Topic: Barani: too good to be true...

[Jasper3012]

Yesterday's passage of precipitation and thunderstorms resulted in large temperature changes. The maximum temperature slightly higher on the Davis FARS24H than the MS Pro. I got a signal from colleagues in Poland that Apogee can be warmer than passive shields lately, and this is not due to overheating of the Apogee TS100 shield. It is most likely due to the response time, while the Barani shield heavily averages the temperature measurement and has a lot of material inside which delays the response to temperature changes. Apogee shields most often depict rapid changes that the Barani Gen III passive shield will not catch. Hence, the maximum measured values may be higher on the Apogee TS100 aspiration and Davis FARS24H.

It’s quite obvious to me that on your first graph, where the Davis is consistently a few tenths above the Barani over a period of hours, the higher temp has nothing to do with better response time. A better response time would give you higher peaks and lower dips, but not the constantly warmer value like on your graph.

[Meteorology fan]

Buy if you want the Apogee TS100 and check it in parallel with the Davis FARS24H. For me, the Barani Meteoshield Pro III is not a reference shield and has quite a few flaws.

[Jasper3012]

And here you are again criticizing the Barani when in both your and my location, it has been the better shield for several days. I don’t buy the theory of Barani averaging out the temp too much and therefore causing higher maxes when I see the pattern of your graph. If this was truly the case and the MS Pro cuts out the peaks and dips too much, you’d be seeing lower maxes and higher mins for any given time interval throughout the afternoon, compared to the Davis. For example, if your theory is correct and you took a random 1-hourly period in the afternoon, you’d see a higher max, lower min and similar avg temp on the Davis compared to the MS Pro, because the MS Pro would be averaging out the values too much and therefore missing the lowest and highest values throughout that interval. Instead, what we’re actually seeing is a higher min, higher max and higher avg temp on the Davis during certain time intervals. This doesn’t match the pattern of better response time/no error at all, but is much moreso an indication of a general (albeight slight) overheating of the Davis. It’s actually quite simple maths. This is the issue I have with your comments, you are very happy to repeatedly point out the flaws in the MS Pro but every time I identify a potential flaw with the FARS, you immediately come up with theories to put the blame on the Barani. You need to have an objective look at both shields, not bash one and ignore flaws on the other.

[mauro63]

I promised not to intervene again, but I am only doing so to clarify a couple of aspects

- I think that, contrary to what I have been accused of, here there is a sort of mystification of reality dictated by commercial interests

- I would like the images of the respective installations to be published, before drawing conclusions on instantaneous data and I REPEAT, in the absence of knowledge of the measurand, do you have the vaguest idea of the unwanted heat transfers to which your systems are subjected?

Metrology is a science, fundamentally exact, the detection of air temperature goes strongly against these rules of accuracy, think about it

M.

[Jasper3012]

Another interesting comparison to add to this thread, Geert has been comparing his MS Pro with SHT-35 temp/hum probe with the RMI Stevenson screen a few metres beside it. The comparison has been running since the 12th of April and in that time, the MS Pro has had an average max that is 0.06C cooler than the Stevenson screen, with the average min 0.03C cooler than the Stevenson. No indication here of a big flaw with the MS Pro, but as we’ve discussed on here, that doesn’t rule out smaller periods (of a few hours for example) where there is in fact a discrepancy between the screen and the MS Pro. On average though, in the Belgian climate, there is very little difference between the two.

[mauro63]

Another interesting comparison to add to this thread, Geert has been comparing his MS Pro with SHT-35 temp/hum probe with the RMI Stevenson screen a few metres beside it. The comparison has been running since the 12th of April and in that time, the MS Pro has had an average max that is 0.06C cooler than the Stevenson screen, with the average min 0.03C cooler than the Stevenson. No indication here of a big flaw with the MS Pro, but as we’ve discussed on here, that doesn’t rule out smaller periods (of a few hours for example) where there is in fact a discrepancy between the screen and the MS Pro. On average though, in the Belgian climate, there is very little difference between the two.

Excellent,
the mistake, which is rather common, is to focus on instantaneous data, which, for technical reasons, can be affected by important differences.
The behavior of a screen, and its qualities, must be evaluated with a broader vision, moving the zoom away, installations that allow you to correctly evaluate an instantaneous data are very rare, the rest is the result of personal opinions, likes and dislikes, interests etc

M.

[Jasper3012]

What do you consider as instantaneous? Actual instantaneous or like daily values for example?

[mauro63]

What do you consider as instantaneous? Actual instantaneous or like daily values for example?

