Topic: Ecowitt wind recordings

[twassmer]

Hi everybody,

I was always suspicious but now almost certain that my WS90 is not recording wind speed properly. I am in a backyard so I have turbulences that make the wind direction unreliable, but I am roughly 5 mph below the wind speed of close-by stations and about 20 mph off gusts. Should I recalibrate?

Tom

[Gyvate]

before you start doubting your readings (and before there will be a long response from our forum member @Arkadiusz_w how and why a WS90 is in his opinion hopeless and some high-end professional high-cost devices are much better - he tends to compare consumer grade devices [what a WS80/WS90/WS68 is]  with real [much more] professional devices which average people cannot pay and don't want to pay - and there is almost no WS80/WS90 wind related post he gives a lengthy commentary of that type):

wind is like rainfall not a homogeneous phenomenon - in order to compare you have to put your reference devices
a) close to each other and
b) at the same height

you have usually no idea how hight the anemomter of your reference station is mounted, what is its exposition etc.
for practical purposes a WS80/WS90 put at 10 m (30 feet) above ground usually provides reasonable results.
When due to the location the wind direction cannot be determined properly, it's not the fault of the sensor but of the location.
You either have to create an appropriate environment (probably rather difficult) or place your sensor in a better place.

Usually at higher exposition also higher wind speed is measured than in lower places (high/low with reference to ground and topology), so your lower readings might be correct for the microclimate your sensor has been put into.

EDIT: removed typos

[Meteorology fan]

Cool that @Gyvate brought me back to the thread. A colleague should place the wind meter at a height of 10 m and let the air flow freely.

As for the WS90 Wittboy, I won't change my opinion, as long as Ecowitt uses 40Khz transceivers that are of poor quality, software updates won't help and wind speed measurement will be inaccurate. Besides, I have confirmation from experts that all reputable manufacturers use 200Khz transducers, which translates into measurement range, response time and accuracy. Ecowitt uses very cheap transducers at the expense of saving money.

The WS80 also suffers from the same thing as the WS90 Wittboy, as it uses the same cheap 40Khz transducers. Already the Ecowitt WS68 is better in terms of measurement stability and features a measurement range of up to 50 m/s. However, it is still far from the linearity and sampling and directional measurements of the Davis VP2.

In the coming days there will be gale force winds in the Polish area and I plan to mount the Hongyuv right next to the Wittboy on a single tripod about 2 m above sea level and do tests on the 1.3.8 software with a working SHT40 sensor in the WS90. Then the problem with high inaccuracy will be clearly visible, as the badly measured values are especially above 30-40 km/h and the error clearly increases with increasing wind speed and is not linear. A professional-grade device based on 200Khz transducers serves as a reference instrument, allowing us to determine the reliability of data from an amateur-grade anemometer.

What Ecowitt engineers apparently don't know is that the WS90 Wittboy and WS80 exhibit temperature drift when measuring wind, which also introduces measurement errors. This was confirmed to me by two anemometer experts while observing footage of a comparison of Hongyuv WDS2E vs Davis VP2 vs WS90 Wittboy.

My goal is to show people who purchase Ecowitt equipment not to write nonsense that their ultrasound-based stations are as precise as those used by professional meteorological services, and unfortunately such deceitful statements are then made and reproduced.

[kheller2]

My goal is to show people who purchase Ecowitt equipment not to write nonsense that their ultrasound-based stations are as precise as those used by professional meteorological services, and unfortunately such deceitful statements are then made and reproduced.

I’ll bite.  Who said something that stupid, and where?

[Gyvate]

My goal is to show people who purchase Ecowitt equipment not to write nonsense that their ultrasound-based stations are as precise as those used by professional meteorological services, and unfortunately such deceitful statements are then made and reproduced.

I’ll bite.  Who said something that stupid, and where?

as far as I can see @Arkadiusz_w cannot (or doesn't want to) distinguish between marketing statements and the factual situation. E.g.  Ambient Weather WS-5000 Ultrasonic Professional Smart Weather Station & Thermo Hygrometer
he considers such a marketing statement as "deceitful" - and similar statements by Ecowitt and resellers.

