&H38 request ID in Evoscan. Multiply by 5/4 or 1.25.

Bear in mind that if you have compensations going on the load zone being accessed may not have the same timing number in that you see in your logs. However, on mine when the octane value is at 255 and the car is warm it fits like a glove. I don't believe there are any other candidates for load in the &H00-3F request ID range having logged them all.

For interest, I believe that &H04,05,06,33 all show timing. We've been using &H06 and subtracting decimal 20. &H05 seems to work without subtracting 20.

There appear to be 2 words in low-high byte order at 2A-2B and 2C-2D.

I'm not running factory boost control, but I expect somewhere in this range would also be boost control data - like wastegate duty etc.

Let us know how you get on.

 

Data.xml in zip file - see the octane value thread for more on how to use this file if you don't know.

 

Good info.. I'll have to try it later..

 

I did some logging with those two parimeters. The octain rating one seems to work as far as I can tell. 99% of the time it stays on 255. Once while doing a 4th gear pull up a hill it dropped to 251 and I got a little knock.

The load figure though I don't think is accurate, at least on my VIII. At WOT I get load readings of 306.75 - 307 which don't jive with where I am tracing my load, between 200 - 240, on my timing maps.

This is on an VIII though, I should have a IX to test it out on this weekend.

 

its possible the formula is off

 

I tested it tonight on my IX, and can tell you it is not the load value. Because when TPS is 100% the byte value is almost always maxed even running only 14-15lbs of boost. Also on idle it is ~120

 

I'll have another look through my logs tonight. On over-run it was about 20, just a bit higher on idle, higher still on cruise, and then followed the torque curve - at 270KPa it hit about 250% load and then rolled off through the revs. I also correlated it with the previous load formula with airflow/RPM*852 and it was reasonable until the airflow signal clipped when the "load" signal continued nicely as expected. Obviously it needs more work, but it was pretty convincing off my logs. I'm happy to be wrong though, we're all learning here!

 

A few more thoughts... when I covered up my timing column I could predict the timing pretty close from using my timing map, RPM and &H38 "load?" value. I'm wondering if this might be different on different ECUs for some odd reason. Mine is JM8857-08 JDM Evo IX GSR. I'll post more when I have my laptop this evening and hope to run a few more logs. I'll also post up my data.xml file that I used to log all 64 items from 00-3F which might help find patterns between the different ECUs. I could just be plain wrong though

 

Yet another idea - sorry, but best to share to get this done... my car has twin wastegate solenoids. If this value I'm logging on my car is "loadboost" as calibrated in ECUflash, then perhaps the address is shifted because the other values nearby might be wastegate duty cycles? It doesn't fit though in that Mitsi seem to usually put the same addresses across most Evos from different countries and years.

 

I'm with ya lets figure this out

I do notice on my logs the only time it would get to a low number ~50 was on a shift or deacceleration.

 

I wonder if you're logging wastegate duty cycle then at the same address? What other items would show this pattern I wonder?

My data.xml file just polls every location from 00-3F and puts them in columns with the same label, someone else needs to log the same I think on the other model variants. I'll post up the file in 5 hours when I'm home as it will save a bit of time (or a macro for you clever people, I just used cut and paste a lot and changed all the numbers LOL). When you have all 64 of them in a table you can get rid of the ones that don't move in a "loadish" manner and we can start to home in on the values?

I love this stuff

 

I wish I had more "daily drive" time on my car, I'm lucky if it comes out twice a week now and most of the time I'm too busy to get the logger set up with a laptop..

What your describing might actually be the map sensor, I'm hoping its the load calculation though, I'd really like to see useful numbers that correlated to cell data.. I do think some of this stuff is two bytes long and not one byte.. if thats the case it may be a shortcoming of this app since you may need to combine two items to produce one larger value..

The DSM ECU does clip at 1600hz also, but they have modified the code in the past to produce two-byte output.. I hope this is something already on the Evo ECU as we search through.. if you log the output, you might want to look for a value that starts to change by a small amount when you hit your clipping value, perhaps thats the other byte..

Its funny though, calculated load in regular OBD-II is unfortunately a max value of 100% because I think thats what the standard is.. I'm hoping that any load value we do find, represents the load cell axis values..

Perhaps we're looking at this the wrong way, and we should be looking for something that represents the "Column number" and not the actual load, this might be more universal and represent a more reasonable and universal value for diagnostics (map tracing)

If that was the case, then you would look for two values that are similar, since the Fuel map and Timing Map have different load axis data locations.

 

I will get some time in today, is it possible that this is a manifold pressure reading?

 

Its certainly possible.. if it gets pinned at anything over a vacuum it likely is..

 

has anyone confirmed the accuracy of the octain number yet?