meteohub.de

Hello together,

the summary might be interesting for others who plan to order a Sheeva Plug to run meteohub on it.

I have a WMR 200 and a RFXCom receiver.
My house has 2 floors, a cellar (30 cm ferroconcrete, which seems to be shielded very good - no GSM mobile signal at all) and a alu made roof (Prefa).
My initially planned setup was to place the WMR and the Sheeva in the entrance area in the ground floor, the wind and temperature sensor on the roof, the rain and UV sensor somewhere in the garden.

As already discussed in another thread (http://www.meteohub.de/joomla/index.php ... 17&catid=8) i had troubles with the reception of some sensors (specially with the wind sensor on the roof).

I really invested a lot of time, but i was not able to fix the problem (i changed the positions of the WMR several times).
As the WMR 200 has no possibility to change the antenna against a external better one (there is no possibility to open the WMR like the WMR 928 and to connect i.e. a ground plane antenna) i put a 43 cm copper wire around the antenna.
This improved the reception a little bit, but not worth to mention.
Then i decided to go for a RFXcom receiver (i anyway need it for the power, gas and water measurements).
The RFXcom has a much better reception.
But even with the RFXCom i had the same troubles.
I discovered, when the RFXcom is connected directly on the laptop i have no problems at all with the signals, so the problem is somewhere in the Sheeva plug.
Also changing the USB cables didnt improve the receptions.
The Sheeva has definetly some problems with the USB port (it produces some radio noise, which influences the signals of the sensors).
It works fine, if the Sensors and the weather station are placed within some meters with direct view on each other.

At the end i sold the Sheeva and ordered a Alix1D.
The quality of the Alix is much better than the Sheeva (plastic cover, very loose USB connectors).
I am also wondering why they put a gigabit lan interface in the sheeva (it will never be able to process the necessary data over 100 Mbit/s).

Nowadays I placed the Alix and the Rfxcom in the cellar in my network rack and i haven´t had any single missed signal :-)

So i really recommend to go for a mini PC like Alix instead of the Sheeva (it costs nearly the same, has the possibility to put WLAN on it, etc. and the quality is much more better (connectors, cover).

@Boris: it´s maybe worth to mention these findings on your hardware overview section to help other people to save a lot of time and nerves.

I bought the sheeva from gdfde (before I read here in the forum about
his problems with it :unsure: )
It runs with a WMR 200 since one week.
But I can't report any of the problems described above.
There are no problems with the reception or the USB connection.

It becomes a little warm, but the power consumption is only 3 W (measured).

I am using the SheevaPlug since September and did not have any real problems until now. One thing is, that the clock do not synchronize with the radio signal, but that is not very dramatic.

gdfde wrote: Then i decided to go for a RFXcom receiver (i anyway need it for the power, gas and water measurements).

What type of sensors are you going to use for your gas and water measurements?

HeinrichH wrote:

gdfde wrote: Then i decided to go for a RFXcom receiver (i anyway need it for the power, gas and water measurements).

What type of sensors are you going to use for your gas and water measurements?

As i do not have so much time right now, the project is stopped.
But my plans are to go for the RFXmeter with the reflective optical sensors from RFXcom for Energy and gas measurements.

For the watermeasurements i need to install a second water meter with a reed contact, as the currently installed water meter hasn´t any possibility to measure the consumption.

@cgn: in my setup, which is a little bit more complicated than the standard setup (sensors in the garden, roof, cellar, etc.) i had some troubles with the sheeva and the reception of the signals.
In an standard environment with short distances to the sensors it works good.

I recently migrated my meteohub from NLSU to the sheevaplug.(weather station is oregon WMR928)

All was going fine( it really was plug and play) until I got to sensor setup stage and noticed that some sensors are not being received. (sensors are about 25m from the base station - but in line of sight).

I had no real reception problems when the NSLU was running. However as mentioned by gdfde at start of thread, it appears the sheeva is inhibiting reception. (The base station was showing blank for some sensors but when I unplugged the sheeva, the offending sensor signal was received loud and clear again).

I am going to try and move the sheeva a further distance back from the base station - but I have my doubts that this will improve reception - as I suspect the interfence is coming down the USB line from the base station.

Anyone have any suggestions that could the improve reception? Can the aerial on the base station be improved upon?

Just a few random thoughts...

I don't have a Sheeva myself, and some of these things have already been mentioned or hinted at by earlier posters, but maybe there might be something here that helps?

StratoQ wrote:as I suspect the interfence is coming down the USB line from the base station.

If that is the source then using a USB cable with RF chokes on it might help?

Anyone have any suggestions that could the improve reception? Can the aerial on the base station be improved upon?

