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

Lost Sensor Contact!!

Discussion specific to Fine Offset and similar rebadged weather stations
gordonwh40
Posts: 20
Joined: Mon 24 Aug 2009 7:45 am
Weather Station: wh1081
Location: Qld Australia

Re: Lost Sensor Contact!!

Post by gordonwh40 »

Hi Gina
It is is difficult to tell from your photographs, but is your mast effectively earthed? Even dry wind over the mast can build up a static charge, and when it reaches a critical level discharges to earth causing a spike of RF interference if the mast is not effectively earthed.
This adventure will rival days of our lives when it is eventually solved.
gordonwh40
EvilV
Posts: 127
Joined: Wed 27 Jan 2010 11:28 am
Weather Station: Watson W-8261
Operating System: XP
Location: NE7 7QE Newcastle upon Tyne UK

Re: Lost Sensor Contact!!

Post by EvilV »

Gina wrote:I'm thinking of the following, from Maplin :-


Does this sound sensible, EvilV? (or anyone else au fait with this subject)
I think I'd have put the capacitors on the other side of the choke, on the grounds that you would be connecting the RF pulse to ground BEFORE any of it was conducted through the choke. However, I was only ever a hobby electronics dabler, whereas you seem to have a proper engineering background.

Actually Gordonwh40's suggestion about static is quite persuasive. You'd better know whether the spikes only take place in low humidity. I don't suppose that DEvon gets that much dry wind, poking out into the Atlantic as it does, but earthing the mast might be a good test before you make your circuit.

Another thing I would do, is uncouple the signal wire from the mast, roll it up and lay it on the ground to see if the spikes go away. There might be an occasional glitch in the socket on the sensor unit, caused by wobbling in the wind or maybe even thermal expansion / contraction. You could simulate movement by manually wobbling the cable.

This thread is turning into a compelling epic. It reminds me of Apollo 13, where brave and resourceful engineers struggle to identify and solve the problems, while a massive team wait with baited breath for the next instalment. :)
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

gordonwh40 wrote:Hi Gina
It is is difficult to tell from your photographs, but is your mast effectively earthed? Even dry wind over the mast can build up a static charge, and when it reaches a critical level discharges to earth causing a spike of RF interference if the mast is not effectively earthed.
This adventure will rival days of our lives when it is eventually solved.
gordonwh40
Hi Gordon, thanks for your input :)

Like EvilV, I think you have a very good thought there. The bottom of the mast sits on the ground but isn't buried. This allows me to put it up and take it down by pivoting at the bottom. We have had a fortnight without measurable rain though there has been some dew. It could be that the base is not earthed well enough. There was a period with no wind but there were spikes. OTOH last night there was no wind and no spikes, much to my surprise.
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

EvilV wrote:I think I'd have put the capacitors on the other side of the choke, on the grounds that you would be connecting the RF pulse to ground BEFORE any of it was conducted through the choke. However, I was only ever a hobby electronics dabler, whereas you seem to have a proper engineering background.
Yes, I do have an engineering background, particularly electrical and electronics. Mains RF filters usually have a capacitor across line and neutral then a choke in each followed by capacitors to earth. I could easily add capacitors at the input too. The output capacitors do the main work though. The chokes provide a high impedance (resistance) at high frequencies while being virtually short-circuit at DC and low frequencies. Capacitors are the opposite - no affect on DC or very low frequencies etc. Thus the choke (inductance) and capacitor provide a high attenuation at radio frequencies while passing DC and LF untouched.
Actually Gordonwh40's suggestion about static is quite persuasive. You'd better know whether the spikes only take place in low humidity. I don't suppose that DEvon gets that much dry wind, poking out into the Atlantic as it does, but earthing the mast might be a good test before you make your circuit.
Agreed! As I said to Gordon, we have had a spell of dry weather for the last couple of weeks. A proper earth is easy enough to implement. I'll do that.
Another thing I would do, is uncouple the signal wire from the mast, roll it up and lay it on the ground to see if the spikes go away. There might be an occasional glitch in the socket on the sensor unit, caused by wobbling in the wind or maybe even thermal expansion / contraction. You could simulate movement by manually wobbling the cable.
I've tried wobbling the connector by hand - nothing.
This thread is turning into a compelling epic. It reminds me of Apollo 13, where brave and resourceful engineers struggle to identify and solve the problems, while a massive team wait with baited breath for the next instalment. :)
:lol: Yes, it is rather :) I hope I'm quite resourceful but I don't know about brave :lol: I enjoy a challenge and I enjoy a puzzle. I like to keep my mind active :) I do the testing but provide ideas as part of the team, of course :) I'm grateful to others for the useful suggestions :)
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

10:50 18 Apr 2010 Connected wind sensors again to check for interference. Once the spikes come back I'll earth the mast. I want to be sure things are still the same. If I were to earth the mast and got no more spikes I wouldn't know if I'd fixed it or just that the interference had stopped.

With the last test - separate cable from TX unit wrapped round sensor cable and mast - we had two spikes then nothing.
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

No spikes yet!
24hrs-Temperature-only-f.png
You do not have the required permissions to view the files attached to this post.
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
Charlie
Posts: 363
Joined: Thu 04 Feb 2010 12:22 pm
Weather Station: 1wire-Cumulus & Fine Offset
Operating System: Windows 7
Location: Whitehorse, Yukon Territory, Canada

Re: Lost Sensor Contact!!

