SPC Express can read samples that are being written dynamically to a text file by other software.   Here's how it works:

• The disk file must be named FileData.txt by default.   If you need to use a different file name, you can over-ride the default by defining the file name in the Special Instructions field for that characteristic. (Click on Edit / EditTheCurrentCharacteristic / Optional Info Tab)  
  Example: *Filedata=whatever.txt   or   *Filedata=C:\Scratch\SPC.txt
  You may also specify the filename globally by typing FILEDATA=C:\SPC\whatever.txt into the Extra Options field in Global Preferences.
  Notes:
  • The File is assumed to be located in the "working path", unless otherwise specified.  
  • The File Name must either end with .txt or have a comma after it.
  • You can only be processing one text file at a time, meaning you can't have two different characteristics processing two different text files at the same time.
  • A common problem is that the file gets named FileData.txt.txt (with two extensions).   Look in your Windows Explorer Tools / Folder Options / View and BE SURE that the checkbox called "Hide extensions for known file type" IS NOT CHECKED!
  
   
• The file should contain records terminated with Carriage Returns just as if they were readings from an electronic gage.   The same formatting rules apply.  
  NOTE: Maximum file length = 1,000 records or 15,000 bytes, whichever is less.
  As of 10-27-2004, these numbers have been increased to 1,200 and 250,000.
   
• To tell SPC Express to read the file, set your Comm Port to 7 (baud rate, parity, etc don't matter).   Get to this set-up screen by clicking Edit / EditTheCurrentCharacteristic / Input Options Tab.   You can specify either a sample starting byte position or a field number.   This works great for Excel-type CSV (Comma Separated Values) files.   To specify a field number, just add 700.   Example: Use 704 to indicate that you want the 4th field.   (It is assumed the commas are used to separate the fields.)
   
• Set the Special Option (use the 'Extra Options' field in File / Global Preferences): FileTimer=ON.   (You must actually type this in.)   This causes the text file to be scanned every 2 seconds for new data.
   
• When a Characteristic is first opened and it uses Comm Port 7, you will be asked if you want to delete the existing FileData.txt file.   Normally you will say 'No'.   You can have this 'No' answer be automatic by specifying the Special Option FileDataStart=KEEP.   You may find that it works out best if you delete the file when you're done processing it (see next bullet).   Whether you delete it or not, SPC Express will start scanning at the beginning of the file (unless you click 'Cancel').   If you do NOT delete it, you may be processing the same samples twice.   That is why you are given the opportunity to delete the existing file.   If you DO choose 'Cancel', the SPC file will be opened but the text File will not be read for processing.   You would typically do this if you just needed to open the SPC file for editing and didn't want to process any new data.
   
• Once SPC Express starts reading the file, it continues until it reaches the end of the file.   When it reaches the end, it stops processing records and waits 2 seconds before checking for new records again.   If the Special Option FileDataDone=DELETE is set, then the text file will be deleted after the last record has been processed.
   
• A typical FileData.txt file might look something like this:

   Dia1  ##Date##09/12/2003##  (optional - see note below)
   Dia1  ##Time##10:45:00##    (optional - see note below)
   Dia1  ##LotNumber##123##    (optional - see note below)
   Dia1   = 1.250
   Dia2   = 2.750
   Length = 4.575
   Dia1   = 1.255
   Dia2   = 2.745
   Length = 4.537
   Dia1   = 1.225
   Dia2   = 2.740
   Length = 4.562
  

  Notes:
  
  • If Date and Time are absent, the current time will be used.
  • You can define "traceability" data using ## as shown above.   If other Traceability fields are absent then that trace field will be blank unless the "P" flag is set.
  • A record can contain sample value(s), or one traceability value, but not both.
  • WATCH OUT! If the record starts with leading spaces, the byte count will start at the first non-blank.
  
  
  
• Another possible file format could be (3 characteristics per record):

  Test 1, 09/30/2003, 09:15:00 AM, 1.25, 2.88, 3.125
  Test 1, 09/30/2003, 09:30:00 AM, 1.26, 2.74, 3.133
  Test 1, 09/30/2003, 09:45:00 AM, 1.28, 2.76, 3.127
  Test 1, 09/30/2003, 10:00:00 AM, 1.22, 2.90, 3.129
  Test 1, 09/30/2003, 10:15:00 AM, 1.24, 2.81, 3.131
  

  For the file format above, you could set Special Options: (requires version 10-09-2003 or newer)
    FileDataDate=9   indicating that the Date - 10 characters long - starts in byte position 9
    FileDataDate8=9   indicating that the Date - 8 characters long - starts in byte position 9 (requires Ver 10-25-2004)
    FileDataTime=21   indicating that the Time - 11 characters long - starts in byte position 21
  
  You can also have SPC Express find the Date & Time fields automatically ....
    FileDataDate=Auto   finds the Date based on the first slash ( / ) in the record.
    FileDataTime=Auto   finds the Time based on the first colon ( : ) in the record.
  
    Note: You can also specify the Timestamp location by Field Number (70n) instead of byte position.
    (requires version 02-15-2006 or newer)
  
  
• Another possible format is if you have several samples for the same feature all in one record:
  (requires version 12-30-2005 or newer)
  

  Width , 09/30/2003, 09:15:00 AM, 1.01, 1.02, 1.03
  Width , 09/30/2003, 09:30:00 AM, 1.04, 1.05, 1.06
  Length, 09/30/2003, 09:15:00 AM, 5.01, 5.02, 5.03
  Length, ##Operator##Bob##
  Length, ##Lot Number##12345##
  Length, 09/30/2003, 09:30:00 AM, 5.03, 5.04, 5.05
  

  (Subgroup size = 3)
  
  SPC Express will read all of the samples in a given record and put them into the same characteristic.   To do this, specify the 'Position of Sample' to be 704 and the 'No. of characters in the Sample' to be 706.
  
  
• Another Method for specifing where your samples are located in the text file is called "FileData Chapter Processing" (requires version 11-14-2008 or newer).   It is especially useful for processing data files coming from CMM machines.
  
  For more information click here.