Instantaneous I mean evaluating a temperature data with sampling every 1 or 5 minutes, it is absolutely correct to have the most frequent sampling time possible, compatibly with the technical limits of the instrumentation we use, but it is their evaluation that must be compared to the situation.
if I have a class 1 test field, according to the WMO classification, I can be reasonably sure that any disturbances or unwanted heat transfers are hopefully cancelled.

here we evaluate instantaneous data with installations that have nothing to do with a class 1, close screens, poles, supports, plants, constructions, roughness of the ground, wind turbulence, their impact on the screens based on the direction of the wind, alignment of the screens, and I could go on for hours

M.

[mauro63]

If you have lower thermal transfer, you will have a better allignement, remove heat transfer and you will have this

 [ You are not allowed to view attachments ]

try to have the same on a usual house installation during the day, with wind, with high solar radiation level, with frequently wind direction changes, with temperatures rapid changes, with system with totally different time constants...

M.

[Meteorology fan]

It's also important to remember that a passive shield like Barani will never be a reference point. Aspiration shields from companies such as Apogee and Met One can serve as a reference point. The temperature changes are often greater than a passive shield shows us, even a class like the Barani Ms Pro, which as a plastic passive shield is the best on the market, but inferior to active ones naturally.

Something that can overheat in winter by 1.5-2.0 degrees from the Apogee TS100 can hardly be considered a reference shield from Barani.

https://www.apogeeinstruments.com/comparison-of-three-fan-aspirated-solar-radiation-shields/

https://metone.com/products/076b-fan-aspirated-radiation-shield/

https://www.apogeeinstruments.com/aspirated-radiation-shield-faqs/

[Jasper3012]

Literally every comment you make, you repeat the same argument, over and over and over again. Getting kind of boring. Perhaps for individual days it cannot be a reference shield, but if the comparison I mentioned a few comments ago maintains the non-existent difference between the MS Pro and the Stevenson screen, why would it not be able to be a reference shield for monthly and annual averages, perhaps even weekly averages? No matter how much you repeat your narrative (and you’ve done so plenty), the data is there for everyone to see…

[Meteorology fan]

God forbid, but why are you defending Jan Barani shield? She has a lot of hidden flaws in this shield! Colleagues in Poland, after more than a year of testing, juxtaposition with Stevenson full-size cages, Apogee shields, Davis FARS24/7, decided that the Barani Gen III cannot serve as a reference. Apogee TS100 can be a reference shield. A colleague who discovered these flaws and worked at WMO stations in the country for several years is of a similar opinion.

Yes most of the time the averages and minimum and maximum temperatures will be what they really are, but in harsh conditions Barani will not work. How will you know when such conditions occur and Barani is no longer “your reference”?

As I've written before. If this cover was so great from the Barani company, meteorological offices would be killing for it. As you can see there are cases where it is used, but it is never on the main measurement at WMO climatological stations. It is only for comparison and that is how I recommend using it. Personally, if I were you, I would take the Apogee TS100 and put it next to the Barani Ms Pro and Davis FARS24/7. Then you will understand how complicated air temperature measurement is and how many variables affect it.  Use the same sensors in these shields. E.g. SHT35 and PT1000 4-wire.

[mauro63]

I believe that the comparison is not possible in a constructive way, send photos of the installations, taken from a distance so that we can proceed with an evaluation of spaces, obstacles and possible disturbances together with the links for verifying the data about the stations where the users has found the Pro's problem.

In the absence it is a total waste of time, that I do not have

M.

[tobyportugal]

Hello Mauro,
I think that snapshots are invaluable information, provided of course that the installation complies with a minimum of rigour.
You know my (imperfect) installation and my particular climate.
The snapshots make it possible (sometimes) to distinguish between the 'installation', 'climate' and 'shelter' factors, because all my sensors are strictly identical.
Remember Antonio's peak in a single shelter at the same times for a few days in May 2022.
We analysed the video images, swapped the shelters and neither you nor Antonio nor anyone else could come up with an explanation.
Someone present here would have shouted out loud that this shelter is a "piece of shit".
This phenomenon reappeared a year later in more or less the same conditions, so with hindsight we can reasonably look for an external, non-climatic explanation.
In all the information that you can read on the left and right, it's mainly the pseudo 'PRO' sensors or the very complicated shelters that are problematic. To operate a TS correctly the bill is at least €1000.
Well, I'm going back to my Belgian dreariness.

[CW2274]

I have no dog in this fight. That said, in all around performance, to think that the best passive shield in the world can compete with the best aspirated shield in the world is a fool's errand.

[Jasper3012]

I posted a pic of the setup earlier in this thread.

[Meteorology fan]

https://www.campbellsci.com/blog/death-valley-collaboration-update

https://www.scottech.net/products/metsensor/874/apogee_aspirated_radiation_shiel/

Location is also an important thing. It should preferably be close to what they presented below in the attachment and preferably take a minimum of one year. However, I realize that not everyone has such a test plot.

