Does anyone know a way of determining the table fill times, other than using the Connect->Station Status->Table Fill Times screens in Loggernet?
The problem with the above is that you need to have uploaded the program to the logger to check the table fill times. If you're writing a program off-site and don't have a logger at your disposal then out comes the pen and paper. I realize it would depend on memory size and allocation etc, but an option to determine table fill times for a standard logger configuration example would be a nifty feature.
maybe someone has figured out a work-around?
thanks,
There really isn't an easy way other than loading it into a logger. Do you have an Applications Engineer you work with? If so, send them a copy and have them load it and get the number for you. Easy-yes; convenient-no.
If you want to do it with pencil and paper you'll need to take a hard look at your data tables. Table 24 in your CR1000 manual http://www.campbellsci.com/documents/manuals/cr1000.pdf shows how many bytes each data type consumes. Take a look at how much data storage your datalogger has available; a 4M logger has about 3.7M, a 2M logger about 1.7M.
Taking the easy case, say you have 0 lapses in your DataInterval instruction, or no DataInterval. You'll consume 16 bytes for a timestamp and record number. CR1000 manual section 9.7.1.2.
Add up all the bytes over a time interval.
Total memory divided by total bytes per time gives time. This is likely a VERY conservative estimate because you probably have lapses in your DataInterval instruction(s). Again, I point you to section 9.7.1.2 of the CR1000 manual.
In a quick test I was storing 22 bytes/minute. I get about 55 days of data storage in a 2M logger when storing a timestamp with each record. Same scenario but with lapses specified and I get almost 200 days.
There are other things that can affect datalogger memory also. Take a look at Table 66 of the CR1000 manual. That shows how competing demands there are on the SRAM.
Hope that helps.
Janet