Page 1 of 1

Strange behaviour after update to build 1016

Posted: Wed 05 Oct 2011 9:39 am
by phklm
I have noticed that after installing 1.9.2 build 1016, we get strange readings in a morning. From around 0715 to 0930 the graphs tend to just 'flat line' on the trend page. This behaviour seems to happen every morning at the same time - on Monday I also noticed the rain rate was showing as -0.2mm during this time.

http://weather.roydshall.org/trends.htm

We are using a Davis VP2. We didn't seem to have this on the previous version (Non-beta).

Any ideas?

Adam

Re: Strange behaviour after update to build 1016

Posted: Wed 05 Oct 2011 10:46 am
by steve
Please zip up the diags folder and attach it.

What version were you using previously?

Re: Strange behaviour after update to build 1016

Posted: Wed 05 Oct 2011 11:51 am
by phklm
Thanks Steve.

It was the current version of 1.9.1 (June 5th, can't remember the build no..92x?).

Re: Strange behaviour after update to build 1016

Posted: Wed 05 Oct 2011 12:34 pm
by steve
It looks like you have a communication problem between the ISS and the console. The inside temperature, humidity and pressure figures were updating OK, and there are no errors indicated in the diags log; only the remote data has flatlined. The fact that this happens every morning at the same time, and a restart of Cumulus is not involved, would tend to confirm this.

Re: Strange behaviour after update to build 1016

Posted: Wed 05 Oct 2011 12:51 pm
by phklm
Thanks Steve.
I have no idea what could be causing a communications issue at that time everyday - I will check that out though.

Re: Strange behaviour after update to build 1016

Posted: Wed 05 Oct 2011 1:11 pm
by steve
It could be that your ISS battery needs replacing. Normally these would last several years, as they are only a backup for when you don't get enough sun to charge up the capacitor enough to last all night. However, the capacitors are notorious for failing; I've replaced mine twice in four years. If you look closely at the capacitor you should be able to see signs of leakage, if the capacitor has failed.

If you can look at the console while the data has flat-lined, I'd expect it to be showing dashes for the outdoor sensors.

Re: Strange behaviour after update to build 1016

Posted: Thu 06 Oct 2011 7:31 am
by phklm
Thanks Steve for your advice, it is much appreciated. I have missed the opportunity to look at the console this morning while it was flat lining, so hopefully I will get a chance again tomorrow.

I guess what you are saying fits with the lengthening dark nights at the moment, and hence why it has only just started. I have noticed this morning that Cumulus says TX-1 battery is Low, would this be the same battery you refer too?

I am going to be popular if I need to get access to the ISS - it is in a difficult to access location!

Re: Strange behaviour after update to build 1016

Posted: Thu 06 Oct 2011 7:52 am
by steve
phklm wrote:I have noticed this morning that Cumulus says TX-1 battery is Low, would this be the same battery you refer too?
Yes, that's it, it's the 3v Lithium CR123 battery in the ISS. That pretty much confirms the reason for the flat lines. How old is the battery? You need to check the capacitor in the ISS to make sure it hasn't failed, as otherwise the replacement battery isn't going to last long.

Re: Strange behaviour after update to build 1016

Posted: Fri 07 Oct 2011 10:34 am
by phklm
Thanks again for the advice.
I am not sure how old the battery is, it has been here longer than me, but an educated guess would probably be around 3 or 4 years so all the pieces seem to fit into place to explain why this is happening.