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 4017) - 17 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

Ecowitt WS80 - winter issue in the lowlands

GW1000 WiFi gateway
User avatar
Carbonara
Posts: 27
Joined: Mon 16 Nov 2020 10:27 pm
Weather Station: Davis and Ecowitt
Operating System: DietPi 7.5.2 ( debian bullseye)

Re: Ecowitt WS80 - winter issue in the lowlands

Post by Carbonara »

can't see the trasducer ... the firmware is update ?? https://osswww.ecowitt.net/uploads/2021 ... %20Kit.zip
User avatar
HansR
Posts: 5870
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by HansR »

I thought already the photo was no good. Can't do better now but it looks to me it is similar as in your photos. The thing is two months old, I don't assume they send out old equipment.

Wrt the firmware : I don't know. I can't check that in the field. Afaik I need to connect it to the USB and run the PC software to install the firmware. It is a mudpool where it stands and I can't get a ladder there so that has to wait until spring when the terrain is accessible again.

But I assume the same argument holds: they would not send it out without recent firmware I guess.
The GW1100 has firmware GW1100A_V2.0.8

But all in all, as I said, seen from the pov that it restarted measuring immediately after I emptied it of the water, I don't think it is a software issue nor can it be solved by software. The problem is water. Afais it, is the water on and below the gauze. That is why I asked: do photo's exist of the cavity below the gauze?

And btw: Ecowitt suspects water is coming in through the top glass which covers the electronics. That is not the case either with my device.

[EDIT:]
The firmware if from 9 july 2021, my order was sent 11/10/21
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
User avatar
HansR
Posts: 5870
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by HansR »

I had a problem again today. Now it obviously was the WS80 itself and not one of the GW1100's. So I took down the WS80, confirmed it was the new design and upgraded the firmware. Not knowing what firmware was inside, now it definitely is 1.2.0. We'll see.
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
Hoeveld
Posts: 58
Joined: Thu 31 Dec 2020 10:18 am
Weather Station: Ecowitt WS80 & WH40 - GW1100
Operating System: Raspbian GNU/Linux 10 (buster)
Location: Newark-on-Trent
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by Hoeveld »

To continue WS80 issues............I too like Hans are having regular wind reading spikes, I have checked with met office local data as a rough cross check and my data is wrong, by over 30mph.

There are underlying issues with the WS80 I believe. I will double check the firmware.

I reported issue to Ecowitt who when down the lines of "water causing issues", I pointed out that as the WS80 is an external device, so they are shipping a replacement.

In the meantime I would like to tidy up my data and remove the spikes.

Today I had a spike recorded at 82mph, my best guess is that the gust was say 52mph.

The system "Locked Up" and flat lined for approx 90 minutes. I was lucky today as previous flat lines have lasted over 12 hours.

What would be an easy, quick fix to clean up my cumulsmx data.??

I would think that wind runs as well as wind averages will be effected ?

Many Thanks

Cordially

Paul
You do not have the required permissions to view the files attached to this post.
User avatar
HansR
Posts: 5870
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by HansR »

Hi Paul,
Hoeveld wrote: Mon 21 Feb 2022 2:36 pm What would be an easy, quick fix to clean up my cumulsmx data.??
If have the following method for correction of values of the WS80:

With spikes I always only correct the following day to see how things develop, flatline I act ASAP.

First I check the environment and the neighbouring stations to see if correction is really needed (e.g. my 131 km/h record recently I did NOT correct). Spikes do not always need correction. It is my experience that the ultrasone wind measurement is much more sensitive / volatile than the cup anemometers. That might be due to the inertia of the last. Not sure though, no doubt this can be a whole other discussion, no doubt most likely at WXForum this has already been done (see e.g. Mark's post).

If the WS80 does flatline - I regard this as the biggest problem, it usually also is on a spike - I reboot the GW1100, if that does not work I go to the WS80 and reset it (push the reset button). Try to have the flatline period as small as possible so it needs checking regularly which makes it a stressful piece of equipment. With the recent storms the droplet theory is bull: droplets do not form because both temperature and wind are too high.

If correction is needed then I shutdown CMX, remove the erroneous lines from the logs and start CMX again.
Then I correct subsequently the record in the Dayfile.
Then as a last step, I edit the records with the record editors.

I rerun my CUtils and that should do it.

The biggest problem is that physical access is often needed which is hard if you mounted it at a pole or a chimney.
It is a disturbing device at the moment, I hope the issues will be solved soon.

[EDIT:] And don't bother about the charts. The values of those are in the database and they will just phase out while not influencing your data.

Regards,
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
Hoeveld
Posts: 58
Joined: Thu 31 Dec 2020 10:18 am
Weather Station: Ecowitt WS80 & WH40 - GW1100
Operating System: Raspbian GNU/Linux 10 (buster)
Location: Newark-on-Trent
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by Hoeveld »

Many Thanks for that,

For clarity "Logs" ...... ( attached data file directory )

Change both the FEB22log & dayfile ?

Thanks for your help.........

