Topic: Ecowitt Wittboy (WS90 + GW2000) initial review

[mcrossley]

But frost melt and dew should be included in your measurements.

[Ra1nman]

But frost melt and dew should be included in your measurements.

Fair enough, I am new to this game in which case a win for the WH40

[Ra1nman]

Well my capacitor hit a record high today of 5.3v due to excessive sun the last few days, it managed to hold 4.2v minimum overnight even. It's interesting, looking back the solar has not been a huge variation this week, big change has been massive reduction in wind speed and gusts. Wonder if the lack of wind has reduced battery drain, since the unit has been reporting at larger intervals, seems to make sense. Previously it would drain to 2.8V overnight and struggle to get above 3.2v during the day.

Having double checked the data, quite convinced the lower wind speeds have directly benefited the capacitor usage and voltage. Interesting to know.

[snubbers]

Great thread, I just signed up having got a WS90/GW2001 (reports as GW2000, advertised as 'upgraded' version on Amazon UK).

Just thought I'd pop in as I have a small airport/airfield close to me, approx 1 mile away as the crow flies, and have friends in the area with weather stations (which prompted me to get one!).
I've mounted mine 1M above the apex on the gable end of the house, it's as tall as I dare for now. I live in a flat area and have good open space infront of the house, quite open space behind and am the tallest house on the street, so hoping to not have to much turbulence.

I wasn't sure the best place to mount it, but the concensus of those with weather stations near me as the higher the better..



Comparing to the local airport:
Barometric pressure - out the box, only 1hPa out, but was 3-4hpa over night when it was cooler.

Wind speed/direction - It's only been 1 day, but checking frequently between the Airports METAR and my station is quitr close, direction is very stable and tracks nicely, windspeed I'll need more time, The weather is atrocious at the moment, but people in work have been checking my station and theirs/METAR this morning and all commented they wish theirs correlated that closely.. I have had some gusts up to 74kph with the METAR showing gusts up to 37knots (68kph) and compared to some stations around me which are reacting badly to high wind speeds, mines very stable.

Precipitation - Too hard to say, it's picked up rain very reliably, but nothing to compare, I've within 10% of surrounding stations, no one is standing out as more or less accurate, it certainly picks up drizzle early and it would wrong to say it's wildly inaccurate, although time will tell, I have 7 stations within 5 miles, so plenty to compare to.

Temp/Humidity - within 0.1C and 1.5% respetively to the METAR data.

Niggles - I wired up a 12v (2A) supply to the unit, hoping it is true that should avoid having to change batteries, I popped in Energizer Lithium anyway, although the 'feature' of the hardware reading weird battery voltages with an external supply connected seems true and how I found this forum!

I could get a WH40 or similar and mount lower down, I have plenty of nice easy to access places I can pop that.. I certainly would not want that so high up, I'm OK to a point with heights, but nearly 9M up was a bit unnerving even with a very sturdy set of ladders tied to the house!

[Ra1nman]

Great thread, I just signed up having got a WS90/GW2001 (reports as GW2000, advertised as 'upgraded' version on Amazon UK).

Just thought I'd pop in as I have a small airport/airfield close to me, approx 1 mile away as the crow flies, and have friends in the area with weather stations (which prompted me to get one!).
I've mounted mine 1M above the apex on the gable end of the house, it's as tall as I dare for now. I live in a flat area and have good open space infront of the house, quite open space behind and am the tallest house on the street, so hoping to not have to much turbulence.

I wasn't sure the best place to mount it, but the concensus of those with weather stations near me as the higher the better..



Comparing to the local airport:
Barometric pressure - out the box, only 1hPa out, but was 3-4hpa over night when it was cooler.

Wind speed/direction - It's only been 1 day, but checking frequently between the Airports METAR and my station is quitr close, direction is very stable and tracks nicely, windspeed I'll need more time, The weather is atrocious at the moment, but people in work have been checking my station and theirs/METAR this morning and all commented they wish theirs correlated that closely.. I have had some gusts up to 74kph with the METAR showing gusts up to 37knots (68kph) and compared to some stations around me which are reacting badly to high wind speeds, mines very stable.

Precipitation - Too hard to say, it's picked up rain very reliably, but nothing to compare, I've within 10% of surrounding stations, no one is standing out as more or less accurate, it certainly picks up drizzle early and it would wrong to say it's wildly inaccurate, although time will tell, I have 7 stations within 5 miles, so plenty to compare to.

