Author Topic: RW 8" tipper  (Read 875 times)

0 Members and 1 Guest are viewing this topic.

Offline CW2274

  • Forecaster
  • *****
  • Posts: 6824
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
RW 8" tipper
« on: January 22, 2024, 07:11:16 PM »
Thought I'd stir up the pot since I'm bored, and looking for actual, factual data. I've had my RW tipper(s) for many years and calibrate my own. The original word I got from RW themselves is that it takes exactly 800 ml of water. Fine. Then a few years back, I was told 801. Fine. Then I see this Novalynx chart     https://duckduckgo.com/?q=novalynx+rain+comparison+chart&t=ffab&iax=images&ia=images&iai=https%3A%2F%2Fnovalynx.com%2Fimages%2F260-2595-calibration-table.jpg   and see it says 824.  Little different, ya think? Some other 8" are the same, some are not. I called RW and asked about the discrepancy, and he stated he had no idea. You'd think the manufacturer would be in the know, but are they? Both my RW buckets always read higher than my two Stratus's...about 4-5 %. So what's what? This stuff is waaay above my pay grade, and those on this board are as good as it gets. So again, what's what?

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 2013
    • Weather at Eau Claire, WI
Re: RW 8" tipper
« Reply #1 on: January 22, 2024, 07:51:45 PM »
The physics or geometry of the situation is that an 8" throat takes the same amount of water in volume to accumulate 1" of rainfall.

What goes on inside has zero to do with the amount needed.

However as you allude to, there are lots of adjustments inside and the gauge itself needs to be very level for the tipper to function. 

One winter I spent a lot of time with the kitchen sink and a very slow dripper to make sure no drops were lost during the actual tip when the divider was vertical. Even with precisely measured (by weight, which any cook knows is more accurate than volume, especially of that amount around 800 cc or ml or grams, all within the range of one another for practical purposes) and I got, from the same gizmo and run the same afternoon, variation of a tip or so, which I thought was very close.



The rain rate makes a difference.  The amount of gunk from the atmosphere washed into the funnel and making the water stick to the surface and not flow into the tipper is a concern (but not for volume calculations).

I am amazed that even using chemistry techniques of volumetric measurement, the amount in a beaker (assuming it is accurately calibrated and almost none are) is way off in percentage.

Using a Pyrex measuring cup?  Forget it.  Get a scale, do a bunch of known weight measurements to see how reproducible they are, then use that to figure out how many ml you need.  Getting it all out of the holding container is another error.

The number of errors adds up in a hurry.

This is all for the lab calibration.  Obviously being out of doors, even within a few feet of one another is also fraught with errors, as one can tell from watching sheets of rain fall during a heavy storm and seeing wind effects, so the fact that two gauges out side measure differing amounts also is a factor.

The pictures of the old National Weather Bureau rain measuring stations with their 12" throats and what I called the "Green Giant Loin Cloth" of  the bigger circle of hanging metal sheets to try to reduced the effect of wind comes to mind.

To me, the biggest error is the mounting of the device out doors.  Inside, of course, should be a highly controlled test, with it being precisely level, and testing the unit to make sure the exact number of drops or calibrated valume causes the tip to occur on each side. Sort of like a pendulum clock being out of vertical when the tick and tock are not the same (for most big clocks) can throw things off.

For the 8" volume needed to give precisely one inch of rain, take 8" diameter, get the area by (pi x radius (4") squared) = 3.14*16 = 50.26 square inches, then times 1" of rain which makes for 50.26 cubic inches, and then convert to cc which is about 1 ml which is about 1 gram (16.39 cc is one cubic inch) = 823.76 cc,   Since one cc of water weighs 0.9982 grams, then it is pretty much within the range of 1 to 1.

Scales we all have are better than they were, but none of this would pass even basic chemistry in high school, say nothing of college and of course quantitative analysis labs, where we could measure the weight of a finger print.  But I doubt we need that accuracy.  Be sure to carefully measure, with an accurate ruler, the throat of your collecting funnel.  It may not be exactly 8", although RainWise is pretty good.  High Sierra and Campbell and Young and Texas Weather Electronics (the Gold Funnel ones) have never been off when I've checked with the best rulers I have.  One old brass official (antique) NWS gauge has an almost knife edge to collect the rain, and assumes no wind.

Check the level.  Check the equality of each bucket tip, check for debris.

But when you have a consistant difference from one type of gauge (Stratus) to RainWise (tipper) then you'd be correct in thinking something, somewhere is a wee bit off.

Oh, don't forget evaporation from collector gauges vs. tippers.

The list goes on and on.