I've heard of arrangements where people have attached an external extension to the antenna (i.e. a length of stiff copper wire with the bottom end wrapped around the plastic cover of the existing antenna), but opinion seems to vary on whether or not this is effective!

If the Sheeva itself is producing RF perhaps some sort of Alu foil RF 'shield'/cage could be arranged (taking care of course to allow adequate ventilation)?

skyewright wrote:Just a few random thoughts...

I don't have a Sheeva myself, and some of these things have already been mentioned or hinted at by earlier posters, but maybe there might be something here that helps?

StratoQ wrote:as I suspect the interfence is coming down the USB line from the base station.

If that is the source then using a USB cable with RF chokes on it might help?

[/quote]This doesn´t help, i also tried a real good shielded (expensive) usb cable.

Anyone have any suggestions that could the improve reception? Can the aerial on the base station be improved upon?

I've heard of arrangements where people have attached an external extension to the antenna (i.e. a length of stiff copper wire with the bottom end wrapped around the plastic cover of the existing antenna), but opinion seems to vary on whether or not this is effective!

If the Sheeva itself is producing RF perhaps some sort of Alu foil RF 'shield'/cage could be arranged (taking care of course to allow adequate ventilation)?

The aerial on the base cannot be approved very easily.
The method with the copper antenna wrapped around the existing antenna doesn´t help as well.

I would not recommend to cover the sheeva with alu foil, as it becomes quite warm.
There is also no possibility to open the base station (without damage) and to replace the antenna with a cable to an external antenna.
I also connected the base station to a RFXcom module with external antenna, but the behaviour was the same.

I purchased an Alix, it´s more or less the same price and i don´t have any troubles anymore (same setup).

gdfde wrote:I purchased an Alix, it´s more or less the same price and i don´t have any troubles anymore (same setup).

ALIX-1D is what I run Meteohub on too. I've been very happy with that arrangement.

My only quibble is that if you have a poor mains electricity supply then with the standard PSU you might get problems with occasional lock ups. I had that problem. It was easily solved by getting a better quality PSU.

PS. When I mentioned an "Alu foil RF 'shield'/cage" I was thinking of a spacious, well ventillated, Faraday cage rather than covering the 'plug' in foil.

Thanks for the suggestions. In hindsight I probably should have gone for the Alix too.

However this evening I did some playing around with the base station and Sheeva. I also bought a USB extension lead so that I was able to position the two over 3m apart. This seems to have done the trick. Firstly two sensor readings came back immediately to the base and later, after more repostioning, the third missing sensor came back - so they are all reporting to base at the moment.

Ill wait and see how it goes over next 24 hrs before Ill make a final decision on it.

I had similar issues (reception became deprecated after switching from NSLU2 to Sheeva). I've experimented quite a bit:
It's definitely not related to RF interference; When I kept the Sheeva powered on in the close vicinity of the NSLU2 (USB still connected to the NSLU2), I was not able to detected any issues.
The moment the USB was connected to the Sheeva I experienced more sensitivity issues. Tried a longer USB cable to no avail. What did help me was to put a passive USB hub in between. After that my problems were reduced. It is still not as perfect as with the NSLU2, but it seems that the USB HUB is filtering some noise.

YJB wrote: It's definitely not related to RF interference; When I kept the Sheeva powered on in the close vicinity of the NSLU2 (USB still connected to the NSLU2), I was not able to detected any issues.
The moment the USB was connected to the Sheeva I experienced more sensitivity issues. Tried a longer USB cable to no avail.

That´s exactly the behaviour which StratoQ and i recognized.
As soon as the USB is plugged into the sheeva, the troubles starts.
So it´s definetely related to the USB connection of the Sheeva as i discovered the same issues while connecting the Sheeva to a RFXcom receiver with very good radio antenna.
I guess, that the USB on the Sheeva is causing some radio noise which impacts the radio perception.

I have mounted my base station on wall just underneath where I got a good reception last evening. But as I half expected, during the day more and more of the sensors have missed reporting back. With the result now I have only reports from 2 of my seven sensors.

I even tried it with a USB hub - but to no avail. I have reached the end of the line with the sheevaplug unfortunately and will be changing to an alix 1D.

I should have paid more attention to your post, earlier, gdfde. I understand now exactly what you mean by "saving a lot of time and nerves".

At least we have this forum to iron out the issues.

This very much sounds like rf noise creeping via USB cable from SheevaPlug to weather station.

Did you try blocking rf by a Ferrite Choke Core, like this?
http://www.radioshack.com/product/index ... Id=2103222#

Yes. I tried the ferrite choke also but no improvement. Overall it has been very intermittent. Maybe Ive got a dud Sheevaplug?

The weather conditions seem to be having effect also with better reception achieved at night and poor during the day. (humidities have been low in afternoon 30/40%).

But I had never seen that kind of reaction with the NLSU.