I've been a tad suspicious my Davis tipping bucket raingauge has been
under recording of late and have noticed it being consistently lower
than Shoeburyness. Some calcuations over the last few weeks have found
it to be about 30% less than Shoeburyness (03693). I also set up my old
manual gauge and found similar discrepancies.

Subsequently, I had the solder out and installed a new 'reed switch' and
using some of the kitchen utensils spent about an hour calibrating.

Fortunately I hadn't lost an email Bernard Burton sent me some years ago
with the following calculation methology:

Example for 5 inch gauge(standard)
Diameter = 5 inches = 127mm
Area = 12667.687 sq mm.
Amount of water needed for 1mm equivilent is 12667-687/1000 - 12.68 ml.
If you pour in say 10 ml = 10000 cu mm, that is equivalent to
10000/12667.687 = 0.789 mm of rainfall.

The Diameter of my Davis bucket measures at 164mm
Area = Pi x r squared
= 3.14159.. x 82 sq'd = 21124.069
10 ml = 10000 cu mm, that is equivalent to
10000/21124.069 = 0.473 mm

Therefo Aprox:
10 ml = 0.47mm (2 tips)
30 ml = 1.41mm (7 tips)
100ml = 4.70mm (23 tips)

Obvious any slight error in fine adjustment will increase the larger the
amount poured in. eg the bucket may have just tipped on the last drop
poured in or almost full on last drop yet not tipped :-0

Do my calculations seem correct ?

Many thanks
--
Keith (Southend)
http://www.southendweather.net
e-mail: kreh at southendweather dot net

Keith(Southend) wrote:

The Diameter of my Davis bucket measures at 164mm
Area = Pi x r squared
= 3.14159.. x 82 sq'd = 21124.069
10 ml = 10000 cu mm, that is equivalent to
10000/21124.069 = 0.473 mm

Therefo Aprox:
10 ml = 0.47mm (2 tips)
30 ml = 1.41mm (7 tips)
100ml = 4.70mm (23 tips)

Obvious any slight error in fine adjustment will increase the larger the
amount poured in. eg the bucket may have just tipped on the last drop
poured in or almost full on last drop yet not tipped :-0

Do my calculations seem correct ?

Yea, they seem about right.

I think the Davis bucket diameter is d = 0.165 m. Using this value we
can easily calculate the area as:

d = 0.165; r = d/2; a = pi*r^2 = 0.021382465 m^2;

This area is about 46.767 smaller than 1 m^2, so if you pour 1 litre
over a 1 m^2 container the level will rise 1 mm, if you pour the same
amount over another container having an area 46.767 smaller then the
level will rise proportionally higher or to 46.767 mm.

To get 0.2 mm we'll need to pour 0.2/46.767 = 4.28 mL per bucket tip.
With this value you can then calculate if there is a systematic error by
slowly pouring say 1 L and see how much it measures.

I'm sure there will also be another error due to the rain rate, i.e.
measurements get lower and lower from the correct value as the rain
rate increases. If you know or calculate this error as a function of the
rain rate then you can post process the rainfall values to compensate
and get more accurate data.

Another error might be introduced by the height at which the rainfall
gauge is placed [1], the higher it is the less rain it will catch (and
measure).

[1]
http://www.onerain.com/includes/pdf/...ageRecords.pdf

On Oct 10, 8:31*pm, ldpc wrote:
Keith(Southend) wrote:
The Diameter of my Davis bucket measures at 164mm
Area = Pi x r squared
* * *= 3.14159.. x 82 sq'd = 21124.069
10 ml = 10000 cu mm, that is equivalent to
10000/21124.069 = 0.473 mm

Therefo * *Aprox:
10 ml = 0.47mm (2 tips)
30 ml = 1.41mm (7 tips)
100ml = 4.70mm (23 tips)

Obvious any slight error in fine adjustment will increase the larger the
amount poured in. eg the bucket may have just tipped on the last drop
poured in or almost full on last drop yet not tipped :-0

Do my calculations seem correct ?

Yea, they seem about right.

I think the Davis bucket diameter is d = 0.165 m. Using this value we
can easily calculate the area as:

d = 0.165; r = d/2; a = pi*r^2 = 0.021382465 m^2;

This area is about 46.767 smaller than 1 m^2, so if you pour 1 litre
over a 1 m^2 container the level will rise 1 mm, if you pour the same
amount over another container having an area 46.767 smaller then the
level will rise proportionally higher or to 46.767 mm.

To get 0.2 mm we'll need to pour 0.2/46.767 = 4.28 mL per bucket tip.
With this value you can then calculate if there is a systematic error by
slowly pouring say 1 L and see how much it measures.

I'm sure there will also be another error due to the rain rate, i.e.
measurements *get lower and lower from the correct value as the rain
rate increases. If you know or calculate this error as a function of the
rain rate then you can post process the rainfall values to compensate
and get more accurate data.

Another error might be introduced by the height at which the rainfall
gauge is placed [1], the higher it is the less rain it will catch (and
measure).

[1]http://www.onerain.com/includes/pdf/whitepaper/InconsistentRainGageRe....

Thanks ldpc, seems to be working fine in this afternoons drizzle ;-)

Keith (Southend)

On 11 Oct, 15:45, "Keith (Southend)G"
wrote:
On Oct 10, 8:31*pm, ldpc wrote:





Keith(Southend) wrote:
The Diameter of my Davis bucket measures at 164mm
Area = Pi x r squared
* * *= 3.14159.. x 82 sq'd = 21124.069
10 ml = 10000 cu mm, that is equivalent to
10000/21124.069 = 0.473 mm

Therefo * *Aprox:
10 ml = 0.47mm (2 tips)
30 ml = 1.41mm (7 tips)
100ml = 4.70mm (23 tips)

Obvious any slight error in fine adjustment will increase the larger the
amount poured in. eg the bucket may have just tipped on the last drop
poured in or almost full on last drop yet not tipped :-0

Do my calculations seem correct ?

Yea, they seem about right.

I think the Davis bucket diameter is d = 0.165 m. Using this value we
can easily calculate the area as:

d = 0.165; r = d/2; a = pi*r^2 = 0.021382465 m^2;

This area is about 46.767 smaller than 1 m^2, so if you pour 1 litre
over a 1 m^2 container the level will rise 1 mm, if you pour the same
amount over another container having an area 46.767 smaller then the
level will rise proportionally higher or to 46.767 mm.

To get 0.2 mm we'll need to pour 0.2/46.767 = 4.28 mL per bucket tip.
With this value you can then calculate if there is a systematic error by
slowly pouring say 1 L and see how much it measures.

I'm sure there will also be another error due to the rain rate, i.e.
measurements *get lower and lower from the correct value as the rain
rate increases. If you know or calculate this error as a function of the
rain rate then you can post process the rainfall values to compensate
and get more accurate data.

Another error might be introduced by the height at which the rainfall
gauge is placed [1], the higher it is the less rain it will catch (and
measure).

[1]http://www.onerain.com/includes/pdf/whitepaper/InconsistentRainGageRe...

Thanks ldpc, seems to be working fine in this afternoons drizzle ;-)

Keith (Southend)- Hide quoted text -

- Show quoted text -

I have a little more confidence in my calibration excersise the other
week as yesterdays 24 hour rainfall total 06 to 06 hours was 15.4mm,
exactly the same as Shoeburyness:
AAXX 22064 03693 16967 /1506 10110 20106 39963 49965 53001 69942 90550
333 20110 3/008 70154 8//99=
May be more luck than judgement, but it looks more probable now :-)

Keith (Southend)
http://www.ssouthendweather.net
'Weather Home & Abraod"