Cordially

Paul
You do not have the required permissions to view the files attached to this post.
User avatar
HansR
Posts: 5870
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by HansR »

Hi Paul,

Yes, erroneous records from the monthly log I do remove to prevent the errors to come back creeping in. Because when e.g. editing records you can choose to read the monthly logs for record examples beside those shown from the dayfile. And yes, differences exist but when time passes by you may get confused. If ever is decided to use the monthly logs for calculations somewhere the old data will haunt you. So although CMX does not reread the logs normally: yes, I do delete the erroneous lines in the monthly log.

If there are many lines to remove a gap may exist of more than a day. That is a pity but it won't break down the system.

Dayfile is the summary where created from all data in the standard logfile but is not recalculated afaik when logs are corrected. This means that data which are evidently wrong, especially when influencing the records and averages (so we're not talking about a tenth of a degree) need to be corrected.

Then when editing the records, you get the value of the dayfile as proposal (and may have the monthly log value as comparison).

NOTE: CUtils utilises the dayfile and the monthly log data for its calculations and does not use webtags or the inifiles to get charts, values or records so it goes back to its origin. Therefore CUtils may be more susceptible to erroneous data. Removal of true erroneous data from the monthly log therefore is important as it may have more effect by its presence than by its absence (e.g. check what an erroneous series of high wind data would do on the wind charts in case of presence/absence. The same holds for solar). For e.g. temperature, CUtils takes the dayfile values as true and does not look elsewhere.

Hope this helps.

Regards,
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
LindaFNM
Posts: 151
Joined: Wed 06 Jan 2021 3:01 pm
Weather Station: Ecowitt
Operating System: Raspberry Pi

Re: Ecowitt WS80 - winter issue in the lowlands

Post by LindaFNM »

I’m afraid this isn’t what you both want to hear, but I had these sort of issues with a WS80 starting last October, when we had prolonged wet and damp weather. Odd spikes that stayed flatlined which progressively got worse. Wind only at first, which was manageable then affected temperature and humidity. The unit became unusable by late December, on retrieval the top was visibly wet inside with heavy condensation. To their credit, the whole unit was replaced by Ecowitt. My data is accurate only in the dayfile for this period till the new unit arrived as it became impossible to edit any logs due to the length of the flatlining. I had previously tried restarting everything and changing from my new GW1100 to the GW1000 and back again. I’m reporting to Ecowitt.net and they were able to see my data.

Mine was water ingress, after a holiday in the warm and dry, and gentle drying with a hairdryer everything started working again fine until it went back outside. I live in a very windy area and rain is often quite horizontal.
User avatar
HansR
Posts: 5870
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by HansR »

LindaFNM wrote: Tue 22 Feb 2022 3:19 pm I’m afraid this isn’t what you both want to hear, but I had these sort of issues with a WS80 starting last October, when we had prolonged wet and damp weather. Odd spikes that stayed flatlined which progressively got worse. Wind only at first, which was manageable then affected temperature and humidity. The unit became unusable by late December, on retrieval the top was visibly wet inside with heavy condensation. To their credit, the whole unit was replaced by Ecowitt. My data is accurate only in the dayfile for this period till the new unit arrived as it became impossible to edit any logs due to the length of the flatlining. I had previously tried restarting everything and changing from my new GW1100 to the GW1000 and back again. I’m reporting to Ecowitt.net and they were able to see my data.

Mine was water ingress, after a holiday in the warm and dry, and gentle drying with a hairdryer everything started working again fine until it went back outside. I live in a very windy area and rain is often quite horizontal.
I don't see why I would not want to hear your observations.
I would like to distinguish between spikes and flattening of the line for data from WS80 and I can tell you that I live in area which might win the competition in wetness and windiness of the English marshes (excluding some isles in the North and west). Check the data of this winter on my site.

Agreed that if it is truly water ingress then Ecowitt rightly replaced the unit and resetting might not help. I don't think that is the case with my unit : it is raining cats and dogs over here now for more that a week with three storms passing by during that time. The device mostly worked well and when it failed (once) a reset worked (And I upgraded the firmware during a short sunny spell).

So in summary: I presume there are more points of failure (water, firmware and I think too there is an issue with the gw1100 but maybe that is related to the firmware). That does not improve the situation.

Your comment says, if I understand it well, that for prolonged periods of bad data (flatlining) it is difficult to remove data from the monthly logs. That I agree because in CMX you can only remove one line at a time. However, the workaround is to stop CMX - the Ecowitt devices can catch up after prolonged stops of CMX - edit the logfile(s) in a suitable editor (e.g. notepad++) and remove whole blocks of lines as needed. After that, save and restart CMX.

Single spikes can always be corrected as they are single entries by definition in a logfile but even then the manual edit of the logfile may be preferred by the user.

This may not be the prescribed procedure but it definitely makes a workaround and life liveable in an imperfect world.
So I agree that longer periods are more difficult to correct but not impossible.

