Author Topic: WH40 vs WH65 Rain Measurements  (Read 18170 times)

0 Members and 1 Guest are viewing this topic.

Offline Rover1822

  • Forecaster
  • *****
  • Posts: 1739
    • Mini Wind and Solar Data project
Re: WH40 vs WH65 Rain Measurements
« Reply #125 on: August 24, 2020, 03:56:41 PM »
I still have to say that nothing provided so far, provides me a reason to purchase a WH40.


Ambient:
  WS-2000
  PM 2.5(2)
  WH31B(2)
  WH40E
  WH31P
EcoWitt:
  GW1100
  GW1000(4)
  WH31(2)
  WH57
  WH51(12),
  WH40
  WH5360B
  WN34S
  WittBoy WS90 + GW2000
  WS90 (other one) + GW1100
Personal Sites: Weather Cam

Offline Platokidd

  • top side land of Lincoln
  • Forecaster
  • *****
  • Posts: 516
Re: WH40 vs WH65 Rain Measurements
« Reply #126 on: August 24, 2020, 07:10:07 PM »
I still have to say that nothing provided so far, provides me a reason to purchase a WH40.

I hear ya and agree but my money is spent.

My main reason for buying was no need for a ladder. Setting my WH40 and it's funnel to offset 1.42 darn near mirrored my wh65 (ws-2902). That's in rain totals and in rain rate, at-least for me. It took many rains of many types for me to zero in on the right offset. This was also done with both sensors side by side within a foot or 2. I found if I got the rain rates close the rain totals got close until finally I zero in in on 1.42. Kept telling myself keep it simple stupid.. :oops:

If nothing comes of this whole wh40 saga , I'll just go back to offset 1.42 and live with it until something else comes along....




« Last Edit: August 24, 2020, 07:15:45 PM by platokidd »
Ambient    
2- WS-2000 2-WS-2902A ARRAYS 1-CONSOLE 2-WH40-RAIN 1-WH31L 1-METEOBRIDGE 2-PM2.5(WH42B)
3-WH-31E 1-SRX100LX 1-WH31SM
ECOWITT
3-HP2552 1-HP2562 1-WH51 10-WH31-2-WH45-3-WH55 1-WH40 2-WS68 1-WS80
5-WH51 1-WH32EP 1-WN30 1-WH41
2-GW2000 2-GW1100
1-HP10

1-Davis 7714
1-Stratus

Offline ramblinche81

  • Senior Member
  • **
  • Posts: 81
Re: WH40 vs WH65 Rain Measurements
« Reply #127 on: August 24, 2020, 07:16:23 PM »
Here is another way to look at it from a 2 axis perspective.

 [ You are not allowed to view attachments ]

Check my math
Funnel 180mm diam 9cm radius
Area = 254 cmcm

477grams water flowed = 18.7 mm expected total precip
Indicated 21.8mm 16% over expected

806grams water flowed = 31.7 mm expected precip
Indicated 36.0 mm 14% over

If it was less we would talk about splash, evaporation, tipper spillage.

It is overage.  Only thing I noticed is tipper holds water via clingage.. first tip is 2.5ml  second tip is 2.2ml. Coincidence that matches the overage % ???

It started to rain so I stopped testing. On Tuesday I will run two more test and if 13-16% overage persists I will scale calibration gain to .85 and see what happens after.
« Last Edit: August 24, 2020, 07:22:21 PM by ramblinche81 »
July 2020
Ecowitt GW1000B gateway @ ecowitt.net
Ecowitt WN51 soil moisture x8 (potted plants and landscape beds)
Ecowitt WH57 lightning
Ecowitt WH31 temp x2
Ecowitt WH40 rain gauge
Ecowitt WS68 wind sun sensor
Dragon Touch tablet display with PWT app

Offline Rover1822

  • Forecaster
  • *****
  • Posts: 1739
    • Mini Wind and Solar Data project
Re: WH40 vs WH65 Rain Measurements
« Reply #128 on: August 24, 2020, 07:20:58 PM »
I still have to say that nothing provided so far, provides me a reason to purchase a WH40.

I hear ya and agree but my money is spent.

My main reason for buying was no need for a ladder. Setting my WH40 and it's funnel to offset 1.42 darn near mirrored my wh65 (ws-2902). That's in rain totals and in rain rate, at-least for me. It took many rains of many types for me to zero in on the right offset. This was also done with both sensors side by side within a foot or 2. I found if I got the rain rates close the rain totals got close until finally I zero in in on 1.42. Kept telling myself keep it simple stupid.. :oops:

If nothing comes of this whole wh40 saga , I'll just go back to offset 1.42 and live with it until something else comes along....

