csv log analyzer (EE and maybe others)

[steveo]

i do a lot of my tuning using a homebrew analyzer for log files in linux.

due to a few requests, i've ported it to work with dos/windows so other people can try it out.

would you like to test it out? let me know how it works!

i've set it up to work out-of-the-box with Scan9495 log files, but can also be configured to work with tunerpro (configs/instructions included).

i think datamaster logs can be used by importing them with scan9495 and saving them. i haven't tried yet.

http://fbodytech.com/files/aldl-analyzer-1.1b.zip

[steveo]

here's what it gives you;

BLM per-cell analyzer, useful for figuring out how to reallocate cell boundaries, or to figure out ranges of AFR within a cell.



**** BLM CELL vs TRIM, MAF, MAP, RPM RANGE/AVG ****
(Ignoring records with records < 20)

* Cell 0 (55 Hits)
BLM: 124 - 130 (Avg 127.3)
RPM: 875.0 - 1425.0 RPM (Avg 1078.6)
MAP: 25.5 - 33.6 KPA, (Avg 30.2)
MAF: 0.0 - 11.5 AFGS, (Avg 2.8)

* Cell 1 (168 Hits)
BLM: 118 - 131 (Avg 124.6)
RPM: 1150.0 - 1675.0 RPM (Avg 1444.6)
MAP: 20.0 - 33.2 KPA, (Avg 27.1)
MAF: 0.0 - 14.0 AFGS, (Avg 3.7)

* Cell 2 (3388 Hits)
BLM: 118 - 134 (Avg 126.1)
RPM: 1450.0 - 2550.0 RPM (Avg 1940.9)
MAP: 16.6 - 34.7 KPA, (Avg 27.5)
MAF: 0.0 - 27.5 AFGS, (Avg 8.6)

* Cell 3 (858 Hits)
BLM: 121 - 129 (Avg 124.6)
RPM: 2250.0 - 4325.0 RPM (Avg 2622.3)
MAP: 13.7 - 34.3 KPA, (Avg 26.0)
MAF: 0.0 - 48.3 AFGS, (Avg 7.1)

* Cell 4 (334 Hits)
BLM: 120 - 129 (Avg 125.6)
RPM: 825.0 - 1450.0 RPM (Avg 1178.1)
MAP: 30.6 - 52.8 KPA, (Avg 38.8)
MAF: 0.0 - 21.3 AFGS, (Avg 4.9)

* Cell 5 (632 Hits)
BLM: 116 - 129 (Avg 125.1)
RPM: 1175.0 - 1750.0 RPM (Avg 1526.5)
MAP: 27.3 - 52.8 KPA, (Avg 36.2)
MAF: 0.0 - 32.0 AFGS, (Avg 4.3)

* Cell 6 (1927 Hits)
BLM: 122 - 131 (Avg 126.5)
RPM: 1450.0 - 2550.0 RPM (Avg 2044.6)
MAP: 27.3 - 55.0 KPA, (Avg 36.6)
MAF: 0.0 - 47.9 AFGS, (Avg 11.9)

* Cell 7 (465 Hits)
BLM: 123 - 129 (Avg 126.2)
RPM: 2400.0 - 3800.0 RPM (Avg 2811.4)
MAP: 27.3 - 53.2 KPA, (Avg 37.7)
MAF: 0.0 - 57.5 AFGS, (Avg 11.2)

* Cell 8 (28 Hits)
BLM: 125 - 131 (Avg 128.3)
RPM: 900.0 - 1450.0 RPM (Avg 1201.8)
MAP: 48.7 - 71.2 KPA, (Avg 57.5)
MAF: 0.0 - 26.4 AFGS, (Avg 2.7)

* Cell 9 (77 Hits)
BLM: 126 - 132 (Avg 128.5)
RPM: 1200.0 - 1750.0 RPM (Avg 1518.8)
MAP: 47.3 - 68.7 KPA, (Avg 57.0)
MAF: 0.0 - 34.7 AFGS, (Avg 5.3)

* Cell 10 (185 Hits)
BLM: 126 - 133 (Avg 128.8)
RPM: 1525.0 - 2550.0 RPM (Avg 2080.3)
MAP: 47.6 - 72.0 KPA, (Avg 58.5)
MAF: 0.0 - 69.8 AFGS, (Avg 10.9)

* Cell 11 (157 Hits)
BLM: 125 - 129 (Avg 127.5)
RPM: 2425.0 - 3775.0 RPM (Avg 2952.1)
MAP: 46.9 - 73.1 KPA, (Avg 57.6)
MAF: 0.0 - 76.0 AFGS, (Avg 15.8)

* Cell 12 (0 Hits) - Not reliable.

* Cell 13 (0 Hits) - Not reliable.