Nevertheless, so far the WS80 is a labour intensive, unsettling device which needs to be daily observed on functioning.
I hope it can be improved and that the new firmware works out and that Ecowitt improves the design of the WS80. I give it two months to prove itself. I like the concept and compactness of the device but will move to the WS68 if it goes on like this.
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
LindaFNM
Posts: 151
Joined: Wed 06 Jan 2021 3:01 pm
Weather Station: Ecowitt
Operating System: Raspberry Pi

Re: Ecowitt WS80 - winter issue in the lowlands

Post by LindaFNM »

Hans,

I guess I didn’t phrase the beginning of my post well!

I think with my unit, there was probably water ingress which was dried out by the sun periodically until the weather became very wet and it just couldn’t dry out.

Hopefully, yours doesn’t have the issues that mine had, it has been somewhat frustrating. I have reasonable dayfiles, and the advantage that I also have a Davis Vue which I download manually, and which I can obtain some missing data from. This is another project which I need to tackle, but I’m doing more fun things at the minute :) .

I agree that the WS80 is a mixed package. I am grateful that it has introduced me to a whole new world of weather interest including entry to the modern way of doing things, but maybe there is room for a lot more real world testing. It does need to be able to be out in literally all winds and weather and withstand this, but also to be relatively easy to manage. It’s not the best thing to dismantle the unit from its pole to have to update firmware or reset it etc, it would help if it could be reset remotely. I do very much like the Ecowitt console which is very informative.
User avatar
HansR
Posts: 5870
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by HansR »

LindaFNM wrote: Tue 22 Feb 2022 8:16 pm I agree that the WS80 is a mixed package. I am grateful that it has introduced me to a whole new world of weather interest including entry to the modern way of doing things, but maybe there is room for a lot more real world testing. It does need to be able to be out in literally all winds and weather and withstand this, but also to be relatively easy to manage. It’s not the best thing to dismantle the unit from its pole to have to update firmware or reset it etc, it would help if it could be reset remotely. I do very much like the Ecowitt console which is very informative.
Linda,
I think we agree on the WS80 which apparently has multiple issues.
I don't have the console but I like the modularity and ease of configuration very much. I sincerely hope the issues can be overcome.
In the meantime, I think corrective actions can be taken and I hope my description makes that clear. If there are errors in it, let me know.
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
broadstairs
Posts: 744
Joined: Thu 14 Aug 2008 7:17 am
Weather Station: Ecowitt GW1003/GW1103/GW2000
Operating System: Windows 7 and Linux
Location: Broadstairs, Kent, UK
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by broadstairs »

Just to add my comments here about the freezing of wind readings. I had this happen on my new WS80 which was supplied under warranty because my original from last March 2021 went through batteries like nobody's business but never froze up. Now this new one froze during storm Eunice last week and it had only been installed about 5 days, I had to leave resetting till the following day! Its been fine since even during the next storm when it measured 67mph gust and regular wind speeds or 30-40mph. I have reported this to Ecowitt but not yet heard back about this. I also found out that it is recommended to reset the device on initial install after putting in the batteries which I had not done.

If you here who have had these issues have not reported them to Ecowitt please do because the more reports they get the more chance there is of improvement.

I have a thread at https://www.wxforum.net/index.php?topic=43510.0 which I will keep updated with any feedback I get from Ecowitt.

Stuart
User avatar
HansR
Posts: 5870
Joined: Sat 20 Oct 2012 6:53 am
Weather Station: GW1100 (WS80/WH40)
Operating System: Raspberry OS/Bullseye
Location: Wagenborgen (NL)
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by HansR »

@broadstairs: yes your point for reporting to Ecowitt I think is well followed up by all I know of.

With your latest observation I can add that I don't think the issue has to do with droplets, water or firmware version (it may). The reset at first install is new for me. It may help, but it should not be necessary. Those are things which can easily be handled by software. And beside that it would be very handy if the firmware could be remotely updated and the device could be remotely rebooted (and that the present firmware could be read on WSView).

Anyway, we'll see... The flattening / freezing being the main problem I think.
Hans

https://meteo-wagenborgen.nl
CMX build 4017+ ● RPi 3B+ ● Raspbian Linux 6.1.21-v7+ armv7l ● dotnet 8.0.3
broadstairs
Posts: 744
Joined: Thu 14 Aug 2008 7:17 am
Weather Station: Ecowitt GW1003/GW1103/GW2000
Operating System: Windows 7 and Linux
Location: Broadstairs, Kent, UK
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by broadstairs »

The issue with any remote action is that the WS80 is transmit only!

Stuart
Hoeveld
Posts: 58
Joined: Thu 31 Dec 2020 10:18 am
Weather Station: Ecowitt WS80 & WH40 - GW1100
Operating System: Raspbian GNU/Linux 10 (buster)
Location: Newark-on-Trent
Contact:

Re: Ecowitt WS80 - winter issue in the lowlands

Post by Hoeveld »

Just to update on the issue with flatlines on WS80..........Yesterday in Notts - UK we had snow and sleet flurries ( first of this year ) the WS80 flatlined at lunchtime, the WS68 keeps on going..........attached screengrabs from same site at same time.........One chart is with WS80 & other is using WS68
You do not have the required permissions to view the files attached to this post.
Post Reply