I know it's been four years since anyone's responded here but I figured I'd give my two cents. If nothing else someone looking for information will find this and hopefully get something out of it.
I am currently constructing a mobile mesonet to research a variety of weather phenomena, but the main purpose is to gather finescale surface observations of drylines. I am basing my design in large part from the information in (Straka et al. 1996). My instruments include:
-RM Young 05103v anemometer
-RM Young 41382 temperature/humidity sensor
-Vaisala PTB 110/CS106 barometric pressure sensor
-Flux Gate Compass M81190
-RM Young pressure port 61002
I do have some questions regarding the construction of the actual structure that would support the instruments. Since the 1990s the concept of mobile mesonet design has remained consistent, but most that are seen have variations in the execution. I was planning on building the structure quite similar to that in (Straka et al. 1996). Are there any issues with such a set up or things I should be aware of?
When it comes to shielding the temperature and humidity sensors, nearly every mobile mesonet has utilized the J (or sometimes called the S) tube design. However, I did read that in 2009 the NSSL began incorporating a U tube design, claiming it to provide better aspiration thus more accurate data. Should I stick to the J tube design or use the U tube? Also, how necessary is it to use a fan to further improve aspiration?
I will be including a GPS unit, but I do not know which unit would be best suited to a task such as this. Any recommendations would be appreciated. I am also unsure of which data logger I will use, again recommendations would be appreciated.
Is there anything else I should keep in mind, alter or be aware of? Any feedback is valued.
Evan: I'm not sure if you finished the rack or are still working on it after all this time but I wish you the best of luck. Something you might need to consider with the older design is the lack of futureproofing. The rack style by todays standards is somewhat outdated and unable to accommodate newer sensors and housings. One of the most notable changes of today is the addition of the U-Tube. This requires a different mounting style than the rack discussed in the publication mentioned prior. That being said, basing your design off of the newer edition of NSSL racks would be more than capable to house any suite of sensors. The older design might not hold up well to the newer suggested siting of pressure ports either as their placement is thought to be best situated farther up front that the original style.
The implementation of the U-Tube was important as it pretty much blew the efficacy of the J-Tube out of the water. Solving all of the problems previously experienced while also maintaining robust construction and better aspiration. I'll talk more about why below. Having a fan is important and ultimately unavoidable because you will not always experience a breeze or driving to move the air through the housing. I'll link the PDF to Waugh's publication below where you can read through for yourself or anyone interested. There are many other things to be discussed but that could probably be made into another thread.
From personal experience and looking around at what others are doing, Garmin makes the GPS-16X-HVS sold by Campbell Scientific. This receiver directly interfaces with CS data loggers and records the slew of GPS data available. The Campbell Scientific CR6 is the logger of choice for the project I am apart of and what the NSSL uses. The only caveat to this is when you are stationary GPS heading gets buggy, hence the supplemental use of a fluxgate compass. The inputs on the logger are all digitally modular so you can configure them to virtually anything of your liking. There's also a cellular and wifi variant as well if you do not want to hard-wire a connection to collect data or monitor in real time. CS also makes a ready to go software suite called LOGGERNET that does things I am still finding out and works flawlessly with their data loggers.
Shawn: The J-Tube was the original mobile vehicle application fan aspirated radiation shield. The namesake came from it's creator Jerry Straka working with the original VORTEX project. As you might've assumed already, "J" from the name came from Jerry and it stuck. It was supposed to utilize the slipstream effect of the vehicle and create an aspiration effect as air was sucked from the top to bottom of the tube due to the pressure gradient. It was made of standard schedule 40 3" pvc and worked decently in low sun environments. Though through further investigation and efforts made by Sean Waugh he determined the J-Tube was actually quite inaccurate and lead a host of problems while taking observations in the real world. The cap at the front becoming a large thermal mass, etc. etc. The cap part at the front of the tube, as Dean stated prior, is a vent cap. I did some searching around and as of right now they no longer make the type NSSL supposedly used in the images you posted above. Another possibility is they made a custom cap that you cannot commercially buy. This wouldn't be surprising as most of what they put out in the field is in house fabricated.
The U-Tube is now the preferred apparatus to housing T/RH sensors as of about a decade ago. Again, like I stated above, a link to Sean's document can be found
here.
And back to Evan: Things you should be aware of. A big one is maintenance and cost of calibration. These are two big ones if you want to maintain reputable data collection and operation. MPG is a big one as well. For a brief period I had a rack atop my vehicle for testing purposes. Needless to say my fuel efficiency went to garbage. Just something else to expect. They are a
lot of work, albeit worth it in the end, but still a
lot of work and much research sunk into them. The anemometer will break in hail inevitably as well as any non-robust plastic parts. Keeping spare replacements of the parts you know are susceptible will save you downtime and headache. Avoiding hail all together is also a known and effective strategy. I could go on for hours about the endless things required so I'll end it here.
I'll probably make a separate thread for anyone who wants to go into mobile observations simply because there is a lot to cover. I'm probably forgetting things in this one but I'll more than likely get to them in a separate post.
Anyways I hope this post finds someone that needs the info.
Cheers