* Cell 14 (26 Hits)
BLM: 127 - 129 (Avg 128.2)
RPM: 1800.0 - 2475.0 RPM (Avg 2170.2)
MAP: 68.3 - 94.1 KPA, (Avg 83.1)
MAF: 0.0 - 0.0 AFGS, (Avg 0.0)

* Cell 15 (31 Hits)
BLM: 126 - 128 (Avg 127.4)
RPM: 2475.0 - 3200.0 RPM (Avg 2796.8)
MAP: 67.2 - 86.0 KPA, (Avg 74.7)
MAF: 0.0 - 89.5 AFGS, (Avg 5.5)

* Cell 16 (1177 Hits)
BLM: 117 - 128 (Avg 120.1)
RPM: 650.0 - 2875.0 RPM (Avg 842.3)
MAP: 22.5 - 48.4 KPA, (Avg 36.0)
MAF: 0.0 - 462.0 AFGS, (Avg 1.1)

* Cell 17 (793 Hits)
BLM: 119 - 130 (Avg 124.6)
RPM: 875.0 - 4150.0 RPM (Avg 1378.6)
MAP: 13.3 - 34.7 KPA, (Avg 26.6)
MAF: 0.0 - 17.8 AFGS, (Avg 1.7)

* Cell 18 (1195 Hits)
BLM: 115 - 129 (Avg 120.9)
RPM: 725.0 - 2250.0 RPM (Avg 960.0)
MAP: 28.4 - 73.4 KPA, (Avg 37.4)
MAF: 0.0 - 32.8 AFGS, (Avg 1.9)

Overall useful BLM Average: 125.91 (0.984 percent)



narrowband VE trim analyzer (this tune is pretty damn good except at idle):




**** Narrowband Trim vs RPM vs MAP ****
(Ignoring cells with counts < 6)
0 10 20 30 40 50 60 70 80 90
0
.... .... .... .... .... .... .... .... .... ....
400
.... .... .... 119 .... .... .... .... .... ....
800
.... .... 127 121 126 127 .... .... .... ....
1200
.... .... 125 125 126 128 130 .... .... ....
1600
.... 125 126 126 127 128 129 .... .... ....
2000
.... 125 125 126 127 128 129 128 128 ....
2400
.... 125 125 126 127 128 128 128 .... ....
2800
.... 125 125 126 126 127 127 128 .... ....
3200
.... 125 125 127 127 128 128 .... .... ....
3600
.... 125 .... 125 127 .... .... .... .... ....
4000
.... .... .... .... .... .... .... .... .... ....
4400
.... .... .... .... .... .... .... .... .... ....
4800
.... .... .... .... .... .... .... .... .... ....
5200
.... .... .... .... .... .... .... .... .... ....
5600
.... .... .... .... .... .... .... .... .... ....
6000
.... .... .... .... .... .... .... .... .... ....



trim analyzer vs maf afgs




**** Narrowband Trim vs MAF AFGS ****

MAF AFGS TRIM MULT COUNT
0 - 5 124 0.971 7064
5 - 10 ---- ----- 0
10 - 15 128 1.000 852
15 - 20 127 0.994 1384
20 - 25 127 0.991 629
25 - 30 126 0.985 251
30 - 35 126 0.981 156
35 - 40 126 0.986 103
40 - 45 126 0.984 37
45 - 50 126 0.985 36
50 - 55 126 0.986 21
55 - 60 126 0.986 19
60 - 65 ---- ----- 6
65 - 70 127 0.992 12
70 - 75 127 0.991 11
75 - 80 ---- ----- 2
80 - 85 ---- ----- 0
85 - 90 ---- ----- 1
90 - 95 ---- ----- 0
95 - 100 ---- ----- 0
100 - 105 ---- ----- 0
105 - 110 ---- ----- 0
110 - 115 ---- ----- 0
115 - 120 ---- ----- 0
120 - 125 ---- ----- 0
125 - 130 ---- ----- 0
130 - 135 ---- ----- 0
135 - 140 ---- ----- 0
140 - 145 ---- ----- 0
145 - 150 ---- ----- 0



o2 volt average during power enrichment (so you can tell if you're going way too lean). it has a wideband mode too, if you run a wideband.




**** Narrowband MV (Average) vs RPM during PE ACTIVE ****
WARNING! This will never get you a good PE AFR,
it's just so you can tell if it's going very lean.

RPM 0 - 400 0.0 0 Counts
RPM 400 - 800 0.0 0 Counts
RPM 800 - 1200 0.0 0 Counts
RPM 1200 - 1600 0.0 0 Counts
RPM 1600 - 2000 0.0 0 Counts
RPM 2000 - 2400 942.0 2 Counts
RPM 2400 - 2800 919.6 4 Counts
RPM 2800 - 3200 931.7 13 Counts
RPM 3200 - 3600 929.2 32 Counts
RPM 3600 - 4000 925.5 36 Counts
RPM 4000 - 4400 918.0 25 Counts
RPM 4400 - 4800 923.9 18 Counts
RPM 4800 - 5200 921.4 4 Counts
RPM 5200 - 5600 930.5 2 Counts
RPM 5600 - 6000 0.0 0 Counts
RPM 6000 - 6400 0.0 0 Counts



knock events per cell. this counts '1' every time the knock count increments in an area (but doesn't care how much it increments)