I'd be interested in hearing what other's experience has been with trying to make sure their gauges are the most accurate they can be.
Dale


And
ECWx.info
&
ECWx.info/t/index.php

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 2013
    • Weather at Eau Claire, WI
Re: RW 8" tipper
« Reply #2 on: January 22, 2024, 08:00:08 PM »
This is the best image I can find for the wind shielding device currently used (note the tipping bucket RM Young device.)
The old Black and White photo was more dramatic, but I can't find it right now.

 [ You are not allowed to view attachments ]
ECWx.info
&
ECWx.info/t/index.php

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 2013
    • Weather at Eau Claire, WI
Re: RW 8" tipper
« Reply #3 on: January 22, 2024, 08:01:19 PM »
For those interested in the effect of wind, here is a start:

https://physics.stackexchange.com/questions/181263/rain-measurement-with-wind-blowing
ECWx.info
&
ECWx.info/t/index.php

Offline CW2274

  • Forecaster
  • *****
  • Posts: 6824
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
Re: RW 8" tipper
« Reply #4 on: January 22, 2024, 08:04:12 PM »
Thank you. Both my RW tippers are screwed into the top of my cinder block wall caps, leveled myself, x and w axis, and are rock solid. My discrepancies are consistent, within reason. All four gauges are the same height and within 15' of the other. I see by your calculation that the Novalynx table is accurate, and RW, the manufacture, is not. All this time I've been doing it wrong? Sure seems that way.

Thank you again for your reply.

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 2013
    • Weather at Eau Claire, WI
Re: RW 8" tipper
« Reply #5 on: January 22, 2024, 08:08:47 PM »
I may have been mistaken on the rain gauge in the wind shield.  It may be a NovaLynx, but essentially the same as an RM  Young.
ECWx.info
&
ECWx.info/t/index.php

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 2013
    • Weather at Eau Claire, WI
Re: RW 8" tipper
« Reply #6 on: January 22, 2024, 08:17:00 PM »
You may have been doing it right, but using erroneous values given to you by someone who should know better.

Not as bad as crashing a multi-million dollar Mars Probe because of a calculation error, but you know the feeling.

My guess is that the best would be to hope for a long slow drizzle rain some time (we've not had one in years here, very disappointing, they are great to sleep to) but only the monsoon deluges that have been short and far between, and see what you gauges show.  A neighbor who is retired had a whole flat area full of cheap and expensive tippers, CoCoRaHs and other high quality stuff that was collectors all within a 15 foot radius.  All summer (rainy_ long he laboriously took and recorded measurements.  To have them within a few 0.01" was rare and one confusing storm he had one of them off by TENTHS of an inch, while they had previously been pretty much spot on.  No sign of a squirrel or dog peeing in one, either.  Strange things happen.

I'm thinking you can have fun this summer between storms running a half dozen volumes through the tippers to see how they come out, based on your measurement of the throat and the recalculation of the volume needed, then weighing the water rather than using a kitchen appliance to get the volume.  The final and likely frustrating thing would be to use an eye dropper or syringe from an animal supply store to count the drops needed to get each side to tip.  Once you start trying to adjust one a little bit, that will throw off the whole calibration, so the other side will need tweaking.  Nonetheless, the Novalynx chart will help.

Good luck.  And pay no attention to the guys who have their unit nailed to the top of an unlevel post and compare to the TV station a couple miles away.

Enjoy the fun.  Dale
As my wife says, it keeps me out of the bars.
ECWx.info
&
ECWx.info/t/index.php

Offline CW2274

  • Forecaster
  • *****
  • Posts: 6824
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
Re: RW 8" tipper
« Reply #7 on: January 22, 2024, 08:33:28 PM »
My data has been collected over many years. I've calibrated both tippers dozens of times just to be sure they are as good as they can be over the years. All my calibrations are done by weight with a scale that is checked for accuracy, every time. It is empirical and consistent, once again, within reason. You answered the question I was most concerned with, what is the correct amount. RW should be taken to the tipper woodshed for this. Unacceptable.

Thank you again for you time.

Offline Cutty Sark Sailor

  • WxElement panel
  • Forecaster
  • *****
  • Posts: 3399
    • Frankfort Weather - TwinHollies WeatherCenter
Re: RW 8" tipper
« Reply #8 on: January 24, 2024, 09:15:14 AM »
As Dale, et.al have mentioned...
A lot of confusion may track back to an error reflecting the exact inside 'diameter' / radius / volume of the collecting cylinder. Just because it advertises 8 inches may NOT reflect the truth. I know fellows who are famous for such.
Also, water is at maximum density at 4C - .99997 gm/cm3. Becomes slightly less dense as temp increases / or falls until frozen. And  variations in air pressure changes decimal slightly. We'll ignore the pressure.

