Author Topic: Anemometer/Windvane/AirTemp implemtation using off-the-shelf hardware?  (Read 542 times)

0 Members and 1 Guest are viewing this topic.

Offline PeteCress

  • Member
  • *
  • Posts: 2
    • Extreme Surfcams
I am seeing plenty of Pi weather station implementations, but all with DIY hardware - which I do not want to get involved with.

The application is a windsurfing shop on a bay that wants to overlay it's surfcam pages (ExtremeSurfCam.DynDns.org) with current wind/airtemp data.

I have done this in the past using a screen scraper to get info from various weather sites on the web, write it to .Txt files, and then put it up on the cam pages.

But site-after-site has changed so that the data is hidden behind .php calls instead of being available in the HTML.

Seems like a good Pi application to me - especially since we have a Pi already in the shop performing another function.

Bottom Line: 
Does anybody have any anemometer/windvane/temp-sensing hardware that they are happy with?    I have a couple of temp sensors hooked up to a Pi in my garden shed that I am not particularly impressed with accuracy-wise.... and I have no clue at all about the wind stuff.

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 1154
    • Weather at Eau Claire, WI
Pete, I don't have a solution to suggest, but a couple comments.
First, when you find a solution that you like, will you please come back to post a summary or some links to the resources you used to solve your problem? I, and I'm sure many others, would love to know how you married the standard sensors to the Pi type input, and how you displayed your information in a way your software can then overlay it on the web site, which by the way is very nice.  Far too many interests to pursuit but fun vicariously enjoying the sport. 

Second, I'm not sure how you can get a standard solution without some degree of DIY involved.  I have used an RM Young Translator 26700 to take in wind direction, speed, and a very hard to find (on the eBay market but easy if you have lots of bucks to throw at the project) temperature humidity sensor that attaches to plug in cards.  The output can be read by Weather Display with one of it's options.  The problem is that these translators, as slick as they are, are also become quite rare on the eBay market, with one now, but it is sold as is, untested and unsure if it will even power up.  I have a couple that work, a couple that don't and I am trying to find out how to use them for parts, and the plug in cards haven't appeared on the scene for a year or two.  I can get an RM Young wind sensor, temp/ humidity and any tipping bucket rain gauge on the air in about a half hour if the sensors are installed.  Just wire up, do some keyboard programming (which takes the longest) and voila you are cooking.  Young's most recent translator will do the job without looking for cards since it incorporates all the stuff mentioned, but I've never seen a used one on eBay.

A Campbell Scientific data logger can do the trick too, with as many or more sensor choices.  This is what many remote monitoring sites, such as mountain or buoy activity may use, with cellphone or satellite telemetry back to the University or NWS or Park Service to monitor their output.  Again the trick here is the programming to record the sensors and how often, but it is pretty straightforward.  Being research and commercial level, the won't be cheap unless you can find a good deal on the used listings.  I would guess that they are pretty hardy, though.  I have a decommissioned buoy setup with an RM Young 5103, temp/humidity sensor and a non-tipping bucket rain gauge and a LiCor solar radiation sensor that I got a few years ago that fed a Campbell Scientific datalogger, but have used some of the sensors in other projects and haven't gotten into learning the programming of the datalogger.  (work keeps getting in the way of hobby time...) 

These all work, but to take a RaspPi, and have it interface is something that I think a lot of folks have wanted, but no straightforward solution has come up.  There was a link in the arduino forum DIY (which your really didn't want )
http://makezine.com/2015/11/20/build-your-own-arduino-weather-station/

This series uses a readily available and low cost Davis wind sensor to accomplish part of what you need,
http://cactus.io/projects/weather/arduino-weather-station

with links to the other sensors and the logic and programming sketches with library links.  At least the wind is a standard rugged component, and forget the rain sensor.  It basically is, from all reports, only good to tell if it is raining or not, but can't really tell inches very well.

The one very last thing I'll mention, hoping to give you some ideas, is the Vaisala system for yachts, which has 3 ultrasonic wind direction/speed sensors, a chrome dome to reliably measure rain, a solar radiation shield to house temperature and humidity and barometer, and some models even have a gps in them.  These are fully integrated and send down a NEMA data stream, but the units are expensive new, and rare used.  Being designed and built by Vaisala, they are again government and research quality but would have the sensor suite all bundled up, working and all you'd need to do is decode the RS232, or NEMA data stream and display.

Somehow I think that all I have mentioned is either bundled/works/expensive or is leaving you with some interface work to do.

Let us know what other options you discover and what you come up with.
Sounds like fun, though. Dale
ECWx.info
&
ECWx.info/wx/index.php

Offline Bushman

  • Forecaster
  • *****
  • Posts: 6496
    • Eagle Bay Weather
I have to ask why the shop does not simply buy a Davis unit?  With a Meteobridge you could everything wanted for ~$500 USD (max).  That said, a lot of surf places use the Inspeed wind devices; easy to hook to the GPIO of a Raspberry Pi (or Pi ZeroW).  Add a high res sensor like a DS18B20 and for a couple hundred and some work you are done.

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 1154
    • Weather at Eau Claire, WI
Thanks, Bushman.  I knew there was more than one way to skin a cat and there are a lot of ways to go about it.  I was hamstrung by the assumption that his assertion of using RaspPi was a must.
ECWx.info
&
ECWx.info/wx/index.php

Offline PeteCress

  • Member
  • *
  • Posts: 2
    • Extreme Surfcams
...his assertion of using RaspPi was a must.
Not a must... just a nice-to-have, since we already have a little experience with the Pi, there is already one in the shop, and I figured that the cost of a anemometer/vane would be modest compared to a turnkey solution.

Offline Bushman

  • Forecaster
  • *****
  • Posts: 6496
    • Eagle Bay Weather
Like I said - doable with not a ton of work and about half the cost of a Vantage Vue etc.  Could be less if you went with an Acurite unit.