Topic: dramatically improve the davis 7714 solar shield

[Buick]

Good to know about the DC fan setup. Sounds like I'll be leaving the FARS stock.

[JCA433]

Buick, I provided some links to some of the mounting hardware I am using for battery and solar panel for mounting on a pole.

https://www.amazon.com/gp/product/B076666P81/ref=ppx_yo_dt_b_asin_title_o06_s00?ie=UTF8&psc=1
https://www.amazon.com/gp/product/B07Y49LSMN/ref=ppx_yo_dt_b_asin_title_o03_s00?ie=UTF8&th=1


Solar panel charge controller








http://www.amazon.com/gp/product/B0B59TMGT9/ref=ppx_yo_dt_b_asin_title_o05_s00?ie=UTF8&th=1

[CW2274]

A common misconception is that an aspirated shield is meant for hotter climates. Not true. Although I'm sure it can make somewhat of a difference, the main reason is to negate lack of wind, not heat. 

[Buick]

It does seem like the amount of airflow required to mitigate the effects of solar radiation would increase with increasing temperature though...

[CW2274]

In this instance, I prefer "insolation" over "radiation", but agreed. Like I said, it does help, just not what the primary duty is. BTW, if you're unaware, I'm in Tucson. 

[CW7491]

In the PNW, for at least 7 mos of the year, if it’s not the constant cloud cover and showers, then it’s the extremely low sun angle and the minimal dewpoint depression that make the benefits of aspiration much less impactful. Wind or no wind there just isn’t much radiation, either incoming or outgoing, to deal with.

[CW2274]

I can't make this any clearer...I get that. That said, if you look at every single ASOS, AWOS across the country, regardless of latitude, they all have fans. Just sayin'.

[CW7491]

No denying it

[CW7491]

I've given it a bit of thought and want to attempt a shield intercomparison with the types discussed in this thread and others prior like many others have done.

Been thinking about how I should go about it. Once I have a tower up *somewhere* I'd like to take the 7714 (modified/unmodified) and compare it to the Young 41003 with similar modifications. Given newfound understanding of the 43408 FARS and the general issues the open intake design encounters, I may consider acquiring a 076B as a reference. The matte black FARS intake is also interesting to me despite the re-radiation issues. I may try to mod the 43408.

I'd probably use fine-wire thermistors for each shield and record with a data logger. Given sensors with filters have problems with gradient dampening the need for fast response sensors is required. Probes will be calibrated before and after their deployment. Since this experiment would be over a significant time span with little if any need for human intervention, remote connectivity would not be necessary.

Setup would probably have the tower at center with two opposite extending tubes (10ft?) from either direction with shields on either side. Don't know if it should be grouped as a control group (unmodified) and the test group (matte black interior) or same shield make/model together on either side (7714, 7714 / 41003, 41003). Regardless of what's decided, the reference shield would remain at the center offset from the tower by a few feet. All measurements taken at 2 meters over grass and generally responsibly sited. Wind speed/direction will be measured using a 3D sonic anemometer. Solar radiation measured with pyranometer.


Would it be helpful to employ the Davis FARS as well? Should I compare that European helical shield as well?

Anything else I missed? Anything to add?


Cheers

I’d really like to see the Davis FARS with a higher CFM fan compared to the MetOne 076b. That seems to be the best shield available. I wish the NWS had gone with it instead of the RM Young for the new ASOS shield. The HO-1088 has some significant design flaws and the RM Young has been shown to have some weaknesses as well. I’m guessing it came down to cost

[Buick]

Be it radiation, insolation, irradiance, or whatever you prefer, like CW7491 said: the PNW/land of perpetual gloom doesn't get all that much of it - I start increasing my vitamin D intake in early October. I’m quite satisfied with the stock FARS and my temperature measurements are generally very close to what my local ASOS reports. I also agree that an aspirated shield is essential if accuracy is important. Even though the 7714 is pretty impressive for a passive shield, it suffers from unacceptably large errors in daily maximum temperature during clear, calm, hot and sunny weather.

