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

[Meteorology fan]

It's physics. A shield that pulls air in from the side will never be as good as pulling air out from underneath like the Apogee TS100 and Davis FARS24H. I have access to the Apogee TS100 in Poland at a good price.

[Jasper3012]

FARS at night isn't necessary with the Barani, the avg min on the Barani over a longer period is less than 0.1C milder than on the Davis. The biggest difference I've ever recorded is a 0.4C difference in the recorded min. The Barani does sometimes lag behind more than this 0.4C value but this is always a temporary thing.

[hmderek]

This is at the station and looking east. The rectangle you drew under the FARS and toward the ground is actually just the dark pole itself (it's a dark green) and the slight shade to the right of it is that pole casting shade on the wooden pole a few metres behind it. The middle rectangle is the "console" casting shade on its supporting pole and the shade on the FARS itself is obviously due to the solar panel. I am a bit of a perfectionist myself but I do think we're taking it a bit too far here 

 [ You are not allowed to view attachments ]

Have I mentioned yet I love your spot? ^_^

[Jasper3012]

This is at the station and looking east. The rectangle you drew under the FARS and toward the ground is actually just the dark pole itself (it's a dark green) and the slight shade to the right of it is that pole casting shade on the wooden pole a few metres behind it. The middle rectangle is the "console" casting shade on its supporting pole and the shade on the FARS itself is obviously due to the solar panel. I am a bit of a perfectionist myself but I do think we're taking it a bit too far here 

 [ You are not allowed to view attachments ]

Have I mentioned yet I love your spot? ^_^

Yes, you have It's a very good spot indeed, can't do better than that here.

[hmderek]

When talking about daily max temps of ventilated vs passive systems, this is also something that can happen. The TS-100 could show a daily max over 0,4 degrees higher than the passive Barani due to stronger responsiveness to temperature changes.

 [ You are not allowed to view attachments ]

[Jasper3012]

Your graph is actually a perfect example of a better response time, it goes higher than the Barani during peaks and lower than the Barani during drops. What I’ve often been seeing isn’t that though (at least not fully explained by it), I’ve often just recorded the Davis being a steady 0.3C warmer than the Barani during sun and wind, with not really a better response time but just a steady overheating. Graph below from a few days ago shows this during the afternoon, you don’t see a more erratic curve due to a better response time on the Davis (the brown line), but rather just a constant 0.2-0.4C warmer.

 [ You are not allowed to view attachments ]

[Meteorology fan]

August is full of days with high weather and light winds in Poland. These conditions do not serve the Barani Meteoshield Pro III shield. On several occasions in the past few days, Barani has overheated by more than 1 degree. In the mornings even 1.5 degrees, afternoons and evenings are sometimes difficult with errors from solar radiation reaching more than 1.5 degrees. The sun angle in August is lower than in July, as can be seen from the frequent Ms Pro III errors in the Polish climate. In autumn and winter with low winds it can be even worse.

Several days in a row the t-max was higher on the Ms Pro III than on the Davis FARS24H.

[hmderek]

Your graph is actually a perfect example of a better response time, it goes higher than the Barani during peaks and lower than the Barani during drops. What I’ve often been seeing isn’t that though (at least not fully explained by it), I’ve often just recorded the Davis being a steady 0.3C warmer than the Barani during sun and wind, with not really a better response time but just a steady overheating. Graph below from a few days ago shows this during the afternoon, you don’t see a more erratic curve due to a better response time on the Davis (the brown line), but rather just a constant 0.2-0.4C warmer.

 [ You are not allowed to view attachments ]

Am I correct to assume that both those sensors are standard Davis issue SHT31s?

[Jasper3012]

No, 2 NTC thermistors.

[JCA433]

August and September are the months in my location when weak winds or no wind calm conditions most often occur.  These weather conditions can be a big problem for passive weather stations.  When the ten-minute average wind speed is less than 1 Knots, large errors are possible.  The worst condition is calm or no winds when very large errors can occur.

[Jasper3012]

August was a calm month here aswell and it was notable in the values, the number of days where the Barani was cooler with the max by like 0.3C were far reduced compared to June/July and I even had quite a few days with equal maxes or the Davis being 0.1C cooler, never more than that though. I suspected before this week that with the reducing solar radiation, the error on the Davis during sunny and windy weather would decrease but to my surprise, in the past few days where there’s been sun and wind, it has still been around that 0.3C warmer than the Barani during the day. The fact the sun angle is significantly lower now has not had an impact so far, seemingly.

I also very recently purchased that new SmartCellino shield from Italy and I will be adding it to the comparison. I only have 2 temp sensors right now so I might have to remove the Davis for some time, until I get a 3rd temp sensor at the station. It’s not ideal as I want to run every shield at the same time, but yeah. I’m hoping to be able to put the Davis back before the winter, so I can see the comparison during consistent low angle sunshine.

