Cumulus Support

I've never seen anything really USB related, apart from the fact I sometimes get 1060 hPa on Cumulus only, the console doesn't have it and I just had 174 km/h lol. It could be related to the USB cable.

Hi
Just for your update in the search for USB problems.......
I have a WH3080 with Solar, and have had many problems with Battery and USB/lost sensor contact communication - but not in the way of losing USB connection (I think).

I have after many, many and many more tests found that "lost sensor contact" is related to the USB cable or connection. I have changed the USB cable serveral times to a more and more RF secure one, without any luck.

The problem only accurs when I have the station USB connected to the computer. After some time The Weather station losses contact to the sensor, but never if I disconnect the USB cable - then the weather station works without any problems. (I have also changed computer - noise!)

When I replace the WH3080 with my previous Weather station without Solar, it never happens.

PROBLEM FOUND:.....Disconnected Solar unit and reconnected rain unit directly unit into the transmitter.....

brdgs
jan

I have had the USB console to PC to Cumulus error happen only twice since March 2012 (when it was installed in Florida).

That said I have removed the batteries and unplugged the USB port; put the batteries back in and plugged it into the USB port and all went well after that.

I have not had a chance but wanted to see what would happen if the FO console was plugged into an externally powered USB hub/port.

I will purchase and install one in the next few weeks and see if that helps with the problem. Around the same time will be installing a weathercam in the vicinity of the weather station.

It could maybe just be some firmware upgrade to the console that could help or some hardware mod to the USB port?

The console firmware isn't upgradeable. It's possibly worth trying other methods of USB connection as from what I've read the problem is related to the USB chip in the station.

I've had the odd problem before but always a restart fixed it...

But since yesterday afternoon all of sudden I will hardly work! I run cumulus and I get this error:

"27/10/2012 10:56:30 : Fine Offset station not found! - check connections and restart Cumulus"

The Base station is still collecting the info but it wouldnt download to the PC. I unplugged the base station from the USB overnight, and this morning I plugged it in and all the data downloaded to the PC perfectly, however it failed as soon as it started to "stream". I leave the base station in during the day to update my weather website. But it's as if it cant find it.

I have reinstalled cumulus, I have tried other cables, usb ports, batteries etc, and will try another laptop later but I was wondering if theres any magic tricks to solve this?

I Bought this last September and I'm guessing if there was a warranty on it, it would have expired after 12months. A bit gutted, as soon as the weather gets interesting again and everyone wants to know, my station dies!

Cumulus is awesome though, Easyweather cant find the station either after the errors occur.

I do have an old house and Where the base station sits gets quite cool, 8C last night so I don't know whether thats not helping so i have sat it above the oil fire to "thaw" lol

One thing you didn't mention is the USB socket on the Display unit? They are known to be badly soldered and will cause intermittent connection problems. You would need to dismantle to re-solder.

Unrelated post about pressure readings moved to here: https://cumulus.hosiene.co.uk/viewtopic.php?f=13&t=8542

Having suffered the USB lockup quite a few times, I thought I would post a question about it - what do I find, it is right at the top of the forum with numerous others suffering

I thought it might be just my system, perhaps the length of USB extension I'm using, but obviously not.

What I'm seeing is a completely flat line set of data, until I spot it has locked up and reset it, so it seems that all of the data is lost between it locking up and the eventual reset.

Harry, isn't your station quite old? I seem to remember you being one of the first Fine Offset users. Wasn't it you who persuaded me to add support for Fine Offset stations in the first place?

steve wrote:Harry, isn't your station quite old? I seem to remember you being one of the first Fine Offset users. Wasn't it you who persuaded me to add support for Fine Offset stations in the first place?

It is quite an old one. My Cumulus directory was created in Dec 2010 which makes me think I bought the station in around June 2010.

I cannot remember whether I did or did not ask you to support the FO station. If you started your support for it in around Dec 2010, perhaps I did.

If it helps, I bought it via ebay and it is the version which includes the radio time receiver. I didn't use the supplied software for long, but I do not recall it ever locking up the USB on that software. It only locks up while the computer is on, I never find it has locked up when I first power up the PC or when I first run Cumulous. The lockups occur with no obvious cause, sometimes once eery couple of months, to twice in the past week. Yesterday evening I added a ferrite torroid to the USB cable, to see if that might make a difference. I also tried waggling the USB connections about which made no difference.

You've been on the forum since August 2008, so I think your station must be older than you think. The theory so far is that the USB lockup only happens on more recent stations. Could you zip your diags folder and attach it, please, so I can confirm it's the same problem?

steve wrote:You've been on the forum since August 2008, so I think your station must be older than you think. The theory so far is that the USB lockup only happens on more recent stations. Could you zip your diags folder and attach it, please, so I can confirm it's the same problem?

Attached!

Not conclusive, but no USB lockups since fitting a torroid on the USB cable.

It looks like a different problem. Windows is saying that the station is disconnected, so it's likely something to do with the cable, or the USB port, or similar.

01/11/2012 16:30:37.931 : 16:30:37 EWUSB: HidControl Device Unplug
01/11/2012 16:30:37.931 : VID = 1941 PID = 8021
01/11/2012 16:30:37.931 : Error - station unplugged - check connections
01/11/2012 16:30:37.931 : EWUSB: HidControl Device Removal

That doesn't happen with the problem that I started this thread for, that only happens with more recent stations. In that case, Windows is quite happy that the USB connection is alive, but the station doesn't respond.

Oh well, thanks for looking at the data. The cable was what I had assumed the problem to be, until I saw this thread - back to experimenting with cables

I've had this issue on three occasions (15-APR, 13-JUN, 21-JUL). In all cases the symptoms were exactly the same:

• Device stops responding to USB communications. The data collection program just locks up waiting for a response that never arrives.
• The hardware console still appears to be displaying live data. Some stuff on it still responds to touch.
• Memory-related functions on the console are non-responsive. When I press on the memory pie chart I don't get any options to view old data or clear the devices memory. This is actually how I check for the issue if my program stops logging data.
• Reconnecting the USB cable makes no difference
• Restarting my program makes no difference. IIRC it just hangs as soon as it tries to communicate with the device.

Hardware is a FineOffset WS-1080 (solar) purchased mid-march new from Jaycar Electronics here in New Zealand. The software I'm running is entirely custom written. The bit relevant here is just a little ANSI C program that reads from the device every 48 seconds (current record pointer, "live" record and any new records that aren't already in the database). It ran on windows from when I wrote it mid-April until about a month ago when I moved it to my Linux server.

My guess is that the problem is caused by the device trying to write to a memory address while the PC is busy reading from it. Perhaps the USB chip and microcontroller are deadlocked. Or the microcontroller trying to write puts the USB chip in some invalid state while the microcontroller writes garbage to memory preventing its own memory functions from working.

I had considered trying reproduce the problem by writing a quick program that continuously reads from the device (probably the record count and current record address fields as well as the "live" record) in a loop logging everything it does until something crashes. I've only got one console though and I don't like gaps in my data.

I'd certainly like to get this bug fixed in my own code but considering how rarely it happens (the way my program schedules its updates may have something to do with that) it has not been a huge priority for me so I've not really looked into it too closely.