare it with the current boost (or load). Since the current value should be the same as the stock setting, I think we should confirm that ECU gets the right target value from newly-defined maps.

Can anyone please check this out?

EDIT: It seems I can get only IDA Pro 4.9 version free, and it does not include a Hitachi CPU model, is there any way I can add-on this to 4.9 version?

 

the new wgdc+bdel maps dont use boost error.

only the TBEC table uses boost error and I dont touch that table...

 

I guess I was not very clear about my previous post or I do not understand how the boost contorl loop works.

I agree TBEC table uses boost error, but what I would like to check is how this boost error is calculated.

In order to calculate the boost error, we need two variables.

Target boost = which is specificed at "Boost Desired Engine 'PSI'" map at 94170715 V7 rom (RPM and gear based)

In comparison, "Base Boost #1 (High gear range)", and "Base Boost #4 (Low gear range)" (RPM based, #1 and #4 is chosen by RPM/MPH crossover value) at 94170008 rom.

Current boost = which is from MAP sensor value.

Boost Error = Target boost value at current RPM and gear - Current boost

Therefore, to get correct boost error value, we should get both variables right.

Since we use a new boost target table at V7 rom, I guess it needs to be reflected to the boost error calculation.

If boost error correction still looks at the same boost target value, I think we can get these strange boost errors.

 

ok i am just looking at the code again

the only thing I can think of is the code is skipping the BDEL - MAP = boost error calculation...

 

can someone log 0xFFFF8CCC and 0xFFFF8CCD?