[JCA433]

One important issue with a DC fan setup for FARS is the siting for the tipping bucket rain gauge.  If the solar panel is mounted directly above the FARS, the rain gauge cannot be mounted above the FARS on the same mounting pole.  This requires an additional transmitter ,indoor receiver./console, and mounting pole.

[TheBushPilot]

I’d really like to see the Davis FARS with a higher CFM fan compared to the MetOne 076b. That seems to be the best shield available. I wish the NWS had gone with it instead of the RM Young for the new ASOS shield. The HO-1088 has some significant design flaws and the RM Young has been shown to have some weaknesses as well. I’m guessing it came down to cost

CW7491:

I would not be opposed to trying higher CFM in the Davis shield. Perhaps comparing stock versus modified fan as well? I do recall a study noting Davis FARS was actually cooler than the reference by an unacceptable margin. Not sure if higher flow would alleviate that phenomenon or not (or exaggerate it).

Yea 076B is probably one of the best and a shame the weather service did not go with it. If I had to guess it was because their probe of choice to replace the HMT337/DTS-1 wouldn't fit. HMP155 is really long, Young already makes a Gill shield for it and the separate add-on PRT for "fast response" temperature measurements easily installs in their FARS. Interchangeability is also way easier with the commercial probes than the HMT. HMT337 warmed probe with separate PRT are hard soldered to the board. Another consideration they may have made was cost/performance. Since ASOS stations are not climate stations, the need for extreme accuracy is simply not needed. That being said my friend who works for NYS Mesonet notes their HMP155s fail at a decently considerable rate due to repeated exposure to high humidity. 0-90% RH is typically fine (±1.0%) but above that drops off to anywhere from ±5% to ±7%. And ultimately leads to premature failure by calibrated specifications.


Cheers

[CW7491]

CW7491:

I would not be opposed to trying higher CFM in the Davis shield. Perhaps comparing stock versus modified fan as well? I do recall a study noting Davis FARS was actually cooler than the reference by an unacceptable margin. Not sure if higher flow would alleviate that phenomenon or not (or exaggerate it).

Yea 076B is probably one of the best and a shame the weather service did not go with it. If I had to guess it was because their probe of choice to replace the HMT337/DTS-1 wouldn't fit. HMP155 is really long, Young already makes a Gill shield for it and the separate add-on PRT for "fast response" temperature measurements easily installs in their FARS. Interchangeability is also way easier with the commercial probes than the HMT. HMT337 warmed probe with separate PRT are hard soldered to the board. Another consideration they may have made was cost/performance. Since ASOS stations are not climate stations, the need for extreme accuracy is simply not needed. That being said my friend who works for NYS Mesonet notes their HMP155s fail at a decently considerable rate due to repeated exposure to high humidity. 0-90% RH is typically fine (±1.0%) but above that drops off to anywhere from ±5% to ±7%. And ultimately leads to premature failure by calibrated specifications.


Cheers

Interesting insight on the HMP155 and RM Young. If you do get your hands on a 076B please let me know. I’d love to see inside design of the sensor chamber.

Not to completely hijack the thread, but as I think I’ve mentioned to you before about the HO-1088, I just can’t believe how the sensor is at the very bottom opening of the shield and completely exposed to the ground. The RM Young has a similar open exposure although much less extreme. In my designs, leaving the sensor exposed at the intake like that does not work well. Even the Apogee shield does the same. I wrote them to ask about it and they claim the exposure to reflected radiation from below is negligible. I think this has been posted before, but here is a link to a sample of their data comparing the Apogee to the 076B and the RM Young. Not surprising which they chose as the reference …

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

[JCA433]

CW7491,  One thing I really like about the DAVIS FARS is reflected sunlight from the bottom is blocked from entering the shield.  The bottom is not open and exposed like other FARS.   Maybe this is why Davis can use such a low powered fan and still get good performance for a consumer weather station.  I bet some of the other FARS with open bottoms would perform much worse if they used the Davis stock fan.