Of course, almost every grown up person knows that with Fine Offset/Ecowitt we are moving in the consumer segment of private or personal weather stations. But @Arkardiusz_w seems to have adopted sort of a "white-knight attitude" to reveal all these "lies" etc. I think every reasonable person is aware that for the price segment we are moving in here noone can expect professional quality sensors etc. - but inside the segment itself, imho, Ecowitt is pretty to very good.
This doesn't mean that obvious errors, mistakes etc. should be denied - so far, in most cases, Ecowitt has provided improvements if such cases occurred.

But asking for more expensive components, as he does, e.g. with the transducers used to build the WS80/WS90 anemometers, will also (possibly significantly) raise the price - and the sensors he refers to are in a very different price segment - one sensor at the price of a full Ecowitt station, and not the least expensive ones. He seems to have a a well-filled wallet what most PWS owners do not have; so asking to go for such equipment is imo not only unrealistic but off-topic in this forum.

I appreciate the findings, but the whole bashing around is neither useful nor helpful.

[Meteorology fan]

Among other things, on Polish forums about meteo stations, such lies were formulated. It was pointed out by other users that with WS90 Wittboy you can have wind measurement on the level of professional stations based on Gill and Vaisala, which is not true.

Also on the wxforum there were people who pointed out that wind can be measured so accurately with Fine Offset sonic wind gauges and can be results comparable to surrounding professional stations.

After every windstorm in Poland, we have a rash of severely overestimated wind gusts relative to the surrounding stations of the national research institute, because Ecowitt and its ultrasound are very inaccurate in strong and turbulent flow. Then such charts that it was blowing 100 km/h at someone's place will appear from Fine Offset station users who have WS80 and WS90, while the surrounding stations based on Vaisala and Gill ultrasound showed max 80-85 km/h.


These windstorm errors are not the size of 3-8 km/h, they are even 10-20 km/h higher on the WS80-90 than it really was, and this is already a serious problem with the quality and accuracy of the equipment.

No reputable company uses 40Khz transducers, because it's the worst kind, and sonic wind meters with them don't cost more than $100-150, hence the accuracy will never be good.

Up to about 20-25km/h you can rely on the WS80 and WS90 Wittboy, because the error, if it happens, is no more than 1-3km/h, but as the speed increases the errors get higher and higher. If I didn't have a WMO-compliant ultrasonic wind gauge based on 200Khz transducers I would believe the unrealistic readings of the WS90 Wittboy and think that the Davis VP2 is poor and inaccurate.

Ambient Weather also relies on Ecowitt ultrasound and I dare say they are just as inaccurate as the WS80/90.

I fully understand that adding better transceivers would increase the price, but also improve the accuracy to get better quality data from the ultrasonic wind gauges, because currently the Ecowitt ultrasonic wind gauges are toys that cannot be trusted. Especially at such crucial times during weather breakdowns as gales and storms.

Ecowitt has a very good cloud and iot app. Hence, more informed Polish users would like more accurate wind meters. WH32EP and WH40 are ok, but these are some of the few products worth considering from Ecowitt that I would recommend to others.

[hiljo]

...before you start doubting your readings (and before there will be a long response from our forum member @Arkadiusz_w how and why a WS90 is in his opinion hopeless and some high-end professional high-cost devices are much better...

LOL

[Rocketman]

maybe follow this thread
https://www.wxforum.net/index.php?topic=46133.0

if you decide to calibrate

[Gyvate]

maybe follow this thread
https://www.wxforum.net/index.php?topic=46133.0

if you decide to calibrate

first of all there has to be a need for calibration - as a rule this is the case when the sensor never shows 0 km/h or m/s or mph.
Then the zero baseline needs to be set ("calibration"). CAL button with the WS80/WS90.
Setting a gain factor = increasing or reducing the readings by a constant factor should be based on reliable reference readings. And the readings of neighbour stations are usually not a reliable source for reference for the reasons explained earlier.

[twassmer]

I think I keep things how they are. My sensor is at about 6 feet/2 m, and there are many trees and houses all around - so it will have lower wind speeds and erratic wind directions due to turbulence. Unfortunately, there is no better place in my backyard. My other station is about 2 miles away on the side of my university's football stadium at 6 feet height and almost open from all sides but from the SSE where there are some buildings.

