Data Logging Results

[PhilT]

OK EVERYONE! First question:

 

What do I need to do data logging? What software/hardware how much does it costs and where do I get it? AP from Cobb part of that?

 

http://legacygt.com/forums/showthread.php?t=20090

 

http://www.tari.co.za/cgi-bin/yabb2/YaBB.pl

[edmundu]

Again, thanks, edmundu! I am attaching a log I took yesterday after an exhaust modification. If I interpreted your input correctly this latest log reflects increased horsepower. Using the time line the 0-60 time dropped from around 5.8 to 5.0, which confirms my driving impressions. Braketorqueing it now produces multiple tire protests This log includes the air sensor voltage you mentioned. Your input is valued. Thanks.

 

The MAFv are good at 4.52v. Your timing is good too, and if you had a little more octane there looks to be room in the tune to accommodate it. Just remember that there is 200ms(0.2 sec) between data points, so it could be closer, but it still is clearly showing gains.

 

Your memory addresses for the IAM and Load are incorrect. For the IAM you should see values no higher than 1.0. For load they should be in the 3.xx range or lower. What hex address are you using? If you search, I put mine up, and I think Kgreb has slightly different ones. This is due to different firmwares being used by Subaru. No biggie, but it does give additional insight into confirming the ECM is happy.

[SeeeeeYa]

Thanks once again. However, over the last several days I have tried every FFxxxx address in EcuExplorer under User-Defined Data Items without success. It reports my ECU as AJ18A(6824) whatever that means. There are other addresses such as (80224A) for example but everyone has said it is an FFxxxx address. Engine Load also eludes me. HELP! Anyone!

[KGreb]

Note this link for the proper addresses: http://www.legacygt.com/forums/showpost.php?p=366904&postcount=41

 

Also, select only the two user-defined data elements that are relevant for your car. Selecting too much data with the default agressive serial port parameters can cause problems. If you run into a lot of corrupted data, change the serial port parameters in EcuExplorer to something less agressive.

[SeeeeeYa]

Thanks KGreb, I will try this again. After no success earlier I searched and found the referenced post, tried the included addresses but none worked. It was after that I began at the bottom and started up, trying them all. I'll try these again, could have been me.

 

I have found that logging the WRONG address can introduce data that will ruin a logfile.

[SeeeeeYa]

Today I tried logging IAM and engine load once again. First I used addresses that KGreb and edmundu suggested. The engine load eventually seemed to work (figures in the 2.xx range) after unpairing it with wrong IAM addresses??!! The address is FF50B0, for engine load in g. However, neither the IAM addresses suggested, nor ANY of those that appear in the EcuExplorer window for selection (I did them all for the 2nd time, 26 in all) worked. I am stumped. I've tried logging a minimum of parameters, no change. I stopped, then started the application - between each address- and selected it anew. Everything else I have tried to log, and as many as I have yet chosen, works fine. One other thing, I never got the load to work until a) I didn't have it chosen at the same time as a WRONG IAM address, and b) it had to be last in order, i.e., if I checked the engine load box after the iam it appeared next to last and didn't work. I know it's confusing. I am open to suggestions.

[edmundu]

Have you tried posting on the Tari board? Calvin, the software creator, might be able to give additional pointer's. So of all the different addresses, what were some of you number's on them? I'm surprised that the 2 given by me and Kgreb didn't work...

[SeeeeeYa]

I'm surprised, too. But one of them worked, finally, the engine load one. Some numbers were 255, strings of numbers, pound signs, etc. all unchanging. It was all so time consuming and complicated while sitting in a running car between logging attempts.

 

I've considered contacting Calvin, now I will. I'm grateful for the free software and all that it has brought me/us. I just don't like to comment until I have my facts in order.

 

Thanks again. History says it could still be me and the answer simple. Meanwhile I am enjoying this. Incidentally, what is engine load g produced by? I think someone said it was calculated from a couple of other values. All the values I have seen are like .22 at idle to 2.8x during a run. It WAS great to finally see that work

[edmundu]

SeeeeeYa,

 

Yeah at idle you'll see small values. Under WOT you can see loads into the 3.xx range as well. This is basically airflow. The higher the airflow the higher the load. This is not a hard number(like the MAf volts). So you can manipulate the MAF Volts curve to rescale it. So that one tune says

Tune A:4.5v=250grams/s

Tune B:4.5v=275grams/s

 

The actual airflow between them is the same, but based on the Grams/s curve, it will have the ecu believe there is a higher load. Hence the ECU will use a different load site, which influences what AFR targets are plus timing/fuel as well.

 

Piggybacks like UTEC and others do this as an intercept or trick the ECU, whereas Protune and Ecutek have the ability to directly modify/rescale the MAF table curves. This is all done so that you can accurately portray the given conditions to the ecu, so it can make the best decision on what values to use.