[Meteorology fan]

Even ChatGPT recognises that a low sun angle does a disservice to Barani Meteoshield Pro and other passive radiation shields. I'm sticking with the Davis FARS24H and the Barani Ms Pro III. I have, however, ordered another radiation shield for myself and there is an aspirational Apogee TS100 to test. I acquired it at a promotional price and it is heading to me from the USA. I have to wait about four weeks.

https://www.apogeeinstruments.com/ts-100-ss-12-v-aspirated-radiation-shield-shield-only/

The comparison will be joined by another PT1000-type sensor in the new version and the SHT35 without filters in the Apogee TS100.

https://termoprodukt.co.uk/Precise-temperature-data-logger-Termio-2

[Meteorology fan]

I use it for a lot of things. I asked him if it was worth using aspiration covers like the Apogee TS100 and he indicated that it was. Especially in low wind conditions.

[Jasper3012]

August was a calm month here aswell and it was notable in the values, the number of days where the Barani was cooler with the max by like 0.3C were far reduced compared to June/July and I even had quite a few days with equal maxes or the Davis being 0.1C cooler, never more than that though. I suspected before this week that with the reducing solar radiation, the error on the Davis during sunny and windy weather would decrease but to my surprise, in the past few days where there’s been sun and wind, it has still been around that 0.3C warmer than the Barani during the day. The fact the sun angle is significantly lower now has not had an impact so far, seemingly.

I also very recently purchased that new SmartCellino shield from Italy and I will be adding it to the comparison. I only have 2 temp sensors right now so I might have to remove the Davis for some time, until I get a 3rd temp sensor at the station. It’s not ideal as I want to run every shield at the same time, but yeah. I’m hoping to be able to put the Davis back before the winter, so I can see the comparison during consistent low angle sunshine.

Funnily enough, just as I wrote this today, I recorded a max 0.3C warmer on the Barani compared to the Davis. This is the first time the Barani has been over 0.1C warmer with the max than the Davis since I started the test mid May. I suspect that this will happen slightly more often in autumn and winter due to the lower sun angle but probably still quite rare, as the calm with sun combination is very uncommon here in those seasons.

[Meteorology fan]

I will report on how the Davis FARS24H compares to the Apogee TS100 and the Barani Ms Pro III. I had a bargain price and purchased the Apogee, as it is however a research grade reference cover. Other people from Poland who operate copies of the TS100 recommended this shield to me and said it was better, but more complicated than the Davis. It worked out so well that the dollar exchange rate had dropped noticeably relative to recent months and I had a price still old for the Apogee TS100. Apogee gives as much as 48 months warranty on the product, which is a testament to the quality of theirs.

One thing I do know is that the response time on the Apogee is certainly much better than on the Davis FARS24H and in harsh conditions the TS100 from Apogee will certainly be slightly better than the Davis FARS24H and even more so than the Barani Meteoshield Pro III, which is a passive shield but one of the best on the market, but yielding in harsh conditions that are not that uncommon in the Polish climate.

I will receive the Apogee TS100 in October 2024 and will still need to organise a few tens of metres of cable for power from the garden and, in addition, a 12V power supply for the permanent power supply. Colleagues in Poland who have the Apogee TS100 will instruct me. The good thing about Apogee is that you can make a fully intelligent FARS using a suitable controller and pyranometer however, this is a high cost that I won't go into right away and will use a simple voltage regulator that was recommended to me by colleagues operating the Apogee TS100 in Poland.

[mauro63]

August was a calm month here aswell and it was notable in the values, the number of days where the Barani was cooler with the max by like 0.3C were far reduced compared to June/July and I even had quite a few days with equal maxes or the Davis being 0.1C cooler, never more than that though. I suspected before this week that with the reducing solar radiation, the error on the Davis during sunny and windy weather would decrease but to my surprise, in the past few days where there’s been sun and wind, it has still been around that 0.3C warmer than the Barani during the day. The fact the sun angle is significantly lower now has not had an impact so far, seemingly.

I also very recently purchased that new SmartCellino shield from Italy and I will be adding it to the comparison. I only have 2 temp sensors right now so I might have to remove the Davis for some time, until I get a 3rd temp sensor at the station. It’s not ideal as I want to run every shield at the same time, but yeah. I’m hoping to be able to put the Davis back before the winter, so I can see the comparison during consistent low angle sunshine.

Hi Jasper,

Greetings and thanks for choosing to add SMarTCELLino by Siap+Micros to your instrumentation, we have been testing it for some time now in various locations in Italy and with excellent results, I will be happy to analyze your reports and provide you with support if needed.

