Our full technical support staff does not monitor this forum. If you need assistance from a member of our staff, please submit your question from the Ask a Question page.


Log in or register to post/reply in the forum.

numbers of records for CR1000


smile Jan 10, 2012 08:54 AM

Hi all,

I have to prepare a program for CR1000 to store some variables of a system composed of 14 tanks, where another system controls the filling and emptying. So in addition to measuring the variables, I have to check the beginning and the end of the events for filling and for emptying of the every tanks, to store the time, with a resolution of 5 seconds. In total I have 14 datatables fills, 14 datatable for the emptying and a datatable for statistics.
With 15 variables stored for each event for filling or emptying, with the program that set the datatables with the “SIZE” parameter automatically (-1), I have a few hours of "Fill Time Table", since the datainterval time is set to 5 seconds (except datatable statistic that is stored every 15 minutes), and a variable is used as a trigger for event in the datatable to discriminates whether to store or not.
The few hours are justified by the fact that the logger calculates the "Fill Time Table" as if all 28 (14+14) datatables are written every 5 seconds.

Having idea of the frequency of filling and emptying events during the day, I set the “SIZE” for a range of about 10 days (45000 records for fills and 5000 records for empty and 15000 for statistics), for a total of 29 DATATABLE with about 715000 records for the 10 days.
But when compile the program the data logger returns error "Out of final storage memory", I can compile without errors by setting the “SIZE” to about one-tenth of the previous attempt, with only about a day of records.
Taking a test to store with only one datatable with 15 variables, automatically “SIZE” every 15 minutes, the data CR1000 calculates approximately 905 days of records for a total of about 1,300,000 records (905 day for 1440 records/day).

Why so much difference between the 1.3 mega records and 82,000 records (for the programm without error), even taking into account the additional memory required to the logger to work with 29 datatables (Appendix H of the manual).

Where am I doing wrong?

I hope I was clear enough.

Sorry for my English
Thanks
Smile


aps Jan 10, 2012 11:11 AM

I think we would have to see the structure of your program to give best advice.

One word of warning about terminology which might be confusing the issue. In our software a "record" is used to refer to the complete set of data written as same time to a data table, which in your case would be fifteen data values, if I understand you correctly.

To explain further, you would can see the number of records reported in various status values you can retrieve from the logger. When you set the size of a table this same definition of records applies too, so when you set the fills table to a size of 45000 records you were making space for 675000 data values, or about 2.6 days or recoding at 5 sec scanning.

* Last updated by: aps on 1/10/2012 @ 5:53 AM *


TweedleDee Jan 25, 2012 08:29 PM

I am assuming you have figured out that your math was wrong:
"Taking a test to store with only one datatable with 15 variables, automatically “SIZE” every 15 minutes, the data CR1000 calculates approximately 905 days of records for a total of about 1,300,000 records (905 day for 1440 records/day). " It is actually 905 days at 96 records/day, or 86880 records. This correlates well with the rest of your description.

Bottom line, you need a CFM100 and memory card.


smile Feb 4, 2012 12:38 PM

Thanks for the suggestions.
Now I understand and test the difference between record and data value, so when I have said "about 1300000 records",
realy mean data values .
However it is normal that I find errors in the "status station summary" like "skipped records in ...."
(the same error is shown for the other 28 files)?

smile

Log in or register to post/reply in the forum.