The past few years, until 2023: using Barani Design...
Since the start of our network, the "Meteohelix IoT Pro" from the Slovak manufacturer Barani Design had been used. In the year of 2023 we counted a total of just under 100 of these weather stations in our network. The stations are compact and affordable for the weather enthusiast, in contrast to the professional measuring equipment that has a considerably more substantial price tag. The specifications given by the manufacturer also seem to be very good. The slogan "that these have been tested by the RMI" is being waved and paraded. For your information, the RMI only uses these for measuring temperature in their climatological network.
For the measurement of precipitation, there is the rain gauge "MeteoRain" (linked via a cable to the MeteoHelix); in addition, there is also the MeteoRain IoT which only transmits rain measurement data via the Sigfox network.
In addition to the Meteohelix and MeteoRain, the manufacturer also has the "MeteoWind IoT" in their range. As the name suggests, the device serves to measure wind speed and direction.
However, so much for the good news show. After three years of use, we now know a lot about the pros, but unfortunately even more about the cons of this equipment. With regret, we have to say that, in our opinion, the negatives outweigh the positives.
Our experiences over the past three years:
review BARANI METEOHELIX - METEOWIND - METEORAIN | |
Positive | Negative |
| The MeteoHelix's shield is compact, easy to install | In the three to four years we have been using Barani weather stations, of the almost 100 stations we have had 60 defects. For us this figure is high, much too high, even for a product that has only been on the market for a short time. Conclusion: Barani's reliability is poor. |
| The radiation shield is passive, which means you don't need a fan to get good temperature measurements. So no wear and tear and no energy consumption. The measured temperature deviations under strong solar radiation and little wind are quite small, so very good. This is also what the 2017 RMI tests showed and what the manufacturer also keeps referring to. | Even after 3-4 years of using these weather stations, we still see the same problems popping up again and again... The manufacturer doesn't seem to find an immediate solution for the defects: - sensors that keep failing because of (insufficient or poorly applied) coating against humidity and/or water, - sensor connection where condensation causes short circuits between different connection pins resulting in corrosion or no sensor data, - defective solar panels or wires on the solar panel that break off after some time due to corrosion (water). We have encountered other issues as well, but the ones listed above occur quite often and somewhat consistently. |
| Temperature deviations remain low even when the MeteoShield itself is dirty. | The legal warranty of 2 years applies. After that, a replacement sensor will cost you 100 euros (+VAT +shipping); experience tells us that these also fail after 1-2 years... |
| No wiring, apart from any rain gauge which is best placed at a small distance. | The specifications given by the manufacturer do seem to us to be "very optimistic", not to say "exaggerated": - Temperature measurement (according to WMO rules) is +/- 0.1°; the manufacturer of the sensor itself guarantees the +/-0.1° accuracy only between 20 and 60°C, outside this range it can be as high as 0.3°. This means that you need to calibrate the sensor at regular intervals. Unfortunately, there is little or no mention of this. - Solar radiation is specified as a Class C pyranometer; this may just be met when first using the MeteoHelix, but not shortly after that. Comparison with 2 calibrated class A pyranometers (i.e. much more accurate pyranometers) shows that several helixes we measured can differ by up to even 20%. In other words, the specification should be taken with a coarse grain of salt. - The specification for air pressure +/-0.5hPa is given at an ambient temperature of 0/50°C. However, the pressure sensor is also outside and experiences large temperature fluctuations. At several stations, we see large deviations appearing over time, which then render the measured value completely worthless. Periodic recalibration is necessary, but who has a device that can do this? |
| Acceptable price for what is promised by the manufacturer in their specification, but... (also read the other column.) | There is the possibility of purchasing an extended warranty; you will of course have to pay extra for this. Unfortunately, sensors are excluded from this extended warranty. Up to you to guess why... |
| MeteoWind Compact sensor (anemometer + wind vane) is compact, but generally works quite well. As for software of the IoT version, see right column. | The filter cap located over the sensor module is plastic and does not fit perfectly with the housing of the rest of the electronics, allowing small pests to reach the sensor with all its consequences. |
