Topic: Barani MS Pro vs SmartCellino radiation shield comparison

[tobyportugal]

Derek, is this your shadow problem? Dank u 

Yes sir. 

That is, low hanging sun came out late in the afternoon, zero wind, trees intermittently shading one shield after another.

Data in the second half of the afternoon is difficult to interpret in the summer, pretty much useless later in the year. Sun is dropping very fast this time of year. Nearing december you can't make heads or tails of the data all day long pretty much.

Right, I'll send you a GLS label for Portugal.
Your problem's no laughing matter.

[hmderek]

Right, I'll send you a GLS label for Portugal.
Your problem's no laughing matter.

The only time I really have my setup in full sun all day is second half of May - July.

I guess most sites in Northern Europe have this problem. You really need a significant area of open fields around your setup if you want to avoid significant shading. I've got a row of trees to the south, maybe 20 meters away. Come December I think the sun is stuck behind the tree tops all day long.

But then winter is more interesting during cold nights anway.

[Kallo78]

Last three's day reports

[Jasper3012]

Something slightly more interesting to report... For a time this afternoon, the Barani jumped to 0.2-0.4C warmer than the SC. This happened during sunshine and wind speeds of 4-6 km/h, gusting to 7 to 12 km/h, so rather weak. It was also generally 0.1-0.2C warmer than the SC for much of the afternoon, as you can see on the graph (green = Barani, brown = SC). To me, this indicates a slight overheating of the Barani.

 [ You are not allowed to view attachments ]
 [ You are not allowed to view attachments ]

[Jasper3012]

The fact the green line is consistently above the brown line tells me this is an overheating, rather than a better response time. I've tried to draw in paint how I see an overheating vs a better response time. In my opinion, the red line on the left is a classic example of a better response time compared to the black line, with a more erratic curve but averaging around the same as the black line. On the other hand, on the right, the red line is just overheating compared to the black line, with no particular difference in response time.

 [ You are not allowed to view attachments ]

[mauro63]

The fact the green line is consistently above the brown line tells me this is an overheating, rather than a better response time. I've tried to draw in paint how I see an overheating vs a better response time. In my opinion, the red line on the left is a classic example of a better response time compared to the black line, with a more erratic curve but averaging around the same as the black line. On the other hand, on the right, the red line is just overheating compared to the black line, with no particular difference in response time.

If this happens consistently, during the hours when there is no solar radiation, that is, every night, I would be inclined to think that the sensors are misaligned.

Was an alignment check done before the tests began?

M.

[Jasper3012]

The fact the green line is consistently above the brown line tells me this is an overheating, rather than a better response time. I've tried to draw in paint how I see an overheating vs a better response time. In my opinion, the red line on the left is a classic example of a better response time compared to the black line, with a more erratic curve but averaging around the same as the black line. On the other hand, on the right, the red line is just overheating compared to the black line, with no particular difference in response time.

If this happens consistently, during the hours when there is no solar radiation, that is, every night, I would be inclined to think that the sensors are misaligned.

Was an alignment check done before the tests began?

M.

I'm talking about this afternoon though, when there was in fact sunshine... Like I've said before, they're identical sensors that were calibrated at the same time. You can even see on the first graph that they were running the same temp, before the Barani started going warmer.

[mauro63]

The fact the green line is consistently above the brown line tells me this is an overheating, rather than a better response time. I've tried to draw in paint how I see an overheating vs a better response time. In my opinion, the red line on the left is a classic example of a better response time compared to the black line, with a more erratic curve but averaging around the same as the black line. On the other hand, on the right, the red line is just overheating compared to the black line, with no particular difference in response time.

If this happens consistently, during the hours when there is no solar radiation, that is, every night, I would be inclined to think that the sensors are misaligned.

Was an alignment check done before the tests began?

M.

I'm talking about this afternoon though, when there was in fact sunshine... Like I've said before, they're identical sensors that were calibrated at the same time. You can even see on the first graph that they were running the same temp, before the Barani started going warmer.

Ok, thanks, my fault, not easy from smartphone
Sorry

M.

[tobyportugal]

Do you have the radiation values for this period? Humidity (I know...  )
It took a long time for the dew to go off today (I was in the garden setting up the Meteorain).
 [ You are not allowed to view attachments ]

[tobyportugal]

Do you have the radiation values for this period? Humidity (I know...  )
It took a long time for the dew to go off today (I was in the garden setting up the Meteorain).
 [ You are not allowed to view attachments ]

At +/-10km
 [ You are not allowed to view attachments ]
 [ You are not allowed to view attachments ]

[Jasper3012]

Do you have the radiation values for this period? Humidity (I know...  )
It took a long time for the dew to go off today (I was in the garden setting up the Meteorain).
 [ You are not allowed to view attachments ]

Radiation or humidity? It was definitely a humid day yes, dense fog this morning and still a humid feel this afternoon. At the time of the error, RH was around 95% and it only just started to decrease from ~100% about an hour earlier. Solar radiation was 250-350 W/m².

