meteohub.de

Sheevaplug - Meteohub 4.9u - OS rainsensor PCR800 RFXTRX433
Distance between sensor and antenna : 15mtr, clear vission, no walls ......

Suddenly strange readings 126.2mm/h ....
Inspect data gives :
20130705090614 rain2 0 0 346
20130705090701 rain2 0 0 346
20130705090748 rain2 0 0 346
20130705090835 rain2 0 0 346
20130705090922 rain2 0 0 346
20130705091009 rain2 0 0 346
20130705091056 rain2 0 0 346
20130705091143 rain2 0 0 346
20130705091230 rain2 0 0 346
20130705091317 rain2 0 0 346
20130705091404 rain2 0 0 346
20130705091538 rain2 0 0 346
20130705091625 rain2 0 0 346
20130705091759 rain2 026502 0 346
20130705091846 rain2 013251 0 346
20130705091933 rain2 08834 0 346
20130705092020 rain2 06626 0 346
20130705092107 rain2 05300 0 346
20130705092154 rain2 04417 0 346
20130705092241 rain2 03786 0 346
20130705092328 rain2 03313 0 346
20130705092415 rain2 02945 0 346
20130705092502 rain2 02650 0 346
20130705092549 rain2 02409 0 346
20130705092636 rain2 02209 0 346
20130705092723 rain2 02039 0 346
20130705092810 rain2 01893 0 346
20130705092857 rain2 01767 0 346
20130705092944 rain2 01656 0 346
20130705093031 rain2 01559 0 346

Does somebody knows what the cause is , and better : how to avoid this.

Thx,

Addendum :
I've tested the RFXTRX433 with de manager-software under windows for 24 hours : no bad readings.
The battery = 'Ok'.

With Meteohub the battery seams 'bad' ....

It seems to me that meteohub :
- get wrong data from the driver
- or translates the data in a wrong way

I saw in 'dmesg' that de RFXTRX433 has a fdti ft232-chip onboard.
With my experience with Raspberry Pi I know that Debian Linux in combination with Raspberry pi and this FT232 chip was a 'pain in the ass ...'.
Recently the problems are solved with a newer kernel and better drivers.

Perhaps the same problem here ..... The meteohub-image is based on a legacy debian (lenny.....) .....

I was hoping to buy a license for meteohub, but this hope has been frozen now, until the problems are solved......

What sensor is delivering the false "low battery reading"?

Meteohub defines a threshold when to signal for low battery,
so that threshold might be triggered or it is a false interpretation of data.
As RFXtrx firmware has changed a lot over the last quarters, that might have passed.

A potential wrong lowbat signal is going to make you decide if you want to use meteohub or not? really?

admin wrote:What sensor is delivering the false "low battery reading"?

All of them, randomly. Most of the time there are two of them with the message 'low'.

admin wrote: Meteohub defines a threshold when to signal for low battery,
so that threshold might be triggered or it is a false interpretation of data.
As RFXtrx firmware has changed a lot over the last quarters, that might have passed.

A potential wrong lowbat signal is going to make you decide if you want to use meteohub or not? really?

Of course it is not only the 'low' battery . The buggy 'rainrate' of the PCR800, mentioned earlier. Reading within Win8 with the RFX-Manager makes good readings, so it's the combination Sheevaplug, RFXtrx433 and Meteohub wiche makes the fault ...

Other sensors :
The readings of the UVN800 seems to be ok, except the low-battery.
The readings of the THGR810 seems to be ok, except the low-battery.
The readings of the BTHR968 seems to be ok, except the low-battery AND recognition as a BTHR918N instead of a BTHR968.
The readings of the WGR800 seems to be ok, except the low-battery.

I hope you can find the error, because the rest of MeteoHub is great.