Problems with station communication: USB locks up completely

[WoodburyMan]

I have a Ambient branded WS-1090. Same issues everyone else has with drop outs. As I was reading this topic I actually had one data drop out, but it seems to drop out in middle of a data read every time. In this instance pressure was 30.02in. It updated some parameters, then all of a sudden I had 30.78in reading, and Error light came on. Next data read 15 seconds later, all back to normal. (I've had to set some very tight Spike Removal values to deal with this).

I have a few things. First I tried switching locations and looking for any other EM interference and no results, I have a powered USB extension (Basically powered hub but only one port) going from my PC to the console, connecting to the original stock usb A-A cable. I haven't replaced the stock cable yet, that was my next step. However after reading this, I realized it could be a memory corruption issue. Luckily I am fortunate enough where I have a extra receiver for my WS-1090. (One in bedroom hooked up to computer for Cumulus, and another for the kitchen downstairs). I just swapped the units to see what happens.

Like any other device the flash memory in the device could be bad. Perfectly would explain why I would go days with it being fine, then many errors in a row if the recording process has happened to find itself in a series of bad blocks/sections of memory space. Then when cumulus reads this data, it goes nuts and either wont read it and display Communication Problem (As it prob sees corrupt data and cant handle it) or it reads the data, with one or two corrupted data points, and either goes on business as usual or ends the read cycle once it gets the corrupt data and displays errors like my random Spike in Pressure I had this morning. (And other values I've had over time).

Will post back with the progress I have had with the other display unit / receiver. if thats the case the broken one can stay downstairs as that's for quick "check the temperature out the door" and not used for anything meaningful.

[Spider-Vice]

I find something very interesting, is the fact that I extended the antenna of my console and if I put it like a centimeter away from where it is currently, it starts losing signal, yet not all times, also when I put it on the same place it might not recover it but when I put it somewhere else then it starts updating constantly every 48 sec (Always changing values instead of repeating), but then it starts going crazy again. Like right not, it's raining, and it was fine, all of a sudden it started updating every 3 minutes just (Failing 3 updates), but always 3 minutes. Sometimes it happens with 5 minutes and so on, but it still updates, even though only in that timeframe.

I honestly don't understand this station.

[ejay]

Not surprisingly, my WH3081 encountered its first error after less than two weeks of continuous operation. Got the dreaded "The station is not initialised" error. I guess it was not a mater of if, but a matter of when.

I took the batteries out and disconnected the USB cable (as per the directions found elsewhere on this forum). After that, the console was reset and weather data could be read by cumulus again.

Interestingly the user manual states weather history data is "stored in non-volatile ring buffer memory (EEPROM) and will not be lost even in case of an interruption of power supply (e. g. change of batteries)". But upon re-inserting the batteries, the memory on the my unit appeared to have cleared, as if there was no EEPROM at all. My console's last two weeks' of stored weather data is gone.

Like any other device the flash memory in the device could be bad. Perfectly would explain why I would go days with it being fine, then many errors in a row if the recording process has happened to find itself in a series of bad blocks/sections of memory space.

My WH3081 appears to be a later model, purchased in March 2012. Maybe FO decided to do away with EEPROM due to the the problems associated with blocks/sectors wearing out? There is no relay in the console (that clicks to disconnect the batteries when plugged into USB). Are FO saving money (and potential problems) by omitting EEPROM and other components?

[Spider-Vice]

The manual isn't very explicit. The Portuguese (or Spanish I don't remember ) version correctly states that only the extremes are saved in a permanent memory and that's correct, those are never lost unless you reset them yourself.

[pete_c]

Moved the Fine Offset from the MW to FL.

I changed the PC to a Lenova Laptop running XP. Its been fine now for over one month or so.

Difference was that the console was in the basement of the house in the midwest; in FL its on the main floor but on the opposite side of the home.

No hiccups. All is well.

[VTHokie74]

My WS 1090 lost contact with the sensor array today. Very conveniently it was 34 days after I received it so no exchange on Amazon. I think I will wait a few days before writing a review. I will see how Ambient does with the exchange on the sensor too. I am very disappointed with the reliability. There is no reason for the lsot signal. I have the display mounted about 3o feet from the sensor inside the basement. The sensor is mounted under the deck (because the factory solar shield is a joke) and the receiver is just across the basement room with a large patio door in between.

[KJ4KTW]

Well, I've had consistent problems with my station losing contact with the computer as well. I've attempted the fix mentioned on page one of this thread with the power management settings. I'll report back on that. Here's what my problem was today: The station (WS-1090) froze up on me so I pulled the USB cable and reset it as normal. No joy. Restarted the computer. Cumulus started up and went into the "Downloading Data" screen for days (literally). Finally, I had to pull the batteries and hard reset the weather receiver itself before it read and communicated to the computer again. This is the first time I've had to pull the batteries on my station in months. Very peculiar.

[pete_c]

Yes; my set up in FL did well up until a week or so ago...it's stuck at "downloading data". I have remote access to the PC but can't do anything about resetting it.

I wish there was a means to communicate to the fine offset device to do a hard reset / soft reset of the device.

[pete_c]

Anyone just leave the batteries out of the Fine Offset?

I never checked; will it work without batteries and just a USB cable?

[mcrossley]

I never checked; will it work without batteries and just a USB cable?

Yes, but of course it restarts if the cable becomes loose, or you PC drops power to the USB port when in standby. A powered hub maybe? Anyway just leave the batteries in as a standby, they are physically isolated by a relay when on USB power so should just have their shelf life before they need replacing.

[Charlie]

That sounds like a good idea - leave the batteries out and then you can reset the station remotely by having the computer briefly remove power from the USB port. The only concern I'd have, is that sometimes getting the remote sensor unit synced again can be problematic and there's no way to power cycle it without building some custom hardware. There are advantages to using wired sensors.

[pete_c]

I'm going to try today as I had a 4 or so hour power failure yesterday night in FL. I had to do the battery / USB cable thing with the help of a neighbor (in FL) yesterday. I'm hoping just a reboot of the laptop I am using will reset the Fine Offset.

[mcrossley]

I'm hoping just a reboot of the laptop I am using will reset the Fine Offset.

I doubt it, laptops do not normally power down the USB ports during a reboot, only on a full power down, they often keep them powered up during standby too (if there is a device present).

[pete_c]

This laptop is an old IBM T-40. I'm trying it out for a bit. The original though was to utilize a 4.5VDC transformer on an x10 appliance module.

There's another problem though. I had a 15 hour plus power outage yesterday there and laptop shut down and there are no provisions to auto restart it when the power came back. Might replace the PC with a mini-itx in a mini case or something similiar.

Do you guys know of a way that I could check if the USB power is being reset remotely? I have heard of odd issues with these old laptops relating to their USB ports / motherboards / BIOS.

I can reboot the laptop remotely but not much else. Is Cumulus polling the time on the Fine Offset device. Would I be able to tell from this?

Might install an IP web camera on my next visit. That said while I am there will most likely replace the laptop with a small Aopen PC that I have sitting around.

[pete_c]

Tried an experiment today with the USB ports. I plugged a hub / CF card reader into my home office PC. I can see the power draw. When I disable it the power draw goes to zero.

Trying to do the same in Florida remotely I cannot disconnect the Fine Offset USB port. I wonder though if it would be different connected to a powered USB hub.

USB-1.jpg

USB-2.jpg

USB-3.jpg