Temp/Humidity - within 0.1C and 1.5% respetively to the METAR data.

Niggles - I wired up a 12v (2A) supply to the unit, hoping it is true that should avoid having to change batteries, I popped in Energizer Lithium anyway, although the 'feature' of the hardware reading weird battery voltages with an external supply connected seems true and how I found this forum!

I could get a WH40 or similar and mount lower down, I have plenty of nice easy to access places I can pop that.. I certainly would not want that so high up, I'm OK to a point with heights, but nearly 9M up was a bit unnerving even with a very sturdy set of ladders tied to the house!

An excellent post and initial review, fits perfectly with the thread title. Sounds like a similar experience to mine, except you were braver than me on the ladders! Although the piezo sensor makes a good effort, fair to say I have conceded the WH40 is required for reliable rain readings, if you look at the monthly comparison of mine below, clear to see. I say to ignore the annual rainfall for the WH40 because it didn't get installed until 9th January and missed around 15-20mm of rain from memory.

It will be very interesting to see how your wind speed compares with airport data, I increased the gain since mine is lower down at the moment, suspect yours will be much more in line straight out the box.

The WH32EP sensor is nice for me, knowing it is calibrated and more critically at industry standard height, I paid about £50 for mine in the UK, they went up a bit after. Still pretty happy with mine for the £200-250 outlay.

Do report back in the weeks ahead with your findings, interesting to know

[snubbers]

An excellent post and initial review, fits perfectly with the thread title. Sounds like a similar experience to mine, except you were braver than me on the ladders! Although the piezo sensor makes a good effort, fair to say I have conceded the WH40 is required for reliable rain readings, if you look at the monthly comparison of mine below, clear to see. I say to ignore the annual rainfall for the WH40 because it didn't get installed until 9th January and missed around 15-20mm of rain from memory.

It will be very interesting to see how your wind speed compares with airport data, I increased the gain since mine is lower down at the moment, suspect yours will be much more in line straight out the box.

The WH32EP sensor is nice for me, knowing it is calibrated and more critically at industry standard height, I paid about £50 for mine in the UK, they went up a bit after. Still pretty happy with mine for the £200-250 outlay.

Do report back in the weeks ahead with your findings, interesting to know

I've read your posts in the thread and can see your own journey with the WS90!

I will say looking at very local weather stations to me, ones I can physically check (drive by on the way to/from work and also walk the dog by them!) today between 5:00 and 6:00 it was most certainly not raining, yet a Davis Vantage Vue and a weatherflow/ambient weather setup (the person had 2 stations next to each other on a garden fence), both read rain for a good 30 minutes (if not the entire hour), pushing the daily totals right up and that seems odd, it's not a good reflection of when it start/stopped raining and makes it look like they had sustained showers which was not the case.

However, I appreciate the WS90 may not be that good over a day, but I'm starting to wonder if it's a case of non are perfect?

[Rover1822]

Basic rule. Do not compare yourself to other stations, other than for maybe a general idea. You have no idea how those stations are being used, maintained, etc. This is very true for things that that change within few dozen feet, temp, humidity, wind, rain .

Get manual gauges to compare against at your location.

Sadly, you cannot assume , just because someone else has a station , well , that it is accurate

[Ra1nman]

An excellent post and initial review, fits perfectly with the thread title. Sounds like a similar experience to mine, except you were braver than me on the ladders! Although the piezo sensor makes a good effort, fair to say I have conceded the WH40 is required for reliable rain readings, if you look at the monthly comparison of mine below, clear to see. I say to ignore the annual rainfall for the WH40 because it didn't get installed until 9th January and missed around 15-20mm of rain from memory.

It will be very interesting to see how your wind speed compares with airport data, I increased the gain since mine is lower down at the moment, suspect yours will be much more in line straight out the box.

The WH32EP sensor is nice for me, knowing it is calibrated and more critically at industry standard height, I paid about £50 for mine in the UK, they went up a bit after. Still pretty happy with mine for the £200-250 outlay.

Do report back in the weeks ahead with your findings, interesting to know

I've read your posts in the thread and can see your own journey with the WS90!

