January was quite windy at times! I am not sure, however, if the 77.3mph gust recorded at 15:03 on the 29th was real or not! There are two samples in the data but everything round about is much lower. I tend to think that it was just a glitch.
March. Reviewing the data for March showed a few times when some data from the Thermo/Hygro unit was missed.The batteries were replaced in November so they should still be okay. Another reminder that I still don't have the battery box moved inside yet! There have been no data drop outs since the 15th so fingers crossed that the problem has gone.
April the first, appropriately?, saw the death of my anemometer. One windy night I went to check the wind speed and found it to be zero. A trip on to the roof revealed that the fan on the TX12 sensor was spinning quite happily but no reading was being sent back to the console. A TX20 is on order to replace it. 5th April, the TX 20 has arrived and been fitted. Wind speed data is back! In a slight modification to the mounting, aided by the options available with the supplied bracketing, the anemometer is not mounted above the supporting mast and TV aerial.
June 3rd. A problem with a kvm switch meant that the PC did not boot up on schedule as Windows didn't see a mouse and waited for a human to tell it what to do. Unfortunately, there was no human present and I didn't notice that the customary email confirmation of the post update data backup hadn't been sent. About 7 hours worth of data was lost as a result.
July 19th. Another silly problem came to light today when I decided to have a quick check of the data list. There was a gap from the 10th until the 18th which I think was caused by the PC clock being about ten minutes fast. As a result, the backup and shutdown routines which were using fixed times were running before HeavyWeather had time to start up and download the days data. To try to avoid this happening in future I have changed the scheduling method so that the backup will run after a delay of twenty minutes followed by shutdown some five minutes later.
July 24th. Updated all 2008 weather files to XHTML 1.0 Strict.
August 7th. Spent the last few days trying to sort out the mess the July data is in. In addition to the comments above, I found that part of the data for the first and also between the third and eighth had not been recorded. In order that the recorded data would be shown in its correct place on the graphs I have inserted dummy data for pressures, humidity, temperature and wind speed and direction. Mostly this has put the data down to the baseline of the graphs. The rainfall has also been a problem. Examining the recorded accumulated totals, none of the missed data appear to have included any periods of rainfall but in the recorded data, the accumulated total is seen to increase at times when no rain has been indicated. To fix this, I have added appropriate rainfall amounts at the appropriate times.
August 19th. Had a bit of a shower today! Recorded 18.6mm of rain in around twenty minutes at around twenty to five.
September 1st. The display unit has again begun to lose the connection to the thermo/hygro unit, the usual indication of power problems. The result of this is the few dodgy samples shown late on the 31st of August on the humidity and temperature graphs. There were aslo a couple of rogue 110mph gusts on the 30th that I have edited out.
A shopping trip followed by a trip up the ladder have hopefully resolved the issue for a while - yet another reminder that I still haven't sorted out the indoor battery box!
October and November. Still having occasional missed samples up until about the 12th of November and then all looks fine again. Wasn't really aware that this had been happening until it was far too late to look for a cause.