Its ok it seems to be fast enought 192 samples a second on my duo core

 

I will check the accuracy of this new formula and get back to you guys.

 

Guys,

I got a chance to check the two new formulas that were posted here, the ones for coolant and air temp. Both of them are still off vs. the pocketlogger.

The coolant formula is improved. With the new formula Evoscan records that the fans kick in @ 200* while the pocket logger records that the fans kick in @ 190*. So it is down from being 22* off to being 10* off. It is an improvement. If anyone knows how to tweak the formula and post it here, then I will test it.

The air temp formula is off. When the fan kicks in the air temp is @ 83.8* with Evoscan but with the pocketlogger it is @ 100* The 83.8 simply does not make any sense given that the temp in my house is 79* F. Someone needs to write a better formula and I will test it.

I did the tests back to back after the car was nice and warm.

I still need to check the NBO2 accuracy, but I have to do that @ WOT to get a good reading. I will post the results when I do the tests.

 

Log Analyzer v2.1 for EvoScan (V0.8 and v0.9) is officially released...

http://www.aktivematrix.com/forum/viewtopic.php?t=41

Test it out, it should work, however I did not do adequate testing on it since I cannot do any logging today nor did I install a current version of Evoscan to verify..

This will be the last release of Log Analyzer for Evoscan, and will be superceded by my new logger.


However I wanted to note that the software was released on the date and time promised (Sunday before 11pm)..

 

nj1266, Assuming that the pocket logger is correct, and if you've got the time, you can gather the data needed to make good temperature correlation formulas.

In EvoScan, change the formulas for coolant and air temp to "x". EvoScan will then just record the ECU output.

For coolant temperature, you need to generate a long EvoScan log as the engine warms up. Start EvoScan, the pocket logger, and a stopwatch all simultaneously, and then write down the time at which the pocket logger reachs a multiple of 10 deg. I'd just write down 100, 110, 120, ..., 200 on a piece of paper, and then record the stopwatch value next to each temperature. Send me the ECU output values from EvoScan and the stuff you wrote down, and I can generate a correlation formula.

You can try to do the same with air temperature.

 

If someone gets a chance to try out the new log analyzer, please PM me if you got it to work okay.. I won't be able to sleep until I know someone did it sucessfully..

 

hehe. I just realized that I mixed up the equations that I listed in my earlier post. The longer equation is for intake air temp and the shorter equation is for the coolant temperature. I have edited that post to reflect this. Try switching the equations and see how EvoScan compares to the pocket logger.

 

Thanks MalibuJack

I did have to add "WidebandAF" to the Evoscan.txt file for DLL to pick it up. I'm not sure if order matters in the file or how the DataItemID is determined.. but making one up seems to work

[Column]
Name=WideBandAF
Description=WideBandAF
ColumnTitle=WideBandAF
DataItemID=54
Type=1
ColumnDecimalPlaces=1
DisplayDecimalPlaces=1

 

Actually I think the order of your columns in your log file do need to be in the same order they are in the Evoscan.txt file for DLL. I changed my columns around and now DLL is showing my AFR's in the 50's

The data looks fine after "processing" it with the analyzer so the change in columns has no effect on this

 

Order of the columns does not matter at all.. but the ID value it uses does matter.. Each unique item needs an unique dataitem ID in evoscan.. It was one of the undocumented glitches I found in that program.

Of course, I didn't add any wideband columns to the definition file because I don't use it.. so I forgot..

 

I was really concerned about if it worked right for everyones logs.. I didn't have enough data to work with the fully test it, and I had not had the time to try it out later with data log lab..

I'm a decent programmer, and rarely make major errors, but I didnt want to release "Microsoft tested code" in otherwords, unverified.. But I wanted to get something out to you guys and I was reasonably confident that it worked to do the job.

 

Yeah I realize this after ordering them exactly the same.

My problem was I gave the Wideband AF the same ID as another column was using...

You can add this to your evoscan.txt (Make sure nothing else has the ID of 55)

[Column]
Name=WideBandAF
Description=WideBandAF
ColumnTitle=WideBandAF
DataItemID=55
Type=1
ColumnDecimalPlaces=1
DisplayDecimalPlaces=1

I wouldn't really consider this a problem.. other than this everything is great man! Good job as usual

 

I did that and equations were almost perfect The Evoscan coolant temp hit 193* when the fan kicked in compared 190* on the pocketlogger. The air intake temp was around 100 just like the one registered by pocketlogger.

So now we know that these two equations are as accurate as possible.

 

Thanks for linking those formulas Fred...

 

So what is the actually formula you need to add to the xml file?