I will say looking at very local weather stations to me, ones I can physically check (drive by on the way to/from work and also walk the dog by them!) today between 5:00 and 6:00 it was most certainly not raining, yet a David Vantage Vue and a weatherflow/ambient weather setup (the person had 2 stations next to each other on a garden fence), both read rain for a good 30 minutes (if not the entire hour), pushing the daily totals right up and that seems odd, it's not a good reflection of when it start/stopped raining and makes it look like they had sustained showers which was not the case.

However, I appreciate the WS90 may not be that good over a day, but I'm starting to wonder if it's a case of non are perfect?

Thanks, I tried to update on here as it helps others, I've got lots of help from this forum, plenty of advice.

I know that some have commented on the Davis VP2 rain gauges as being the weak point, interesting I thought, but suspect in general they are pretty good. I have a few semi local Davis stations and have access to the data, it normally reads in line with mine, except for humidity which I think my Ecowitt should be better at now, with the WH32EP in service.

I can confidently say that the Piezo is not a long term match for the WH40, some days it seems to work well, others not so much. Early showers or drizzle I think it can beat the WH40, but then it goes off track with normal rain. My monthly total says it all really, manual gauge and WH40 matched and the Piezo on WS90 is 10mm under. There has been some frost and snow which thawed, the Piezo would miss this, likewise there could have been some drizzle which the WH40 missed of course.

If you can pick up a WH40 on Amazon when discount is available I think you would benefit, Cumulus has just added option to allow WS90 to show early rain alert but still use the WH40 for main readings. This I feel is a benefit for the Piezo potentially, heavy dew can trick it into false readings though.

Technically, someone will correct me otherwise, but, with a WH32EP attached the temperature and humidity? should be more accurate than a Davis VP2, assuming adequate radiation shield. At least that is how I have read it.

Just ordered another soil moisture sensor today!

[Ra1nman]

Basic rule. Do not compare yourself to other stations, other than for maybe a general idea. You have no idea how those stations are being used, maintained, etc. This is very true for things that that change within few dozen feet, temp, humidity, wind, rain .

Get manual gauges to compare against at your location.

Sadly, you cannot assume , just because someone else has a station , well , that it is accurate

This is fair comment, although if there are enough of them and within proximity to your site, good basis for comparison. A manual rain gauge will be useful though.

[snubbers]

Basic rule. Do not compare yourself to other stations, other than for maybe a general idea. You have no idea how those stations are being used, maintained, etc. This is very true for things that that change within few dozen feet, temp, humidity, wind, rain .

Get manual gauges to compare against at your location.

Sadly, you cannot assume , just because someone else has a station , well , that it is accurate

Thanks, sound advice indeed! The only comparisons for technical accuracy I've made are the barometric pressure (relative) from my (very) local airport, I was really just anecdotally observing local stations as it was fascinating how many report rain persisting for 30-60 mins after a heavy downpour.

Thanks, I tried to update on here as it helps others, I've got lots of help from this forum, plenty of advice.

I know that some have commented on the Davis VP2 rain gauges as being the weak point, interesting I thought, but suspect in general they are pretty good. I have a few semi local Davis stations and have access to the data, it normally reads in line with mine, except for humidity which I think my Ecowitt should be better at now, with the WH32EP in service.

I can confidently say that the Piezo is not a long term match for the WH40, some days it seems to work well, others not so much. Early showers or drizzle I think it can beat the WH40, but then it goes off track with normal rain. My monthly total says it all really, manual gauge and WH40 matched and the Piezo on WS90 is 10mm under. There has been some frost and snow which thawed, the Piezo would miss this, likewise there could have been some drizzle which the WH40 missed of course.

If you can pick up a WH40 on Amazon when discount is available I think you would benefit, Cumulus has just added option to allow WS90 to show early rain alert but still use the WH40 for main readings. This I feel is a benefit for the Piezo potentially, heavy dew can trick it into false readings though.

Technically, someone will correct me otherwise, but, with a WH32EP attached the temperature and humidity? should be more accurate than a Davis VP2, assuming adequate radiation shield. At least that is how I have read it.

Just ordered another soil moisture sensor today!

I'll definitely get a rain gauge just out of interest, and the WH40 is cheap enough to not worry about cost, but I can see that there is no perfect answer, but interesting to know Cumulus is trying to make good use of both sets of data!