PlatoKidd, I think you have opened up a lot of eyes, including mine. I was preparing to buy the unit for the same reason you did.
Ambient:
  WS-2000
  PM 2.5(2)
  WH31B(2)
  WH40E
  WH31P
EcoWitt:
  GW1100
  GW1000(4)
  WH31(2)
  WH57
  WH51(12),
  WH40
  WH5360B
  WN34S
  WittBoy WS90 + GW2000
  WS90 (other one) + GW1100
Personal Sites: Weather Cam

Offline hdtvluvr

  • Senior Contributor
  • ****
  • Posts: 214
Re: WH40 vs WH65 Rain Measurements
« Reply #129 on: August 24, 2020, 07:27:41 PM »
Does the rain sensor on the HP2551 have the same issues?  Size of the cup looks the same.
Acurite Atlas
Acurite Access
WeatherUnderground KTNMUNFO2

https://www.wunderground.com/dashboard/pws/KTNMUNFO2

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: WH40 vs WH65 Rain Measurements
« Reply #130 on: August 24, 2020, 07:30:03 PM »
Does the rain sensor on the HP2551 have the same issues?  Size of the cup looks the same.

HP2551 uses the WH65 and so no it doesn't have the issues of the WH40 which is the stand alone rain gauge. Yes they look the same internally....because...well they use the same parts. The problem with the stand alone WH40 is that it is fitted with a larger funnel that seems to present problems. You would think that larger is better but some things that have been noticed is the the outer rim of the WH40 funnel is too flat (not enough slope), walls are not tall enough, and because of the larger size the spoon tipper likely is not keeping up with increased water collection volume.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: WH40 vs WH65 Rain Measurements
« Reply #131 on: August 24, 2020, 07:42:30 PM »
Here is another way to look at it from a 2 axis perspective.

 [ You are not allowed to view attachments ]

Check my math
Funnel 180mm diam 9cm radius
Area = 254 cmcm

477grams water flowed = 18.7 mm expected total precip
Indicated 21.8mm 16% over expected

806grams water flowed = 31.7 mm expected precip
Indicated 36.0 mm 14% over

If it was less we would talk about splash, evaporation, tipper spillage.

It is overage.  Only thing I noticed is tipper holds water via clingage.. first tip is 2.5ml  second tip is 2.2ml. Coincidence that matches the overage % ???

It started to rain so I stopped testing. On Tuesday I will run two more test and if 13-16% overage persists I will scale calibration gain to .85 and see what happens after.

Yeah that is great. I personally am not interested in any correction factor to the WH40 with the 7 in / 180 mm diameter original funnel. The WH40 funnel is flawed in design. Even if you can figure out what correction factor yields adequate values...there is still spillage in more than one place....from bounce out and from tipper not keeping up with water flow rate. In either case these seem like factors that will most definitely be both different with different types of rain. Therefore arriving at some average correction value is not acceptable to me at least. Because sometimes it will be more and other times it will be less. I feel the better approach is to use the WH65 funnel that fixes both the bounce out and flow rate overrun. A correction factor with the WH65 funnel fitted to the WH40 is an adjustment that has to do with only changing funnel size. This correction factor has nothing to do with correcting for design problems of the WH40 funnel.

Perhaps someone else is interested in a correction factor to the WH40 with its original funnel...I'm just not even interested in hearing about it. I just don't think it is the best approach. I'm sure that anyone can buy the WH40 and Ecowitt will probably throw in a free WH65 funnel if you ask for it at time of order. To me this is the best approach. All I think is the objective of this thread is to make the WH40 perform at least equal to the WH65. If the WH65 can be improved ....well then I'm sure the same fix should be possible for the WH40 with the WH65 funnel attached.

What I do think might have been a novel approach would be a totally new funnel for the WH40 with taller walls and/or a greater slope. But Ecowitt engineers have already looked at this and determined that a better funnel design would not fix the problem....likely because there is still a flow rate overflow problem with the larger size funnel. Therefore...the solution falls back to using the WH65 funnel on the WH40. It is either that or a totally completely new sensor ...Lets call it the WH4000 to signify its greatness. Yes I just made that name up...sounded good.

If the Ecowitt engineers felt that a correction factor is all that was required....well I think they'd be pushing that over a firmware update. I just don't think this is the right approach, and I'm glad the Ecowitt engineers have not gone down this path either.

