I have submitted this question to engineers at Campbell but they are on Pioneer Day Holiday today. Can anyone else help?
Can someone provide me with the specific specs on the 5 resistors that are on the 05103 Junction Box board? I purchased a few 05103 units direct from RMyoung a while back and they do not have all the resistors that the units that Campbell supplies do. I think this is done by Campbell to make the 05103 output compatible with Campbell instructions and loggers?
Thanks for any help here!
Hi furban,
There are 2 transzorbs in the box for surge protection. There is a 1K ohm resistor in series with the wind direction signal output and also a 1M ohm resistor to the wind direction reference let to pull the wind direction output of the sensor to ground when the sensor is in the gap of the potentiometer. Your sensor is still compatible with the program instructions. You could go to radio shack and get the 1Meg ohm resistor and install it between the wind direction input on the logger and a ground terminal to prevent any flaky readings when the direction sensor is in the gap of the pot.
If you ask RM Young for the current manual it has the wiring in the back of the manual and shows the components. I believe that this is now standard with all the wind monitors but please check with them.
Regards,
IslandMan
The 1M resistor is not standard in all sensors as it adds a very small error to the measurement and also makes the sensor incompatible with some RM Young displays that deal with the open circuit situation in the gap another way. RM Young fits the 1M resistor to order as a special sensor variant.
If you are unable to find or fit the resistors there is another way of measuring these sensors with the logger that deals with the gap condition. This only works with the CR800, 1000 and 3000 loggers with newish operating systems (dated mid 2007 on) and will only work if the 1M pull-down resistor is NOT fitted. In many ways this is a better solution than the pull-down resistor (it is quicker and has less noise susceptibitly) but is not generally used as it will not work with all our loggers.
The technique uses the open circuit detect range code (mv2500C) plus careful selection of excitation voltages and settling times. The open circuit condition can then be detected with one extra line of code:
BrHalf(WD,1,mV2500C,1,Vx1,1,2000,False,450,250,357,1.5)
If (WD > 360) OR (WD = NaN) Then WD = 0
APS, thanks for your helpful post. However, I'm wondering why the multiplier of 357 and offset of 1.5? Why would changing the code to look for an open input change the multiplier and offset from 355 and 0?
Thanks
Sorry for the confusion, the example was cut out of a user program who is rather fussy about the multiplier and span they use - they have studied various sensors to the nth degree and use an average value with a bias for their alignment process...read on.
For general use, if you do not want to check each sensor, you should use the manufacturers multiplier of 355. However, the offset you use depends on how you align the sensor and whether you want to take out the correction for true north mathematically or by physical alignment.
With regards to the alignment my personal preference is to align the sensor to a target at a known bearing (corrected for magnetic declination if you wish) with a target near South, i.e. well away from the deadband of the sensor. You point the sensor at the target and rotate the sensor base until the logger outputs the exact direction of the bearing to the target. This is with the offset set to zero. That will be as good as you can get.
The alternative approach is to find a target at North, point the sensor at the target and rotate the base till logger outputs zero degrees. The problem with this approach is the logger will read zero over 5 degrees (the deadband). For practical reasons, most people find the mid-position in the deadband and then lock the sensor base. If you do this you should use an offset of 2.5 degrees (half the deadband) because you will have introduced that offset by using the mid-position of the deadband.