[Jasper3012]

Radiation or humidity? It was definitely a humid day yes, dense fog this morning and still a humid feel this afternoon. At the time of the error, RH was around 95% and it only just started to decrease from ~100% about an hour earlier. Solar radiation was 250-350 W/m².

 
If you could tell us the make and model of your dataloggers and pt/probes again, it would be a lot clearer 
What recovery method do you use for your data ?
What smoothing is used to produce your graph ?
Is it a graph taken every minute, or an instantaneous graph taken every 1 minute ?

The probe is a NTC probe containing a (calibrated) thermistor 10K3A1A sensor (highest accuracy available in this series). The data is transmitted via sigfox and uploaded to the site, I can’t give you more specifics since Geert made the weather station itself and the site, he knows all about it. The first two graphs (first graph = Barani, second = SC) (https://meteo-be.net/nl/grafieken/male-brugge/last_24h) show the 10 min average, 1 min min and 1 min max for every 10 min interval. The first graph with the green/brown (Barani and SC) only shows the 10 min average, the second one only the 10 min min.

[tobyportugal]

Radiation or humidity? It was definitely a humid day yes, dense fog this morning and still a humid feel this afternoon. At the time of the error, RH was around 95% and it only just started to decrease from ~100% about an hour earlier. Solar radiation was 250-350 W/m².

This isn't a brawl! 
I find your arguments and deductions a bit flimsy.
This is a shelter comparison, and all quality shelters will give very similar Tn and Tx.
In a shelter, two parameters (for our technical and financial level) are possible: one is absent (humidity).
In the immediate environment, two parameters are possible: one is absent (radiation), the second is debatable (wind).
We may have our grievances against Barani, but I think Jan knows what he's talking about:
"it's better to have the values as close as possible to the shelters".
This type of test set-up is found in other manufacturers, and I seem to have seen it at Apogèe.
 [ You are not allowed to view attachments ]

From memory, your WeatherWise system has several channels, so why not fit an SHT?
A simple WS68 on your wooden stem will give you wind and radiation, ok it's not a pyro but the aim is to know the evolution of the radiation and not to know if it's 300w or 320w.
A simple GW1000 and you have enough to analyse with real data, a GW works with solar energy (I did the test for the Italians).
We all know that Belgium is full of water and we're in the autumn, so it's a shame to get bogged down in these habits.
For example, yesterday I realised that the FARS was going to have problems with the humidity coming from the ground, and I remember Jan's advice about a radiant shelter, so I'm probably going to 'recycle' my Rad14 into a radiant.
To minimise errors, I'll talk to Jan about determining the right height.
Have a good Sunday.

[Jasper3012]

Radiation or humidity? It was definitely a humid day yes, dense fog this morning and still a humid feel this afternoon. At the time of the error, RH was around 95% and it only just started to decrease from ~100% about an hour earlier. Solar radiation was 250-350 W/m².

This isn't a brawl! 
I find your arguments and deductions a bit flimsy.
This is a shelter comparison, and all quality shelters will give very similar Tn and Tx.
In a shelter, two parameters (for our technical and financial level) are possible: one is absent (humidity).
In the immediate environment, two parameters are possible: one is absent (radiation), the second is debatable (wind).
We may have our grievances against Barani, but I think Jan knows what he's talking about:
"it's better to have the values as close as possible to the shelters".
This type of test set-up is found in other manufacturers, and I seem to have seen it at Apogèe.
 [ You are not allowed to view attachments ]

From memory, your WeatherWise system has several channels, so why not fit an SHT?
A simple WS68 on your wooden stem will give you wind and radiation, ok it's not a pyro but the aim is to know the evolution of the radiation and not to know if it's 300w or 320w.
A simple GW1000 and you have enough to analyse with real data, a GW works with solar energy (I did the test for the Italians).
We all know that Belgium is full of water and we're in the autumn, so it's a shame to get bogged down in these habits.
For example, yesterday I realised that the FARS was going to have problems with the humidity coming from the ground, and I remember Jan's advice about a radiant shelter, so I'm probably going to 'recycle' my Rad14 into a radiant.
To minimise errors, I'll talk to Jan about determining the right height.
Have a good Sunday.

I don't want to spend my money on further wx equipment, to be honest. I could ask Geert to equip me with a T+H probe but that one has a time constant of 1 min for 1 m/s and it doesn't fit in the FARS, so that's not an option.