Look, I'm not saying that things need to be my way. I'm just saying this is my opinion and given what we now know. If anyone is pleased with their WH40 and some correction factor...okay then...so be it, I'm happy for you. You are welcome to share what that factor is like Platokidd shared in what works for him based on the rain he sees. Everyone can learn from all these experiences and take the approach that they want.
« Last Edit: August 24, 2020, 08:10:26 PM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline ramblinche81

  • Senior Member
  • **
  • Posts: 81
Re: WH40 vs WH65 Rain Measurements
« Reply #132 on: August 24, 2020, 09:04:52 PM »
Has anyone run a standard volume through 65 and determined if it matches ?  Not vs another Guage vs a real standard.  Back to the accuracy vs precision. You calibrate a scale with known weights, not another scale. You calibrate a ph meter against a std solution, not another ph meter.
July 2020
Ecowitt GW1000B gateway @ ecowitt.net
Ecowitt WN51 soil moisture x8 (potted plants and landscape beds)
Ecowitt WH57 lightning
Ecowitt WH31 temp x2
Ecowitt WH40 rain gauge
Ecowitt WS68 wind sun sensor
Dragon Touch tablet display with PWT app

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: WH40 vs WH65 Rain Measurements
« Reply #133 on: August 24, 2020, 09:19:29 PM »
Has anyone run a standard volume through 65 and determined if it matches ?  Not vs another Guage vs a real standard.  Back to the accuracy vs precision. You calibrate a scale with known weights, not another scale. You calibrate a ph meter against a std solution, not another ph meter.

You make things too complicated. All I do is take a WH65 and compare it to a Stratus manual gauge. If the Stratus says it rained 0.43 inches and your WH65 says it rained 0.41 inches....sounds good enough to me...and these are the sort of numbers that I see. Sometimes they match perfectly.
« Last Edit: August 24, 2020, 10:44:48 PM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline Mandrake

  • Forecaster
  • *****
  • Posts: 1102
Re: WH40 vs WH65 Rain Measurements
« Reply #134 on: August 25, 2020, 03:20:45 AM »
I am afraid I disagree with you Galfert on this one and side with Rambinche81
There are far too many variable at play here to argue about absolute precision on a rain gauge.
I have chosen to stay mute on this subject until now.

If you consider that rain drops vary in size, the atmospheric conditions also vary and that the manual gauge almost certainly will have a slightly different profile to the WH65 and you immediately get a error rate. You can only prove the accuracy by being scientific in your approach.

You could have an array of rain gauges, manual or otherwise and they will all vary to a degree depending on the conditions unless you get perfect rain in perfect conditions.
Until folks recognise the folly of what they are seeking this topic will not die.

Dont get me wrong, it might well be that the design of the catching saucer for the WH65 might be sub optimal in some conditions and that is the real question that needs establishing to see if its possible to improve the design and feed that back to Ecowitt.
G1ZFO

Ecowitt HP2551A + WH65 Tri-Wing (Wunderground: IGUILDFO67)
Ecowitt GW1000 (Wunderground: IGUILDFO68)
Ecowitt GW1000 (Mk2) test environment driving CumulusMX on a RPi 3b
Ecowitt GW2000 (Test)
Ecowitt WS90 Wittboy - Test
Ecowitt WH51 (x6) Soil Moisture Sensor
Ecowitt WH41 PM2.5 AQM Sensor
Ecowitt WH31 (x8) Thermo/Hygro Sensor
Ecowitt WS80 Ultrasonic Anemometer (pre-prod test)
Ecowitt WH57 Lightning Sensor -test
Ecowitt WH32-EP (SHT35) + Davis 7714 Screen
Ecowitt WH45 CO2/PM2.5/PM10 -Test
Ecowitt WN34 Soil Temp Sensor -Test
Ecowitt WN34 Water Temp Sensor -Test
Ecowitt WN35 Leaf Moisture

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: WH40 vs WH65 Rain Measurements
« Reply #135 on: August 25, 2020, 07:05:10 AM »
I'm not claiming that the WH65 is perfect nor optimal. I'm absolutely in agreement that a better rain gauge than the WH65 exists.

Facts:
- The WH40 is not as accurate as the WH65 pretty much consistently in all situations, heavy or light rain.
- The internal parts of both rain gauges are identical. Only funnel size is different.
- Change out the the WH40 funnel for the WH65 funnel and you've physically made a stand alone WH65. All that is now needed is a correcting factor so that it matches the WH65 gauge numbers.