[SeeeeeYa]

edmundu,

 

Being a glutton for punishment, can you direct me to some literature/website that will enable me to understand the schematic architecture of the ECU? Some of the things you tell me make my brain itch. I'm pretty computer/electronics literate, but no programmer.

 

Having now logged, used it to evaluate and compare modifications, maps, tunes, etc. has changed my relationship with cars forever. No way but forward. Thanks.

[edmundu]

edmundu,

 

Being a glutton for punishment, can you direct me to some literature/website that will enable me to understand the schematic architecture of the ECU? Some of the things you tell me make my brain itch. I'm pretty computer/electronics literate, but no programmer.

 

Having now logged, used it to evaluate and compare modifications, maps, tunes, etc. has changed my relationship with cars forever. No way but forward. Thanks.

 

 

Hahaha, this stuff sucked me right in as well. I don't have any specific locations, but, alot of info was cultivated over years of screwing with project cars of mine. Thinking back, I could have figured out some of my past tuning issues sooo much easier had I had the knowledge I do today! As for some more Subaru ECU help, check Accessecu.com, Ecutek.com.

[sandwood]

OK, so this is my first time, so bear with me. This is First gear, into second, then a little way into third. What sort of values am I looking at.... Seems like logging intake temps is not very useful when it is so cold out? knock correction looks good. comments? Can someone remind me the most useful things to log? TIA.

 

RPM I.temp. Timing KC Volt MAP(PSI)

2585 29.00 40 4 2.680 12.769

2781 29.00 37 4 2.820 13.929

2896 29.00 36 4 2.860 14.220

3126 29.00 37 3 2.920 14.365

3266 29.00 35 4 3.100 15.380

3507 29.00 33 4 3.200 15.961

3766 29.00 30 3 3.380 16.831

3972 29.00 28 4 3.520 17.847

4161 29.00 18 3 3.740 19.008

4545 29.00 17 3 3.940 20.604

4757 29.00 17 3 4.000 21.620

5059 29.00 21 3 3.980 21.329

5379 29.00 21 2 4.060 21.910

5603 29.00 22 2 4.000 20.314

5860 29.00 30 2 3.760 16.976

5840 29.00 45 0 3.020 10.592

5880 29.00 38 0 1.660 4.788

5697 29.00 32 0 1.540 3.627

5401 29.00 27 0 2.740 3.773

 

3918 29.00 14 5 3.920 24.086

4307 29.00 16 6 4.140 26.118

4353 29.00 18 7 4.260 28.149

4552 29.00 19 7 4.420 30.325

4808 29.00 22 9 4.660 32.937

5029 29.00 21 7 4.660 34.824

5319 28.00 21 6 4.640 33.663

5502 28.00 21 4 4.640 32.792

5713 28.00 21 3 4.700 32.502

5931 28.00 21 3 4.720 32.357

6129 27.00 22 2 4.740 32.357

6312 27.00 22 1 4.740 31.922

6480 27.00 22 0 4.740 31.631

6677 26.00 23 0 4.720 31.486

6853 26.00 23 0 4.700 29.890

 

6465 26.00 17 0 4.020 8.125

6123 26.00 19 0 2.140 5.514

5814 25.00 19 0 2.520 4.353

5571 25.00 27 -1 3.700 12.478

 

4838 25.00 19 3 4.120 20.894

4894 25.00 -4 9 4.400 25.537

5013 25.00 23 9 4.580 30.035

5118 25.00 21 7 4.720 34.243

5194 25.00 20 5 4.660 34.678

5334 25.00 20 4 4.640 33.663

5455 25.00 19 3 4.680 32.792

5578 25.00 19 2 4.680 32.647

5665 25.00 19 1 4.700 32.937

5746 25.00 19 1 4.740 32.792

5775 24.00 19 1 4.380 27.278

5813 24.00 35 0 3.560 5.949

5527 24.00 30 0 2.700 4.933

5287 24.00 22 0 3.200 4.063

4966 24.00 19 0 1.920 4.063

4679 24.00 19 0 2.380 3.773

4386 24.00 19 0 2.280 3.482

4144 24.00 19 0 2.500 3.337

[edmundu]

Useful Parameter's:

 

Rpm

MAP

Baro Psi

Fuel Inj Duty

Ign Timing

KC

Load

IAM

Wastegate Duty

IAT

Throttle Opening

Maf Volts

Relative Manifold PSI - (easier to read boost, rather than subtracting Absolute from Barometric)

 

Log looks good Sandwood. Strong timing, holding steady til redline. Mid-upper teens thru trq peak then riding into the mid-upper 20's. Nice MAf V, definitely flowing some air there! +KC values, although trailing off up top, it looks like there is some room to use better fuel, and have some benefit from it.

[sandwood]

now trying this converted to pdf with a few more values...

