Page 1 of 2
Interesting MX observation.
Posted: Sun 05 Jun 2016 10:53 pm
by ICP963
Right now in Chandler Arizona it is HOT. 114.0 with a Dew Point of 21.4 Humidity of 4%.
My cloud base gauge is rolling from 0 to just over 21,000 and back starting at 0 and back up over and over again. I've never noticed that behavior before, but then again it is HOT!!!
Re: Interesting MX observation.
Posted: Mon 06 Jun 2016 5:21 am
by ICP963
Same day, the temp is now below 90 (10:15PM) and the dew point is 45.5 and Humidity is at 22% with the Cloud Base sitting at a steady 10,000 feet. I'll have to check again tomorrow afternoon and see what the cloud base readings are since we are expected to be set to broil again tomorrow.
Gotta love living in the desert.

Re: Interesting MX observation.
Posted: Mon 06 Jun 2016 7:30 am
by steve
Which gauge - the one in the UI or on a web site, or both? If one on a web site is doing it, please give a link.
You'll need to look at the actual data being supplied to the gauge to determine whether the value in the data is flipping between zero and the actual value, or whether the data is OK and it's some kind of glitch with the gauge.
Re: Interesting MX observation.
Posted: Mon 06 Jun 2016 9:49 am
by mcrossley
I think it will be the gauge code, a quick way to check if it is what I think it is would be to change the following parameter in gauges.js...
Code: Select all
niceScale : true,
//to
niceScale : false,
This may have other effects, so it is not a solution to the problem if it does fix the cloud base scale.
Re: Interesting MX observation.
Posted: Mon 06 Jun 2016 8:48 pm
by ICP963
steve wrote:Which gauge - the one in the UI or on a web site, or both? If one on a web site is doing it, please give a link.
You'll need to look at the actual data being supplied to the gauge to determine whether the value in the data is flipping between zero and the actual value, or whether the data is OK and it's some kind of glitch with the gauge.
I'm guessing the difference between the temp and the dew point was too great and that the reading was over 20,000 feet and it got lost.
I will try the .js change and see what I get later this evening when I get home. (If I can figure out how

)
Roy
Re: Interesting MX observation.
Posted: Wed 08 Jun 2016 6:16 am
by ICP963
mcrossley wrote:I think it will be the gauge code, a quick way to check if it is what I think it is would be to change the following parameter in gauges.js...
Code: Select all
niceScale : true,
//to
niceScale : false,
This may have other effects, so it is not a solution to the problem if it does fix the cloud base scale.
I just now made this change and I will have to wait and see if it happens again. Everything worked fine today and the Cloud Base stayed below 20,000 Ft today.
Will keep watching, but thankfully the temp is not predicted to be ashot as the last few days. 105 for us is a cold spell this time of year.

Re: Interesting MX observation.
Posted: Tue 14 Jun 2016 4:51 am
by ICP963
ICP963 wrote:Will keep watching, but thankfully the temp is not predicted to be as hot as the last few days. 105 for us is a cold spell this time of year.

Predicted high for this weekend is 119F so I will keep an eye out for the rolling guage over the weekend. That is if I can keep from melting.

Re: Interesting MX observation.
Posted: Sun 19 Jun 2016 11:07 pm
by ICP963
quote="ICP963"]
ICP963 wrote:Will keep watching, but thankfully the temp is not predicted to F (Highest I have recorded with Cumulus since I started running it.)be as hot as the last few days. 105 for us is a cold spell this time of year.

Predicted high for this weekend is 119F so I will keep an eye out for the rolling gauge over the weekend. That is if I can keep from melting.

[/quote]
Current Temp is 117.9F with a Dew Point of 29.3F which would put the cloud base at 2013 ft. The gauge reads to 25,000 Ft and it continues to roll up from 0 to about 20,000 and then starts over at 0.
My high temp today so far has been 119.1 (Highest I have recorded since I started running Cumulus).
Re: Interesting MX observation.
Posted: Mon 20 Jun 2016 8:35 am
by mcrossley
I would need to see the page to take this diagnosis any further.
Re: Interesting MX observation.
Posted: Mon 20 Jun 2016 8:29 pm
by ICP963
mcrossley wrote:I would need to see the page to take this diagnosis any further.
It is the page that you get when you log into MX via the internal ip address, mine is
http://192.168.0.106:8998/gauges.html so I am not sure how you would get to it.
I use Teamviewer to get into another machine in the house and then remote desktop into the pi if I am not at home and need to get to it.
Re: Interesting MX observation.
Posted: Mon 20 Jun 2016 8:33 pm
by mcrossley
Ah, OK, do you have a web page and see the same behaviour on that?
Re: Interesting MX observation.
Posted: Mon 20 Jun 2016 11:00 pm
by ICP963
mcrossley wrote:Ah, OK, do you have a web page and see the same behaviour on that?
No webpage at this time. I have been thinking about setting one up, but haven't had the time lately.
It is a rare enough happening that it is not a huge deal. Not sure if the old Cumulus version had the problem, and I am usually away from the Pi during the day when it is blazing hot so I can't easily switch the USB port to another computer to see if the old one does it too.

Re: Interesting MX observation.
Posted: Mon 20 Jun 2016 11:07 pm
by water01
Cumulus 1 cannot have the same problem, as the gauges in Cumulus MX dashboard are a slightly modified version of the Steel Series gauges created by Mark Crossley, and didn't exist in the Cumulus 1 dashboard only on the website if you installed them yourself as a replacement of the standard web site gauges.
Mark I think was asking to see your web page as he assumed you had used the standard MX web site where his gauges are installed and used as the standard MX gauges and it would have been easier to debug what was going on.
Re: Interesting MX observation.
Posted: Tue 21 Jun 2016 12:03 am
by ICP963
water01 wrote:Cumulus 1 cannot have the same problem, as the gauges in Cumulus MX dashboard are a slightly modified version of the Steel Series gauges created by Mark Crossley, and didn't exist in the Cumulus 1 dashboard only on the website if you installed them yourself as a replacement of the standard web site gauges.
Mark I think was asking to see your web page as he assumed you had used the standard MX web site where his gauges are installed and used as the standard MX gauges and it would have been easier to debug what was going on.
I thought that might be the case with the old Cumulus so I will not waste my time running the line over to the old laptop that ran it before the Pi and MX. And, like I mentioned, I am sure that the temp extreme and LOW dew point are not a common happening I am not that concerned about it. Everything seems to get logged correctly, just not displayed well in a VERY extreme condition that is very rare.
I do appreciate the help and comments though.
Re: Interesting MX observation.
Posted: Tue 21 Jun 2016 7:26 am
by mcrossley
OK, I just took a look at the console - it is a "niceScale thing".
Change line 137 of gauges.js to...
fixes it.
I have changed the main gauges code base to apply a permanent fix.