Topic: WittBoy = GW2001 (station) = WS90 + GW2000 on Kickstarter & Beta test

[box]

Likewise there will be quite a few of the seasoned folks on this forum in different parts of the world (aka different weather scenarios) testing the WS90.
Please be aware part of this testing will involve fine tuning the calibration etc
This is very much not a finished product yet so please don't jump to any conclusions from early comments posted about observations.
Its all about product development to make it the best it can be.

Isn't the weather station factory calibrated?

I assume this is more about tuning/training the various algorithms in a wide variety of situations

It wouldn't surprise me if this continues after the beta phase with retail customers

A bit like Tesla data collection from customers to train their autodrive software

[Mandrake]

I shall be surprised if Ecowitt did not continue to fine tune the hardware and software.
This is something that they do as standard and do so in conjunction with their customers.
Better that they do this than throw out hardware and software and leave it as is sold for customers to lump or like!
I much prefer an evolutionary approach!

[tweatherman]

Does anyone have any idea other than rain what Ecowitt/Whitboy might try to manipulate via software algorithm(s)?

Thanks,
tweatherman

[Mandrake]

Be careful what you mean by software algorithms.
As far as I know Ecowitt have no intention of going down a manipulation path like WeatherFlow for its data in the cloud.

When I talk of improvements I mean that Ecowitt will constantly be looking at the firmware code in their devices to more effectively measure the raw data.
They will be undertaking this inhouse and via feedback from testers and forum members alike.
This might also imply hardware revisions and we have observed iteration changes in their products where they have learned from the field about performance.

[Gyvate]

Be careful what you mean by software algorithms.
As far as I know Ecowitt have no intention of going down a manipulation path like WeatherFlow for its data in the cloud.

When I talk of improvements I mean that Ecowitt will constantly be looking at the firmware code in their devices to more effectively measure the raw data.
They will be undertaking this inhouse and via feedback from testers and forum members alike.
This might also imply hardware revisions and we have observed iteration changes in their products where they have learned from the field about performance.

Examples for such improvements are the firmware update of the WS80 sensor array and a hardware revision of the WH65/WS69 array (here a special filter foil is now applied to the solar sensor).

And if algorithms serve better results then this "manipulation" is called calibration or correction etc.
E.g. there are meanwhile algorithms which can tell pretty exactly the sunshine duration from the raw solar radiation/luminosity readings which come pretty close to the results provided by a pyranometer. (Not talking about the rule of thumb that everything > 120 W/m² is considered sunshine time).

[mauro63]

Perfect explain !
Ecowitt will NOT manipulate any data  on server sides

M.

[mcrossley]

(Not talking about the rule of thumb that everything > 120 W/m² is considered sunshine time).

It's not a "rule of thumb", it is the definition of the required direct normal irradiation level for sunshine by the WMO.

[olicat]

Hi Mark,

do you have proof, a link or something?
I also used this definition of sunshine for a long time.  But all I could say was "because everyone does it that way".
Do you have more?

Oliver

[mcrossley]

The WMO standard is hard to find since they updated their web site - read all my links are broken! But the jist from the standard is...

"Sunshine duration during a given period is defined as the sum of that sub-period for which the direct solar irradiance exceeds 120 W/m-2 "

To measure the measure the direct normal irradiation you will need a tracking pyrheliometer - good summary of them here: https://www.hukseflux.com/applications/solar-energy-pv-system-performance-monitoring/what-is-a-pyrheliometer

[mcrossley]

Found a copy of the WMO measurement publication (definition at 8.1.1) - https://www.weather.gov/media/epz/mesonet/CWOP-WMO8.pdf

[broadstairs]

All weather stations I know of in the price ranges we normally work with that measure solar use global irradiation even Davis stations. There are some separate solar measuring devices which get very close to the direct measurement. For years now I've been correcting folks who say 120 w/m2 is sunny, glad to see it corrected here.

Some years ago I wrote a script to calculate expected solar max values for a location using global irradiation measurements and it formed the basis for what is used in a Weather Display for calculating when it is sunny.

Stuart

[Rover1822]

120 W/M^2 , is kind of sunny, probably also depends on location and season.
It is not full sun.

For instance , solar panels are rated at 1000 W/M^2 , which is like lab condition, (I know useless information ).
I always , for fun, not for metrics, look at whatever my stations say, vs my solar output, an do some weird mental math. (Efficiency coe , etc).

I am so off topic...