Conclusion: Unless there is something I missed, we can 100% recreate the WH65 with a WH40 by just swapping out the funnel and changing the calibration gain.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: WH40 vs WH65 Rain Measurements
« Reply #136 on: August 25, 2020, 07:08:20 AM »
Mandrake,
I don't understand what you disagree about.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline Mandrake

  • Forecaster
  • *****
  • Posts: 1102
Re: WH40 vs WH65 Rain Measurements
« Reply #137 on: August 25, 2020, 10:16:04 AM »
I guess what I was simply trying to say that simply comparing rain gauges (manual or tippers) is fraught with issues as unless the funnel design is identical you will always get slightly different results in different ambient conditions.

Scientific measurements of tipped water as stated will prove the accuracy (which is being disputed here in this thread)

Getting a design of funnel that works well in all conditions is hard and if I recall correctly the really accurate systems are about a square foot set at ground level.
I am just concerned that there is a frenzy of comments building here that the rain gauge is rubbish which I think is unfair given that its accuracy is more a issue of environmental conditions.
Some folks may find the sensor is fine for 99% of the time.

I would highlight the fact that some folks myself included have issues with the WH65 with its smaller funnel is less effective in light drizzle and if the weather is warm as the light precipitation evaporates before it can collect enough to trip the bucket. The bigger funnel is more effective in that scenario though of course just as impacted by warmth.
G1ZFO

Ecowitt HP2551A + WH65 Tri-Wing (Wunderground: IGUILDFO67)
Ecowitt GW1000 (Wunderground: IGUILDFO68)
Ecowitt GW1000 (Mk2) test environment driving CumulusMX on a RPi 3b
Ecowitt GW2000 (Test)
Ecowitt WS90 Wittboy - Test
Ecowitt WH51 (x6) Soil Moisture Sensor
Ecowitt WH41 PM2.5 AQM Sensor
Ecowitt WH31 (x8) Thermo/Hygro Sensor
Ecowitt WS80 Ultrasonic Anemometer (pre-prod test)
Ecowitt WH57 Lightning Sensor -test
Ecowitt WH32-EP (SHT35) + Davis 7714 Screen
Ecowitt WH45 CO2/PM2.5/PM10 -Test
Ecowitt WN34 Soil Temp Sensor -Test
Ecowitt WN34 Water Temp Sensor -Test
Ecowitt WN35 Leaf Moisture

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: WH40 vs WH65 Rain Measurements
« Reply #138 on: August 25, 2020, 11:34:35 AM »
I guess what I was simply trying to say that simply comparing rain gauges (manual or tippers) is fraught with issues as unless the funnel design is identical you will always get slightly different results in different ambient conditions.
Yes I agree. But using a manual rain gauge that is trusted like a Stratus is a good way to get an idea for how trusted the results are from an automatic rain gauge. I don't expect any amount of fine tuning to ever result in perfect results compared to a manual rain gauge. But this comparison is a good way to determine if your automatic rain gauge is still performing as expected...within its capabilities...or if it has fallen into requiring maintenance.

Quote
Scientific measurements of tipped water as stated will prove the accuracy (which is being disputed here in this thread)
I'm all for more scientific testing to determine accuracy. I think though we are beyond the need to determine that the WH40 is clearly not performing the the level of accuracy of the WH65. Even if the WH40 is acceptable to anyone in its current form, for many the concern is valid. Even the manufacture has agreed that there is an issue. Any additional testing will only confirm what we already know.

Quote
Getting a design of funnel that works well in all conditions is hard and if I recall correctly the really accurate systems are about a square foot set at ground level.
I am just concerned that there is a frenzy of comments building here that the rain gauge is rubbish which I think is unfair given that its accuracy is more a issue of environmental conditions.
Some folks may find the sensor is fine for 99% of the time.
All valid, I agree.

Quote
I would highlight the fact that some folks myself included have issues with the WH65 with its smaller funnel is less effective in light drizzle and if the weather is warm as the light precipitation evaporates before it can collect enough to trip the bucket. The bigger funnel is more effective in that scenario though of course just as impacted by warmth.
I can see exactly what you are saying. Yes the WH65 is not perfect by any means. The WH40 for light rain where bounce out would not occur, and where it would not overrun the rain rate capacity of the tipper would potentially be one circumstance where the WH40 in its unaltered form to perform better than the WH65. It would also be important in this light rain situation that the funnel be both coated by means to reduce water adhesion and the conditions would need to be so that it isn't too warm so that evaporation off the funnel does not occur.

I think we are on the same page. I agree with everything you've said.