[Meteorology fan]

I am also of the opinion that you should not do gain calibration for wind speed on the WS80 and WS90, because they are already calibrated at the factory. They measure correctly in the lower range up to about 20-30 km/h, as confirmed by a wind meter based on 200Khz transducers and calibrated in an accredited tunnel. The problem increases with higher speed, which we are unable to correct with a simple correction factor for the entire range of measured speed by the WS90 Wittboy.

Certainly, whoever hasn't yet upgraded from 1.1.9 to 1.3.3, it's worth doing a software upgrade to 1.3.8, because the WS90 Wittboy is more stable on this software, although far from perfect and the manufacturer's declaration of 5% accuracy above 10 m/s. Below you can see that the WS90 Wittboy indicated an unreal 112 km/h during the gale from 1.2.3, when it was blowing 60-75 km/h on the surrounding stations. Something was going wrong on this software during heavy rainfall and wind gusts around the aforementioned values.


Ecowitt has been struggling for several years to improve the stability of wind speed measurements. Of the things I have also observed on different versions of the software, it can show the wind in a room during silence. At the same time, the Hongyuv WDS2E placed next to it showed 0.0 km/h, or at times for a few minutes 0.04 km/h, when the WS90 Wittboy, but also the WS80 after calibration fell into the drift of the wind direction, but also appeared values around 1.8 km/h - 2.2 km/h away from other objects.

It looks as if the electronics in the WS90 are not stable, which is probably due to the poor 40 Khz transducers, which are not very stable relative to the 200Khz transducers.

Here a sample auction with these converters 200Khz so that everyone can see the price as well as photographs of 40Khz transkuders taken out of a damaged WS90 Wittboy.

https://www.alibaba.com/product-detail/200Khz-Piezo-Transducer-Ultrasonic-Sensor-for_62002447234.html?spm=a2700.galleryofferlist.normal_offer.d_title.289f3763kKYQ0N

I additionally add a video of how the WS90 Wittboy behaves on the 1.3.8 sofet against the Hongyuv WDS2E and Davis VP2. It was placed a few meters from the car, on a tripod anchored to the ground: https://www.youtube.com/watch?v=Kx0uhwch2PU

[broadstairs]

Arkadiusz_w  to use an English colloquialism I believe you are "flogging a dead horse" here on this forum. (Google it)

I am one of the WS80 users and am happy with my decision for a number of reasons. Firstly I am an amateur weather enthusiast and as such have to select equipment according to my budget and therefore am not interested in the high end equipment you are promoting. The main reason is that it will not improve my weather monitoring because my location limits where I can place my equipment significantly as I live in a residential environment. OK the readings themselves may be perhaps 5% better but that to me is immaterial 5% on wind is neither here nor there. What matters is that I have an approximate record of my conditions in my location which gives me some idea of the conditions prevailing at the time. It is well known by people with better knowledge that wind in particular measured in a resident location is normally at least 10% lower than expected in an open field environment.

I think your time would be better spent trying to get a job with Ecowitt or Fine Offset development so you can really see "from the sharp end" how things get designed and built in order to make a profit for the company in its market place.

Please stop peddling your ideas here as I for one am fed up with your interventions in every thread about the WS80 and WS90, we know your opinions and do not need to have them repeated again and again.

Stuart

[Meteorology fan]

You may be fed up, but I'm writing the truth and making you aware of what I've discovered new. You want to live a lie, then live it. The truth always hurts. From the information I've received, I know that in the future we should expect better wind measurement products based on better class transducers.

I am acting in the interest of customers from Poland, among others, who want sensors that are reliable and comply with the manufacturer's specifications in the product sheet.

After the windstorms in Poland you will see an interesting comparison on the forum how the WS90 Wittboy with software 1.3.8 performed and what differences we are talking about relative to the reference instrument.

[broadstairs]

I am not living a lie and I find that suggestion offensive.  I do understand the limitations of the equipment and my location and am quite happy with my decision.

Please do not reply to my post, this is my last update in response to you.

