Mitsubishi Lancer Evolution 4/5/6/7/8/9 EvoScan Project v0.8

[the-moss2]

ok, I'm going to try plugging both in tonight and see if it makes a difference.

[andenbre]

is there any like pdf on how to get this data software going and use all the options in it. like a faq or a starter up

this is currently a prerelease so there is no documentation on it other than feedback that users post in the forums.

[hexman]

I do not think that evo4mad if anyone knows what the load algorithm is. The closest I have seen is the Subaru formula:

g/4 cylinders charge = 120 * maf (g/sec) / rpm (rev/min) <-- Just as an FYI. That's the fomula used by the Denso ECU's and the SSM (Subaru Select Monitor) protocol, to calculate engine load.

Jorge (RiftsWRX)
www.ProjectWRX.com

Above is from EvoM.net http://forums.evolutionm.net/showthread ... ost3199606

I agree we need to know how this is calculated and be able to log it. This would help the tuning process immeasurably.

Up until now I hear a lot of speculation about load...

[Blak94GSX]

Evo 9 people, what sample rate are you getting?

I only plugged the ODB-II part of my tactrix cable into the car (not the prorammer part) and was getting around 15samples/s indicated on the software.

Should I be connecting the other connector also?

Run both cables connected. That puts the ECU into Diag mode instead of just regular OBDII mode. The switch only happens at startup, so you don't need the second plug attached once you are sampling data from the ECU, although I don't know why you would want to disconnect the plug...

[mike1023]

is there any like pdf on how to get this data software going and use all the options in it. like a faq or a starter up

this is currently a prerelease so there is no documentation on it other than feedback that users post in the forums.

sounds good to me also how come the log is does not keep rpm in one colum , and then timing in othere and tps and so on so its easyer to read the log.

is there someway to make it do that

[andenbre]

I am hoping there is an actual value for load that the ecm will spit out like it has for rpm, ignition timing, ect... I think that this would be more accuate than a value calculated by the software its self.

[andenbre]

is there any like pdf on how to get this data software going and use all the options in it. like a faq or a starter up

this is currently a prerelease so there is no documentation on it other than feedback that users post in the forums.

sounds good to me also how come the log is does not keep rpm in one colum , and then timing in othere and tps and so on so its easyer to read the log.

is there someway to make it do that

what I do is tag the first set of samples then cut and past each set over to the right of the previous set over and over. it takes some time but you get a timeline of concurrent samples that you can follow better.
If you need a screen shot I can post to clear up my explanation

[galvitron]

Paid, downloaded

Tested successfully with Windows 2000 Professional

Build 5.00.2195
Service Pack 4
.NET framework 1.1 installed (No 2.0 installed)
USB driver works fine

Evo4Mad, I just thought I'd mention this since you don't list it under supported OS list.

[andenbre]

I think I figured out the timing issue when the ecu reports back 20 its actualy hex for 32 then subtract 10 and you get the 22 which is approx 10 degrees to high so if that factor is changed to "x - 20" that might be more accurate. what we need is someone with ecu+ to run a parallel log and compare the two logs for evoscan inaccuracys and find out exactly how far things are off

[Blak94GSX]

I have a variety of loggers at my disposal. I will compare the results and see if there is some consistency on the difference in timing reported.

[andenbre]

what would be hot is if you (or any body else) could compare the other data logged that you can and let Evo4mad and the rest of know what is on target and how far anything else is off

here is a log that I edited out most channels and corrected the ignition timing to x - 20 and graphed it

[Blak94GSX]

Yeah I'll post a bunch of log files from various systems for comparison tomorrow.

[Blak94GSX]

The logger portion connects and samples data pretty well. I'm getting 120ish samples a second.

My only complaint so far is the data file format. Instead of writing out separate rows for each data element, can you change the data file to write out a field list at the top when a new log file is created, and then each of the data elements in a column. Something like this format which every other logger follows:

[Evo4Mad]

Thanks I'll add this format as an option... but the timestamp is different between the first column and the end column which is the reason for the current format. I will cross tabulate the results and add use 2 columns StartTime / EndTime when using this format.

The logger portion connects and samples data pretty well. I'm getting 120ish samples a second.

My only complaint so far is the data file format. Instead of writing out separate rows for each data element, can you change the data file to write out a field list at the top when a new log file is created, and then each of the data elements in a column. Something like this format which every other logger follows:

[c-spec]

any release date on the fix for the timing table and other stuff?