 

 

 

relevant info: 93 octane BP (not my regular gas), around 32 degrees, 20g with FMIC (showing about 20 psi peak), K&N intake, custom AP tune....

log 2-9-06.pdf

[TDC Tuning]

I think that your car has alot more power hidden away in that tune

[sandwood]

Yes, I think that might be true. Since tuning it, I took out a 2.5" neck/restriction in the CObb DP. Also, I specifically asked for a very safe tune (from Agile). I am eager to take another crack at it, and to get a dyno run with the cutout open wide....

 

how do you read wastegate duty cycles again? (what do they mean etc.)

[edmundu]

Yes, I think that might be true. Since tuning it, I took out a 2.5" neck/restriction in the CObb DP. Also, I specifically asked for a very safe tune (from Agile). I am eager to take another crack at it, and to get a dyno run with the cutout open wide....

 

how do you read wastegate duty cycles again? (what do they mean etc.)

 

It looks like the boost is tapered off up top, so their is room for more there, and if you asked for a safe tune, it is likely running into the 10's on AFR. You can safely lean it out some more, being that you have a fmic an all...

 

Wastegate Duty: The higher the % duty, the more the air is diverted away from the wastegate actuator. Thereby keeping the wastegate closed longer.

 

IF the duty % is lower, it allows more boost to push on the actuator. Thereby blowing open the wastegate, and controlling the turbine speed which is how boost is controlled.

[mccorry]

Well.... I finally got around to datalogging my Cobb Stg. 2 flash.

The car has a catless up pipe, Cobb catted downpipe, and is running Cobb Stg 2 91 oct.

 

Can you guys take a look at these and make sure everything looks alright? All 4 were done in 3rd gear. Pull 1 is 3rd into 4th... while pull 2 is only a partial pull..... I had to let out for traffic.

 

 

---------------------

deleted datalogs... reposted just below......

[mccorry]

OK... did the pulls yesterday with Stg 2 and the stock air box....

Then today... made a few more with Stg. 2 and the K&N. Only other change was the ambient temperature (it is much warmer today).....

 

I didn't get even 1 - KC value...... Maybe the KnN Typhoon isn't that bad to run with Cobb Stage 2.

 

Please let me know what you guys think....

Stage 2 pull 1.pdf

Stage 2 pull 2.pdf

Stage 2 pull 3.pdf

Stage 2 w KnN 1.pdf

Stage 2 w KnN 2.pdf

Stage 2 w KnN 3.pdf

Stage 2 w KnN 4.pdf

[mccorry]

Oh... and all pulls were approx 1 minute apart (both days....)

[Mblock66]

I am not sure the knock is the only thing you should be concerned about. You can't get the correct AFR's with the software so you could still be running like super super lean without knowing it

[PhilT]

From the logs, the first thing i see is that your MAF is reduced considerably since fitting the K&N, but your boost pressure is up. This indicates that the MAF sensor isn't reading correctly, presumably de to the turbulence in the air, exactly the same as what I experienced. The boost is still much lower than I would have expected to see with Stg 2, 18 psi is the normal figure I think.

As Matt points out, the stock AFR sensor is maxed out at 11.025, but I'm not sure that will give you a reading at all once you go open loop. It's possible you could be very lean, but as you have all positive KC I don't think that's the case.

If you still have your EGT sensor in place it would be a great help to see what temps you are getting.

I think you could extract a lot more power with a tune, although I don't think you will be harming anything by leaving the K&N on, if only you could see your EGT's to be sure.

If you have any intention of a Protune in the future, I would remove it, and put it back on when you get tuned, but that's just my €0.02

[Jon in CT]

I understand why people attach their logs as .pdf files. But, please people, don't convert your .csv files to real PDF documents before attaching them. It is impossible to do any serious analysis of the data in a spreadsheet or graphics program when the data are in a PDF document.

 

Simply rename your .csv files to .pdf prior to attaching and include the following as the last line in your post:

The following are .csv files.

[Beanboy]

Attached are two logs in .csv format... One in 2nd and one in 3rd. Also attached "real" pdf versions. Curious why wastegate duty cycle isn't 100%? Also didn't log AFR...doh! Might give it another shot later in the week, but I'm getting Pro Tuned Monday, so no huge deal. Just wanted to give it a shot!

 

2nd gear was flat ground, so didn't hit full boost...

 

Everything look normal to folks?

2ndgear.pdf

thirdgear.pdf

2ndgear_real_pdf.pdf

thirdgear_real_pdf.pdf

[autoeng]

Hi all,

 

Just looking through this thread, but I notice a critical data channel that's missing in your logs: coolant temp. It is really hard to understand what's happening if you can't see if the coolant temp was the same from file to file. While MAF is a critical input for base maps, ECT is generally used to correct those base maps.

 

Likewise, since you're looking at MAF, I would also log intake cam position duty cycle and tumble valve DC as well, just for comparison.

 

Not trying to be critical, just trying to help you guys make the best comparisions possible.

 

autoeng

05 LGT 5MT