Welcome to the Cumulus Support forum.

Latest Cumulus MX V3 release 3.28.6 (build 3283) - 21 March 2024

Cumulus MX V4 beta test release 4.0.0 (build 4018) - 28 March 2024

Legacy Cumulus 1 release v1.9.4 (build 1099) - 28 November 2014 (a patch is available for 1.9.4 build 1099 that extends the date range of drop-down menus to 2030)

Download the Software (Cumulus MX / Cumulus 1 and other related items) from the Wiki

Pressure Graph Wonky

Discussion specific to Fine Offset and similar rebadged weather stations
Post Reply
User avatar
akasonny
Posts: 232
Joined: Mon 15 Jun 2009 4:43 am
Weather Station: Zephyr TD-1000
Operating System: Windows 10
Location: Sahuarita, AZ, USA
Contact:

Pressure Graph Wonky

Post by akasonny »

A week ago, I noticed that, on the graphs, the pressure reading was flat. I assumed, after 10 years, the Fine Offset unit I'm using failed and I replaced it. The problem persisted. I, then, uninstalled Cumulus, deleted the directory folder entirely, and re-installed. The problem still persists.

I'm not a newbie but I guess I'm a neophyte (still haven't figured out CumulusMX). I've researched everywhere for this problem but I can't seem to get a handle on this. I did a screen capture of the pressure part of the trends screen for clarification and attached it here.

Anyone with a clue?
You do not have the required permissions to view the files attached to this post.
AllyCat
Posts: 1124
Joined: Sat 26 Feb 2011 1:58 pm
Weather Station: Fine Offset 1080/1 & 3080
Operating System: Windows XP SP3
Location: SE London

Re: Pressure Graph Wonky

Post by AllyCat »

Hi,

You appear to have just one "rogue" zero in the data file at about 2.30 pm. Correct that (in the September file) using a text editor or the built-in Editor and you might find that the remaining data is not "flat" (just swamped by the autoscaling).

Pressure is measured (only) by the Console, did you replace the whole station? Similar spikes with two totally different stations suggest a USB or PC problem. FO stations do occasionally produce "spikes", are they in the Console's data log if you repeat the download (e.g. after installing the backup files from an earlier date)?

Cheers, Alan.
User avatar
akasonny
Posts: 232
Joined: Mon 15 Jun 2009 4:43 am
Weather Station: Zephyr TD-1000
Operating System: Windows 10
Location: Sahuarita, AZ, USA
Contact:

Re: Pressure Graph Wonky

Post by akasonny »

Hi Alan,
...and thank you for your response. I searched the Sep17log.txt file in the data folder but couldn't see it.

I didn't replace the whole station but just the transmitter which monitors pressure, rain and temp. The rain and wind speed mechanical items remain untouched. When I re-installed Cumulus, I only saved the Cumulus.ini file and the data folder and deleted everything else (probably not the correct way to do that, in hindsight). Then I re-inserted them back into the Cumulus folder after re-installing Cumulus. The control unit (in the house) remains the original.

Today I see the pressure graph functioning once again...but it, ironically, looks much like the indoor temperature graph. You can view this at www.cactusjunction.info if you have a mind to.

As I said, I'm a neophyte here but I've been using Cumulus for something like 7 years. I could just dump the Sep17log.txt file I suppose...?
sfws
Posts: 1183
Joined: Fri 27 Jul 2012 11:29 am
Weather Station: Chas O, Maplin N96FY, N25FR
Operating System: rPi 3B+ with Buster (full)

Re: Pressure Graph Wonky

Post by sfws »

Have a read of http://wiki.sandaysoft.com/a/FAQ#Cumulu ... my_station. You say you only kept the "data" folder, but what is your oldest folder in the "backup" (or "backup\daily") folder? You can stop Cumulus, then copy all the files from that oldest folder into your current data folder to (as Alan says) repeat the download from your console when you restart Cumulus.