Any efforts to improve on the WH40 even if they involve keeping the larger original rain funnel is okay by me. It feel that that would not be the approach that I would take....but I can understand that some people may not want to run the WH40 with the smaller WH65 funnel....especially for those that don't experience the heavy rains that I experience here in Florida.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline K7RMK

  • Senior Member
  • **
  • Posts: 52
    • East Wilson Weather
Re: WH40 vs WH65 Rain Measurements
« Reply #139 on: August 25, 2020, 11:53:29 AM »
I received a small funnel yesterday and tried to take more accurate measurements. I got 7.015" diameter for the large funnel and 4.440" diameter for the small funnel. When we run the math, we get a correction factor of 2.496 which is close enough to 2.5 (my measurements aren't perfect).

(pi(7.015/2)²)/(pi(4.440/2)²)=2.496

2.5 makes sense because the resolution spec for the WH40 is 0.004" and the WH65 is 0.01".

0.01/0.004=2.5
Ecowitt GW2000 + Meteobridge on RPI 4B
Ecowitt HP2551-C Display Console
Ecowitt WH32-EP Outdoor Temp/Humidity Sensor + MeteoShield Pro Radiation Shield
Ecowitt WS68 Anemometer
Ecowitt WH40 Rain Gauge
Ecowitt WH57 Lightning Sensor
Ecowitt WH41 Outdoor PM2.5 Air Quality Sensor
Ecowitt WH43 Indoor PM2.5 Air Quality Sensor
Ecowitt WH31 x4 Temp/Humidity Sensors
Ecowitt WH51 Soil Moisture Sensor
Ecowitt WH34S Soil Temperature Sensor
Ecowitt WN35 x2 Leaf Wetness Sensors

Offline ramblinche81

  • Senior Member
  • **
  • Posts: 81
Re: WH40 vs WH65 Rain Measurements
« Reply #140 on: August 25, 2020, 12:56:02 PM »
Hoping one of you more familiar with hardware and settings can weigh in. Or maybe ecowitt is better suited to explain.

Since there is a calibration factor shown for rain gauges, I can only assume there is a possible need. Has anyone adjusted rain gauge gain and for what purpose ?

The calibration page for my gw1000 shows a gain factor available for the rain gauge. If my gauge is consistently over measuring vs a standard, i would reduce this value below 1.0 to the appropriate level...0.9 or 0.82 based on measured calibration. If under reading, then factor up. My goal would be deliver 25mm equivalent volume as a controlled test, and system displays 25mm.

Will the rain gauge gain factor do that ? Or is that a Lucy question ?

While we expect items out of box to be pre calibrated to perfection, every qc job I ever had involved recalibration of sensors, gauges and related periodically.

Regarding scientific systems, the pictures I see show arrays of collectors. 6x6 or even 10x10. That tells me there is a lot of variability across adjacent cross sections. If one device was accurate and precise, scientists wouldn't put 36 or 100 in an array. If scientists are planning on variability, they add multiple sensors gauges.

I think the wh40 can be managed and provide a reasonably accurate reading. There are conditions which errors will occur due to design elements., and deluge type rains are what comes to mind.

Thanks for input in advance as I learn more about this hobbyist hardware.


July 2020
Ecowitt GW1000B gateway @ ecowitt.net
Ecowitt WN51 soil moisture x8 (potted plants and landscape beds)
Ecowitt WH57 lightning
Ecowitt WH31 temp x2
Ecowitt WH40 rain gauge
Ecowitt WS68 wind sun sensor
Dragon Touch tablet display with PWT app

Offline Rover1822

  • Forecaster
  • *****
  • Posts: 1739
    • Mini Wind and Solar Data project
Re: WH40 vs WH65 Rain Measurements
« Reply #141 on: August 25, 2020, 01:12:10 PM »
Really all I was saying , is that I'm quite content with the WH65 which I have, and I am not content with the reports being put out about the WH40, as I was thinking of getting one. This is a pre-buy determination and solely my thought process.

I don't expect either to be world class accurate,  I did expect the WH40, to well, be more or at least as accurate than the WH65.  Now, as I don't have a WH40 , it may well be in certain circumstances, I'm just going by the reports of others. Well, which is pretty much what you do when thinking about buying something.

Now that the manufacturer/vendor has acknowledged an issue, I will hold off till something comes out of this.