Stuart

[Rover1822]

Arkadiusz_w,

I can appreciate the depth of your investigation into these units. However, it is not like you are bringing up something new.  I have , in prior posts, described my dissatisfaction with the WS90 (I don't have a WS80 so I don't comment on those), but I do not do so when someone is looking for support on an already purchased unit. There is no point in just informing someone they have a unit you think is a bad unit to buy after they have bought it. The object of their post is to find help, guidance on making what they have work, or work better.

Deal is , if someone has already bought the unit, try to help if they have an issue. Not bring forth why they should not have bought it. Now if someone is asking to for guidance on buying a unit, yes, that would be an appropriate time to discuss potential issues.

As always, just my opinion

-Rover

[Rocketman]

Arkadiusz_w,

When do you think Hongyuv will come up with a wireless ultrasonic weather station version ?

[Meteorology fan]

@Rocketman As for Hongyuv, they are targeting the professional market and don't have much in the way of wireless stations. There is one option that an amateur might consider for the Davis VP2 and that is plugged into the Davis ISS transceiver, which allows reading on the Davis console and Davis gateways. The 200Khz transceivers are impregnated against rainfall so that it does not cut out the measurement in heavy rainfall. The durability of such a coating is about 2-3 years, it can be quickly renewed afterwards. This is a better ultrasonic anemometer than the LCJ offered by Davis for the ISS. Wider measurement range, lower tripping threshold, plus optional heating of transducers.

One downside is that it refreshes every 2.5 s, but you gain wireless data transmission. Mine is on a cable and refreshes 5 times per second, which doesn't let any short bursts escape. In addition, the console gives a taste of 3 s, as recommended by the WMO.

https://www.alibaba.com/product-detail/HY-WDS2DVSE-0-60m-s-Ultrasonic_60803585058.html?spm=a2700.shop_plgr.41413.34.7bb67ee9f2iXkc

For this anemometer I have a HY-DISPLAY console with transmission over RS485. This allows you to save data to an SD card, send to the iot Hongyuv cloud, but also to the WU or your own server. My WDS2E is practically the maximum variant of this anemometer. It has a measurement range of up to 60 m/s, in addition it is extended with EMI + frequency at 5Hz output, but also GPS and compass, which I deactivated, as they are unnecessary and interfere with me in the field. The console is equipped with a WIFI module.

Optionally, the WDS2E can be extended with a wider measurement range up to 75 m/s.

Below you can see how such a kit with an ultrasonic wind meter based on 200Khz transducers performs against the Davis VP2 and Vue, which was right next to it. The appearance of red numbers on the Hongyuv display means that the wind speed has exceeded the threshold of the user-set alarm for high speed.

I use such a set mainly for storms where strong downwelling currents are interacting, and Davis provides me with data on temperature, precipitation, pressure and other meteorological variables on the 6313 console.

A video showing how the kit works: https://www.youtube.com/watch?v=gyTwjdCesxg

Perhaps in the near future Ecowitt will release a product with higher accuracy, which will be based like Hongyuv on 200Khz transkuders, because there are such inclinations on their part after discussions with me. In my opinion, the user should have a choice. A cheaper and much inferior WS90 Wittboy based on cheap 40Khz transducers and a more expensive one based on 200Khz transducers with higher accuracy and wider measurement range. Several people in Poland are waiting for such an Ecowitt ultrasonic windmeter with 200Khz sensors, because the accuracy of the WS90 Wittboy does not satisfy them. I myself have stations high above the city and the sensor in the wind gauges from Ecowitt measure the wind badly, although the soft from version 1.2.3 to 1.3.8 is a big improvement, but some hardware limitations with 40Khz transducers I can not jump over and here, however, 200Khz transducers are required.

[Platokidd]

After reading this thread I decided to buy a new weather station and drop all my Ecowitt stuff. 
My new Davis weather station.
https://www.amazon.com/dp/B00FD22WKQ?psc=1&ref=ppx_yo2ov_dt_b_product_details
 [ You are not allowed to view attachments ]

[broadstairs]

 

Perfect!

Stuart

[giusCB]

other things... the wind value displayed on the ecowitt console is the average of the last 8.8 seconds (transmission time of the ws90), so it is impossible to have identical real-time values. The acquisition systems of Arkadiusz_w are completely different from each other; it is a test completely without sense.

