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

WS3083 USB Connection problems

Discussion specific to Fine Offset and similar rebadged weather stations
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: WS3083 USB Connection problems

Post by AllyCat »

Hi,
steve wrote:The Fine Offset stations with solar sensors do seem to be much more susceptible to the 'lock up' problem, and the 'synchronise' facility in Cumulus doesn't seem to help so much.
I haven't looked in detail at how the "synchronise reads" operates, but the additional transmissions from the Solar Pod (with an exactly one minute period) are a considerable complication. Presumably the original idea was that the two transmitters could operate independently (i.e. at different locations) without any risk of one transmitter repeatedly "blocking" the other.

But in practice the two transmitters are linked (by a 6-core cable) and appear to synchronise to each other, so that the two transmissions are "interleaved". That means that there is no time within the "cycle" of either 48 seconds (main TX) or 60 seconds (Solar TX) that is more than 6 seconds away from a transmission from the other sensor/transmission (at some time within each 4-minute period).

I thought that Steve was aware of this and it's (somehow) taken into account, but perhaops not?

Cheers, Alan.
User avatar
steve
Cumulus Author
Posts: 26702
Joined: Mon 02 Jun 2008 6:49 pm
Weather Station: None
Operating System: None
Location: Vienne, France
Contact:

Re: WS3083 USB Connection problems

Post by steve »

This complication did occur to me, and I wondered if it was why the synchronisation didn't work so well on solar stations; I keep meaning to ask Jim Easterbrook (who devised the scheme) for his thoughts. Cumulus does the 'avoid' thing around the time of the 48 second update and also every 60 seconds, but the latter is to avoid the period where the console updates the age of the entry. I'm not at all sure that this coincides with the solar update - I suspect not, as presumably the transmitter and the console are not synchronised in any way. But does the console do the solar update as soon as it receives the data?

If there is a thrd update which needs avoiding, it gets to the point where there is no 'safe' period at all, allowing for clock drift.
Steve
vr34
Posts: 32
Joined: Wed 13 Jan 2010 12:43 pm
Weather Station: Bresser WLAN 7-in-1 (7003300)
Operating System: Raspberry OS Bullseye, CumulusMX
Location: South of France

Re: WS3083 USB Connection problems

Post by vr34 »

Thx steve for this information. I will contact greenfrogscientific.co.uk (the EU contact of Aercus Instruments) for more help, if they can do (? :?) I will inform here about their answer.
I do not use gauges and short realtime and do not care if the website update is long (10min) but at least if I cannot remove a battery evry 3-4 days, it would be great otherwise I will try to exchange to a WS without that solar sensor !
My parents in law have the same station but without solar sensor (ws2083) and I will install their website next weekend. We will see if they hare less usb problem with their WS.
Regards,
Vincent
jim-easterbrook
Posts: 111
Joined: Thu 09 Jul 2009 10:47 am
Weather Station: WH1081, Elecsa AstroTouch 6975
Operating System: openSUSE 13.1, Raspbian, OpenWrt
Location: Epsom, UK
Contact:

Re: WS3083 USB Connection problems

Post by jim-easterbrook »

steve wrote:This complication did occur to me, and I wondered if it was why the synchronisation didn't work so well on solar stations; I keep meaning to ask Jim Easterbrook (who devised the scheme) for his thoughts.
I came here today to see if you had any thoughts... I've got pywws users with 3080 type stations who are getting frequent lockups. I'd like to find a solution but I'm reluctant to shell out time and money on an inferior station. :)
Jim
vr34
Posts: 32
Joined: Wed 13 Jan 2010 12:43 pm
Weather Station: Bresser WLAN 7-in-1 (7003300)
Operating System: Raspberry OS Bullseye, CumulusMX
Location: South of France

Re: WS3083 USB Connection problems

Post by vr34 »

Hello,
For information, since I completely reinstalled Cumulus (same build as before 1.9.4 build 1099), suppressed any time synchonisation in Win7 and added FOReadAvoidPeriod=10 in cumulus.ini and applied any recommendations about USB presented here, my WS3083 did not lock for more than one week :D Fingers crossed and hoping that it will continue...
mgrainger
Posts: 2
Joined: Tue 08 Sep 2015 12:37 pm
Weather Station: WS3083
Operating System: Windows 10
Location: Omokoroa

Re: WS3083 USB Connection problems

Post by mgrainger »

I would like to "fly a kite" on this subject which has hassled me for several months. I have concluded (rightly or wrongly) that it is an issue with Windows 10 and USB and not the WS3083 or the Cumulus software. I run two screens and the Cumulus displays on screen two, so if it is running on screen 1 it has shutdown and auto-restarted as per the setting [Station settings > Program Settings > Restart if data stops]
I have tried various USB ports on the computer - there are USB3 and USB2 ports - that may be a factor??????? It "seems" more stable on the current USB port, which I think is USB2.
I have logged these errors for months and noticed that there was a commonality about the time, so I started inspecting Eventvwr to find out what else happened at the times the "Fine Offset station not responding. You may need to reset it (remove batteries and USB cable)" error was logged.
This is where I make a massive leap.
I went back and checked the Eventvwr logs for the last few occasions and I found that about the same time as the Cumulus shutdowns there is a "The VSS service is shutting down due to idle timeout. " 8224 event ID.
This could, of course, be a total coincidence but something has to be triggering this very annoying event and I am very tired of shutting down Cumulus, unplugging the USB, removing a battery, replacing the battery, resetting time and date on station, resynching with remote sensor, reconnecting USB, restarting Cumulus.
I have made a registry change to set the VSS service idle timeout at 136 years, but is is early days yet.
Could be totally off my rocker on this. Thoughts anyone?
mgrainger
Posts: 2
Joined: Tue 08 Sep 2015 12:37 pm
Weather Station: WS3083
Operating System: Windows 10
Location: Omokoroa