If you look at your "dayfile.txt" file http://wiki.sandaysoft.com/a/Dayfile.tx ... n_the_file you can quickly see which dates might have the odd pressure values.
Then in your "Sep17log.txt" http://wiki.sandaysoft.com/a/Monthly_lo ... ion_1.8.5: you only need to look at the pressure values on those dates. Delete the line from the monthly log that contains the rogue value, or change the value at that time by averaging values on the line before and line after.

Look at cumulus\web\monthlyrecordT.htm, 3 lines up from the bottom, delete "//" at the start of the line. That will make http://www.cactusjunction.info/monthlyrecord.htm display September. That shows you that it looks like your pressure units got changed at some point between 12th and 18th, that might have contributed to your problem, but it also shows you need to correct some of your extreme figures.


It is not the transmitter (what you replaced) that measures pressure, but as Alan says the console (that you say you did not replace). Inevitably there will be some correlation between the pressure and (indoor) temperature.
User avatar
akasonny
Posts: 232
Joined: Mon 15 Jun 2009 4:43 am
Weather Station: Zephyr TD-1000
Operating System: Windows 10
Location: Sahuarita, AZ, USA
Contact:

Re: Pressure Graph Wonky

Post by akasonny »

Thanks for chiming in here.

I will follow your instructions. I got a little lost in the Sep data.txt file. The dayfile will surely give me a better clue where to look.

Obviously, I wasn't aware of the correlation between the inside temp and the barometric reading. All this time, I had assumed the barometer was located out on the sensor. I've learned a lot in just this short exchange.

I'll post again, once I've completed these steps.

Many thanks!
User avatar
akasonny
Posts: 232
Joined: Mon 15 Jun 2009 4:43 am
Weather Station: Zephyr TD-1000
Operating System: Windows 10
Location: Sahuarita, AZ, USA
Contact:

Re: Pressure Graph Wonky

Post by akasonny »

Interestingly enough, the pressure being reported is accurate but the graph doesn't reflect that.

An exam of the dayfile.txt and the Sept files don't reveal any errors..the pressure range is well within reason on all data lines. I had "48 hours" checked on the Cumulus console, in the past, for graph use. When I switched to 12 hours, the graph for pressure returned to something fairly normal now. Switching back to 48 hours, the line was flat on the graph. Though there's still a fairly close correlation there with the inside temperature, its finally doing something.

I find myself more confused than ever.
sfws
Posts: 1183
Joined: Fri 27 Jul 2012 11:29 am
Weather Station: Chas O, Maplin N96FY, N25FR
Operating System: rPi 3B+ with Buster (full)

Re: Pressure Graph Wonky

Post by sfws »

I mentioned some correlation (at higher temperatures, molecules of air have more activity; remember pressure is measuring force over unit area), one would not expect close correlation.

The graphs are calculated from frequent measurements coming from your station providing you leave Cumulus running. The log only includes the measurements at the logging interval. You would need to examine the Cumulus recent history web tags http://wiki.sandaysoft.com/a/Webtags#Recent_History to see the figures being used to produce the graphs.
AllyCat
Posts: 1124
Joined: Sat 26 Feb 2011 1:58 pm
Weather Station: Fine Offset 1080/1 & 3080
Operating System: Windows XP SP3
Location: SE London

Re: Pressure Graph Wonky

Post by AllyCat »

Hi,

There's quite an art to reading graphs, particularly if they are "auto-scaling" as here. You need to look at the actual figures along the vertical axis. Also, using a longer horizontal axis (time) will tend to include a wider range of data and reduce the "sensitivity" of the vertical scale.

Thus, the pressure graph looked "flat" because the bad data point had compressed the scale very considerably. Conversely, your internal temperature was hardly changing at all, but it was stretched out to "fill" the axis. The "staircase" nature of the line is another clue that the scale is highly expanded.

There have been some reports that the indicated barometric pressure from the Fine Offset Consoles does vary a little in correlation with the internal temperature, but the current change of 0.2 inch Hg with a 4 degrees F change of internal temperature is probably just a coincidence. The outside temperature is changing by much more, but those graphs don't appear to correlate particularly well.

Cheers, Alan.
Post Reply