LTB Weather Station

by RobWLakes, published

LTB Weather Station by RobWLakes Apr 4, 2018


Design Tools


Liked By

View All

Give a Shout Out

If you print this Thing and display it in public proudly give attribution by printing and displaying this tag.

Print Thing Tag

Thing Statistics

11999Views 1168Downloads Found in Engineering


Weather Stations are a very useful devices, and as we increase our knowledge of the global weather, quantifying our own local experience is also popular. Integrating aspects of the weather to IoT is increasingly important.

This project does not invent anything much new, but should give a working set of plans for a weather enthusiast to build their own station. The design is somewhat double-brick out-house in most places and has not been designed for hi-res printing with wafer thin walls. However it will produce a robust system that people may like to refine themselves. The OpenSCAD files are provided to allow people to customise the designs. The system was designed and built for an Arts/Science Exhibition, FLOAT, as a demonstration station, but has been wired up and demonstrates the usual functions. A cutaway funnel is also provided, which was used with the Rain Gauge for this exhibition.

The system also has the Arduino program included to illustrate how the the weather station can be interfaced, processed and results displayed. The Arduino program, as it stands, does not provide any definite calibration for rain and wind speed (if it is accurate, that has happened only by chance :-). Wind direction, Temperature and Barometric pressure are accurate. If you build it, and use my Arduino program, you will have to spend some time calibrating wind speed, and fine tuning rainfall.

The non-printable extras I used are listed below, and in the OpenSCAD and Arduino scripts. Plus the Arduino gear.... (I added a Barometric/Temperature sensor, but an obvious extension would be to add Humidity as well.)

Some suggestions for mounting the sensors are provided, along with glue on brackets and fittings to construct a simple "T" frame combined with aluminium tubing, however this can be adapted as circumstances dictate.


Extra hints

The thrust bearings should be a tight fit and not require glue. The 5mm brass tube for the axles though will benefit from some cyanoacrylate on the ABS to hold them in place. Rough the tube up a bit with sandpaper or a file to help adhesion. The Temperature and Barometric pressure does not need calibrating. However Rainfall (it is fairly close) and wind speed will need calibration. As long as the magnet in the wind direction sensor is close enough to trigger two adjacent reed switches when half way between the two reeds, it will allow 8 reed switches to reliably indicate 16 directions.

The reed switches in the direction indicator are vertical and are not trimmed, just the top end curled over to allow easy soldering to the common earth wire ring. Extra spacing maybe required, eg a small ring of heat shrink tubing to keep the moving parts of the anemometer and wind speed separated and seated on the bearings in the stationary base. This was too fine to print.

All the magnets N-S poles should be aligned along the line of the reed switch. The magnet lines of force between N-S have the best switching effect, not one of the poles, N or S, on its own. This also helps eliminate bounce, or multiple triggering.

Non-printable Parts List

Arduino Uno Board + Prototyping Shield
10 of Magnetic Reed Switches, glass 14mm x 2mm
4 Thrust Bearings 8mm OD x2.5mm by 5 mm ID
3 Neodymium magnets 3mmOD x 2mm
Scrap brass tube OD 5mm (eg Car Radio Antenna)
Hook up wire
BMP085 Barometer/Temperature breakout (I2C)
PCF8574 I2C to LCD interface board
4x20 characters HD44780-compatible LCD
2mm steel axle for Tipping Bucket flip flop
Plastic tubing for spaces eg Heat shrink


The Arduino system as described has a major shortcoming, it is assumed the sensors will be close to the CPU and there will be no long runs of cable. If longer cables are contemplated then larger pull-up resistors than the internally programmed pull-ups maybe required on all the reed switches. Very long cables may not be possible or recommended.

Mounting the computer close to the sensors and building a radio link maybe the best alternative, and using some other board than a UNO with at least 10 pins I/O free and some Wi-Fi could give a good combination. eg

Battery power poses even more challenges, but a larger battery charged by a solar panel is a viable option these days. Designing ultra low power electronic gear is a specialist area.