Re: WS3083 USB Connection problems

Post by mgrainger »

Well that was a bust! Broke again, 3 days later while we were on holiday.
ABZ-Dave
Posts: 2
Joined: Mon 08 May 2017 4:44 pm
Weather Station: WS3083
Operating System: Raspberry Pi

Re: WS3083 USB Connection problems

Post by ABZ-Dave »

I'm currently running Weewx on a Raspberry Pi to gather data from my WS3083 but looking at potentially moving to Cumulus and am getting these lock ups every week or so. For now the solution I've come up with is to remove the batteries from the console and power it via the USB cable. I then have the Raspberry Pi plugged into a network enabled power socket which means that when I get a lock up I can remotely power down the Raspberry Pi (and therefore the console) for 30 seconds or so and then power it up again solving the problem.

This is obviously far from ideal but is the best work around I've come up with and means that 99% of the time I can quickly get my weather station back up and running again even if I'm away from home. I did experiment with doing an auto reboot on loss of comms but the reboot is too quick and doesn't power down the USB port so doesn't help resolve the problem and leaves you with a constantly rebooting computer due to comms never being re-established.
vr34
Posts: 32
Joined: Wed 13 Jan 2010 12:43 pm
Weather Station: Bresser WLAN 7-in-1 (7003300)
Operating System: Raspberry OS Bullseye, CumulusMX
Location: South of France

Re: WS3083 USB Connection problems

Post by vr34 »

Thx ABZ-Dave and mgrainger for your messages.
My WS3083 gets these bad lock ups even after my complete reinstallation. But it looks a bit better : only 2 lock ups since my previous message dated Feb 21.
For information, the computer connected to the WS and running Cumulus 24/24h is a netbook with windows 7.
Windows 7 and 10 with Cumulus, Linux with Weewx: same problem!

I have noticed than sometimes I only need to restart Cumulus (I use Teamviewer to do that from outside even from my smartphone)

Some one reading and working wth Cumulus MX has the same problem?

My conclusion/solution :
As the way is to remove batteries, unplug the USB cable, boot the pc and wait few seconds to replace/restart them...
As I equipped my home with a domotic box and several commandable power plugs, I will replace the batteries by an AC adapteur, will put both the PC and the WS on the power plug in order to restart them if any look up occurs when I am several days out of my home.
Maybe in the future: make a simple utility (in visual basic) that detects a constant value (external temperature?) for several hours in the current log txt file and order the domotic box to switch off then on the power plug...

Regards

Note: 3x 1.5V = 4.5V. I can not find any information about the power plug close to the USB one in the manual. I am surprised by the message " DC 8.0 V " written close to the plug!?!?
ABZ-Dave
Posts: 2
Joined: Mon 08 May 2017 4:44 pm
Weather Station: WS3083
Operating System: Raspberry Pi

Re: WS3083 USB Connection problems

Post by ABZ-Dave »

Providing your computer has enough power at the USB port which a laptop should have then you don't need to worry about a separate power supply as the USB will be able to power your WS3083. That's how I've got mine running and it seems to work fine other than the occasional lock up.

From the research I've done there seems to be a school of thought that suggests the USB lock ups are caused by the computer attempting to communicate with the weather station while the weather station is trying to communicate with its sensors. Playing around with frequency of reading the weather station may therefore make the lock ups worse or better but a lot of the time it would appear that there is a good portion of luck involved.
vr34
Posts: 32
Joined: Wed 13 Jan 2010 12:43 pm
Weather Station: Bresser WLAN 7-in-1 (7003300)
Operating System: Raspberry OS Bullseye, CumulusMX
Location: South of France

Re: WS3083 USB Connection problems

Post by vr34 »

Thx ABZ-Dave. I've never tested it before. Now, no batteries, as the USB of my old netbook (plugged on the power supply) is sufficient to power the WS display :D
User avatar
odintheterrible
Posts: 110
Joined: Wed 16 Jan 2013 8:18 am
Weather Station: Brandson 304577 & GW1000
Operating System: Windows 10 64-bit
Location: Torre-Pacheco, Murcia, Spain

Re: WS3083 USB Connection problems

Post by odintheterrible »

Greetings to one and all :)

Now this situation of power supply vs auto update timings is turning into a bit of a lucky dip, given that 'Fine Offset Electronics Co., Ltd' are being totally tight lipped about the whole thing ! I know from previous contacts I've had with them that they are well aware of this problem and have previously offered a DIY FIX (https://cumulus.hosiene.co.uk/viewtopic.php?f=16&t=10510 ... scroll down the page to the post with the pictures) of sorts ... but there has been little other feedback to suggest that they have a proper solution.
Given this, and all the comments that are circulating on this forum, I'm just wondering ... and here I'm fishing for comments before diving in, if it would be feasible to cut the power feed cables (pos & neg) in a USB cable, leaving the data cables in tact, and then powering the console via an external mains adapter plugged into the power socket on the side of the console (removing the batteries from the console too) ... then it would be a relatively simple task to insert a sensitive regulator into the power supply to accurately control the voltage, given what it says in the user manual, about using non-rechargeable batteries (1.5v Vs 1.2v for rechargeable) which I take to understand implies a voltage requirement of 4.5v (3 X 1.5v) for the console.

Any input greatly received ;)
"Qui Custodit Ipsos Custodes ? ..."

Duct Tape is like 'The Force' ... it has a 'Light Side' AND a 'Dark Side' and it binds the Universe together !
Post Reply