[Jasper3012]

  We need an absolute humidity sensor that is completely independent of a shelter and the wetting of the sensor! and at an affordable price it is far from being won 

I don't see it as necessary at all for the scope of this comparison. Quite frankly, I don't care about these little effects that have been getting mentioned in the past few days, like the effect of a bit of condensation on the sensor, a slight difference in time constant, etc. I'm happy with the setup and would much rather be discussing the results and the main reasons for these results than to get stuck in an endless loop of trying to look at and improve on every little miniscule detail that might have the slightest of influence.

[mauro63]

As for me, I will not post anything here anymore, every post is subject to attacks and considerations that have nothing to do with the post itself

We continually refer to secondary aspects, even if not in terms of importance, and we are unable to have an analysis, an opinion, an idea with respect to the pure comparison of temperature that is the pourpose of this topic

I have already said that whoever wants can open a specific post that deals with advanced thermodynamics and that addresses aspects such as time constants, wet-bulb effects, condensation supercooling etc and we continue to dirty this post with continuous messages

good continuation to all

M.

[Jasper3012]

As for me, I will not post anything here anymore, every post is subject to attacks and considerations that have nothing to do with the post itself

We continually refer to secondary aspects, even if not in terms of importance, and we are unable to have an analysis, an opinion, an idea with respect to the pure comparison of temperature that is the pourpose of this topic

I have already said that whoever wants can open a specific post that deals with advanced thermodynamics and that addresses aspects such as time constants, wet-bulb effects, condensation supercooling etc and we continue to dirty this post with continuous messages

good continuation to all

M.

Why do you constantly “threaten” to stop posting as soon as there’s a few comments that aren’t to your liking? We are all adults here and we can respectfully disagree, it doesn’t mean you have to back out immediately… I agree that there have been a number of off-topic comments and I have certainly contributed to this, but it is my intention to stop this and return to the actual topic of the thread. I’ve answered pretty much all the questions I can think of with regards to the sensor I use, the data transmission, smoothing, etc. and I’ve made it clear I don’t want to engage any further in discussions regarding secondary aspects, because I simply don’t care about them. I’ve tried to stay on topic but when people like bianconero and toby ask questions, I’m not just going to ignore them. Enough has been said about these secondary aspects now though.

[mauro63]

As for me, I will not post anything here anymore, every post is subject to attacks and considerations that have nothing to do with the post itself

We continually refer to secondary aspects, even if not in terms of importance, and we are unable to have an analysis, an opinion, an idea with respect to the pure comparison of temperature that is the pourpose of this topic

I have already said that whoever wants can open a specific post that deals with advanced thermodynamics and that addresses aspects such as time constants, wet-bulb effects, condensation supercooling etc and we continue to dirty this post with continuous messages

good continuation to all

M.

Why do you constantly “threaten” to stop posting as soon as there’s a few comments that aren’t to your liking? We are all adults here and we can respectfully disagree, it doesn’t mean you have to back out immediately… I agree that there have been a number of off-topic comments and I have certainly contributed to this, but it is my intention to stop this and return to the actual topic of the thread. I’ve answered pretty much all the questions I can think of with regards to the sensor I use, the data transmission, smoothing, etc. and I’ve made it clear I don’t want to engage any further in discussions regarding secondary aspects, because I simply don’t care about them. I’ve tried to stay on topic but when people like bianconero and toby ask questions, I’m not just going to ignore them. Enough has been said about these secondary aspects now though.

You see, it's not about what I like or what I don't like, it's simply about all the aspects that are not inherent to the post but it digresses on secondary aspects, extremely interesting and important to the point of deserving a specific post, end

When I hear this topic addressed correctly, that is, the comparison between Barani and Siap, I will be happy to intervene.

M.

[Jasper3012]

As for me, I will not post anything here anymore, every post is subject to attacks and considerations that have nothing to do with the post itself

We continually refer to secondary aspects, even if not in terms of importance, and we are unable to have an analysis, an opinion, an idea with respect to the pure comparison of temperature that is the pourpose of this topic

I have already said that whoever wants can open a specific post that deals with advanced thermodynamics and that addresses aspects such as time constants, wet-bulb effects, condensation supercooling etc and we continue to dirty this post with continuous messages

good continuation to all

M.

Why do you constantly “threaten” to stop posting as soon as there’s a few comments that aren’t to your liking? We are all adults here and we can respectfully disagree, it doesn’t mean you have to back out immediately… I agree that there have been a number of off-topic comments and I have certainly contributed to this, but it is my intention to stop this and return to the actual topic of the thread. I’ve answered pretty much all the questions I can think of with regards to the sensor I use, the data transmission, smoothing, etc. and I’ve made it clear I don’t want to engage any further in discussions regarding secondary aspects, because I simply don’t care about them. I’ve tried to stay on topic but when people like bianconero and toby ask questions, I’m not just going to ignore them. Enough has been said about these secondary aspects now though.

You see, it's not about what I like or what I don't like, it's simply about all the aspects that are not inherent to the post but it digresses on secondary aspects, extremely interesting and important to the point of deserving a specific post, end

When I hear this topic addressed correctly, that is, the comparison between Barani and Siap, I will be happy to intervene.

