This wiki-40 page should serve as an introduction to what is the weather station and what were main problems with it.

What is 'Weather Station'?

Weather station consists of commercially available 'Weather Monitor' (made by Davis company), temperature/pressure/humidity sensors that are located outside of the building (on the roof above the north-west entrance to the 40-meter lab) and a junction box that allows the signals from the roof outside sensors to be seen on the 'Weather Monitor' screen. In addition to these three components (Weather Monitor, Sensors and Junction Box) there is the fourth component, 'WeatherLink' , which is responsible for communication between 'Weather Monitor' and UNIX computers which are located in the control-room. Pictures of main components of 'Weather Station' are below.

If everything works correctly, the information about temperature/pressure/humidity/wind speed should be available in two places. Firstly, this information is indicated on the screen of the 'Weather Monitor' (a small black box hanging on the north wall of the laser interferometer room, approximately near the beam-splitter chamber), and secondly, the same information should be available on the EPICS screen "c0Checklist.adl" of the control UNIX computers.

Using only outside sensors, junction box and 'Weather Monitor', but not using the 'WeatherLink' device, one can read inside/outside temperature, humidity, wind speed/direction and pressure on the screen of 'Weather Monitor'. But if one does not use the 'WeatherLink' device, it is not possible to access this information from the control UNIX computers. In order to be able to transfer signals from the 'Weather Monitor' to the UNIX control computers, one needs to use a separate additional electronic device, 'WeatherLink', which enables the communication between 'Weather Monitor/ and the control UNIX machines through the processor 'c1pem1'. In this case, data about inside and outside temperatures, amount of precipitation, outside pressure, wind speed and direction is indicated on the "C0Checklist.adl" EPICS screen of MEDM, and one can see trends of these data in 'Dataviewer'.

Short history of 'Weather Station' in 40-meter lab

The Weather Station was assembled together in the summer of 2001 by a SURF student Victor Tsai, and then according to Alan Weinstein, it soon stopped working, and it was in a nonfunctional status for several years before Andrey Rodionov made an attempt to fix it in April 2008.

As of the 1st of April 2008 the only information that seemed to be reasonable was the indication of the inside temperature on the screen of 'Weather Monitor'. This means that there were two separate problems with the weather station.

* Problem 1: No data for outside temperature, pressure, humidity were available even on the screen of 'Weather Monitor', so the signal from outside sensors was lost before getting into the 'Weather Monitor'.

* Problem 2: No indications on the MEDM "c0Checklist.adl" screen on UNIX control computers. This means that processor 'c1pem1' (located in the laser interferometer room near other processors in a DAQ rack near the middle of Y-arm) that should communicate with control UNIX computers did not do his job, and as it turned out later, the long blue ethernet Cat5 cable that goes from "WeatherLink" to the processor 'c1pem1' was defective.

In April 2008 both problems were resolved.

* Firstly, communication between outside sensors and 'Weather Monitor' was restored by re-plugging cable endings into the cable connectors located half-way between 'weather sensors on top of the roof' and 'Weather Monitor' (see picture below). These cable connectors were moved inside the 40-meter interferometer room. They should not be outside where unfavorable weather conditions can shorten their life and the wind can break the connection. For your information, these are not cat-5 ethernet cables.

* Secondly, communication between 'Weather Monitor' and UNIX control computers was revived by rebooting the processor 'c1pem1' and making the new internet connectors at the end of the long blue Cat5 cable that links the 'WeatherLink' and 'c1pem1'. Rebooting of the processor 'c1pem1' turned out to be a nontrivial problem because 'c1pem1' was attempting to reboot from a nonexisting computer op140m, so it was necessary to convince 'c1pem1' to boot from 'Linux 1'.

Pictures of the main components

Scheme of the 'Weather Station' components (from the manual)

<<ImageLink: execution failed [Cannot have arguments without name following named arguments] (see also the log)>>

Weather Monitor (front view)

attachment:Weath_Monitor.JPG

Weather Monitor Box (rear view) and "Weatherlink" device which allows communication with computers

attachment:DSC_0500-W_Mon-Cables.JPG

Sensors on the roof of the building (overall view and "zoomed-in" view on the next picture)

attachment:DSC0503-OutsideView.JPG

Wind Sensors on the roof "zoomed-in", rain collector and external temperature sensors are nearby but not shown

attachment:1.JPG

Cable connectors half-way between the roof sensors and the junction box, which should be inside of the building

attachment:3.JPG

Rack with: (1) computer 'c1pem1' and (2) network switch board. Signal from "Weather Monitor" goes through the "WeatherLink" to the "c1pem1" processor, and then a signal goes out from "c1pem1" to the "Network Switch Board" and further to the control UNIX computers. This rack with "c1pem1" is located near the DAC rack not far from the MC2 mirror of the mode-cleaner

attachment:Computer_Rack.JPG

RJ-45 cable ending for Cat5 8X8 cable that links 'WeatherLink' and 'c1pem1'. This connection near 'c1pem1' was flaky, so it was replaced in April 2008

attachment:DSC_0505.JPG