The RM Young shield has good performance despite the open bottom and has very strong fan.   The RM Young FARS may suffer from wet bulb effect in thunderstorms as rain is pulled into the FARS by the strong fan.

[CW7491]

CW7491,  One thing I really like about the DAVIS FARS is reflected sunlight from the bottom is blocked from entering the shield.  The bottom is not open and exposed like other FARS.   Maybe this is why Davis can use such a low powered fan and still get good performance for a consumer weather station.  I bet some of the other FARS with open bottoms would perform much worse if they used the Davis stock fan.

The RM Young shield has good performance despite the open bottom and has very strong fan.   The RM Young FARS may suffer from wet bulb effect in thunderstorms as rain is pulled into the FARS by the strong fan.

Yes, I use the Davis FARS modified with a fan and it really is a great design

[CW2274]

I’d really like to see the Davis FARS with a higher CFM fan compared to the MetOne 076b. That seems to be the best shield available. I wish the NWS had gone with it instead of the RM Young for the new ASOS shield. The HO-1088 has some significant design flaws and the RM Young has been shown to have some weaknesses as well. I’m guessing it came down to cost

Since ASOS stations are not climate stations, the need for extreme accuracy is simply not needed.

What is a "climate station"?

[CW2274]

   The RM Young FARS may suffer from wet bulb effect in thunderstorms as rain is pulled into the FARS by the strong fan.

I still have a hard time conceptualizing this. Do you not think that they thought of such a thing? I mean, this is RMY, one of the premier whether instrument manufactures in the world.

[TheBushPilot]

Interesting insight on the HMP155 and RM Young. If you do get your hands on a 076B please let me know. I’d love to see inside design of the sensor chamber.

Not to completely hijack the thread, but as I think I’ve mentioned to you before about the HO-1088, I just can’t believe how the sensor is at the very bottom opening of the shield and completely exposed to the ground. The RM Young has a similar open exposure although much less extreme. In my designs, leaving the sensor exposed at the intake like that does not work well. Even the Apogee shield does the same. I wrote them to ask about it and they claim the exposure to reflected radiation from below is negligible. I think this has been posted before, but here is a link to a sample of their data comparing the Apogee to the 076B and the RM Young. Not surprising which they chose as the reference …

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

Haha yes to no ones surprise the reference shield of choice is the 076B. I will have to document the shield when I inevitably acquire one.

We have to remember that with the HO-1088, it was originally designed as a dew point sensor. Not temperature. That is until it was phased out in favor for its more reliable capacitive chip counterpart. And now it's used purely for temperature if I'm not mistaken. But like you mention it's wide open at the bottom with the little sensor very near the opening. The shields assume siting over grass, but do not take into account transient reflective surfaces like snow or ice which introduce the re-radiation error. You can clearly see that with the graphs on Apogee's website Young versus their own shield. If I recall correctly Met One actually made a more similar design to the Young and Apogee shield but had a small plate at the intake - I will try and look to see if I can find the document.

And as JCA433 mentions with ingesting precipitation, it is more common than originally thought. I reached out to one of the gentleman at NSSL who work on the mobile mesonets and inquired about rain ingestion. During his capstone project developing the U-Tube which replaced the J-Tube, a mobile application FARS. He ran anecdotal tests to determine if the Oklahoma Mesonet experienced similar issues of sensor wetting in high rain/wind events with the Young 43408 because they were constantly running into wet bulb error when driving in and out of thunderstorms. TL;DR prior to the new shield the combined use of the J-Tube and 43408 collocated on top of the vehicle worked "well enough" but significant errors were still encountered with rain and reflected radiation.

What is a "climate station"?