| Price of wind sensor MeteoWind is not too bad, although it has gone up quite a bit over the past few years. | Many of the Barani rain gauges (MeteoRain, including also the IoT version) show large deviations when compared to other rain gauges (including the RMI manual rain gauges). We are talking about an average 10-20% under-measurement, but some weather conditions cause deviations up to almost 50%. The reason for this deviation is still a big question mark. Hence, we have our reservations about this rain gauge, even if they would have you believe that it outperforms all other devices. |
| The experience with the wind meter is slightly better, but here too there are a few things to note: - The original cups of the wind-gauge are rather fragile, but they have been replaced by a somewhat sturdier successor, so this problem seems to have been solved. - The software of the IoT version sometimes decides it's suddenly no longer necessary to send wind speed data for a while; after notification and confirmation of the problem by the manufacturer, we are still waiting for the promised update of the software... a year and a half is quite a long time despite several promises. To this moment we are still waiting… - During frost, the anemometer freezes relatively quickly; several anemometers from other manufacturers also suffer from this, but in our opinion a lot less quickly. | |
Conclusion after 3-4 years of experience with Barani equipment:
On paper everything seems to be nice and good. The marketing department of Barani has done a good job: good website where the equipment is presented with nothing but advantages compared to other competitors on the market.
However, practice seems to prove that much cannot be delivered from what is touted on the website. Much is promised, but there appear to be many empty promises...
Barani's MeteoShield is one of the few products with which they put a good product on the market. But once electronics and software are involved, things go seriously wrong and solutions take a long time or don't come at all. The initial ideas are good, but in our experience the implementation much less so.
Customer service is limited to the delivery of replacement parts, but we miss real solutions to all the frequent problems we encounter. We are often told that we are the only ones who experience problems, the defects are due to the Belgian climate, and so on. In other words, for every problem there is a cause that is not theirs to find and/or resolve.
Our experiences are not based on a small number of devices but on almost 100 of them, which in our opinion represents a more than representative number and gives a good picture of what the Barani products exactly are and how they behave after several years.
Consequently, as far as the BMCB network is concerned, we can draw only 1 conclusion: the end of the Barani era... for us, it was a waste of money.
More details about the defects can be found here
The alternative: the new weather station developed by BMCB
Because of all that we have described above, the idea to look around for an alternative was raised in early 2023 already.
The new weather station has to meet a number of requirements:
- of course measuring according to the valid rules of the WMO
- no defects would be ideal, but we have to remain realistic: far fewer defects than Barani's weather stations
- more flexibility, supporting more types of sensors and combinations
- also use IoT networks such as Sigfox and LoRa, and to be energy efficient
- calibration should be possible of sensors in the weather station itself, not afterwards on the server
- have memory on board to store data
- be affordable, but not at the expense of the issues mentioned above
Development was soon thereafter started and the first 4 test weather stations were installed in the Ardennes in early November 2023. The Ardennes was chosen because of more extreme weather conditions that can be expected there compared to Low and Central Belgium, especially during winter. The purpose of the test stations was to quickly identify any design shortcomings.
The testing bore fruit: two problems surfaced and have since been corrected. The new version will, of course, undergo further testing and also be done in larger numbers to maximize reliability.
The new weather stations also have greater flexibility: multiple versions will be made that have different combinations of sensors. A "weather station" (location) can consist of 1 or more (measurement) modules measuring different parameters:
- a module with 2, 3 or 4 temperature sensors + connection for rain gauge
- a module with a temperature/humidity probe + 1 or 2 additional temperature sensors + connection for rain gauge
- a module for wind measurement: currently Barani's MeteoWind sensor is supported (i.e. not IoT!), but the necessary provisions are in place to connect instruments from other manufacturers
- a module that acts as a barometer where atmospheric pressure is accurately measured; this station or module is intended for indoor use to avoid large temperature fluctuations; indeed, temperature has a great influence on the accuracy of pressure sensors
- scheduled: module for solar radiation measurement with different types of pyranometers
- scheduled: module for measuring soil moisture and leaf wetness
- in planning: module with 8 temperature sensors
- on the planning: module for interfacing with all kinds of sensors that support the Modbus RTU protocol (to give some examples: weighing rain gauges, smart-pyranometers, etc...).