If you want should be useful to open a specific post

M.

[Jasper3012]

Hi Mauro, thanks to you aswell for the discount! Looking forward to setting it up and seeing the results. Like I said, it will be a Cellino vs Barani comparison at first and then hopefully expanding to Cellino vs Barani vs Davis FARS 24/7 within the next month or two, once I get the 3rd temp sensor setup.

[hmderek]

This puts the 'advantages' of a fan aspirated shield into perspective. Foggy morning today. Typical Apogee behavior when moisture starts to evaporate. As soon as the sensors dry up (doesn't happen at the same point) the readings jump up. Effectively showing up to 1.5 degrees overcooling on the Apogee sensors. (Same happens earlier with the passive sensors too, but only to a very limited degree)

 [ You are not allowed to view attachments ]

[tobyportugal]

 [ You are not allowed to view attachments ]

Hello Derek,
Personally, I'm sticking with it:
1 shelter + 3 sensors = 3 results
That puts into perspective some people's hasty conclusions. 

[hmderek]

Hello Derek,
Personally, I'm sticking with it:
1 shelter + 3 sensors = 3 results
That puts into perspective some people's hasty conclusions. 

I don't follow, that's the graph I just shared but without the other sensors?

[tobyportugal]

Yes, it shows that a shelter with different sensors can have different reactions.
It's a good example for people who often forget to take precautions when carrying out their analyses.

[Meteorology fan]

I found some time and generated charts from August for my location. You can see perfectly the moments when this Ms Pro III Barani fails. The two shields being compared side by side are Davis FARS24H and Barani Meteoshield Pro III. Soon the Apogee TS100 will join the comparisons with the SHT35 Sensirion without filter and PT1000 4-wire and calibrated kit.

[Meteorology fan]

bianconero57 - I am removing the company filters from the SHT35 from Ecowitt because it is a poor option on the SHT35. It's a good sensor from Sensirion, but that doesn't mean there will only be an Ecowitt WH31EP in the Apogee. A colleague uses such a kit and it's ok, but we do redundant measurement on a resistance thermometer, 4-wire compensated along with a logger. I have such a set up in the Barani Meteoshield Pro III, Davis FARS24H and so will add a second sensor for comparison with a higher accuracy sensor from the SHT35, which he uses without a filter for better response time.

We use this, among others, with a logger under the computer: https://termoprodukt.co.uk/Precise-temperature-data-logger-Termio-2

Below is a chart for the past 24 hours. You can see how messy the Barani Meteoshield Pro III behaved in the morning when it was blowing weakly against the Davis FARS24H. This is certainly not a research grade shield and cannot be relied upon. I don't know if a wood or plastic Stevenson-type cage isn't a better choice, as it certainly generates less radiation error in these conditions than the Meteoshield Pro III.

I can also see this with a colleague who has a Stevenson WMO Screnn and in the mornings and late afternoons/evenings, when the light is lower and the wind is weak, the errors induced by Meteoshield Pro III reach 1-1.5 degrees. This is therefore not a new standard of accuracy.

[Jasper3012]

You can repeat this same exact point another million times but it’s not going to stick if others have a different opinion… After 5 months of testing, I’ve got a pretty good idea of the pros and the cons of the MS Pro and for me, the pros far outweigh the cons. In my climate and my weather station site (open and exposed), the Barani is simply a better option than the factory Davis 24/7 FARS. This might change with a stronger fan, but the factory one isn’t good enough.

[Meteorology fan]

The wooden, WMO Screen is a better choice than the Barani Ms Pro III. We have a full-size Stevenson in Poland and I can see how the Barani strongly overestimates the WMO Stevenson Screnn right next to it under certain conditions and these are differences of more than 1-1.5 degrees. This is no successor to Stevenson and will not send it down the dinosaur road.

The reason why I chose the Apogee TS100 is the response time, which is better than the Davis FARS24H and the Barani Meteoshield Pro III. Of course, I take into account that there is a risk of liquid overcooling of the sensors under certain conditions, but this is certainly rarer than the Ms Pro III overstating the air temperature.

More and more people are noticing that the Barani Ms Pro III overestimates the actual air temperature by a large margin, and are choosing the counterparts vented in Poland for comparison.

The Davis FARS24H could use a more powerful windmill, it's true, and I miss the faster response as in the Apogee TS100, hence I chose the Apogee aspirated cover for comparison with the Barani Ms Pro III, Davis FARS24H on one mast and with analog sensors. He wants to check the long-term differences and will certainly publish them publicly. Redundancy of air temperature measurement is key, as passive counterparts will not always work in a given situation.