Sent a PM... not sure if you still need someone.

I have a 2006 IX GSR 5-speed with JDM MAP sensor. ECU 88590015.

Mods currently are...

- JDM MAP sensor
- Modified Jestr tune
- Full cat-less TBE
- K&N Panel drop-in

The Evo is my daily though, so I don't wanna hack it up too much. Let me know if I can help out in any way.

 

thats extremely intersting!!! great job on the finding!

 

Sorry I haven't updated this much. My fourth gear exploded a couple of day ago and I am waiting for the new transmission to come in.

So far, I've been able to tune WOT areas of the A/F map fairly successfully. The low load/boost portion is a little different. Once its done, it will look nothing like a stock map at all.

 

So Matt,

Is it likely that this mod will be possible with only a JDM MAP sensor ?

Mine's installed and working now, let me know if you need any testing. I have a version 15 ROM on my IX

 

cant wait to see the final results of this "project"...

BTW where can i find a JDM 3 bar map sensor? can some tell me or pm me? quite interested of acquiring the unit...

 

SC,

You can try it out while leaving the stock MAF in place and plugged in for BARO and IAT reference until I find the best way to replace those signals.

First I need some data logs from you.

Add this table to your current ROM .xml:

<table name="MUT table" category="Misc" address="3ee84" type="3D" level="1" scaling="Hex16">
<table name="X" type="Static X Axis" elements="32" scaling="Hex16">
<data>-</data>
<data>0</data>
<data>-</data>
<data>1</data>
<data>-</data>
<data>2</data>
<data>-</data>
<data>3</data>
<data>-</data>
<data>4</data>
<data>-</data>
<data>5</data>
<data>-</data>
<data>6</data>
<data>-</data>
<data>7</data>
<data>-</data>
<data>8</data>
<data>-</data>
<data>9</data>
<data>-</data>
<data>A</data>
<data>-</data>
<data>B</data>
<data>-</data>
<data>C</data>
<data>-</data>
<data>D</data>
<data>-</data>
<data>E</data>
<data>-</data>
<data>F</data>
</table>
<table name="Y" type="Static Y Axis" elements="30">
<data>MUT0X</data>
<data>MUT1X</data>
<data>MUT2X</data>
<data>MUT3X</data>
<data>MUT4X</data>
<data>MUT5X</data>
<data>MUT6X</data>
<data>MUT7X</data>
<data>MUT8X</data>
<data>MUT9X</data>
<data>MUTAX</data>
<data>MUTBX</data>
<data>MUTCX</data>
<data>MUTDX</data>
<data>MUTEX</data>
<data>MUTFX</data>
<data>MUT10X</data>
<data>MUT11X</data>
<data>MUT12X</data>
<data>MUT13X</data>
<data>MUT14X</data>
<data>MUT15X</data>
<data>MUT16X</data>
<data>MUT17X</data>
<data>MUT18X</data>
<data>MUT19X</data>
<data>MUT1AX</data>
<data>MUT1BX</data>
<data>MUT1CX</data>
<data>MUT1DX</data>
</table>
</table>


Change values in that table to match the ones outlined in red:


Use this as the requestID.xml or mitsulogger and take some logs of idle, closed loop driving/cruising and some WOT:

<MitsubishiLogger>
<config>test</config>
<Functions>
<!-- Custom Functions will go here, these will be available as functions and formulas
to be available to the Request Eval tag -->

</Functions>
<ECUItems>
<!-- If you use custom requests, don't disable or remove AFRMap or InjPulseWidth values
Please note that all keywords are case sensitive, and the Hex requestID values must be in
capital letters (6F instead of 6f, stupid minor bug I found - Fixed 3/14/07)
Response="2" (or some value of response) now a required parameter
-->

<Request LogReference="RPM" RequestID="21" Eval="int(31.25*x)" Unit="rpm" Logged="y" Response="2"/>
<Request LogReference="AirTemp" RequestID="3A" Eval="-0.00000003166*x^5+0.00001425*x^4-0.002490*x^3+0.2143*x^2-10.279*x+361.01" Unit="deg F" Logged="y" Response="2"/>
<Request LogReference="Baro" RequestID="15" Eval="x/2" Unit="kPa" Logged="y" Response="2" Notes="0.00486*x Bar"/>
<Request LogReference="MDPRAW" RequestID="38" Eval="x" Unit="Int" Logged="y" Response="2"/>
<Request LogReference="JDMMAP" RequestID="38" Eval="0.19347*x-14.6" Unit="psig" Logged="y" Response="2"/>
<Request LogReference="LOAD_hi" RequestID="00" Eval="x" Unit="Int" Logged="y" Response="2"/>
<Request LogReference="LOAD_Lo" RequestID="01" Eval="x" Unit="Int" Logged="y" Response="2"/>
<Request LogReference="MAFSOURCEMAIN_hi" RequestID="02" Eval="x" Unit="Int" Logged="n" Response="2"/>
<Request LogReference="MAFSOURCEMAIN_lo" RequestID="03" Eval="x" Unit="Int" Logged="n" Response="2"/>
<Request LogReference="TBAF_hi" RequestID="BB" Eval="x" Unit="Int" Logged="y" Response="2"/>
<Request LogReference="TBAF_lo" RequestID="BC" Eval="x" Unit="Int" Logged="y" Response="2"/>



<!-- Test Requests These look to be bitmapped for actuators or other flags
but are often reallocated for different load mods, Optional Unit parameters
"Hex" Results converted to Hex
"Raw" force bracketed Hex output of the response to the request (used only for regular requests)
"Int" is Integer, its basically the integer output of the Eval function
"Bin" is a binary representation of the integer value of the response -->