I'd consider it any station that rigorously follows AASC and or WMO siting standards. i.e. Kentucky, Oklahoma, Kansas, NYS, etc. Mesonets. Unlike ASOS which can have questionable at best siting, not suitable for area climate representation. It's kind of like accuracy versus precision. ASOS is extremely precise with its measurements but lacks areal accuracy.

   The RM Young FARS may suffer from wet bulb effect in thunderstorms as rain is pulled into the FARS by the strong fan.

I still have a hard time conceptualizing this. Do you not think that they thought of such a thing? I mean, this is RMY, one of the premier whether instrument manufactures in the world.

You'd think, and it is a little ironic that the best of the best still have issues with seemingly no solution. I'm sure they have thought about it but it is such an infrequent phenomenon that it's a *shrug*. The Young FARS have a notoriously strong intake velocity and as such fine mist particles will be ingested. That's why understanding the performance and limitation of each instrument in use is so important. Seemingly normal data may actually turn out to be absolutely wrong. Par for the course, mother nature don't care.


Cheers

[JCA433]

CW, this problem could be prevented or greatly reduced by slowing down the fan speed during thunderstorms. 

[CW2274]

Thank you both for your replies. Seems like someone would have solved this "issue" ages ago. I mean, it ain't exactly rocket surgery... 

[JCA433]

I remember years ago when I taped the Ambient Weather temperature sensors to the bottom of the 7714 shield and despite shading from direct sun light, they still read temperatures about 6F above the temperature reading of the sensor inside the 7714.  The errors from reflected sun light can be very large even over grass.

[CW7491]

I’d really like to see the Davis FARS with a higher CFM fan compared to the MetOne 076b. That seems to be the best shield available. I wish the NWS had gone with it instead of the RM Young for the new ASOS shield. The HO-1088 has some significant design flaws and the RM Young has been shown to have some weaknesses as well. I’m guessing it came down to cost

Since ASOS stations are not climate stations, the need for extreme accuracy is simply not needed.

What is a "climate station"?

In short, USCRN is the national climate station. It’s NOAA’s gold standard climate station.
https://www.ncei.noaa.gov/access/crn/overview.html
Instrumentation and photos for every site are available at this link.

You have one nearby in Tucson Mountain Park at the Sonora Desert Museum.
https://www.weather.gov/wrh/timeseries?site=SDMA3&hours=72

[CW2274]

I’d really like to see the Davis FARS with a higher CFM fan compared to the MetOne 076b. That seems to be the best shield available. I wish the NWS had gone with it instead of the RM Young for the new ASOS shield. The HO-1088 has some significant design flaws and the RM Young has been shown to have some weaknesses as well. I’m guessing it came down to cost

Since ASOS stations are not climate stations, the need for extreme accuracy is simply not needed.

What is a "climate station"?

In short, USCRN is the national climate station. It’s NOAA’s gold standard climate station.
https://www.ncei.noaa.gov/access/crn/overview.html
Instrumentation and photos for every site are available at this link.

You have one nearby in Tucson Mountain Park at the Sonora Desert Museum.
https://www.weather.gov/wrh/timeseries?site=SDMA3&hours=72

Well how 'bout that! I never knew I had weather measuring royalty directly under my obs! Seen it in person many times. Thanks!

https://www.weather.gov/wrh/LocalWeather?zone=AZZ504

[JCA433]

I have some temperature data to share comparing the Davis 7714 shield with insides painted flat black to the Davis 24 hour FARS for the period AUG 10,2023 to AUG 31, 2023.   The temperature and humidity sensors in both the 7714 shield and the FARS are the Davis SHT31.  The average wind speed is measured with the Davis  Vantage PRO 2 anemometer with wind cups and wind vane with the wind cups about 88 inches above the ground.  The temperature and humidity sensors are mounted 70 inches above ground and over grass. Wind speed is measured in Knots.  I am using a 12V DC fan (44cfm)  in the Davis FARS.


