Can Cumulus damage the Instromet Data Logger.

[fdhugo]

In case it's useful...
Yesterday on rebooting the PC and restarting Cumulus it took a long time to start transferring data, and the transfer seemed to take a long time when it did happen.
I left it alone and did other things, then found that it had succeeded.

When I get access I'll see if this morning's data is accessible, and reset the logger if not.

[steve]

The transfer apparently taking a long time to start is probably because it's having to read off and discard all of these older entries that it has no use for. Note that download will take quite a while, as it takes 2 seconds to read each entry with the 'wait time' set to 1000ms.

[fdhugo]

Thank you, I shall be patient.
I have just cleaned up the July data for Associated British Ports, who use it.
Latterly many gaps, and data duplicated and inserted out of chronological order.

[steve]

I've reported back to Instromet that the apparent fix from doing the reset didn't last long, and explained what I've seen from the diagnostic logs. I don't think there's any more I can do for now.

[fdhugo]

OK.
Thank you.
I have emailed the files, too big for the Forum.
I shall attempt another reset.

[fdhugo]

Diags.zip is too big for your email, 28 megs.

[steve]

I filter all my email through Gmail, which has a 25MB limit on attachments.

[fdhugo]

Progress?
Cumulus closed, logger reset, opened, within ten minute period.
Data recorded.
Closed across ten minute period, opened.
Data transferred.
Closed across ten minute period, ImetWaitTime set to 200ms, opened.
Data transferred.
Closed now for a longer test.

[fdhugo]

Worked.
Closed for about four hours.
Opened.
Data transferred with normal pause and at normal rate.
I still don't trust the PC not to sleep, so have closed Cumulus again.

Trivial puzzle.
Recovered data is not at round ten minute times, and drifts if over a longish period.
Not serious, but untidy.
Any ideas?

[steve]

The archive entries are timestamped by the logger, so perhaps the clock drifts - or perhaps it's just the way the logger works, the interval may be approximate. I really have no idea how the hardware/firmware works.

[fdhugo]

OK, trivial.

Seriously...
This morning.
PC accessible, not Sleeping.
Opened Cumulus.
Data transfer appeared normal.
The log has duplicated entries and missing data.
Log and cumulus attached.

[steve]

It's the same as the previous runs where there were problems - the logger starts off sending the correct logs in sequence, and then suddenly switches to earlier logs. That spurious "E" is there again in the response when things start going wrong, so it's obviously related in some way to the problem.

[fdhugo]

That's related to the logger?
I'm getting a different response to resetting.
I'll email Richard.

[steve]

That's related to the logger?

I don't understand the question. Everything I have said is related to the logger.

[fdhugo]

Sorry.