Edit: I just tried to find a statement on the Ecowitt site  for the WH40 that said the WH40 is more accurate than the WH65 , I couldn't find one. I deleted a statement that I made that  said they marketed the WH40 as being more accurate, as I can't substantiate that.
« Last Edit: August 25, 2020, 01:32:45 PM by Rover1822 »
Ambient:
  WS-2000
  PM 2.5(2)
  WH31B(2)
  WH40E
  WH31P
EcoWitt:
  GW1100
  GW1000(4)
  WH31(2)
  WH57
  WH51(12),
  WH40
  WH5360B
  WN34S
  WittBoy WS90 + GW2000
  WS90 (other one) + GW1100
Personal Sites: Weather Cam

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: WH40 vs WH65 Rain Measurements
« Reply #142 on: August 25, 2020, 01:48:30 PM »
Hoping one of you more familiar with hardware and settings can weigh in. Or maybe ecowitt is better suited to explain.
Okay I'll bite. I hope others also weigh in to your questions.

Quote
Since there is a calibration factor shown for rain gauges, I can only assume there is a possible need. Has anyone adjusted rain gauge gain and for what purpose ?
Platokidd mentioned his need for adjustment based on his observations in comparing to another trusted rain gauge. There are many others whom have done the same.

Quote
The calibration page for my gw1000 shows a gain factor available for the rain gauge. If my gauge is consistently over measuring vs a standard, i would reduce this value below 1.0 to the appropriate level...0.9 or 0.82 based on measured calibration. If under reading, then factor up. My goal would be deliver 25mm equivalent volumeadmin as a controlled test, and system displays 25mm.
Yes, it works exactly as you have stated. But keep in mind that your controlled test is something that is often done with other rain gauges here on this forum. Davis comes to mind. Dumping a set amount of water works for them because their design doesn't lose water and doesn't seem to have tipping rate issues given the collection diameter, at least not to any significant extent. But in this particular situation with the WH40 there are factors that need to be taken into consideration. If you only account for rain accumulation and you manually poor a measured amount of water through the system then if you base your calibration off of only that then you are going to be disappointed in the results. Because you would likely then not be accounting to funnel design and bounce out and deluge rain rates that overcome the rate capacity of the tipping spoon.

Quote
Will the rain gauge gain factor do that ? Or is that a Lucy question ?
You can ask Lucy to get an answer from their engineers. But at this point based on your questions so far I don't see that there is anything to ask Lucy. The basic usage of the system is to adjust the rain gauge gain to a factor that you are comfortable with. The factory setting of 1.0 gain is just an average determined by the engineers during the design phase of the product.

Quote
While we expect items out of box to be pre calibrated to perfection, every qc job I ever had involved recalibration of sensors, gauges and related periodically.
Yes, correct. But in this case with the rain gauge I think that calibration is mostly a factor of its environment rather than a function of manufacturing tolerances. All spoons are the same size and the springs probably give off the same force within what is measurably relevant. Temperature and friction and use ..etc....could affect the tipper I suppose. And for that reason and other small factors it is nice to see the ability to have a manual calibration setting.

Quote
Regarding scientific systems, the pictures I see show arrays of collectors. 6x6 or even 10x10. That tells me there is a lot of variability across adjacent cross sections. If one device was accurate and precise, scientists wouldn't put 36 or 100 in an array. If scientists are planning on variability, they add multiple sensors gauges.
You lost me there. Not sure what this means.  But if you are saying that a scientist doesn't go by one measurement.....then yes that is true. Experiments require multiple tests to determine that the results are able to be reproduced and are thus then verifiable. I don't think though that most weather hobbyists are looking to have an array of rain gauges just to account for variability. Sure some have multiple systems just to see how they each perform and for the fun of it too. I think most of use with automatic rain gauges are willing to accept some level of accuracy for the convenience of automation. If not then we would all only have manual rain gauges. Which is why many of us have both manual and automatic. Even manual is not perfect...but we do know that good manual rain gauges are less affected by intrinsic errors compared to automatic rain gauges.

Quote
I think the wh40 can be managed and provide a reasonably accurate reading. There are conditions which errors will occur due to design elements., and deluge type rains are what comes to mind.
I'm not so hopeful. The WH65 is not a spectacular rain gauge. Given its price and component build and overall design characteristics it is what it is. The WH65 isn't trying to pass itself as anything spectacular. But the WH40 does have claims to be superior. From various user experience of those that have both a WH65 and WH40 the WH40 is a disappointment in comparison.  If the WH40 has design aspects that make it inferior to the WH65 then there is a big problem here, because it is marketed as being superior. It doesn't live up to its claim. On paper it seems as it should just because of the larger diameter...but the other components funnel and tipper fall short of the overall design goals of this product. I just don't think the solution is to find some magic calibration offset. My only hope for the WH40 is that at least it can be made to perform equally to the WH65, and for me that means using the WH65 funnel with the proper calibration offset to account for that collection volume. I know that are those that disagree with this sentiment of needing to replace the WH40 funnel for a smaller diameter. And to those people I say ...have at it. Figure it out. Find out what offset is required to account for bounce out and for tipper overflow. Platokidd has already worked that out with some empirical testing of merely comparing one rain gauge to another. That is awesome! Two different approaches. Neither perfect.....but the WH65 isn't perfect either and I do keep that in mind. My goal is not to make the WH40 acceptable to some great standard...rather my goal is to make the purchase of the WH40 a good purchase as compared to the WH65. Further into the future sure I do expect Fine Offset / Ecowitt to come out with a totally new solution that is better than both the WH40 and the WH65....something that lives up to the WH40 claims of superiority.
« Last Edit: August 25, 2020, 02:22:53 PM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline Mandrake