Date                 MAX7714       MAXFARS     Difference      Average wind speed
8/10/2023          95.7F            95.4F           -0.3                   1.7K
8/11/2023          91.5F            90.6F           -0.9                   0.2K
8/12/2023          92.9F            92.7F           -0.2                   2.0K 
8/13/2023          91.9F            92.2F           +0.3                  3.0K
8/14/2023          93.6F            94.0F           +0.4                  2.8K
8/15/2023          93.5F            92.5F           -1.0                   0.3K
8/16/2023          90.0F            90.2F           +0.2                  2.8K
8/17/2023          93.0F            93.2F           +0.2                  3.0K
8/18/2023          89.2F            88.0F           -1.2                   0.4K
8/19/2023          90.7F            90.7F             0                     2.2K
8/20/2023          88.2F            88.6F           +0.4                  4.8K
8/21/2023          89.4F            89.6F           +0.2                  3.2K
8/22/2023          90.8F            90.5F           -0.3                   1.6K
8/23/2023          90.6F            90.6F              0                    2.5K
8/24/2023          92.3F            90.3F            -2.0                  0.7K
8/25/2023          91.3F            91.0F            -0.3                  0.6K
8/26/2023          92.2F            92.5F           +0.3                  1.8K
8/27/2023          91.5F            90.0F           -1.5                   0.7K
8/28/2023          91.9F            91.9F              0                    3.2K
8/29/2023          91.3F            91.8F            +0.5                 5.9K
8/30/2023          90.6F            90.7F            +0.1                 5.2K
8/31/2023          92.3F            92.6F            +0.3                 3.1K




 Overall, the performance of the 7714 shield is good with largest differences occurring when average wind speed is less than 1Knot.  The largest difference -2.0 occurred on August 24, 2023 because the maximum temperature time was early 11:58 AM . The maximum temperature time for the other dates with average wind speed less than 1Knots : AUG 27 11:27AM; AUG 25 12:26PM ;AUG 18 9:57AM; AUG 15 11:43AM;AUG 11 11:49AM.  The average difference with wind speeds less than 1 knots   is -1.2 F, not bad for a passive shield.

[JCA433]

Here is the average temperature data between 10AM and 3PM for the period AUG 11, 2023 to AUG 31, 2023.



Date             AVG7714         AVGFARS      Difference    Average wind speed
8/11/2023      87.7F               86.4F           -1.3               0.4K
8/12/2023      90.7F               90.0F           -0.7               1.5K
8/13.2023      89.3F               89.4F           +0.1              2.9K
8/14/2023      91.6F               91.4F           -0.2               2.7K
8/15/2023      86.9F               85.3F           -1.6               0.5K
8/16/2023      87.9F               88.0F           +0.1              3.0K
8/17/2023      90.6F               90.6F              0                2.5K
8/18/2023      82.1F               81.5F           -0.6               0.8K
8/19/2023      88.5F               88.4F           -0.1               2.4K
8/21/2023      85.7F               85.9F           +0.2              2.8K
8/22/2023      88.4F               88.0F           -0.4              1.5K
8/23/2023      88.3F               87.6F           -0.7              1.7K
8/24/2023      89.4F               88.3F           -1.1               1.2K
8/25/2023      88.8F               87.5F           -1.3               0.9K
8/26/2023      89.7F               88.7F           -1.0               1.4K
8/27/2023      87.0F               86.1F          -0.9               1.2K
8/28/2023      88.9F               89.0F          +0.1               3.0K
8/29/2023      89.4F               89.6F          +0.2               5.9K
8/30/2023      88.7F               88.9F          +0.2               5.9K
8/31/2023      90.1F               90.2F          +0.1               2.9K

The average difference for the period is only 0.44F above the FARS average.  This is excellent relative performance for a passive shield considering low average wind speed and very intense sunlight.     The average temperature is very close to the FARS when the average wind speed is 2K and higher.