Adding time stamp to table/array

[mbeenen]

Hi,

In my posted VI I am displaying some collected temperature data on a table.  What I am having trouble with is getting the time data to work properly on the table.  When I set the include time data to true, it reads out the time as 1904 etc. I'm not sure if it's possible for me to include a time stamp in my array because when I try to convert a time stamp to a double precision float, the output is still a cluster, and thus cannot be added to the array.  I've also tried to convert the array to a string and concatenate the time stamp on, but the signals input for the table does not accept strings, and because I would eventually like to export this table to a spreadsheet in a file, I don't really want to link the strings to the indicator directly.  I've tried to work with the instrument as well, but when I read time data from it it gives it back in the form of multiple numbers, such as 2008, 8, 28, 13, 30, 31.12 and I can't consolidate it to one number.  Needless to say, I'm a little stumped and am open to suggestions on how to fix this problem, whether its a quick fix or a change in the way I'm trying to record the data.

Thanks in advance,

Mike

Message Edited by mbeenen on 08-28-2007 03:59 PM

[tbd]

Hi Mike,

      Generating some meaningful timestamps is probaby no big deal (see attached VI) but your diagram leaves me with a question or two...

You're acquiring an array with every loop - should all the elements in each array get the same timestamp?

The Loop executes periodically, so some time elapses between the last element of the first loop, and the first element of the next loop - yet the arrays are being concatinated.  This means the time-difference between elements isn't uniform in the resulting array (making it more trouble to calculate correct times later.)

Perhaps one place to start is: what would you like the resulting spreadsheet to look like?  Perhaps:

    T0      ,     T1      ,     T2      ,     T3

Sample(T0,0), Sample(T1,0), Sample(T2,0), Sample(T3,0)

Sample(T0,1), Sample(T1,1), Sample(T2,1), Sample(T3,1)

Sample(T0,2), Sample(T1,2), Sample(T2,2), Sample(T3,2)

If not, please clarify your intention!

Cheers.

[mbeenen]

tbd,

Sorry if I was a bit unclear about what I wanted to do.  I think it will make more sense how I want to implement the time stamp with my updated VI.  You can see where I put my timestamp into the array, and everything works fine with the exception that the time is in the form of seconds.  This is where I have my problem, because when it is formatted into the date/time format, it naturally becomes unable to be converted into a double because of the special characters.  And I can't convert the array into a string before I create the table because the Build Table VI is unable to accept strings as a parameter.  This is the main issue I have with your solution, because although the table has the proper time stamps, I am unsure how I would export it to my spreadsheet VI.

In essence, the problem has two solutions:

1:  Find a way to get the time stamp into the array inside the while loop in the proper format

2:  Change the way I build the table and export it to the spreadsheet, so that I can convert the array into strings for the timestamp to be inserted.

I would prefer the former over the latter, but any advice would be great.

Thank you,

Mike

P.S Thanks for mentioning the build array function inside the loop, I'm pretty new and forgot that I could use auto-indexing.

[Dennis_Knutson]

If you want to use the Express VI for the table, you should be creating a waveform data type and passing this. The way to create a waveform data type is with the Build Waveform function. Wire the current date time to the t0 input of the Build Waveform, wire the array you get from the 34970 measure function to the Y Array  input, and wire 1/measurement interval to the dt input of the Build Waveform.

What are you trying to do with the inner while loop? I also don't quite understand the purpose of the decimate array. You are removing every other element for display in the table but then it looks like you want to write everthing to Excel. You've not posted some of the subVIs including the Excel one so I don't know what other mods would need to be made.

[Elizabeth_S]

Hi Mbeenen,

Please take a look at the following discussion forum:

http://forums.ni.com/ni/board/message?board.id=170&message.id=179804&requireLogin=False

[mbeenen]

Here are the sub VI's that I'm using.  The inner while loop is designed so that the loop will halt immediately when stopped, and not have to finish its current iteration, as the iteration period is intended to be rather long.  I may modify the program to have a programmable start/stop time later, but this is at least my temporary solution.  The purpose of the decimate array function is to seperate out the channel label from the actual data so they can be placed as headers in the Excel function.  The intent was likely not clear with the lack of the SubVI's.  Overall, the display table on the front panel is no longer necessary, it was just for me to see how the table was working in the early stages of development.

The one issue I had with your suggestion is the fact that I the build waveform function only accepts a 1-D array for the waveform input, and with multiple and variable number of channels being used I'm not sure how I could easily manipulate the data I gather into multiple 1-d arrays.

As always the help is appreciated,

Mike

[mbeenen]

I will also attach the main VI again, because of minor changes and alterations of some of the Sub VI names

[mbeenen]

I managed to fix my issue.  It turned out the best solution for me was to create a seperate array of strings in my while loop and then wire that to my spreadsheet VI to be put in as the row headers, similar to what I did with the channel headers.  I'm not sure if this is great style, but I guess my primitive knowledge of LabVIEW sometimes forces me to use "backdoor" solutions.  Thank you for all the suggestions, I think they really helped stimulate my thinking to get to my solution.

Mike