Cumulus Support

I think I might experiment with a diode pump circuit for counting anemoneter pulses which will convert a pulse rate to an analogue voltage which can then be read by a D/A converter such as the DS2438 with either 10v or 250mv A/Ds. The 250mv A/D would give better linearity but OTOH a non-linear response to the wind speed would not necessarily be a bad thing. I think a wind speed reading represented by the Beaufort Scale would make more sense than a purely linear scale and I was planning to use this sort of windspeed display anyway.

For anyone who doesn't know, here's the Beaufort Scale courtesy Met Office :-

The digital alternative to the above would be a CMOS 4000 series type 4520 dual 4 bit binary up counter, giving 8 output bits which could be read by a DS2408 1-wire device. But this would require more time to build. Using a reading every 3 seconds I could read the wind speed in increments of 1/3rd mph with max of 255/3 = 85mph which should be plenty

I'm having considerable trouble trying to get the DS2408 to work and it's doing my head in so I'm leaving it for a while. I think I could do with coming back to it afresh later. I'm now looking at other ideas.

Hi Gina,

Sorry, I don't have much to contribute to this thread. My own design aims are very much around wireless communications (i.e. battery operation) and "getting the best out of the Fine Offset sensors". IMHO there's not too much wrong with the FO anemometer (at £2.50 from Maplin) except it's calibration may be doubtful, but that applies equally to almost any home-built anemometer.

Certainly the FO Vane has its weaknesses (some of which can be improved), but it only consumes an average of around 1 uW from the battery! My "planned" improvement is to actually take advantage of its "lively-ness" and take the average of multiple "local" measurements to produce a average for the transmitter (at full / higher resolution). For this I'll be using a "PICaxe" microcontroller, which, although they cost about the same as a full Arduino board, can easily run at sub-miiliwatt (i.e. alkaline battery) power levels. It can also natively read a DS18B20 which is indeed an excellent temperature sensor.

Cheers, Alan.

Hi Alan

Thank you for your reply I understand your wanting to keep power down for battery use. My weather station is and will continue to be near my observatory which has mains power backed up be a sealed lead acid battery, so I always have power available. So keeping power to a minimum is no longer that important to me and I admit, I've forgotten that at times.

1-wire is handy for communication up to 100m on a single twisted pair but there are other, simpler methods that people like me sometimes forget. If I'm going to use CAT5e cable for comms there isn't just one pair of wires, there's four. So I can use 1-wire where it's convenient and the other wires as well for other commumication methods.

I have a DS2408 working now after debugging the Arduino library for it But I'm not using that for my wind instruments - I've decided to just get the now obsolete DS2423 and DS2450 circuit working with new optical sensors and hardware. The anemometer is working with the DS2423 counter and the wind vane direction with Gray optical encoder almost working using the DS2450. I'll post diagrams and photos shortly. I'm hoping to have the direction encoder working properly tomorrow, luck permitting