Post by Charlie »

Hi Gina,
If you do decide to build the filter, you could likely increase either L or C by an order of magnitude for better filtering and still only have unmeasurable negative effect on switching times of the counter relays. What were you thinking of using as a ground reference to route the offending currents to?
I might be tempted to try common mode chokes if there is still some spiking once the filter is tried. It's a pity that a spike couldn't be captured on a storage scope - it's a lot easier if you know exactly what you're dealing with. :P
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

Charlie wrote:Hi Gina,
If you do decide to build the filter, you could likely increase either L or C by an order of magnitude for better filtering and still only have unmeasurable negative effect on switching times of the counter relays. What were you thinking of using as a ground reference to route the offending currents to?
I might be tempted to try common mode chokes if there is still some spiking once the filter is tried. It's a pity that a spike couldn't be captured on a storage scope - it's a lot easier if you know exactly what you're dealing with. :P
I can certainly increase C but 1mH is the largest value of choke sold by Maplins. For the ground for the Cs I was thinking of using a piece of copper pipe driven into the ground near the base of the post holding the Stevenson screen and TX unit etc. Then wiring that to the junction of the Cs.

I do have a hand-held digital storage oscilloscope but the problem is setting it up to catch the interference. The frequency of occurrence is rather low! Also, the LCD screen is difficult to see outdoors.

Current situation is that in spite of re-connecting the wind sensors there are no spikes. Well, not as yet. So I have just re-connected the rain gauge and fan. It rather looks as if the cause of the spikes has ceased. Many might say "that's good" but I want to find the problem. Intermittent problems that just "go away" I can't stand!!! There is always the feeling that they'll come back!
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

EDITED I thought I had magic with the rain plot but it was the way I was plotting the data. All the same the rain gauge or it's cable appeared to be picking up interference and reading feet of virtual rain.
8hrs-Rain.png
Now had two spikes on the temperature plot.
24hrs-Temperature-only-h.png
You do not have the required permissions to view the files attached to this post.
Last edited by Gina on Sun 18 Apr 2010 6:02 pm, edited 2 times in total.
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

18:00 18 Apr 2010 Earthed mast. Just wind sensors connected - no rain gauge, no fan

EDIT The earth wasn't good enough - got a spike. On checking the conductivity from mast to earth with a multimeter I got <2Mohm (off the scale). So did it properly...

19:55 18 Apr 2010 Re-did mast earth. 9" copper pipe hammered into ground and connected to mast.

Resistance using just 3/4" tip of MM probe in soil to mast showed 1.2Mohm. The resistance mast to earth is probably a lot lower but, in any case should prevent any build-up of static charge.

Temp plot showing spikes without mast properly earthed :-
24hrs-Temperature-only-i.png
You do not have the required permissions to view the files attached to this post.
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

Here is a live plot of temperature only. This will be no more than 10 mins out of date as the data is uploaded every 10 mins.Image

At time of posting we have one spike, just before 21:00. Wind sensors are connected, rain gauge and fan are not. The mast was earthed as from about 20:00 on 18 April. This would appear to indicate that earthing the mast has not cured the problem.
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

Plenty of data spikes overnight so earthing the mast has not helped. Next stage will be to lower the mast, remove the sensors and replace the sensor cable with another cable to the mast top, then stick the mast back up again. Can then see if it's RF pick-up by cable or mast. While I've got the anemometer and vane units down I may take them apart and see if there are any dry joints or suchlike. I'd like to see how they're made, anyway. I'll take photos.
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

10:45 19 Apr 2010 Lowered mast, removed wind sensors (unplugged from coupler half way up mast), extended cable to TX unit and taped it to the mast. Excess cable length fed down inside the mast. See above (top of page) for live temperature plot.
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
EvilV
Posts: 127
Joined: Wed 27 Jan 2010 11:28 am
Weather Station: Watson W-8261
Operating System: XP
Location: NE7 7QE Newcastle upon Tyne UK

Re: Lost Sensor Contact!!

Post by EvilV »

I've opened the 24 hour temp chart in a new window and saved the url into my bookmarks bar so I can open it any time irrespective of whether I am on this forum.

At the time of posting, there seem to be no spikes. Looks like its the wind gadgets.


Image

Wish we had 15C. It's 7.5 here!


LATER EDIT:

This is getting a bit weird because I was under the impression that you had previously isolated the wind apparatus and were getting spikes with a new extension cable plugged into nothing but strung up the mast. Now it seems you have a new cable plugged into nothing but no spikes. You already ruled out the static build up theory (except that I wouldn't count a 9" copper pipe as a good earth myself. In my Radio Ham days I used at least a one metre copper rod hammered down into the ground and watered it regularly in dry weather) so why have the spikes gone now? Is it because you are now shielding the cable by running it up the inside of an earthed tube, perhaps? And what will happen when you re-connect your wind gear at the top?

I may have missed something here in your current arrangements.... It si getting complicated to keep up for an 'auld git'. :)
Gina
Posts: 1885
Joined: Sat 21 Feb 2009 12:41 pm
Weather Station: Nothing working ATM - making one
Operating System: OS X, Linux Mint, Win7 & XP
Location: Devon UK

Re: Lost Sensor Contact!!

Post by Gina »

EvilV wrote:I've opened the 24 hour temp chart in a new window and saved the url into my bookmarks bar so I can open it any time irrespective of whether I am on this forum.
Good idea :)
At the time of posting, there seem to be no spikes. Looks like its the wind gadgets.
Looks like it. I shall be leaving things as they are for a few hours, so we'll see.
Wish we had 15C. It's 7.5 here!
Yes, lovely day here, continuous hazy sunshine and warm. Unfortunately, I have indoor jobs I have to get done! Sorry you haven't got the same weather.
Gina

Sorry, no banner - weather station out of action. Hoping to be up and running with a new home-made one soon.
Post Reply