From searching the web for problems with LaCross weatherstations I found out that other owners had been experiencing the same problem of spurious wind speed readings. It was thought to be caused by the unshielded cable between the wind instrument and the thermo/hygro unit picking up interference and triggering a rogue reading. Severel owners had replaced the cable with a length of 8 core shielded network cable (CAT5e) and reported success.

The existing cable between the sensors is a flat 4 core telephone type cable. It has a plug at one end which plugs into the thermo/hygro unit but the other end is soldered directly to the PCB inside the wind vane housing. The options were to either open up the unit and unsolder the old cable or to cut the cable close to where it leaves the housing and join the old cable and new one together. I decided to do the job properly and replace the entire cable.

Rather than using network cable I had a length of 4 core shielded cable intended for CCTV cameras. This had a bare shield wire and a foil liner encasing the 4 cores of stranded wire. It was a similar diameter to the original wire and so would fit into the connector correctly, something the CAT5 cable would not. However the individual cores are not configured as twisted pairs as they are in CAT5 cable.

Anyway, I took the opportunity of yet another communications breakdown to do the modification. As I had lost 24 hours worth of recordings another couple of hours wouldn't make much difference. I took the windvane down off it's pole and brought it inside for disection. The anemometer is removed from it's spindle by a single small screw.The case can then be opened by removing 3 small screws to reveal the PCB. The cable is stuck to the PCB with hotmelt glue which had to be carefully prized off before the connections could be unsoldered. The legend on the PCB lists the connections as pin1=serial output - black wire, pin2=VDD ( power +3V ) - red wire, pin3=regulator CE ( don't know what this is for? ) - green wire, and pin4=gnd (0V) - yellow wire. Interestingly these are the same colours in a standard 4 core telephone extension wire but they are wired up exactly the opposite way round.

Having released the old cable my new one was soldered in, not forgetting to first feed it through the support post and then through the hole in the case. The white wire was joined to the screen wire & foil & sleeved with heatshrink. As I didn't have any hotmelt glue I used a ty-rap for strain relief. Before screwing everything back together again I took the bits outside and plugged the cable into the thermo/hygro unit and then checked the base unit to see if it was working. Much better to find out now than when it's back up the pole.

Since the modification was done I have not had a single rogue wind speed reading. However, I am now getting a problem with the thermo/hygro unit. It is recording rogue 'minimum' temperature readings of -19.3 degrees C several times a day so it looks like the next step is to replace the cable between the thermo/hygro unit and the base unit with a screened one to see if it makes a difference. At least this cable has a plug at both ends so no disassembly of the units will be required.