Weather Station Hardware > Other Weather Station Hardware

Proof of Concept - AWS Indoor Display

(1/1)

TheBushPilot:
For the past two or three years I've been planning a home AWS. One of the bigger issues that plagued this process was the need for an indoor display. While using commercial grade instrumentation has its perks when considering validity of measurements, one of the major downsides is lack of easily presentable data. For that, most stations use an LCD screen which reads directly from the remote sensor suite and presents it in a user friendly manner. The problem with my design is that data loggers don't come with pretty displays ready to go out of the box. (Or at least the retired ones I'm employing.) The 24 x 2 LCD display on the face of Campbell Scientific's CR23X logger unfortunately won't suffice in this instance.

The solution? An R. M. Young 26700 Programmable Translator. This device can read incoming serial data and display it with relative ease. I ultimately went with this device because of how much I liked it. Not only this but price for performance is hard to beat in my opinion. Layout, interface, programming, it's a solid piece of tech.

The idea was to have the data logger output select data in serial for the display to read and present independently of the logger or support software requiring a computer. The example program I put together to test this idea can be found below. Where INLOC values for PANELT_C and PANELT_F are the internal temperature readings in Celsius and Fahrenheit of the temperature probe within the CR23X. The program is executed each second to test stability while running a subroutine - subroutine when executed will send serial data of each selected intermediate data location to the 26700. Reason I mention stability is because when the data logger executes the program, if the execution time takes too long, it will create overrun errors. These overruns cause the program to skip scans creating gaps in data. Typically a second or two will be missing during the subroutine execution. Often times all one needs to do is modify the program to fix it. While doing early tests, I found overruns occurred when the execution interval for the serial send subroutine was less than five seconds per each interval. Or 0 Seconds into a 5 Second interval. A bit of a balance to achieve quasi-fast readout while also not creating errors within the program. Found that one minute output intervals sufficed for my purposes. Then, main program saves the INLOC data to final storage with instruction 84 (data table).

 [ You are not allowed to view attachments ]

The test achieves two things simultaneously: Successfully send serial data to the display; send multiple data strings to the display. With instruction 15, parameter 5 selects the INLOC that will be sent via serial. Then with parameter 7, is each successive INLOC after the first selected. In this case, because PANELT_C is chosen, the location immediately following is PANELT_F.

 [ You are not allowed to view attachments ]

The real time output of the logger as seen from LoggerNet above and the read serial data from the CR23X to 26700 below.

 [ You are not allowed to view attachments ]

And the CR23X requiring only TX/RX and GND to the remote display. In doing so, the eventual setup could be classified as a "one wire weather station." (With extra steps and a little more involvement, of course.) :roll:

 [ You are not allowed to view attachments ]

Then with the 26700 translator, reading the data is as simple as setting an INP SERIAL and DISPLAY VAL within the program instruction. Ensuring the number of #flds equals the correct number of data strings input from the output device. And after getting it to read correctly, setting up the screens to ones liking for operational use.

(When the time comes I'll probably make an in depth thread on the process of this "home AWS".)

Cheers

Navigation

[0] Message Index