M.

I do agree that we need to stay on topic more, I just wanted to say you don't have to leave the thread if there is some off-topic talk. Anyway, like I've been saying, I think enough has been said about the relevant (and not so relevant) secondary aspects, so hopefully we can fully get back to the comparison now. I don't have anything particularly interesting to report at this time, but obviously I'm keeping an eye on the data.

[Kallo78]

Yesterday’s report and graphic image that also contains the modified MetPro screen with inside wh31ep without filter.
Grey days with little radiation. From today perhaps a little sun should see it.

[ypsinine]

Mauro,

About the filter on the WH32EP when used in the SmartCellino, do you recommend removing it or keep it in place?
As I understand you are running sensors both with and without a filter in your station? Any key takeaways you have learned about it?

I have been running without the filter for about three weeks now in the Barani to test and the temperature is noticably quicker to change both up and down.

I assume there is some sort of standard here or guidelines from WHO regarding response time?

[mauro63]

Mauro,

About the filter on the WH32EP when used in the SmartCellino, do you recommend removing it or keep it in place?
As I understand you are running sensors both with and without a filter in your station? Any key takeaways you have learned about it?

I have been running without the filter for about three weeks now in the Barani to test and the temperature is noticably quicker to change both up and down.

I assume there is some sort of standard here or guidelines from WHO regarding response time?

You see, the filter issue is complex and fascinating

As a general rule, if the filter is in a passive shield and we have a multi-parameter sensor, therefore temperature and humidity, it should always be used.

Filters have different types of construction, porosity and different hygroscopic characteristics, they also have a different impact on the sensor's time constant.

In the tests I am doing, I note that the sensor without any protective filter, which is conceptually incorrect, is the one that reacts fastest to temperature variations and he record the largest transients, it remains to be seen whether they correspond to reality, but that is another matter.

Big differences are noted with humidity, especially in the early morning, and after a night with air close to saturation, the sensor without any filter is the first to "dry out" and therefore the differences in humidity are appreciably higher than the instrumental tolerance.

Personally I recommend leaving the filter but, given the low quality of the one supplied with the EP sensors, to provide for its replacement no more than every 2 years, much then depends on the place where they work.

EDIT

Please excuse me, I only just read the second part of your message, yes the WMO guidelines call for a τ63 time constant of 20 seconds

M.

[mauro63]

me too !! and so your SHT35 with its filter in place returns a constant of how much according to you ???

Certainly significantly superior to the WMO indications.

In this regard, WMO itself has released studies and official statements that tend to create some confusion on this aspect.

The time constant, speaking of temperature, is not a problem in modern sensors, most of them respect it, the problem is that the sensor is inserted in a SRS and the time constant is that of the complete system, so the solar screen introduces its own time constant, much greater than that of the sensor inside it, with the result that the WMO indication is not respected.

Fortunately, WMO does not close the door, "suggesting" but not obliging to use PRT sensors of modest diameter and screens that can guarantee a certain internal flow with ventilation of 1 m/s

At an amateur level, I fear that few can boast such a time constant, on the other hand WMO itself admits that it does not particularly like the use of an active ventilation screen as a reference for other types of error in certain conditions and to always use a passive screen for comparison.

Personally, on a professional level, I would have imposed another solution, in my opinion definitely more performing.

we are off topic again, I would close this interesting aspect here

M.

[tobyportugal]

Off-topic issues are inevitable if you want to broaden your scope.
Copy more than 2 years old:

Hello Jan,
To inform you that I will remove the filter from the original Ecowitt probe (SHT35) which is in the old shelter.
If you enlarge the graph on page 4, you can see the delay in the reaction of the Ecowitt filter (old shelter SHT35).
By keeping this filter operational, on test probes, we lose part of the benefit of the tests.
Thank you for confirming or refuting my deduction.
Have a nice day.

Hello Toby,
The filter will increase the reaction time from a few seconds to a few minutes: 
https://www.baranidesign.com/faq-articles/2022/7/25/how-to-test-an-air-temperature-sensor-response-time
https://www.baranidesign.com/faq-articles/2019/5/6/difference-between-sensor-response-time-and-sensor-time-constant-tau

Value of the micro-perforations in the filters:
Hello Toby,
The filter is only important to keep the humidty sensor membrane clean, for longterm stability. The value you should look for is 20 micro-meter pore size.  For very dusty dirty environments, 12 micro-meter pore size.

Edit: To be more complete, I'm currently moving towards another SHT45 model offering filter certainty without affecting performance.

[mauro63]

My data from yesterday, unfortunately not very useful, a day characterized by high humidity and low levels of solar radiation, today seems to start a little better, we'll see 

 [ You are not allowed to view attachments ]

M.