Cumulus Support

Does anyone know why I am getting these strange temperature spikes?
shown hightlighted in the following graphs?

Could this be a sign that my transmitter is on the way out?

Solorize wrote:Could this be a sign that my transmitter is on the way out?

Hi,

Probably not, nor even that the batteries may need changing. Are there any spikes on the other external data (the barometric pressure is measured within the Console)?

The FO 108x models use the same Analogue-Digital converter for their external temperature measurement (a simple thermistor) as is used for the wind direction (vane). Furthermore, in their "cheap and cheerful" design, the vane cable is continuously connected to the A/D converter input (it is "selected" by a voltage applied to the other vane wire to make the measurement), so even a tiny amount of electrical interference (if it occurs at the instant each 48 seconds when the temperature measurement is made), or worse, a leakage current in the wind cable/connectors due to moisture, can upset the external temperature measurement. See around page 21 (yes, page 21) of Gina's enormous thread concenring spikes from her electric fence a few years ago.

Your spikes seem quite minor, but perhaps Steve's excellent software is catching (masking) the worse ones?

Cheers, Alan.

I have the same spikes intermittently and I put it down to the crap design as when there is a bit of wind blowing the spikes are more common. Don't they share a channel somewhere?
At present we are getting 1 Anticyclone High after another and very little wind and I haven't seen a spike for a few weeks now.
You'll notice that the temperature line is often more bumpy when there is a reasonable wind, at least it seems so on my system.
I just edit the spikes out when they occur.

Super-T wrote:Don't they share a channel somewhere?

Hi Terry,

Not intentionally, but the wind vane and the anemometer do have conductors in a common sheath, so (for the reason I mentioned above) "crosstalk" in the cable (particularly if it's been extended) can cause the pulses from the anemoemeter to be coupled into the shared A/D converter.

Cheers, Alan.

Alan

Would the wind direction vane moving - i.e. switching the reeds - during the temperature measurement affect the reading? As most FO don't measure the intermediate 'two relay' positions very well if at all, I can also see positions that are 'open circuit' being possible (I think there is a base resistance in there is well though iirc?). If the vane is blowing around vigorously in the wind this is more likely to happen during the temperature reading than when it is calm.

mcrossley wrote: I can also see positions that are 'open circuit' being possible (I think there is a base resistance in there is well though iirc?).

Hi Mark,

I don't believe that there is a parallel resistance in the sensor (it would need to be many 100s of kohms) but there might be a high value (1Mohm?) in the transmitter module. Certainly if you unplug the wind cable, the "pointer" disappears (or should do) from the Console LCD display. That is what I would also expect to occur if no reeds are closed.

I don't think that movement of the vane will affect the A/D (temperature) measurement because at that time there is no voltage applied to the "driven" end of the vane resistors. But my comments are only theoretical and I'm actually quite surprised that the FO technique works as well as it (generally) does.

Cheers, Alan.

My 3080 had lots of small spikes showing up on the Internal Temp / Humidity / Barometer graphs in Cumulus, and the console was fluctuating in the same three readouts. The batteries in the console were a year old so I changed them. Spikes now gone, readouts on console OK. Not sure if the removing the batteries cured the problem or the batteries were in need of a change.

Colin

Hi,

I have had another big spike today that hit my Dew Point Temperature and Humidity at the same time.
The wind was only about 9mph at the time.

I will change over the batteries this weekend to see if that helps, if not I may
buy a replacement transmitter from Maplin and see if that sorts it out.

This has only started happening the past couple of weeks and has been fine before then.

Just to confirm I had extended the cables to the windvane and anometer early last year
and did not notice any problems. It only seems to have started very recently.


New Temperature Spike today:




Humidity spike today

Large spikes should be avoidable using a spike removal setting.

It will just be the Humidity value - that is also used in the dewpoint calculation.

Hi Steve,

Thanks for the info.

Could you suggest what value I should put into the Humidity TextBox on the Spike Removal section,
in relationship to the spike shown on my previous graph? Mine is currently set at [999.0%]
but am a bit unsure what value to enter.

Thanks in advance.

Mark

btw.
It could be that my transmitter is on the blink as I have just remembered that I was having a
problem with my humidity values not showing correctly, i.e. being at a higher value than it
should, against other local stations. So it coul be that the Humidity sensor is on its last legs.
Therefore I may be better off getting a replacement Transmitter.

You need the smallest change in value that would normally never occur over a 48-second period. For humidity, you should be safe with a value of 10, say, or even less.

Thanks Steve.

I will try that tonight and see if it helps.

Mark

Solorize wrote: I may buy a replacement transmitter from Maplin and see if that sorts it out.

Hi Mark,

Beware that a new transmitter from Maplin may not work with an older Receiver/Console. See for example this recent thread.

IMHO, it's probably better to buy a complete station when on "Special offer" (c. £50) from Maplin or Clas Ohlson, which will then give you "spares" for all the sensors.

Cheers, Alan.

Hi AllyCat,

Thanks for the link. I had a good read through all the information there
and does look like there are differences in the transmitters now.

I bought my WH1081 around 12th February 2011 and on the rear of
the LCD base station it states that it is a WH1081 and the Frequency
is 433.9MHz.

BTW.
Does anyone know the rough date that the wireless protocol changed
from old to new?


Looking on the Maplin website at:
http://www.maplin.co.uk/n96fy-and-n96gy ... ter-676616
it does not state what frequency the transmitter is.

As there are no "special offers" on the WH1081's at the moment. I may
just try my luck and just buy a replacement transmitter and hope that
it pairs up with the LCD base unit. If it does not work then I will have
to hold out until a "special offer" is on and just buy a complete station.

BTW.
Has anyone had any luck with buying a new transmitter from Maplin
and have it pair successfully with their original base unit?

If so, can you let me know when you purchased your original weather station
and was it from Maplin's.
As if it was around the same time as when I purchased mine it may
mean that it will work.