Absolute timing error in Read/Write measurement VI

[jesboy]

Hi

My problem is quite weird... I have some data which I have set attributes to with absolute times. I then save the data with Write measurement and open it in another VI with Read Measurement... Somehow the absolutte time of ALL the data has moved 3sec forward (I have sampling rate of 10Hz which is ok)!! I really cant see where the error is, it´s the basic read/write express VI I use.. And it does it no matter what data I use.. never had this problem before please help!

I use LabVIEW 2011 32-bit

Best

Jesper 

[A.Rohde]

Hi Jesper

Could you please post a VI that replicates this problem, then I will be happy to take a look at it.

Best Regards

Anders Rohde

Applications Engineer

National Instruments Denmark

[jesboy]

Hi Anders

I have attached a VI with describtion of "how to run to see error" in front panal together with one of the data-signal file I use. (datafile is about 8hours of 10Hz recording with 11signals + 3empty signals).. If you also know an easy way to delete empty signals in waveform without using "Select Signals" I would be glad to know (dont always 3empty signals, sometimes 0-2 or 4....). But most importaint is to fix the weird 3 sec. timing issue..

PS: I now have problems posting this reply with attachments, I´ll try send it bit by bit.

Thanks for helping

Best Regards

Jesper

[jesboy]

what the .... cant you attach a .lvm file?!?!?!

[RavensFan]

Certain file extensions are not allowed.

Zip up your file and attach that.

[jesboy]

way ahead of you;-)

[RavensFan]

I don't actually see the problem you are talking about.   When I run your VI with your attached LVM file, the output file seems to match the input file.  At least there are no changes in the time stamp on the order of 3 seconds.

But I do see a few things.

1.  The timestamp in the very first column of one of the header rows went from 2012/09/25 07:46:30 to 1903/12/31 16:00:00 (though all the other timestamps in that row stayed the same.)

2.  You use a decimal comma.  I had to change the Read Express VI from using the normal decimal point to use a comma so that it can read your data file.

3.  The first column when from having fractional parts with intervals of 0.1 to showing all zeroes.  But this may be a symptom of #2.

4.  Your timestamp shows "07:46:30,7" in that time row.  When it is written out for me (besides the problem listed in #1), it drops the decimal part of the number and shows just "07:46:30".  But that may just be another symptom of having a decimal "comma" in your data file. (#2)

First thing I would try if I were you would be to fix your VI and your data files so that they use decimal points instead of commas and see if that fixes the problems.

[jesboy]

@RavensFan: Have you saved the file and run the new copy as described in VI?

I actually read the data with "dot" (in read measurement), but it works fine, think the reason is that I have a Danish/European version of Labview and windows where "comma" is "normal". If I read the file with "comma" I also get time "0" on first colomn, but also still get +3sec to al other signals, when I save and run the new kopi2. Time goes to 07:46.33,7 on signals, when it should be 07:46:30,7 .

I dont have oppotunity to chance the comma to dot in the data, sorry..

Maybe Anders can read th data like me in Denmark?

Best

Jesper

[RavensFan]

I did save your file and run it.

I saw odd things about the resulting file, but nothing like a 3 second offset.  I am attributing some of the odd things I saw to the comma vs. decimal point problem.  (It may not be that, but having different a different decimal point means so many things have to line up perfectly for things to work right:  the file, the Express VI settings, the OS settings.)

I did not see a 3 second offset.  I don't have a clue as to why you would see that.  It just seems too odd.  Why 3 seconds and not 4?  Why not 1 or 2?  Why not an hour off, or a day?

[jesboy]

Hi Anders

Did you have time to look at the problem?

Best

Jesper