I'm doing a test to monitor how many skipped scans/records occur in my application in a certain amount of time. My scan rate is one second and the interval to store to the data tables is one second as well. Therefore, I would expect to have a skipped record every time I have a skipped scan since I'm skipping a data storage interval but that does not seem to be the case. Sometimes, I see 5 or 6 skipped scans and only one skipped record in my data table. Why is it that the skipped scan is not causing a skipped record as well? It's what I would expect.
Thanks.
新对话如下：
Welcome to the forum Bonilt,
I believe each skipped scan increments that counter, but the SkippedRecord counter only increments the first time a group of records gets skipped. In your example lets say you skip 5 scans in a row. SkippedScans will equal 5 but Skipped Records will only equal 1.
Is that what you are observing?
Regards,
Janet
新对话如下：
Hi Janet,
Thank you for your response.
This is not what I'm observing. The skipped scans do not occur in a row. Sometimes I even have 5 or 6 skipped scans that produce no skipped records.
Is there any explanation as to why this is happening?
Thanks,
Bonilt
新对话如下：
Bonilit,
Can you post a copy of your DataTable structure? Include everything from DataTable to EndTable. And, are you calling your table every Scan?
What OS is in your logger? You can see that in the Status Table.
Janet
新对话如下：
DataTable(Measurements_OneSec,True,-1)
DataInterval (0,1,Sec,25)
CardOut (0,432000)
Sample (1,WindSpeed,IEEE4)
Sample (1,WindDirection,IEEE4)
Sample (1,AirTemp,IEEE4)
Sample (1,RH,IEEE4)
EndTable
My scan rate is one second and I'm calling the table on every scan. The OS is 17.
新对话如下：
What are you doing to force a skipped scan?
新对话如下：
I'm not forcing it. It seems like it's happening whenever I have a clock set.
新对话如下：
What is causing the clock set? Are you doing that manually with your KD or LoggerNet? When I do a manual clock set using LN or the KD my SkippedRecords increments by one, SkippedScans do not increment. This is expected behavior but not what you are reporting.
We may have to talk through this in person. I suspect there is more than one thing going on. Send your complete program and phone number to training [at] campbellsci [dot] com
Janet
新对话如下：
Janet,
The synchronization of my clock is done automatically through a synchronization schedule. I've discussed this with technical support already since it seems that it is very likely to have skipped records/scans whenever the clock is set given the fact that I'm scanning and storing every second.
Since there does not seem to be a solution to prevent the skipped scans/records whenever the clock is set, what I'm trying to do is figure out how many skipped scans/records I get within a certain amount of time considering how many times the clock has set during that time frame. I don't get a skipped scan every time the clock is set but every once in a while I do and I would expect the skipped record to increase as well but it does not. In some cases, I just get skipped scans but no skipped records.
What I'm trying to understand is why no skipped records are reported when I get skipped scans? I'm not sure if the program has anything to do with this. But how does Loggernet handles the time stamping of the data whenever there's a skipped scan/record?
Thanks
* Last updated by: bonilt on 11/10/2009 @ 2:38 PM *
新对话如下：
Section 9.7.1.2 of the latest CR1000 manual (http://www.campbellsci.com/documents/manuals/cr1000.pdf) goes into detail about timestamps and lapses (Skipped Records).
I'm not seeing what you are so I think there is more than one issue at hand. Everything I'm seeing is expected behavior based on the descriptions given in replies above. SkippedRecords increment by one when a record is skipped or several in a row. SkippedScans increment when the program scan takes longer to execute than the given time. There are other things that can cause a SkippedScan as well. Power cycling the logger should not increment either the SkippedRecord or SkippedScan.
I'd recommend you talk to the tech support person you were working with originally and figure out why your results are inconsistent.
Regards,
Janet