**** Knock Increment Records vs RPM vs MAP ****
0 10 20 30 40 50 60 70 80 90
0
.... .... .... .... .... .... .... .... .... ....
400
.... .... .... .... .... .... .... .... .... ....
800
.... .... .... .... .... .... .... .... .... ....
1200
.... .... .... .... .... .... .... .... .... ....
1600
.... .... .... .... .... .... .... .... .... ....
2000
.... .... .... .... .... .... .... .... 1 ....
2400
.... .... .... .... .... .... .... .... .... 2
2800
.... .... .... 2 .... .... .... .... .... ....
3200
.... .... .... .... .... .... .... .... .... 1
3600
.... .... .... .... .... .... 1 .... .... ....
4000
.... .... .... .... .... .... .... 1 1 ....
4400
.... .... .... .... .... .... 1 1 1 1
4800
.... .... .... .... .... .... .... .... .... ....
5200
.... .... .... .... .... .... .... .... .... ....
5600
.... .... .... .... .... .... .... .... .... ....
6000
.... .... .... .... .... .... .... .... .... ....

[bybyc5]

Lots of new things to look at.
Downloading scan9495 and installing tonight. Looks like an easy program to get to know. And since your log analyzer is set up for these logs without conversion, this is the way to go for me. I'll do some data logging and see what I can get figured out.

I hope one of the cables I have is compatible with scan9495. I have an old Andrew Mattei serial to ALDL and one with Craig Moates APU1.

I have never reallocated cell boundaries, this should help a lot.

All these functions are amazing, and can see how it will dial in my tune.

The knock cell function should be interesting in my set up. I basically have my knock system disabled for now. I have false knock issues ( solid urethane mounts, large link single row timing chain, valve train noise, Long tubes, etc.. ) that drove my knock sensor nuts, wrecking havoc on timing tables.

This should be fun! :thumbsup:

Thank you for your hard work!!

Keith

[steveo]

I hope one of the cables I have is compatible with scan9495

it should 'just work' with pretty much any serial cable. it doesn't do anything that tunerpro can't do with a bit of work, but it's a nice program just because it's LT1 specific, it was basically made to replace the old gm TECH handhelds.


I have never reallocated cell boundaries, this should help a lot.

it's really common to see an entire row or column of the blm table go pratically unused; definitely a waste. but do remember that you want more cell boundaries in areas with more varied fueling, so don't try to split the table up across the entire rpm range.. lower RPM is generally where fueling is all over the place.

[bybyc5]

Having some trouble getting scan9495 to load without errors. Tried on both my laptops. Thought it may have been because I loaded the program before plugging into the aldl. Tried several ways. I'll try a fresh download, and reinstall. Wish there were a way to check integrity of a download, like an md5 comparison. If all else fails, I will do the data logging in TunerPro Rt, or datamaster.

[steveo]

what kind of errors?

[bybyc5]

Got it working, and made a short road test to see if all is working. I am guessing it takes a good amount of driving to populate the cells, however I do have a question or 2?

BLM CELL vs trim,maf,map,rpm range/avg. I see ignoring < 20, but what is the cells registering 0 hits- Not reliable?

Narrowband MV (avg) vs rpm during PE active. Why all zero's, and what is the counts counting?

The knock counts is interesting. I have a lot of harmonics in the car due to rigid mounts and not stock stiff suspension. Right at 3k rpm there is a body resonating that is hard to describe, but it is interesting to see the knock sensor pick up on this. The car was to be taken to the track often, so I thought it better to have everything rigid. Turns out the car ended up being a street "sleeper", and I regret all the rigid stuff. Been threatening to go back to stock stuff. :rolleye:

One of the first mods I did was to put a set of Hooker Super Competition Long tubes on. ( we all know what a pia that is on our f-body LT1 cars :mad1: )
My first road test after the install was disappointing. The car was a lot quicker with stock manifolds and full exhaust ??? Data logging showed my knock sensor was wrecking havoc on my tune. And its even worse with the headers uncapped. The LT4 knock module was a waste of $ for me, it did zero. So until I can find a way to have some knock protection that wont respond to false knock...my knock tables are zeroed, and My timing tables are a little conservative, and my fueling a little rich, and my operating temperature a little on the cool side @ 170*F.

Thoughts on my first recording?
I will do more driving over the next week to populate better cell filling. Can I save and reopen a log in scan9495?

Or maybe this analyzer is not for VE table tuning? Just looking to see what the PCM has for learned data?

I had to zip the file to get it to upload here. There is no .csv allowed.

Also I have not changed my target afr yet. I wanted to get some reading on it the way it is now. I "can" get it into closed loop, but the motor does not like it at all. Open loop it actually drives really well, even for a stick car. Guys with automatics can get away with allot of cam, and minimal drivability issues. A loose converter, and a little more gear nets a really nice street car.

Thank you!

[yoheer]

how i can view L32 3.4 log files? csv file saved with Gary Doug's scan9495 for v6 engines

[steveo]

how i can view L32 3.4 log files? csv file saved with Gary Doug's scan9495 for v6 engines

the program mentioned in this thread has been replaced by trimalyzer

to “view” a csv file without analyzing it use a spreadsheet program like excel or openoffice calc