Printing times with 0.4mm nozzle, 30% fill and Support

3 Anemometer Cups and arms 1h21m (x3) = 4h3m
Anemometer Arm Hub 3h31m
Anemometer base 2h
Anemometer Bracket 16m
Wind Direction Vane 4h21m
Wind Direction Base 3h30m
Wind Direction Bracket 16m
Tipping Buckets 1h43m
Rain Gauge base 6h18m
Rain Gauge Funnel (full) 12h30m (Invert to print to avoid Support printing)

Estimated total 35 hours of Printing

More from Engineering

view more

All Apps

Auto-magically prepare your 3D models for 3D printing. A cloud based 3D models Preparing and Healing solution for 3D Printing, MakePrintable provides features for model repairing, wall thickness...

App Info Launch App

Kiri:Moto is an integrated cloud-based slicer and tool-path generator for 3D Printing, CAM / CNC and Laser cutting. *** 3D printing mode provides model slicing and GCode output using built-in...

App Info Launch App
KiriMoto Thing App

With 3D Slash, you can edit 3d models like a stonecutter. A unique interface: as fun as a building game! The perfect tool for non-designers and children to create in 3D.

App Info Launch App

Hi Rob
Thanks for the additional infos about printing. One of my colleagues would like to build this station but has no 3d printer. Perhaps
I will try to print it for him, starting with the wind direction vane. I am not sure whether it would be better to add a helping break away part instead of slicer generated support. I am used to openscad, so this would not be a problem. will see (perhaps)...

I found the auto support generated by the Robox Automaker slicer worked quite well, but not all slicers are equal, and some work better on some things and not others. It is a linear type of support on either side, and constant width, apart from the very tip of the pointer, so it should build the support reliably on most slicers. I will try to add a photo I reconstructed from what I could find of the original support material to give you an idea of how it went. Great to hear of an actual build. Brilliant!

Hi FunLab,
Thank you for interest in my project. I have probably made an error in not exporting all my .stl files in the preferred position to print. For example I print the Rain Gauge funnel upside down and build it up from the outermost top section first. My slicing software allows me to rotate the object before printing. My printer will do the funnel without support, however even then it takes 12 hours! I think if if the sides were thinner it would print quicker, but then may not print without support. The Wind Direction vane does need support and looks quite strange ot begin with, but prints quite well. If I can find the old supports, I will publish a photo. I take support printing now as just part of the deal, it works so I use it. I try to avoid it, but not at the expense of not doing anything at all.

I have used 0.4 nozzle and the quality of the print is quite good. On my printer (CEL Robox) the flat areas print up quite nicely. I have deliberately made the thickness of most of the parts more on the excess than too thin. This should make it accessible to a wide range of printers. The only really thin part is the tipping bucket mechanism as it needs to be light for the weight of the water to work.

Your estimate of time to print is pretty good, I estimate approximately 35 hours in all. I have taken your concern and published the printing times on the Thingiverse site to help others make up their mind. I suspect many will only print one of the three sensors, rather than do all three. Ultimately the OpenSCAD files are there for people to optimise their own designs. The overall concepts are not rocket science, so even just designing one from scratch in another CAD system should not be too hard.

This is a very nicely designed thing but I do not see how some of the parts would print nicely with a single extruder FDM type printer. Many parts need massive support structures or must be cut in half and turned around or need custom support parts which break away later to be able to be printed. Also there are many flat roofs which are also problematic to print nicely. To me it seems, that the parts are made for printing with an SLA type resin printer. How should the wind director be printed for example? Not sure if it will print ok with support structures....
Also I am missing some estimate of how long printing of all required parts may take. Of course it depends on the printer but I assume it to take something between 20 and 25 hours.
I think this thing is missing important printing guidelines/hints/tips how to make one...