[Wizza]

PCB Assemblies

While it was back at Christmas time that I decided to start my WeatherDuino project, it took a good couple of months for some of the final parts to come in from eBay China. And then work and life got in the way for a while, however I've finally managed to get the bones of my system built and running over the past few weeks.

I built the RX station first, and all went to plan and started up first time. At the moment I only have a DHT22 temp / humidity sensor and a BMP180 Baro module installed. As I'm running a v5.02 board with eeprom, I went with a 20x4 LCD display with blue back-light. Initially I ran some test sketches to test the individual sensors and the LCD display one at a time, and then a RTC sketch to set the time/date.

The TX station was built next, and again all went to plan and started up first time after programming. At the moment its running on test with a DHT22 module, and I have a SHT11 coming for initial outside tests. I'll probably follow this with a SHT25, and / or a SHT15 once I find a suitable radiation shield that I am happy with.

I've ordered Davis wind speed and direction sensors from the US via Amazon, and have managed to do a deal on a TB1-0.2 Tipping Bucket Rain Gauge, and have that on the way.

The Wireless Display was the last module to assemble. As a few will have read, it didn't start up first time, and after some checking and fault finding with the CRO and a Spektrum analyser it was discovered that a faulty RX module was to blame. Swapped it out and all came to life.

Well that's the basic build completed and the station has been running on constant test for a couple of weeks now, and all good. I'll be sure to add more info and photos as my system comes together.

Cheers for now, from Brisbane Australia.

Mark R

---

Here are a few photos of the finished boards.

           
{click on photos to enlarge}

[uncle_bob]

Looking good!

[hornychz]

Congratulations

[Wizza]

The Wireless Display

Over the last couple of nights I finished off my first wireless display. Mounted the PCB and the LCD display into an IP-rated enclosure with a transparent cover. I figured that since this first wireless display was going in my workshop, it seemed fitting to see the inner workings. I mounted the LCD display on some metal PCB standoffs, and simply drilled a hole in the top of the enclosure for the antenna.

I had to figure a way to connect the USB lead and exit the enclosure. I didn't really want it exiting from the top, so I used one of the Mikado "right-angle" USB leads, which allowed the cable to curl around the exit out of the rear. A grommet pushed tightly into an appropriately sized hole makes an exit point for the USB lead.


               

---

Managed to find a nice rain gauge in good condition, thanks to JT, and have it on the way.

       

[werk_ag]

Hi Mark,

Thank you for sharing your experience in building the WeatherDuino Pro2. Is good to know that the assembling is going well.

That rain gauge seems really good, when I saw it, I was tempted to place a bid.

Rgds

[Wizza]

OTA Rain Gauge

I received the rain sensor yesterday. Opened it and all looked ok after transit. Did some initial testing this afternoon and discovered that the reed switch was faulty. Thanks to the design of this equipment, replacemet was pretty straight forward, and didn't take too long. Have posted a photo below showing the terminal block with the reed switch and the connections. Had to do a mechanical calibration to balance the see-saw, and all is now balanced and silky smooth. Did a quick test and confirmed that 33 tips of the buckets showed as 10ml on the WeatherDuino station. Over the weekend I'll do some calibration testing with measured water, per JT's advice and instruction - many thanks....

Shall keep you updated as build progresses.

Mark R

               

{click to enlarge photos}

[JT118]

Ok 33 tips shows that you have not changed the config file in the TX sketch.

Go to the TX sketch and click on config file tab: scroll down and you will see this:

// Define Rain Gauge tip bucket value (mm)
// --------------------------------------------------------------------------------------
/*
Use 0.300 for standard FineOfset rain gauges (Datasheet value = 0.2794)
Use 0.200 for standard Davis rain gauges
*/
#define COLLECTOR_TYPE 0.300 // Please note: Values must be in metric units. Max. resolution: 3 decimal digits


// --------------------------------------------------------------------------------------

Change the #define COLLECTOR_TYPE 0.300 to 0.200 which I am fairly sure your gauge is. Your readout should now be 50 tips resulting in 10mm on screen ( and this should take 314ml of water to do ). This will, I think, be correct for your gauge assuming it is correctly calibrated.

[Wizza]

Thanks JT... I'll have a look at that in the morning... I loaded Cumulus software tonight and managed to get it configured and talking to my WeatherDuino station. Long term plan is to run it on a Pi and have standalone updates to Internet.

   

[JT118]

In reply to above and pm's, I thought doing this on the forum would open it up;

The first thing that you need to do is establish the cross sectional area of the gauges collector. Please see "NEMS" document in your Fine Offset Thread.

http://www.meteocercal.info/forum/Thread...264?page=3

On Page 44 (of document, p58 of pdf) there is a table which states that calibration is done using 653mm of water.
This means that the diam of the TB3 is 203.89mm (remember this is sharp edge to sharp edge: or the "watershed" of the gauge, one of the many factors that contributes to the "vagueness" of FO gauges is the lack of a sharp edge.)

653 ml
65300 / 2 = 32650
32650/Pi= 10392
Square root of 10392 = 101.945mm radius x 2 = 203.89mm diameter.

Check your gauge diam accurately using a vernier. It is probably the same as the TB3, but you need to check. You cannot do any other calibration until the cross sectional collector area of your collector is known. All rain gauge calibration, whatever type it is, must start with this. If the diam is different to this you will need to do the arithmetic on the FO post:

Assuming it is the same 203.89mm then the rest of the calibration should be done using the NEMS table on page 44. (99 to 109 tips per 653ml etc.etc.). It also gives acceptable validation ranges. If your gauge is not 203.89mm diam, then one will need to draw up a similar table after establishing a new cross sectional area.

I suspect that your gauge has been very well calibrated given the seals on all the adjusters. So I would be very loathe to start changing anything until you are absolutely sure of the basic starting point sums of cross sectional area as outlined above.

If the TX config is changed to 0.200 it will take 326.5ml of water to register 10mm on the readout after 50 tips of the bucket. (not the 314ml mentioned earlier which assumes a 200mm diam.)

I attach a pdf from the Australian BOM with further ideas and numbers.

Tipping-bucket-rain-gauge.pdf Size: 702,94 KB  Downloads: 995

[Wizza]

Hi JT, I measured the collector this afternoon, and it is 200mm across. From yesterday's testing I measure that it was about 6.5ml to tip the bucket once. I re-ran the calibration test this afternoon, based on the 653mL test outlined above and I registered 18.2mL of rain fall on the WD. I confirmed the gauge was level. As you mentioned, this gauge looks pretty calibrated and set with all the red locktite, so I was pretty loathed to start adjusting it. But, if it isn't quite right, I guess I need to do something. Do you think the WD software just isnt matched to the gauge..?? Can I put a decimal place in the TX config for the rain gauge, and make it 2.2 or something like that (instead of 2.0 or 3.0)..??? Or do you think I should crack the locktite and start mechanically adjusting the stop screws in the gauge..???
Cheers for now...
Mark R