Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5

Near Real-Time Windspeed Tests. Part 3 - DIY Anemometer Build

Following on from Part 2 (adding 4 Hall sensors to Fine Offset anemometer) I also wanted to build an anemometer from scratch for near real time wind speed monitoring, cheapness being the objective here. The attached photos are worth a thousand words, so the saying goes, but I'll try to summarize it too. The main constituents are:

1 x section of 40mm diameter drainpipe
1 x 635zz stainless steel ball bearing ( OD = 19mm, ID = 5mm, thickness = 5mm)
1 x length of M5 stainless steel studding
1 x OPL550A  light sensor
1 x red LED
3 x plastic Easter eggs
assorted pieces of 1.6mm thick printed circuit board (PCB) material with copper layer removed
assorted M3 fixings (screws, nuts, washers)

One of the shields on the 635zz bearing is prised off using a small headed screwdriver, its internal grease removed with IPA/Acetone, and replaced with a drop or two of light oil. The bearing shield is refitted, and the bearing itself is located into a piece of plastic (bearing housing) containing an 11mm diameter through hole and a 19mm diameter blind hole (matching the outside diameter of the bearing) overlaid over the 11mm hole. The bearing is kept in place with a piece of PCB material. A small length of M5 studding (acting as a spindle) is attached to the bearing with 2 nuts. On the bottom of the studding an optical chopping disk containing 8 slots and made from 1mm thick plastic is fitted. The bearing housing has 3 x M3 tapped holes around its circumference to allow it to be screwed to the plastic drainpipe. It has further holes in it to mount a small piece of veroboard to it that has an OPL550A light sensor and red LED soldered to it on either side of the optical chopping disk. The OPL550A has built in hysteresis and produces nice clean logic signals with my 30 metre long cable run. There is a 2W heater made from power resistors also on the veroboard that switches on at 2 degrees C, and off again at 4 degrees C. The heater is designed to warm the bearing in cold conditions although what effect it actually has on anemometer performance in these conditions is unknown - it may well do nothing at all.

On top of the M5 studding, and sandwiched between 2 circular pieces of PCB material for additional rigidity is attached a can cap to act as a rain shield. A 2mm thick bead of silicone sealant around the top circumference of the drain pipe, together with another bead around the inside of the can cap act as additional rain baffles. Attached to the top of the can cap are 3 lengths of PCB material to which the bottom of 70mm diameter plastic Easter eggs are attached. For further rigidity the middle of the Easter eggs are attached higher up on the 5mm studding spindle. Silicone sealant fills in any remaining gaps where water may penetrate. Mechanically the Easter eggs attachment to the studding is not ideal - top heavy so to speak with respect to the bearing position, but it has lasted 18 months so far in all kinds of weather conditions (including numerous snow falls) with little sign of deterioration.

The finished anemometer produces 8 pulses ('clicks' in Weatherduino terminology) per revolution at a 50% duty cycle, and, if counting the 16 logic transitions (both logic high to low, and low to high) it makes per revolution gives a 0.5 mph resolution at all wind speeds in a 0.33 second measuring gate period. Calibrating it on a virtually windless day in a car up to 60 mph gave a linear response of output frequency versus wind speed, with a 2.7 mph wind speed producing 1 rev/sec of the cups. Its starting speed is somewhere between 1 and 2 mph.

Part 4 will show the 0.33 second updated anemometer output being sent between two NRF24L01 2.4GHz RF transceivers, and displayed on a 0.96" OLED screen as an auto-scaling scrolling line/bar wind speed graph.





Nice job! At this rate you'll be able to DIY a complete weather station.

(04-05-2018, 23:55)uncle_bob Wrote:  Nice job! At this rate you'll be able to DIY a complete weather station.

Thanks. I think I'll stick with just a DIY anemometer build for now - not enough free time at work Big Grin

Hopefully I'll be able to get a full Weatherduino system up and running at some point in the future. For now the anemometer is just going to be wirelessly linked to a dedicated display housed in a IKEA RIBBA deep photo frame. The front facia is just a piece of paper with cut outs in it.

A4 paper with no cut-outs:


A4 paper with circular and rectangular cut outs:


plastic sheet holding LEDs, OLED, LCD, 7-segment displays:


All this is another story though, and I'm not sure it'll ever get finished....

Good job, hibs1!

Best Regards

[Image: banner.php]
My outdoor AQM-I: here

Hi hibs1,

I must say that you have an impressing ability to build things. Please keep posting your work, DIY things are always inspiring for many people.



Users browsing this thread: 1 Guest(s)