[Stefan]

@Arkadiusz_w
I know you are here.
Most consumers would consider Ecowitt's product as satisfactory one taking its affordable price into account.
However, few people would be naive enough to believe its accuracy is comparable to professional ones.
Ecowitt is good at developing display console and IoT platform.
Anyone who subjectively doubt accuracy of reputated manufacturers like Davis, he is tended to be brain-washed in most condition.

Talking about calibration, only sensors with consistent performance under full scal can be calibrated.
Complicated algorithum is needed to ensure accuracy of 40KHz ultrasonic anemometer.
Our company release our first 40KHz HY-WDC2E ultrasonic anemometer in 2008.
Though we keep improving our algorithum, it will never be as good as our 200KHz HY-WDS2E ultrasonic anemometer.

[Meteorology fan]

@Stefan Thanks for being here. Maybe this will talk some sense into them. For me, WS90 Wittboy is a totally unreliable equipment and I take the data from it very carefully. The worst problem that I and people from Poland see is that the error in the WS90 Wittboy is non-linear and cannot be corrected by simple correction by the user. I have seen many times its big accuracy problems. One thing I do know is that if I didn't have the Hongyuv WDS2E, I would believe the results from the WS90 Wittboy. Anyone can see how the Hongyuv WDS2E vs Davis Vantage Pro 2 fared in the recent convective snowstorm:

https://www.youtube.com/watch?v=TrY6KolXSIk

You can see perfectly how close the Hongyuv WDS2E + 5Hz readings are to the Davis VP2, which is a rotating anemometer. Ecowitt's marketing brochures claim that the anemometer is highly accurate and that there is no problem in readings during heavy rainfall. I checked this thread with the Hongyuv WDS2E and Davis VP2 and found this to be untrue. I have evidence of this.

https://shop.ecowitt.com/products/wittboy

[Stefan]

@Arkadiusz_w
Sometimes, we are easily misguided without comparison.
Revealing truth may sometimes hurt people's feeling.
There is no doubt that they will use better ultrasonic probes some day if they carefully listen to your advices.

I can list some famous brands as followings:
VAISALA,LUFFT,GILL,RM YOUNG etc...
Davis mostly provide professional weather station for agriculture.
If you have interest, you may check exhibitor list below:
https://www.meteorologicaltechnologyworldexpo.com/en/exhibitor-list.php
Most reputated manufacturer attend this expo every year.

[Meteorology fan]

@Stefan I am very familiar with the site you mentioned. I even own products from selected manufacturers on the expo list.

For my part, I can say that the reliability of the wind measurements of the WS90 Wittboy would not have been verified by me if it were not for the fact that many Polish friends have reported inaccurate wind speed measurements relative to other measuring instruments they own. Having a reference instrument from you, the Hongyuv WDS2E ultrasonic windmeter, we began to verify compliance with the manufacturer's specifications, which writes to us about the exceptional accuracy of the WS90 Wittboy in its brochures. Next to it was a comparative Davis VP2 to see if it came close to indicating the Hongyuv based on 200Khz transducers.

Now we know that what the WS90 Wittboy shows is a lottery and is less stable than the WS68 rotary anemometer, which is closer to the Hongyuv WDS2E when it comes to measuring wind speed. Logically, through inertia, it will miss ultra-high wind gusts. It also measures direction worse, even relative to the Davis, which has a much more accurate direction measurement than any Fine Offset rotary windmeter.

I'm counting on Ecowitt to take these words to heart and stop writing untruths in public that can be challenged by having, as I do, the right kind of equipment.

In addition, they will improve the accuracy of their products to comply with the specifications in the product sheet. What they themselves write about below, their weather station WS90 Wittboy suffers. Subsequent updates, including the latest 1.3.8, have not solved the problem.

[broadstairs]

Can we now put this discussion to bed? There are plenty of threads folks who if they are interested can find on this topic by using search.

I for one am completely fed up with this continuous discussion about the WS90/80 and how bad you believe it to be. You are never going to persuade many people here to buy the professional equipment you believe we all should have because this is a hobby. I believe the majority here understand the limitations of the budget options which we choose (yes choose) to purchase many probably because it is what fits our wallet.

Please now drop all this and leave us in peace.

Stuart