[davidefa]

According to eq 8.3, page 1.8-4 of the above wmo document ( in the absence of a sun-tracking pyrheliometer, as we could compare a fixed horizontal solar sensor ) should be:

Code: [Select]

(G-D) >= 120*cos(z) W/m2i.e. we should take in consideration solar zenith angle too

[mcrossley]

According to eq 8.3, page 1.8-4 of the above wmo document ( in the absence of a sun-tracking pyrheliometer, as we could compare a fixed horizontal solar sensor ) should be:

Code: [Select]

(G-D) >= 120*cos(z) W/m2i.e. we should take in consideration solar zenith angle too

That assumes you have a diffuse irradiation sensor as well as a global though. Sorry this is going way off topic!

[Gyvate]

120 W/m² may be a standard, but still a ridiculous one in my opinion. It's still a rule of thumb.
You can easily have 200 W/m² under cloud cover, so where is the sunshine ???

You want to be exact, you have to use either a pyrheliometer for direct solar irradiation or a pyranometer for global solar radiation - whatever you are looking for.
But the costs of these instruments are easily in the range of a Davis Vantage.
I prefer the data from my Blake-Larsen SunRecorder which mimics the reaction of the human eye and gives even Sunshine time readings during dust and dawn (when applicable).
And it is comparatively affordable for such an intelligent solution (160 EUR/135 GBP/180 USD). (And which CumulusMX supports  ).

For me it is interesting when I as a person experience sunshine and not when a "stupid" formula defines it.
A number on a display alone is not going to enlighten my day.

[davidefa]

That assumes you have a diffuse irradiation sensor as well as a global though. Sorry this is going way off topic!

Yes, but assuming ( even if not true ) the diffused component can be neglected, we should consider the cos(z) factor
Or in other words if we not consider the cos(z) we underestimate the sunshine duration measurement.

[mcrossley]

120 W/m² may be a standard, but still a ridiculous one in my opinion. It's still a rule of thumb.
You can easily have 200 W/m² under cloud cover, so where is the sunshine ???

You are confusing the values from global irradiation (what most amateurs measure) and direct normal irradiation (what a tracking pyrheliometer measures). They are very different.

[Gyvate]

120 W/m² may be a standard, but still a ridiculous one in my opinion. It's still a rule of thumb.
You can easily have 200 W/m² under cloud cover, so where is the sunshine ???

... You are confusing the values from global irradiation (what most amateurs measure) and direct normal irradiation (what a tracking pyrheliometer measures)*. They are very different#.

* Do I ? "...You want to be exact, you have to use either a pyrheliometer for direct solar irradiation or a pyranometer for global solar radiation ..."
# Really ... ?

Anyhow, what counts for me is the human eye ...

[havtrail]

It's interesting that the solar panel on the weather station I run charges the battery at times when officially, by that WMO definition, there is not a "sunshine" condition.

Rich K.

[tweatherman]

Just curious if anyone has insight to what temp/hum. sensor will be in the new station?

Thanks,
tweatherman

[hiljo]

Delivery scheduled for today 10 am (in about 2 hours). But 10 days out for holidays 
So I'll have to wait before I can see/test it..

[mauro63]

Just curious if anyone has insight to what temp/hum. sensor will be in the new station?

Thanks,
tweatherman

Sensirion sht30 but ready for future upgrade

M.

[Gyvate]

It's interesting that the solar panel on the weather station I run charges the battery at times when officially, by that WMO definition, there is not a "sunshine" condition.

Rich K.

How do you tell at which amount of solar radiation exposure the super-capacitor starts being charged ?
If it's a really good one (S-C), then it charges at low wattage/area.

What outdoor array do you have ?
Because, in fact, the batteries of the WH65/WS69, WS80, WS68 arrays (the Ambient WS-2000/5000/2902 outdoor arrays) are not charged. What is charged is a super-capacitor which can run the "station" (the outdoor sensor array - station = console + sensors) up to two days when not being reloaded after having been fully charged. It's only then that the batteries come in as a backup. This only occurs at special conditions (long[er] time not sufficient sun exposure, very cold - like during polar night as an extreme case).

Still, it would be interesting to know what minimum amount of wattage/m² of global radiation is needed for the S-C to start being charged.

[giusCB]

Just curious if anyone has insight to what temp/hum. sensor will be in the new station?

Thanks,
tweatherman

Sensirion sht30 but ready for future upgrade

M.

Ciao Mauro. Are you sure?
The newest ws80 has the sht40, there is also a firmware update for this sensor https://osswww.ecowitt.net/uploads/20211223/WS80%20Upgrade(v1.2.0)%20Tool%20Kit.zip

[henry]

Just curious if anyone has insight to what temp/hum. sensor will be in the new station?

Thanks,
tweatherman

The standard sensor will be sht40 ( the upgrade version of sht30). However it will be a sensor with sht35 sensor available for accessory part: it can be plug and play type, and it is easy to make this upgrade or change at user side if they want to have a better sensor.