<Request LogReference="Request00-bin" RequestID="00" Eval="x" Unit="Bin" Logged="n" Response="3" Notes="Load Ignition"/>

<Request LogReference="Request00-hex" RequestID="00" Eval="x" Unit="Hex" Logged="n" Response="5" Notes="Load Ignition"/>
<Request LogReference="Request01-int" RequestID="01" Eval="x" Unit="Int" Logged="n" Response="3" Notes="Load2 AFR" />
<Request LogReference="Request02-raw" RequestID="02" Eval="x" Unit="Raw" Logged="n" Response="9" Notes="Load1 AFR" />

<!-- Place all Complex requests below this line

You can place custom requests anywhere in the file, however if you want to use values
from current samples and not from the previous pass, it should be placed after the most
recent calls that your using. All LogReference labels are available as variables for
evaluation, therefore be careful how you name them (no spaces and whatnot VBScript rules)

-->

<Request LogReference="Load" RequestID="FF" Eval="((256*LOAD_hi)+LOAD_lo)*10/32" Unit="INT" Logged="y" Response="2"/>
<Request LogReference="MAFSOURCEMAIN" RequestID="FF" Eval="((256*MAFSOURCEMAIN-hi)+MAFSOURCEMAIN_lo)" Unit="INT" Logged="n" Response="2"/>
<Request LogReference="2byteAirFlow" RequestID="FF" Eval="((256*TBAF_hi)+TBAF_lo)*6.29/64" Unit="INT" Logged="y" Response="2"/>

<!-- Below are the two currently available custom load calculations, Both make assumptions
about Injector Latency, one uses the 14v value, one is a pre-set value fudged by battery voltage
both are thrown off when you have values for injector voltage latency other than stock values
I will be replacing these formulas with a linear regression for InjectorLatency11v and InjectorLatency14v
Which will result in a much more accurate starting point that can be tweaked for MAF error
-->

<!-- <Request LogReference="Load" RequestID="FF" Eval="5*InjectorScale*(InjPulseWidth-InjectorLatency14v)/AFRMAP" Unit="Site" Logged="n" Response="2"/> -->

<!-- <Request LogReference="Load" RequestID="FF" Eval="5*InjectorScale*(InjPulseWidth-InjectorLatency14v)/AFRMAP" Unit="Site" Logged="n" Response="2"/> -->

<Request LogReference="Latency" RequestID="--" Eval="((Injectorlatency11v - Injectorlatency14v)/(11.72-14.06)* battery +(Injectorlatency11v - (((Injectorlatency11v - Injectorlatency14v)/(11.72 - 14.06 ))* 11.72 )))" Unit="" Logged="n" Response="2"/>
<Request LogReference="Load" RequestID="--" Eval="5*InjectorScale*(InjPulseWidth-Latency)/AFRMAP" Unit="Site" Logged="n" Response="2"/>

<!-- <Request LogReference="LoadAlt" RequestID="FF" Eval="5*InjectorScale*(InjPulseWidth-(-0.1026*[Battery]+1.8741))/AFRMAP" Unit="Site" Logged="n" Response="2"/> -->

<Request LogReference="LoadAlt" RequestID="--" Eval="5*InjectorScale*(InjPulseWidth-(-0.1026*Battery+1.8741))/AFRMAP" Unit="Site" Logged="n" Response="2" Notes="Evoscan Load"/>


</ECUItems>
</MitsubishiLogger>

 

I take it all this information will only work on a specific rom ID for the evo 9, does anyone have this info for an evo 8 rom id, preferrably 94170014, or 94170008 which is the original

 

This all comes down to the same question I asked years ago in the DSM community......

Why do you want to get rid of the MAF?

I got two answers.....

1. Ricers can VTA
2. Power freaks were outflowing stock MAF

For the majority of folks running less than 450whp the stock maf is great and doesn't have the driveability issues that running mafless tends to have.

 

Another reason why people need to ditch the MAF is because it becomes an intake restriction when you start to move A LOT of air like with bigger turbos.

I almost sucked my screen thru the MAF into the turbo with a 35r on only 20psi.





With some tuning to smooth things out I've been running with no screen in my MAF for about a year now with no problems.

sorry for the big pic.

 

Always in boost is the power feeling the same. Well the reason I am interested in a map over maf setup is because in every scenario I have seen on this site, a switch to speed density = more hp at the same boost level. And if you are trying to be efficient as possible meaning less boost more power, why not switch to map if it will yield an increase in power.

 

I'd like to be able to drive the car safely even if I blew IC piping or blew my turbo... with the stock MAF sensor, this isn't possible.

 

I don't really know how to quanitify if the power is the same. shortly after I pulled the screen in my MAF, over a year ago, I pulled the 35r off while trying to diagnose another problem I was having.

 

Matt,

I've made the requested changes to my ROM & Mitsulogger, and have recorded some logs as you requested. The logs are attached to this post.

Please let me know what else I can do for you.

 

Sweet. What I will do is try to match the VE table as close as I can to your current setup. Then you can run the patch with only minor fine tuning. You'll have to leave the MAF housing plugged in and in place until I am done testing other input methods.

 

UPDATE:

Car is back on the road with a new 5-speed transmission. While it was down, I pulled the intake manifold off and drilled out the bung for the IAT sensor. I'll be wiring the GM IAT sensor directly to the ECU and compensating for any temp differences in the ECU. As for the baro, I think I am going to try to use the stock MDP sensor and rescale it to match the typical voltage output of the stock baro sensor. Hopefully I'll get a lot of testing done this weekend.