  • Forecaster
  • *****
  • Posts: 1102
Re: WH40 vs WH65 Rain Measurements
« Reply #143 on: August 26, 2020, 03:20:08 AM »
I will add that I know Ecowitt are working on a tipperless rain gauge but its still very early in design and so a long way from reality yet.
G1ZFO

Ecowitt HP2551A + WH65 Tri-Wing (Wunderground: IGUILDFO67)
Ecowitt GW1000 (Wunderground: IGUILDFO68)
Ecowitt GW1000 (Mk2) test environment driving CumulusMX on a RPi 3b
Ecowitt GW2000 (Test)
Ecowitt WS90 Wittboy - Test
Ecowitt WH51 (x6) Soil Moisture Sensor
Ecowitt WH41 PM2.5 AQM Sensor
Ecowitt WH31 (x8) Thermo/Hygro Sensor
Ecowitt WS80 Ultrasonic Anemometer (pre-prod test)
Ecowitt WH57 Lightning Sensor -test
Ecowitt WH32-EP (SHT35) + Davis 7714 Screen
Ecowitt WH45 CO2/PM2.5/PM10 -Test
Ecowitt WN34 Soil Temp Sensor -Test
Ecowitt WN34 Water Temp Sensor -Test
Ecowitt WN35 Leaf Moisture

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: WH40 vs WH65 Rain Measurements
« Reply #144 on: August 26, 2020, 06:32:37 AM »
Yes but if the results from the efforts by WeatherFlow on a tipperless haptic rain gauge are any indication, then I don't fancy that idea. I'm okay with Ecowitt offering this haptic rain gauge but it better not be their only game plan.
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline Mandrake

  • Forecaster
  • *****
  • Posts: 1102
Re: WH40 vs WH65 Rain Measurements
« Reply #145 on: August 26, 2020, 07:27:00 AM »
back to my earlier comments that its very hard to make a consumer rain gauge (automatic/manual) that is going to be effective in all conditions.
There will be likely flaws in most designs resulting in some form of compromise.
G1ZFO

Ecowitt HP2551A + WH65 Tri-Wing (Wunderground: IGUILDFO67)
Ecowitt GW1000 (Wunderground: IGUILDFO68)
Ecowitt GW1000 (Mk2) test environment driving CumulusMX on a RPi 3b
Ecowitt GW2000 (Test)
Ecowitt WS90 Wittboy - Test
Ecowitt WH51 (x6) Soil Moisture Sensor
Ecowitt WH41 PM2.5 AQM Sensor
Ecowitt WH31 (x8) Thermo/Hygro Sensor
Ecowitt WS80 Ultrasonic Anemometer (pre-prod test)
Ecowitt WH57 Lightning Sensor -test
Ecowitt WH32-EP (SHT35) + Davis 7714 Screen
Ecowitt WH45 CO2/PM2.5/PM10 -Test
Ecowitt WN34 Soil Temp Sensor -Test
Ecowitt WN34 Water Temp Sensor -Test
Ecowitt WN35 Leaf Moisture

Offline Platokidd

  • top side land of Lincoln
  • Forecaster
  • *****
  • Posts: 516
Re: WH40 vs WH65 Rain Measurements
« Reply #146 on: August 28, 2020, 10:12:44 AM »
Had a brief rain yesterday that didn't last long. Yes I know it's only one read to compare, I'm just sharing data.

