[spudstrawb]

The rain sensor is now working. I tested the reed switch and it seemed okay. Further investigation showed that I needed to change the pins that the relevant wires were attached to in the RJ11 plug for it to work with WeatherDuino. A new plug was fitted with the relevant wires connected to the appropriate pins and it is now working. I did a quick and dirty calibration and it suggests that the sensor is not consistent with either Davis or FO standards. I will have to do a proper calibration. Making progress though.

[werk_ag]

The rain sensor is now working. I tested the reed switch and it seemed okay. Further investigation showed that I needed to change the pins that the relevant wires were attached to in the RJ11 plug for it to work with WeatherDuino. A new plug was fitted with the relevant wires connected to the appropriate pins and it is now working. I did a quick and dirty calibration and it suggests that the sensor is not consistent with either Davis or FO standards. I will have to do a proper calibration. Making progress though.

Do you know the value corresponding to each tip?
You may have to set it on the weatherduino software, and very important, don't forget to adjust the Cumulus multiplier accordingly.

Check this topic: http://www.meteocercal.info/forum/Thread...figuration

[werk_ag]

One other question - the file "data_structs.h" has a list of Davis functions and a note that not all of them are functional in WeatherDuino. Out of curiousity, are "uint16_t monthRain;" and "uint16_t yearRain;" operational in WeatherDuino?

Yes, they are. However Cumulus seems to don't use those data. It does a even better job, calculating them from the dayfile.txt

The output comes about every 5 seconds and the above is 2 lots of output (each lot starts with "a" and then the water depth value in mm, and then the rest of the string "b0c0d0e0f0g0h0i"). So there are 20 characters in the string. If there was less than 1m of water depth, then the string would have a "0" after the "i" in order to keep the 20 character string. I guess that it's a voltage output but I really don't know.

Unless you have direct access to the sensor output, I don't see how to use it. Just for curiosity, do you have any reference or link to the sensor manufacturer?

[spudstrawb]

Yes, they are. However Cumulus seems to don't use those data. It does a even better job, calculating them from the dayfile.txt

If I can't get my water tank level to show on the WD unit, then I thought it might be useful to show the monthly and yearly rainfall in place of the Srad and UV lines (which I have no use for at this stage). I'll have a go with the software and see what I can come up with.

Unless you have direct access to the sensor output, I don't see how to use it. Just for curiosity, do you have any reference or link to the sensor manufacturer?

When you talk about "direct access to the sensor output", what is required? With the connection to my PC, I wrote a little script in AutoIt which talks to the depth gauge, takes the 20 character string and strips away everything that isn't part of the depth value. Then it calculates the water tank level as a percentage and reports the depth and % values. I got the gauge from http://www.electrosense.com.au/liquid-level.htm but mine is hardwired via the serial cable, rather than being the wireless version (hardwired was cheaper).

[werk_ag]

When you talk about "direct access to the sensor output", what is required? With the connection to my PC, I wrote a little script in AutoIt which talks to the depth gauge, takes the 20 character string and strips away everything that isn't part of the depth value. Then it calculates the water tank level as a percentage and reports the depth and % values. I got the gauge from http://www.electrosense.com.au/liquid-level.htm but mine is hardwired via the serial cable, rather than being the wireless version (hardwired was cheaper).

Supposing your sensor is one of these:

The sensor consists of a pair of absolute pressure transducers, one immersed in water and the other exposed to the atmosphere above the surface of the water. The differential pressure enables water level to be computed. with an output of 0-5V DC proportional to depth, or as a serial digital data output in ASCII form

What I want to mean with "direct access to the sensor output", is having access to the 0-5V DC signal. If you have access to it, with some changes in the software, you can use for it, the ADC pin used by the solar sensor.

[spudstrawb]

Supposing your sensor is one of these:

The sensor consists of a pair of absolute pressure transducers, one immersed in water and the other exposed to the atmosphere above the surface of the water. The differential pressure enables water level to be computed. with an output of 0-5V DC proportional to depth, or as a serial digital data output in ASCII form

What I want to mean with "direct access to the sensor output", is having access to the 0-5V DC signal. If you have access to it, with some changes in the software, you can use for it, the ADC pin used by the solar sensor.

I only have a single pressure transducer but it looks like it might output "serial digital data output in ASCII form" rather than an output of "0-5V DC proportional to depth". I'll put my multimeter across the serial plug/socket pins and see what comes up. But the "a2055b0c0d0e0f0g0h0i" type string that I get in hyperterminal makes me think that I'm getting the "serial digital data output in ASCII form".

[spudstrawb]

I'm still working on things. I haven't gone live yet because I'm waiting for my SHT31 sensor to turn up to replace the existing HTU21D sensor. I'm getting ridiculous readings for outside humidity at the moment. I'm waiting on some parts to turn up from ebay to finish putting it all together in the case as well.

In the interim, can someone tell me if it is possible to get data from the Raspberry Pi to the WeatherDuino Pro2? Currently the time and date is pushed across from the Pi to the RX unit, but I'm wondering if I can also push my water tank level value across as well. I guess Werk_AG is the most qualified to talk on this but input from anyone with any knowledge or suggestions would be most welcome.

[Wizza]

spuds, I think the passing of the time and date will be a function of the CumulusMX software. Therefore, I recon the CumulusMX software would have to be re-written to be able to read your tank levels and pass them across. And then the WD software would also have to be re-written to receive and display the value. I guess nothing is impossible, but that one sounds like a bit of work...
Mark R

[spudstrawb]

Thanks Wizza. CumulusMX uses an exe file so I don't think I can get MX to do the tank level value transfer to the RX unit. I'd most likely need to use something independent of MX to do it.

[werk_ag]

Hi spuds

As Wizza said, the sync of date and time is part of the Davis protocol. Each time the Cumulus starts (and by default each day at 4h00), it syncs the console date and time (in our case the RX unit), with the PC.
I really can't see a way to make the same PI that is running Cumulus send the data that you want to the RX unit. The COM port is open by Cumulus and it should be keep open to listen for Cumulus requests (loop requests).