WH1080 rain gauge versus graduated cylinder.

[dionaea]

I have been concerned for some time that my weather station rainfall readings were on the low side. I have constructed a Graduated cylinder gauge using an 80mm. diameter perspex tube 1 metre long. A funnel fits the top of the tube perfectly with no overlap & I have a 1 metre rule on the tube. To my mind this should be pretty accurate. During the period July to today my weather station shows rainfall of 99.7mm whereas the cylinder shows 110mm. I assume that the cylinder is more accurate as I have read about the tipping bucket not tipping if the rainfall is a little short etc. I regularly check for spiders etc. Is there any point in of adding some sort of multiplying formula (& I have no idea how to do this) to my Cumulus 1 program to compensate for the shortfall? Have other users found this sort of discrepancy?

Cheers,
Chris

[steve]

TBRs are known for under-reading for lots of reasons. Yours seems to be under-reading by about 10%, so you could apply a multiplier of 1.1 for rain in Cumulus, and see how that goes.

[beteljuice]

If you are happy that you have a genuine 'shortfall' of approx. 10 %, then you can compensate for this by:

Configuration => Calibration => Multipliers

Set rainfall to 1.10

Oops - cross posted with Steve

[dionaea]

Thanks Steve & Beteljuice for that - the only problem is - exactly what do I do & in what file? Didn't I mention that I am a noddy?

Cheers,
Chris

[beteljuice]

Err .........

That would be open Cumulus, look at the menu at the top of the screen, and:

Configuration => Calibration => Multipliers

Set rainfall to 1.10

... but I did forget to mention to press the 'OK' button when done

[dionaea]

Thanks Beteljuice - I hadn't noticed that part of Cumulus before - told you I was a noddy!

Cheers,
Chris

[Repairman77]

We had 2.5mm of rain this morning and the console read 1.2mm.
The calibration was set to 1.21 and seemed to work previously.
I've had to set it to 2.00 now to get a comparable reading with the rain sensor.
I recently made a sock shaped cover that fitted over the rain sensor out of dressmaker's net.
I was wondering if this was the reason for the lack of calibration from the EasyWeather sensor.
Today I have put a small brass weight on top of the net on the top of the sensor box to make sure the net is held firm on the top and this creates a dip in the net as well.
I wonder if anyone else who has used net around the rain sensor has had a similar problem.
Mike.

[Spelve]

Hi Mike,
I bought a second hand w8681 in August and had a similar problem in that the rainfall only recorded 50%(approx) of that shown in graduated cylinder.
Investigated and found that the tipping buckets only recorded in one direction.
Thought this must be a fault with the reed switch and considered getting replacement. (Maplin supply these)
In the end it was easier to use a 2X multiplier and settle for less accuracy on the rain readings.
Best wishes
Richard

[Repairman77]

Well yesterday and in the early hours of the morning we had 13mm of rain as measured by the conventional rain gauge.

The WH1080 remote sensor reported 7.4mm and that is with the multiplication factor set to 1.3 in Cumulus.

The remote sensor seems to have a mind of it's own; some days it will agree with the rain gauge and other days it will be wildly out.
I've seen various mods on these forums for improving it but I wonder if it's really worthwhile.

I think the whole design is flawed to some extent. Unfortunately all but the very expensive remote units seem to be designed along similar lines. I think one of the fundamental reasons for failure is the size of the collecting funnel; logic denotes that the larger the funnel the more accurate the reading will be.

I think I'll just continue to measure and record rain data with the rain gauge, and give the remote readings a miss.

Mike.

[gemini06720]

I think the whole design is flawed to some extent. Unfortunately all but the very expensive remote units seem to be designed along similar lines. I think one of the fundamental reasons for failure is the size of the collecting funnel; logic denotes that the larger the funnel the more accurate the reading will be.

Mike, I think you might be right on this point. I now have (or had for the past 5 years) a Vantage Pro2 ISS (integrated sensor suite) and the rain collector cone is at least 15cm (6in) in diameter with a depth of 25cm (10in) ... and I will not give more accurate measurements until spring...

I still have an old La Crosse WS2310 weather station and the rectangular rain collector is 12*5.5cm (4.75*2.25in) with a depth of 5cm (2in).

That is quite a difference in rain-collection area between the 2 weather stations...

[Repairman77]

Yes Ray; I'm sure that is the case.

The professional rain gauges seem to be a minimum of 6" diameter, and the better ones 8" diameter. That largest size must have been chosen for a reason; perhaps an even larger funnel would not bring any appreciable improvement in accuracy.

I suppose it would be possible to make a large funnel that sat over the top of the remote sensor box that would collect a lot more rain than the existing rectangular funnel and make it a little more accurate. You could compensate for this in the calibration section of Cumulus with a negative number.

Yesterday for instance we had very light rain in the morning, around 2mm but nothing was registered on the remote unit as there appeared to not be enough to measure over the long time period of a few hours. A larger funnel would probably have measured something. However the down side of a large funnel would be ascertaining as to whether the system could cope with a downpour; possibly not.

My existing rain gauge in the garden is of the conical type with a 4" funnel; I intend to make a 6 or 8" version later and calibrate that; I should be able to get more accurate readings then.

What amazes me is the price of pro rain gauges. After all it's only a cylinder, a funnel and a measuring glass. Up to $200!
http://www.novalynx.com/260-2520.html

Mike.

[gemini06720]

...However the down side of a large funnel would be ascertaining as to whether the system could cope with a downpour; possibly not...

Mike, the rain collector cone of my Vantage Pro2 station has, so far, been able to handle all the downpours we had over the past 5 years ... and at times, the amount of rain we receive (here in our valley on the east coast of Vancouver Island, British Columbia, Canada) can be ... pretty incredible...

The only time (for a couple of days) I had problems with the collector cone was ... when it got plugged ...

[Repairman77]

Ray, I was wondering if the much smaller WH1080 cone and bucket system could handle a large amount in one go; perhaps there wouldn't be a problem.
However another problem came to mind; a large funnel would collect even more raindrops before they started falling into the hole, in very light rain conditions. Coating the funnel with the acrylic window cleaning stuff may help though.
Think this is a 'no win' situation.
Mike.

[beteljuice]

Repairman, If I remember correctly, your rain guage is 'on high'.

Generally speaking the higher off the ground you go the greater the velocity of any wind, and therefore windshear !

There is a reason why there are different recommended heights / situations for the different sensors.

Mentioned elsewhere recently, I modifed my La Crosse rain bucket x10 (rain guttering proved to be the same 'width' and made for a very simple mod ).
I had concerns about its ability to cope with a downpour, but over three years there have been no obvious shortfalls - only snow problems and bird seed

[Repairman77]

Thanks Beteljuice,

Yes my rain sensor is with the other sensors around 10M up on the tower but with very little wind today 8-18mph I wouldn't have thought that was the problem of the low reading.
Actually in the past when it has been pretty well the same as the rain gauge it has been far more windy.

It would be very difficult to actually move the sensor to a lower place that was in the clear at my location. Where the other manual rain gauge is would be far too far away from the transmitter.

I may try a larger funnel when the weather improves but at the moment climbing over a wet roof at 3 degrees is not on the cards.

I think we have plenty of rain to come over the next few days so I'll monitor the difference then.

I did notice that the pro 6" and 8" tipping bucket units are rated at 1% accuracy but they do have a precision made movement, not a piece of cheap plastic.
http://www.novalynx.com/products-rain-gauges.html

Mike.