26 replies to this topic

[mullender83]

Snowstorm1, on 06 August 2011 - 09:54 , said:

Mine is reading 89% when others are 93%+ in my area actually that seems preety okay that doesnt it?

Ive literally just dont know what to do with it when i know for a fact its wrong ive just lost all ideas that screen one seemed okay till the reply was it may not work completely but id just be happy for it to achieve correct constantish readings

If your reading is all ways a constant 1/2/3c off what it should be and you are sure of that then maybe its worth calibrating your weather software to adjust temperature down -1c/-2c?

[Snowstorm1]

mullender83, on 06 August 2011 - 11:51 , said:

If your reading is all ways a constant 1/2/3c off what it should be and you are sure of that then maybe its worth calibrating your weather software to adjust temperature down -1c/-2c?

Its not only that though it varies between 1c to 3c but when it rains heavily it can get to around the right temp but as soon as it stops it goes back up also and when the sun is on it can go 6c out but in winter it does give better readings


Ive adjusted temp down by 1 degree and humidity increased by 4% see how this works now

Edited by Snowstorm1, 06 August 2011 - 19:03 .

[reef]

The humidity sensor in the WMR100 is notoriously poor. Its peak range is 40-80% where it is accurate to +/- 5%. Below 40% and above 80% that accuracy decreases to +/-7%! Essentially this means that you could be reading 86% when an 'official' station is reading 93% and it'd still be within specification.

The reason your temperatures are more accurate during the daytime when its raining and during winter is because at those times solar radiation is at a minimum. You can set an offset (temperature adjust) but then you'll under-read during the night.

I think you'll probably just have to shield it the best you can and accept its in an environment thats slightly warmer than other stations elsewhere.

[Snowstorm1]

reef, on 06 August 2011 - 21:05 , said:

The humidity sensor in the WMR100 is notoriously poor. Its peak range is 40-80% where it is accurate to +/- 5%. Below 40% and above 80% that accuracy decreases to +/-7%! Essentially this means that you could be reading 86% when an 'official' station is reading 93% and it'd still be within specification.

The reason your temperatures are more accurate during the daytime when its raining and during winter is because at those times solar radiation is at a minimum. You can set an offset (temperature adjust) but then you'll under-read during the night.

I think you'll probably just have to shield it the best you can and accept its in an environment thats slightly warmer than other stations elsewhere.

Oh right is putting the temperature down by 1 degree and humidity up by 4% can help my problem during spring/summer/autumn?

Then maybe in winter half that so its only decreased by 0.5c and 2% and what months would you suggest to do this in winter when solar radiation is at minimum?

[reef]

Snowstorm1, on 06 August 2011 - 21:10 , said:

Oh right is putting the temperature down by 1 degree and humidity up by 4% can help my problem during spring/summer/autumn?

Then maybe in winter half that so its only decreased by 0.5c and 2% and what months would you suggest to do this in winter when solar radiation is at minimum?

I wouldnt change the temperature 'live' in your software, but rather do an adjustment in your records if its been a sunny day. So for example on a very sunny day if you had a max temp of 24C and min of 12C, take 2C off when you record the temperature (so 22C/12C). On the other hand, if its been overcast and had a max of 18C with little solar radiation, then leave it be. Its very crude though, I wouldnt really recommend it. Humidity is a tricky beast, Id be tempted to leave it be. The problem with a blanket adjustment is that your sensor might not necessarily be out by the same amount throughout its range. You could end up pushing it even further out quite easily. Plus if you're so determined to have your results match other sites then it sort of defeats the point of having your own station.

At the end of the day though you're recording the temperature in your garden which is actually warmer than sites nearby. No amount of tinkering will change that fact unfortunately. Id concentrate on finding the absolute best location for the sensor and shield it as best as possible from solar radiation. Most of us have to do this, unless you're lucky enough to have a massive garden that is!

[Snowstorm1]

reef, on 06 August 2011 - 21:24 , said:

I wouldnt change the temperature 'live' in your software, but rather do an adjustment in your records if its been a sunny day. So for example on a very sunny day if you had a max temp of 24C and min of 12C, take 2C off when you record the temperature (so 22C/12C). On the other hand, if its been overcast and had a max of 18C with little solar radiation, then leave it be. Its very crude though, I wouldnt really recommend it. Humidity is a tricky beast, Id be tempted to leave it be. The problem with a blanket adjustment is that your sensor might not necessarily be out by the same amount throughout its range. You could end up pushing it even further out quite easily. Plus if you're so determined to have your results match other sites then it sort of defeats the point of having your own station.

At the end of the day though you're recording the temperature in your garden which is actually warmer than sites nearby. No amount of tinkering will change that fact unfortunately. Id concentrate on finding the absolute best location for the sensor and shield it as best as possible from solar radiation. Most of us have to do this, unless you're lucky enough to have a massive garden that is!

Its above the temp on a night anyways normally so ill see how it works and see if it gives constantish readings and I dont keep records? That is my best location I think however I am going to try and move it below the level of the wall instead of above as that will mean its shaded more

[Snowstorm1]

Is that a good idea?