2B is actually injector pulse width again.

I also wondered about running speed density. I wondered how it would run if the MAF was unplugged - I only did it once by mistake, saw the CEL and plugged it in again.

 

2JCSBANKS:
I'm almost sure u'll NEVER find 2 bytes LOAD value as a high byte and low byte. Or they'll have to be send as 2 bytes for 1 request.

Imgine -U have a 2 bytes value of load - It take time to process a request - so if we can request the low and the high value of the load- we'll see high byte of the load value - the second request gives us a low bute of the load value - but LOAD changes between requests.

Imagine U are cruising and it changes around 250-260 value. U made the low byte request and it gave u 254 value. While making second request load changed to 257- so the high byte will be 1 and low byte will be 1. U asked high byte and received 1 - and your total load is 254+1*256= 510 ! value...And vice versa.

The ecu doesn't use floating points it it's calculations-it uses uint8 and uint16-so the load we are seeking is from 32 to 960 calibrated originally in your ecu .
If u'll 1 byte value find value changing from 128 to 226-than it is AFR which is used in our load tables.

Btw- why do we think that injector PW is not a load. I tried to change the injector scalings by 40 % but didn't noticed changes in so called injector pulse width - even at high rpm -where closed loop is disabled.

 

Good point on the split bytes.

I've not found the AFR value so far up to 7F. Next time I log I'll go up to 9F. I believe A0 on is AYC/ABS, but the values didn't work out right on previous attempts. F0 is actuator, Cx is involved in resetting DTCs.

IPW could indeed be the load, but I think I have seen it changing, but then I was probably also changing the load...

 

My IPW is changing on the same boost level - the one that says "old" has an original richer fuel map and seems to run higher IPW.

http://john824.fotopic.net/p31039777.html

http://john824.fotopic.net/p31038755.html

Also, changing my lean spool disable makes it leaner at 7000 RPM+ with the same boost and timing. It shows lower IPW up there and leaner on NBO2 and WBO2.

 

FOUND !!!!
AFR most likely= 0x33 (or 0x32) My log is a bit spoiled so its difficult ot read request headers-but the value is found...
Very easy to check-i'll do it on evo8 and evo7 ecus once i'll come to my car.

 

&H32 would fit on mine. Spookily it is a bit leaner than the map readings.. 11.2 rather than 10.2, but this is where lean spool is active on mine which leans it out by 1 AFR so it would appear to fit.

I wonder about now working out load from AFR target and IPW, but we have the dead time in the IPW calculation to contend with...

 

&H58 looks similar. And &H35 might still be a load candidate...

 

Enter EcuFlash ->high octane fuiel MAP->Edit->Edit MAP definition->Put UINF8 Instead of afr and uint 16 instead of load. AFR raw value should match the map exactly. From idle to 5000 on a pressed clutch (stopped car) U should get 128 if u use the same flash as evo9 jdm.

 

My AFR raw value doesn't match under boost - but as I say lean spool is in operation. Where it is not it seems to match. On mine, lean spool actually modifies the AFR value, so I think the value you've found is after this modification but before any others.

&H35 seems to clip at 161 unfortunately.

 

This is an indirect way of tracking the load zone we're in of course - but only if the values are different in different zones!

 

No-of course we wont track load like this! We just found the value.

Strange - in my logs i see at high load h35=188
H35 can be MAF/RPM raw value.. The number of MAF pulses happened at last 5 (or so) teeth of flyweel. I think ECU has few maf counters - 1 is checked every ~158ms and gives us airflow (HZ) .the second is cheched every N flyweel teeth passws - it gives ECU airflow/RPM value. Its more close to load - but have to be rescaled using MAF scale table.

 

MAF HZ value is more laggy than raw MAF/RPM value -so if you'll try to make the the corellation-it will be a straight line only if u increase the rpm slowly.

 

Can't find load and I give up!

But I think I've found vehicle speed at &H2F

 

I'll have to verify the vehicle speed. That's actually a lot more important than most people think.

Short of having a dyno, vehicle speed is the best parameter to tune with to see if your tuning is making more power. I remember in my Eclipse, I would do third gear pulls and log how long it took to get from 60-90mph. I'm not sure if the Evo ratios would allow 60-90, but maybe a 50-80 or 60-80, etc.

Keep tuning until you can get the fastest time without any knock. I think I remember starting at like 4.7s 60-90 on my Eclipse and just with tuning got to 3.5s 60-90.

Thanks,
Eric

 

I've already been doing this but with RPM on a known section of road in a particular gear. See the MIVEC thread.