PDA

View Full Version : What's wrong with BBL to cause this?



gmh308
July 11th, 2011, 11:41 PM
Ran up some logs today, but they seem all corrupted.

The attached log is all over the place...rpm's at 10,000 plus, airflows of 4g/cyl, inj PW's of 207ms.

Export to CSV and the time stamps are all over the place and the PID values are all over the place too, the readout on the V2 is normal when logging, but this is what comes up on the [normally fantastic and without peer] scan tool.

11386

Anyone have any ideas? Known issue that I missed?

:(

joecar
July 12th, 2011, 01:34 AM
Hi Ian,

What V7/V8 build versions do you have installed...?
What V2 firmware...?

gmh308
July 12th, 2011, 09:35 AM
Hey Joe, its been working fine in the past, Scan is 7.5.6/139, and V2 is Boot 2.6.2 & 2.6.16.

Cheers Ian

swingtan
July 12th, 2011, 09:41 AM
try these....

http://forum.efilive.com/showthread.php?12679-why-does-this-happen-alot-log-reading-prob

http://forum.efilive.com/showthread.php?7919-Beta-Releases-and-Log-Files-Full-Of-Notes

http://forum.efilive.com/showthread.php?9409-Strange-log-file-problem

Blacky
July 12th, 2011, 12:12 PM
Ran up some logs today, but they seem all corrupted.

Anyone have any ideas? Known issue that I missed?

:(

Can you send that *.efi file to me at paul@efilive.com, thanks
Paul

gmh308
July 12th, 2011, 03:00 PM
try these....

http://forum.efilive.com/showthread.php?12679-why-does-this-happen-alot-log-reading-prob

http://forum.efilive.com/showthread.php?7919-Beta-Releases-and-Log-Files-Full-Of-Notes

http://forum.efilive.com/showthread.php?9409-Strange-log-file-problem

Thanks Simon. Looked through those and went in search of errant other programs. In the end, the Windows 7 PC had done over 900 hours of uptime with at least a hundred hibernates. Probably 200.

So I shut everything down and did a restart. No change. Still displays bad.


Can you send that *.efi file to me at paul@efilive.com, thanks
Paul

I did however try the same log on a a Vista PC, and it works fine. Have emailed the log Paul. Am sure it will be fine, though a number of parameters are missing from the screen view selection, as well as the file export function, that were being logged on the V2...i.e. g/cyl.

Thanks kindly, Ian

swingtan
July 12th, 2011, 04:02 PM
Well, thinking some more on this.

I used to get this all the time under W-XP. Now under W7, I've never had this problem even when I've tried to replicate it. I don't think it's the OS change that has directly fixed this though, as I had one other rather bad issues under XP.

My user profile had gone rouge, seriously large actually. My profile under Xp was 14MB in size where is should have been more like 3MB. I think this is where things really went bad. I had a few other issues and all were resolved with the OS change ( which of course resulted in the new profile ). So maybe try creating a new user with a fresh profile and see if the problem remains or goes.

Simon

gmh308
July 12th, 2011, 04:48 PM
Well, thinking some more on this.

I used to get this all the time under W-XP. Now under W7, I've never had this problem even when I've tried to replicate it. I don't think it's the OS change that has directly fixed this though, as I had one other rather bad issues under XP.

My user profile had gone rouge, seriously large actually. My profile under Xp was 14MB in size where is should have been more like 3MB. I think this is where things really went bad. I had a few other issues and all were resolved with the OS change ( which of course resulted in the new profile ). So maybe try creating a new user with a fresh profile and see if the problem remains or goes.

Simon

Thanks Simon, tried a new W7 user. Same result. Scan seems to munch the log.

swingtan
July 12th, 2011, 04:58 PM
Post the log, I'll rip it open in my "log-tool" perl script and have a look. If it's really the log it's self, then we should be able to see what is going on and may even be able to recover some of the data.

Simon

gmh308
July 12th, 2011, 08:39 PM
Post the log, I'll rip it open in my "log-tool" perl script and have a look. If it's really the log it's self, then we should be able to see what is going on and may even be able to recover some of the data.

Simon

Thanks Simon. Here it is. Am sure it will check out fine and the issue comes down to something else wierd.

11393

swingtan
July 12th, 2011, 09:17 PM
Good news... it's not your PC or setup. I just opened it and it's as per your image in post one.

From what I can see....

You have a channel count of 75 and a frame size of 83 bytes. Total frame count is 40430 and the meta data is 2309 bytes. So all up the files size should be around 3357999 bytes ( give or take some sundry bytes ).

The header and meta data looks fine, but the frame data looks to have lost some data and skewed the frame data. This has messed up the frame data completely, but think it's still all there. I'll see if I can nudge it back into alignment.

kangsta
July 12th, 2011, 09:42 PM
log opens fine here...

Win7 x64
EFILive v7.5.6 Build 157

swingtan
July 12th, 2011, 10:33 PM
That's interesting kangsta, That's an older version of Scan tool, current build is 170.

The log data indicates that one or more PID's are possibly not adding the correct number of bytes. This is making the actual frame length less than the indicated 83 bytes, causing a shift in the data and the subsequent scrambled charts. It doesn't explain why it opens in build 157 though. Here's a couple of images showing a "normal" log file and the one that doesn't work in build 170.

11394

11395

You can see the difference in the data pattern of the frame data, the first is uniform while the second has an obvious data shift involved. I'll try opening the log in an older version I have on a different computer.

Simon.

swingtan
July 12th, 2011, 10:53 PM
I just tried to open the log in builds 155 and 164 and both were no good. I think that kangsta may have opened the tune file and not the log file.

Simon.

gmh308
July 12th, 2011, 11:05 PM
Good news... it's not your PC or setup. I just opened it and it's as per your image in post one.

From what I can see....

You have a channel count of 75 and a frame size of 83 bytes. Total frame count is 40430 and the meta data is 2309 bytes. So all up the files size should be around 3357999 bytes ( give or take some sundry bytes ).

The header and meta data looks fine, but the frame data looks to have lost some data and skewed the frame data. This has messed up the frame data completely, but think it's still all there. I'll see if I can nudge it back into alignment.

Thanks Simon. I am confused now. What would enable the log to open ok in Vista?

kangsta
July 12th, 2011, 11:24 PM
yeah i know im running an old one, a few of my autocals are still in active duty so i like to keep the same versions with the customers while logs are being taken.

I'm pretty sure I opened gmh308's log file but i was playing with another file at the same time so i might have been mistaken. I'm at the airport now but i'll check again later when i get to the hotel.

swingtan
July 12th, 2011, 11:57 PM
I'm not sure why it opens in vista, but the frame data changes near the end. Here's an image of where the data changes...

11396

kangsta
July 18th, 2011, 06:55 AM
not sure what happened before, i cant get it to open properly on my system must have gotten the file confused :/

swingtan
July 18th, 2011, 10:39 AM
lol... that's good news than as I was wondering how it could have been opened on any version.... I must admit I'd forgotten about this, I'll see if I get some time later on to have another look.