So it is perfectly possible to use a module with temperature sensors in combination with a module for wind measurement. The two modules then form a weather station.
Each module has its own transmitter on board to send the data. Where necessary, functionality is in place to perform calibration of sensors (e.g. temperature sensors) and store the calibration directly in the module. The great advantage of this is that the calibration can be performed much more precisely than afterwards on the server (where the data is delivered with a resolution of 0.1°). Moreover, every recipient of this data immediately receives the correct data.
The individual modules of the weather station use a solar panel for power supply. If sensors are used that have higher energy consumption, the size of the solar panel can be increased.
Each module has an on-board memory that allows storage of data for 1 year of observations. In addition to the data sent over the wireless network, additional data can be stored, either with better time resolution. Via the wireless network the data are sent every 10 minutes, in the memory data can be stored, for example, every 2 minutes.
The development of the software of these modules is done entirely in-house. We are no longer dependent on a manufacturer. Although we strive not to create a proliferation of different versions, we can consider everything. When we discover a problem, we can also intervene and solve the problem much faster.
So doesn't the weather station have any drawbacks?
When you compare it to the MeteoHelix, it is bigger without being "really big." The electronics box measures 20x10cm and is 8cm high.
There is also more wiring; it's hard to get away from that when you want to measure temperature at 1.5m height and also the temperature on the grass and/or in the soil... the solar panel can be mounted separately, but it has a cable connection to the electronics box. The connection of the rain gauge is also via a cable connection.
But no worries, our team will be happy to help you with the installation...
And the price?
First of all, we would like to make it clear that the prices we mention below are only indicative and calculated based on the prices currently known and for a given lot size. Prices of parts may increase, exchange rates may fluctuate to our disadvantage, ... The price applies to people who contribute (i.e. add value) to our network. We ourselves do not need to earn from it, it was no different in the past. We are not a commercial company and do not need to make a profit. We only increase the total parts cost by 15% that serves as a buffer:
- for fixing owner defects, call it a kind of warranty;
- to pay for prototypes during development;
- for the cost we have when a module would not work, call it "failure during startup".
- NOT for working hours or wages, and
- NOT for things that do not advance our club and measurement network.
A lot of course depends on what you want to measure and with what sensors. You can make it as expensive as you want. We do not make sensors ourselves (with the exception of the temperature/humidity probe). We use sensors that are on the market and have already proven their soundness. Due to the great flexibility of the new design, we can easily deal with a large range of sensors and what does not work by default, we can make extensions for that fit in our electronics box (which we have already foreseen).
The cost price of the electronics box (including solar panel) is around 250 euros; this is the price without sensors, no rain gauge and no 'shelter' or MeteoShield.
Count on about 25 euros for any good temperature sensor that can be perfectly used for measuring grass and soil temperatures. The measuring probe for temperature/humidity that goes into the shelter/MeteoShield is about 60 euros and uses the same sensor element as the MeteoHelix.
The choice of rain gauge is up to you, all models of 'tipping bucket' rain gauges can be connected. We are currently testing which types show the best performance and will be updating this document during the coming months. After all, we want to find a better instrument than the Barani MeteoRain, which we are not at all satisfied by.
Measurement of solar radiation is more expensive: the cheapest pyranometers cost quickly several hundred euros, up to thousands of euros for professional and more accurate models.