[Jasper3012]

God forbid, but why are you defending Jan Barani shield? She has a lot of hidden flaws in this shield! Colleagues in Poland, after more than a year of testing, juxtaposition with Stevenson full-size cages, Apogee shields, Davis FARS24/7, decided that the Barani Gen III cannot serve as a reference. Apogee TS100 can be a reference shield. A colleague who discovered these flaws and worked at WMO stations in the country for several years is of a similar opinion.

Yes most of the time the averages and minimum and maximum temperatures will be what they really are, but in harsh conditions Barani will not work. How will you know when such conditions occur and Barani is no longer “your reference”?

As I've written before. If this cover was so great from the Barani company, meteorological offices would be killing for it. As you can see there are cases where it is used, but it is never on the main measurement at WMO climatological stations. It is only for comparison and that is how I recommend using it. Personally, if I were you, I would take the Apogee TS100 and put it next to the Barani Ms Pro and Davis FARS24/7. Then you will understand how complicated air temperature measurement is and how many variables affect it.  Use the same sensors in these shields. E.g. SHT35 and PT1000 4-wire.

I’m not defending Jan Barani, I’m defending his product, because I think you are going way over the top with your obsessive criticism. I said in my previous comment that I believe it can definitely serve as a reference in monthly and annual averages, I’d not put it as a reference shield for daily values, due to the error that can technically occur (although it will only actually occur on maybe 2-3 days out of 365).

[Jasper3012]

Pic of my setup again:

 [ You are not allowed to view attachments ]

[Meteorology fan]

2-3 days a year is a big averaging, maybe in Belgium it is so, where it is windier than in Poland. In the continental climate, for example, in Russia, such days can be as many as 50-100 days with a large overshoot during, sunny high weather and low wind. In the Polish climate it is about 5-15 days on average, and this depends on the location and wind conditions. In my case, the Barani shield, as in the case of colleagues in Poland, remained as a comparison to the active one. It lost the main measurement, because in low wind conditions in my valley it is often not in balance with the air temperature on the aspiration shield. Yes the minimums coincide, similarly the maximums, but not always either.

@Jasper3012 You have tall grass next to the station, which can project much higher maximum values on aspiration than on passive shielding.

[mauro63]

I have no dog in this fight. That said, in all around performance, to think that the best passive shield in the world can compete with the best aspirated shield in the world is a fool's errand.

Very true, there are conditions in which only a screen with an intubated sensor and ventilation can significantly reduce the radiative error in a way that a passive one will never be able to do.
Likewise, it is crazy to think of entrusting air temperature detection to a screen which, in the event of operating problems, returns junk data

The installations are important because, I repeat, only conditions of lack of unwanted heat transfer can allow a reliable comparison.

if two screens are too close together and, purely by way of example, the wind blows in the same direction, the windward screen will provide better performance, which will vary with the rotation of the wind.

Any pole, structure or support present in the immediate vicinity will have the same effects, which will vary the mode and impact on the data an infinite number of times with each slight variation in conditions.

therefore, I don't have the truth in my hand, and none of you here do, so if we want to compare each other in a friendly manner and increase our knowledge, fine, otherwise there's no point in continuing, at least for me

[mauro63]

Pic of my setup again:


Thanks,
reduce and level the height of the turf, if possible move the pole with the solar panel a couple of meters forwards or backwards so that the two screens "see" each other without obstacles
the distance between the two screens seems sufficient to me, it eliminates those poles that can be seen around them

[CW2274]

Pic of my setup again:

Curious, why are the two solar panels not aligned? If I see what little shadowing there is, looks like the Davis is facing east, no? If so, obviously less fan speed average from less direct sun.

[Jasper3012]

2-3 days a year is a big averaging, maybe in Belgium it is so, where it is windier than in Poland. In the continental climate, for example, in Russia, such days can be as many as 50-100 days with a large overshoot during, sunny high weather and low wind. In the Polish climate it is about 5-15 days on average, and this depends on the location and wind conditions. In my case, the Barani shield, as in the case of colleagues in Poland, remained as a comparison to the active one. It lost the main measurement, because in low wind conditions in my valley it is often not in balance with the air temperature on the aspiration shield. Yes the minimums coincide, similarly the maximums, but not always either.

@Jasper3012 You have tall grass next to the station, which can project much higher maximum values on aspiration than on passive shielding.

I very much doubt that this grass is going to cause any sort of difference between the shields, but sure, I will go and cut it tomorrow and see if there’s an immediate effect.

[CW2274]

I have no dog in this fight. That said, in all around performance, to think that the best passive shield in the world can compete with the best aspirated shield in the world is a fool's errand.

therefore, I don't have the truth in my hand, and none of you here do, so if we want to compare each other in a friendly manner and increase our knowledge, fine, otherwise there's no point in continuing, at least for me

Best comment of the thread.