1 inch = 25.4 mm, and since the density of liquid water at varying "livable" temps might be considered insignificant, then "1 inch equals 2.54 cc=2.54ml= 2.54gm (per temperature: 50C= 0.998cm3 or 2.535 gm).
An 8 inch inside-diameter cylinder 1" tall should have a volume of 50.265 in3, or converting to metric,
it is a 20.32 cm diameter 2.54cm tall cylinder ( a radius of 10.16 cm) = 823.7 cm3 volume x .99997(4C) = 822.975 gms

So, (KISS), if my 8 inch gauge does NOT decant 823 gm per inch at 4C (822gm @ 122F / 50C) ... then
1. it isn't an 8" inner diameter gauge.
2. my rain isn't pure water (ridiculous example: 'Heavy Water' masses about 11% greater)
3. the scale isn't accurate
4. the rain is extremely warm
5. the air pressure is absurd
6. I goofed up the decant
7. I assume that a gauge marking of one inch is accurate, when in fact it may not be, and eyeball and advertised dimensions don't correlate with math.

If a manufacturer specified an 801gm per inch reference, then that reflects an inner diameter of about 7.89". And if their measurement markings are actually calibrated for a (7.9") cylinder, then what's the problem?....
...there is none on the eyeball method... or on the 'weight' method... IF the right baselines are observed.   

Properly marked and calibrated gauges of real 8" compared to a misrepresented but accurately calibrated 7.89" gauge will SHOW identical collection values, but WEIGH different masses!

 This of course completely ignores calibration or accuracy of any tipping / electronic device recording such input. Collect it with bad advertising, then you get what you get.
« Last Edit: January 24, 2024, 10:24:07 AM by Cutty Sark Sailor »
 


Offline CW2274

  • Forecaster
  • *****
  • Posts: 6824
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
Re: RW 8" tipper
« Reply #9 on: January 24, 2024, 06:35:32 PM »
If a manufacturer specified an 801gm per inch reference, then that reflects an inner diameter of about 7.89". And if their
measurement markings are actually calibrated for a (7.9") cylinder, then what's the problem?....
Ummm, what? What measurement markings? It's a tipper. All I wanted to know is the figure of 801 or 824 accurate for this tipper, which I measured the collector diameter, and it's exactly 8", maybe a hair over. Your over-the-top explanations are nothing more than subterfuge AFAIC. I'm going to redo the calibration at 824, as Dale and NovaLylx state.

Offline Cutty Sark Sailor

  • WxElement panel
  • Forecaster
  • *****
  • Posts: 3399
    • Frankfort Weather - TwinHollies WeatherCenter
Re: RW 8" tipper
« Reply #10 on: January 24, 2024, 08:49:58 PM »


hmmm, what? What measurement markings? It's a tipper. All I wanted to know is the figure of 801 or 824 accurate for this tipper, which I measured the collector diameter, and it's exactly 8", maybe a hair over. Your over-the-top explanations are nothing more than subterfuge AFAIC. I'm going to redo the calibration at 824, as Dale and NovaLylx state.
WEEELLLL... EXCUUUUUUSSSSEEEE ME! Thought you wanted to stir the pot?

First of all... where do you come off basically calling me  'liar, con-man , deceiver,' etc--referring to my 'over-the-top' explanations as 'subterfuge'? KMA.
'Redundant' post maybe, unwanted, superfluous, whatever, but done in good faith.

Your 'measurement marker' IS the tipper... and its counting circuitry and software... isn't it?

BTW...congratulations for the conclusion that 824 would be the most reliable figure for a collector that is"... exactly 8", maybe a hair over."   
I suppose 824 could be equivalent to 823 + a hair.


 

 
« Last Edit: January 24, 2024, 08:53:39 PM by Cutty Sark Sailor »
 


Offline CW2274

  • Forecaster
  • *****
  • Posts: 6824
    • Conditions @ CW2274 West Tucson-Painted Hills Ranch
Re: RW 8" tipper
« Reply #11 on: January 24, 2024, 09:08:22 PM »


hmmm, what? What measurement markings? It's a tipper. All I wanted to know is the figure of 801 or 824 accurate for this tipper, which I measured the collector diameter, and it's exactly 8", maybe a hair over. Your over-the-top explanations are nothing more than subterfuge AFAIC. I'm going to redo the calibration at 824, as Dale and NovaLylx state.

BTW...congratulations for the conclusion that 824 would be the most reliable figure for a collector that is"... exactly 8", maybe a hair over."   
I suppose 824 could be equivalent to 823 + a hair.
Not my conclusion. That said, better than anything you offered.