[Rover1822]

You may also want to avail yourself of a manual  rain gauge. Yes , one of those that fills up and has very accurate markings . Here in the US the goto is  a Stratus https://www.amazon.com/gp/product/B000X3KTHS/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1.

Other words, not something from the local store that says "rain gauge:" and costs a few dollars

[Ra1nman]

The £5-10 rain gauges off Amazon also do the trick if you want to save the other £40 for a WH40

[Ra1nman]

Prime example of the WS90 and WH40 actually tracking each other quite well, medium to light rain. Around 0.3mm over reading on the WS90, the problem of course is that 0.3 here and 0.3 there soon adds up, over a month that could soon become a negative 10mm (like this month) or a positive 5mm.

[Gyvate]

that's in my observation and imho not a good and representative example - it's just a spot observation
in my observation this overreading is not constant over time (if it were, you would be correct, but that's not a very scientific approach to take spot-on observations as "proof") but it even goes in the other direction and interestingly irons out over longer periods (weeks, months).

If it were a constant overreading, it could be corrected by the linear calibration - what in my observation doesn't work - at least not with the lower rain rates. Sometimes it overreads, sometimes it underreads (calibration and firmware unchanged !!). Once you change firmware, you can dump all your earlier observations. You don't change parameters during a test.

I'm not saying that the WS90 is a very exact instrument - it isn't (yet) and needs a non-linear correction in its characteristic line especially in the lower rain rate area.

I've said this several times here in this thread, and my observations are based on WS90/WH40/WS69 plus two manual rain gauges with no changes to firmware and calibration.

A proper field test of 5-10 stations with a constant calibration and no change in firmware should provide over 1/2 year with hopefully enough different rainfall events a sufficient amount of data to find the proper characteristic line - the remaining inaccuracies could then be handled by the linear tier calibration.

[Ra1nman]

that's in my observation and imho not a good and representative example - it's just a spot observation
in my observation this overreading is not constant over time (if it were, you would be correct, but that's not a very scientific approach to take spot-on observations as "proof") but it even goes in the other direction and interestingly irons out over longer periods (weeks, months).

If it were a constant overreading, it could be corrected by the linear calibration - what in my observation doesn't work - at least not with the lower rain rates. Sometimes it overreads, sometimes it underreads (calibration and firmware unchanged !!). Once you change firmware, you can dump all your earlier observations. You don't change parameters during a test.

I'm not saying that the WS90 is a very exact instrument - it isn't (yet) and needs a non-linear correction in its characteristic line especially in the lower rain rate area.

I've said this several times here in this thread, and my observations are based on WS90/WH40/WS69 plus two manual rain gauges with no changes to firmware and calibration.

A proper field test of 5-10 stations with a constant calibration and no change in firmware should provide over 1/2 year with hopefully enough different rainfall events a sufficient amount of data to find the proper characteristic line - the remaining inaccuracies could then be handled by the linear tier calibration.

My thoughts as noted before I posted the examples - in summary and as highlighted in my quote below:

Prime example of the WS90 and WH40 actually tracking each other quite well, medium to light rain. Around 0.3mm over reading on the WS90, the problem of course is that 0.3 here and 0.3 there soon adds up, over a month that could soon become a negative 10mm (like this month) or a positive 5mm.

[Gyvate]

My thoughts as noted before I posted the examples - in summary and as highlighted in my quote below:

Prime example of the WS90 and WH40 actually tracking each other quite well, medium to light rain. Around 0.3mm over reading on the WS90, the problem of course is that 0.3 here and 0.3 there soon adds up, over a month that could soon become a negative 10mm (like this month) or a positive 5mm.

you were missing or a +/- zero mm  - hence my comment,
because that's in my observation the more likely scenario based on about one year of observation

[Ra1nman]

My thoughts as noted before I posted the examples - in summary and as highlighted in my quote below:

Prime example of the WS90 and WH40 actually tracking each other quite well, medium to light rain. Around 0.3mm over reading on the WS90, the problem of course is that 0.3 here and 0.3 there soon adds up, over a month that could soon become a negative 10mm (like this month) or a positive 5mm.

you were missing or a +/- zero mm  - hence my comment,
because that's in my observation the more likely scenario based on about one year of observation

I'm lazy like that keeps people guessing

[Ra1nman]

Well it has been around 5 months since the station was setup, now with the additional sensors things have been mostly ticking over just fine. Had one minor glitch with humidity getting stuck at 99% and the GW2000 dropped out once, requiring a power cycle. Otherwise been very happy with this setup, data is easily sent to various sources from the GW2000, makes life easy to receive weather information on my iPhone, iPad and watch. Continue to use an old iPad for displaying 'live' data via PWS_Dashboard, mainly due to speed of updates and ability to customize the display tiles. Would be nice if either Ecowitt or another App would allow for basic summary data in large text to be displayed, on old tablets or even larger phones this gives us options for viewing data around the house, also makes use of potentially discarded devices.