As you can read, we still have many ideas, and in the coming months and years we want to realize them. We will gladly keep you informed about everything that happens and what we can recommend and reject... To actually prove ourselves, we want to compare ourselves with professional weather stations. In the coming period you will certainly read more about this...
First we will test before actually offering. After all, we don't want weather station owners to serve as beta testers...
Above article was posted on March 10, 2024
Update April 24, 2024
Installation of the latest version of WeatherWise weather stations began on Friday, April 12. Two of our newest weather stations were installed on the grounds of the Royal Observatory of Belgium at Humain (near Marche-en-Famenne). This is the same site where one of the synoptic weather stations of the RMI is located and where the RMI also conducted the tests of Barani's Meteoshield.
The measurement setup is in collaboration with the CRA-W (Centre wallon de Recherches Agronomiques). Other weather stations for agricultural purposes are also being tested at this site.
The BMCB weather stations - fraternally side by side - can be viewed live via the RMI's webcam at the Humain weather station: https://www.meteo.be/en/weather/observations/webcams/webcam-humain
2 weather stations were installed to test and compare the two different measuring sensors we are using. These weather stations can be further compared for performance with the other weather stations present at this site, including the official RMI weather station.
The weather stations measure air temperature (2 versions of sensors), relative humidity, grass temperature at 10 and 5cm height and soil temperatures at a depth of 5, 10 and 20cm.
A second location already received a new weather station. It was placed at Losheimergraben (Büllingen) 100m away from the German border. Here it concerns him a temperature station (in the shield at 1.5m and also at 50/30/10cm above the ground). The intention is to measure the grass minimum, but with snow cover the lowest temperature is measured just above the snow, hence also measuring sensors at 30 and 50cm height.
A third location that is equipped with a new weather station is at the very eastern side of Bruges (Male - Sint-Kruis district), a stone's throw from Sijsele (Damme). Again, it is a temperature station (shield at 1.5m and at 10cm above grass).
Weather permitting, several more locations will be added in the east of the country during the weekend of April 28.
Update May 8, 2024
The rollout of WeatherWise weather stations continues steadily....
On Sunday, April 28, 3 locations in the east of our country were provided with the latest type: Grünen Kloster / Elsenborn and also Mürringen received a measuring station with temperature at 1.5m and above grass, at Braunlauf an update was done from the first test station (which had survived the winter without problems) to the newest version with measurement of temperature in a radiation shield at 1.5m, at 2 levels above grass and also humidity and precipitation.
On May 6, a new WW station was installed 100m over the border with the Netherlands at Sint-Kruis (part of Sluis) with measurement of temperature and humidity; two more stations followed the day after: in the north of the province of Antwerp at Wuustwezel, yet another defective Barani Meteohelix was replaced by a WeatherWise station (temperature in radiation shield and above grass, humidity, precipitation) and a new location at Stekene (East Flanders) also received a measuring station for temperature and humidity.
So far, none of the WeatherWise stations show signs of problems (and of course we try to keep it that way...).
Update May 17, 2024
Today was another tour of the province of East Flanders.
o
The first stop was in Merelbeke where a weather station was installed at a new locatin to measure temperature at 1.50m and above the grass as well as relative humidity.
The second stop was in the Flemish Ardennes where our former weather station in Brakel received a thorough update. The Barani weather station - defective since a long time - was replaced by a WeatherWise version; the same parameters are now measured as in Merelbeke, but also the precipitation.
Today, to our great annoyance, we have to note once again that measuring precipitation with the Barani MeteoRain is also problematic. In the east of our country, large amounts of precipitation fell in the past 24 hours (30-40 mm in some places) and certain rain gauges only managed to register a few millimeters. Time to replace that Barani equipment with a better working alternative as well.
Update May 23, 2024
In the province of West Flanders, the coastal area was visited today and the faulty Barani-Meteohelix station in De Haan was replaced by a new WeatherWise measuring station.
An additional weather station was placed at Westende so that in addition to De Haan, we have a second measurement location less than 1 km from the sea.