WH40 at 1.0 offset with wh65 funnel  = 0.02   (Guessing 5 tips of the spoon? resolution .004x5= 0.02)
Ws2902a at 1.0 offset with stk funnel = 0.07   (Guessing 7 tips of the spoon? resolution .01x7= 0.07)

Now, if .004 +/- = a 1.0 offset on the wh40.. Then the 2.5 offset talked about earlier in this thread would be much closer. (.05)
Ambient    
2- WS-2000 2-WS-2902A ARRAYS 1-CONSOLE 2-WH40-RAIN 1-WH31L 1-METEOBRIDGE 2-PM2.5(WH42B)
3-WH-31E 1-SRX100LX 1-WH31SM
ECOWITT
3-HP2552 1-HP2562 1-WH51 10-WH31-2-WH45-3-WH55 1-WH40 2-WS68 1-WS80
5-WH51 1-WH32EP 1-WN30 1-WH41
2-GW2000 2-GW1100
1-HP10

1-Davis 7714
1-Stratus

Offline K7RMK

  • Senior Member
  • **
  • Posts: 52
    • East Wilson Weather
Re: WH40 vs WH65 Rain Measurements
« Reply #147 on: August 28, 2020, 10:28:17 AM »
Had a brief rain yesterday that didn't last long. Yes I know it's only one read to compare, I'm just sharing data.

WH40 at 1.0 offset with wh65 funnel  = 0.02   (Guessing 5 tips of the spoon? resolution .004x5= 0.02)
Ws2902a at 1.0 offset with stk funnel = 0.07   (Guessing 7 tips of the spoon? resolution .01x7= 0.07)

Now, if .004 +/- = a 1.0 offset on the wh40.. Then the 2.5 offset talked about earlier in this thread would be much closer. (.05)

The WH40 could have had anywhere from 4 to 6 spoon tips to round to 0.02" total.

4 tips: 0.016" (0.02" rounded)
5 tips: 0.020" (0.02" rounded)
6 tips: 0.024" (0.02" rounded)

If it was 6 tips, the numbers are pretty close but this is such a small sample size.

Also, the rain gauge uses gain (multiplication) for calibration, not offset (addition). I'm sure you know that and simply misspoke, but I just wanted to clarify gain and offset are not synonymous.
Ecowitt GW2000 + Meteobridge on RPI 4B
Ecowitt HP2551-C Display Console
Ecowitt WH32-EP Outdoor Temp/Humidity Sensor + MeteoShield Pro Radiation Shield
Ecowitt WS68 Anemometer
Ecowitt WH40 Rain Gauge
Ecowitt WH57 Lightning Sensor
Ecowitt WH41 Outdoor PM2.5 Air Quality Sensor
Ecowitt WH43 Indoor PM2.5 Air Quality Sensor
Ecowitt WH31 x4 Temp/Humidity Sensors
Ecowitt WH51 Soil Moisture Sensor
Ecowitt WH34S Soil Temperature Sensor
Ecowitt WN35 x2 Leaf Wetness Sensors

Offline galfert

  • Global Moderator
  • Forecaster
  • *****
  • Posts: 6822
Re: WH40 vs WH65 Rain Measurements
« Reply #148 on: August 28, 2020, 10:30:31 AM »
WH40 with WH65 funnel could have experienced 6 tips and still result in 0.02 total rainfall...because that would have been 0.024, which would still round out to 0.02. Meaning that the WH40 with WH65 funnel were only about 1 tip off from each other. That is within the margin of acceptable error to be 1 tip off. If you have a WH40 with WH65 funnel next to a stock WH65 then I would expect them to experience the same number of tips. The spoon mechanism is identical.

EDIT: K7RMK, looks like we had the same conclusion. I was still composing my post as you had just completed your post.
« Last Edit: August 28, 2020, 10:36:00 AM by galfert »
Ecowitt GW1000 | Meteobridge on Raspberry Pi
WU: KFLWINTE111  |  PWSweather: KFLWINTE111
CWOP: FW3708  |  AWEKAS: 14814
Windy: pws-f075acbe
Weather Underground Issue Tracking
Tele-Pole

Offline Feek

  • Member
  • *
  • Posts: 12
    • My amateur radio blog
Re: WH40 vs WH65 Rain Measurements
« Reply #149 on: August 28, 2020, 02:26:29 PM »
This is a very interesting thread to read, thanks everyone.

I've been in email exchange with Lucy and am about to pull the trigger on a GW1003 which I believe includes the WH40 rain gauge.  I've asked her if they can throw me in a WH65 funnel as well (or instead of).

My installation won't be perfect, the main wind gubbins will be about 25ft up in the air and the rain gauge will go on top of a 9ft washing pole so the entire system will be a bit of a compromise.  The reason I want it is mainly for wind speed and direction so that when the wind kicks up, I can automate my rotator and ensure my ham radio aerial is always pointing into the wind.  Everything else is a bonus.

 

anything