Overall I would easily recommend this setup, hope it continues to be reliable for a few years yet. I will try and update this thread periodically when anything useful crops up.

[Gyvate]

Well it has been around 5 months since the station was setup, now with the additional sensors things have been mostly ticking over just fine. Had one minor glitch with humidity getting stuck at 99% and the GW2000 dropped out once, requiring a power cycle. Otherwise been very happy with this setup, data is easily sent to various sources from the GW2000, makes life easy to receive weather information on my iPhone, iPad and watch. Continue to use an old iPad for displaying 'live' data via PWS_Dashboard, mainly due to speed of updates and ability to customize the display tiles. Would be nice if either Ecowitt or another App would allow for basic summary data in large text to be displayed, on old tablets or even larger phones this gives us options for viewing data around the house, also makes use of potentially discarded devices.

Overall I would easily recommend this setup, hope it continues to be reliable for a few years yet. I will try and update this thread periodically when anything useful crops up.

Are you looking something like this ?
 [ You are not allowed to view attachments ]
It's a version of the Weather34 template - one of two used with Meteobridge
(but in could be hosted [the template] on a NAS or web hosting service with FOSHKplugin - would need some configuration work)
they were mainly meant for Davis stations only by the developer, but can also be used with Ecowitt hardware  -  if you know how)
this view, updated via Custom server/FOSHKplugin or Meteobridge, could come to your mobile device via a normal web browser

[Ra1nman]

Thanks, yes, that is heading in the right direction, ideally I would love to get something with text twice that size in app format, without too much configuration. PWS_Dashboard was enough for me and my relatively small brain Am hopeful the 'Smartmixin' app or maybe MyPWS will come up with something to fill this gap, must be others who would like a dashboard with very large basic information, something you can glance at across the room.

[Gyvate]

you could post in the SmartMixin thread suggesting to Vincent a "basic view" or "simple view" option for the app with wind, rain, T/RH, solar/UV, pressure only - but then in bigger letters/numbers
https://www.wxforum.net/index.php?topic=43629.0

[Ra1nman]

Thanks, I literally posted a reply this morning and mentioned it, I do like the Smartmixin app quite a bit. If a basic large display can be added that would be superb, often I just want to see temperature, wind and rain, maybe solar for the summer months. Having tiles which you could reposition and prioritise by text size, well that would just be perfect, probably asking a bit too much

Even if the current Smartmixin main display just had a 200% text size option it would bridge the gap. MyPWS is another candidate for this option, I like certain things from both, good for getting my actual Ecowitt temperature on my apple watch also.

[mitschke]

Thanks for the review. Maybe I missed it in here, but how is your experience with the Wittboy's anemometer accuracy in moderate to heavy rain?

I have mounted the Wittboy and a classic anemometer on the same pole, an as soon as the rain started to come down heavy, the Wittboy's wind readings dropped dramatically compared to the other Anemometer.

Above: classic anemometer, gusts 40-50km/h, the dropping in the marked area. Wind: > 20km/h
Below: Wittboy, after the rain started, gusts almost immediately drop to ~30km/h. Wind < 15km/h
 [ You are not allowed to view attachments ]

I never experienced such a consistent deviation in that amount between these Anemometers before.

[Gyvate]

I can't share or confirm this experiences. That said, I may not have had similar conditions yet - i.e. strong rainfall combined with super strong winds. In my cases they were up to 50 km/h in combination with strong rain, but I couldn't see a drop in readings compared to my cup-wheel anemometer WS69 sensor during these periods.
But I can't exclude that this may change once the rain comes in horizontally in bigger amounts ....

[mitschke]

Less rain, stronger winds, more consistent readings:
 [ You are not allowed to view attachments ]

But still a litte drop, after it was getting wet. (Above Wittboy, below cup-wheel)