Dear J.Mark and Bob, Sorry to be the bringer of bad news but... Houston, we have a problem. I took a look at the test data Bob sent from the JWD in the new wind-sheltered site. We are missing a lot of small drops, the very opposite of the intention of the quiet, wind-sheltered site. I agree this defies logic but it is what we have. I have included a ppt file with some examples from KWAJEX, Bruce's trailer site in 2000 and the new data. We have only a few days worth of the data from the new site so the lines should not match but I think the trend is clear regarding the absence of the small and most numerous drops. In looking at the raw data, small drops occasionally register (6th and 7th columns) but not as many as should be there. For example: 2002 5 7 4 27 0 1 27 96 138 250 402 204 74 40 9 0 0 0 0 0 0 0 0 0 2002 5 7 4 28 0 1 20 58 65 112 149 51 20 3 0 0 0 0 0 0 0 0 0 0 2002 5 7 4 29 1 2 15 28 26 11 6 0 0 0 0 0 0 0 0 0 0 0 0 0 Also there appears to be occasional noise in the smallest size category: 2002 5 7 20 8 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 11 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 12 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 13 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 14 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 15 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 17 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 18 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 19 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 20 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 20 21 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 21 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 21 1 8 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 21 2 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 21 3 12 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 21 4 7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 21 5 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 21 6 11 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 21 7 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 21 8 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2002 5 7 21 9 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I have also attached the entire file of raw counts with the date stamp if you want to peruse it. If time is 99 99 then the time did not get recorded, this is a not infrequent problem with the instrument (mine does it as well) but usually only occurs when it is not raining. It is a puzzle. I am not sure what the problem is but I have a few suggestions to try to diagnose the problem. 1) Check small drops can be detected in conjunction with large drops [I think this is ok based on the raw data but might as well double check]. Take something like a plant mister out to the instrument and see if it can register small drops. Hold the mister about a 4 in from the unit and gently spritz. [gently in that you can see individual drops rather than a spray as you would use to clean a window]. If the unit can see the drops they show up on the display. Look for counts in the first 2 columns of the display after the timestamp, these are the smallest drop counts. The closer the mister is to the instrument, the shorter distance the drops have to get to their terminal fall speeds and the "smaller" they will appear in diameter (since the instrument deduces the diameter from the momentum of the drop). I tried this on my JWD and I got about 5 drops each in the smallest two categories after some trial and error with ever more gentle spritzing and moving the mister from 1 foot to about 4 in from the instrument head. The point is to get more than 1 or 2 drops in the smallest categories. 2) Check for vibrations. If the unit is vibrating it will have a similar effect as noise. 3) Check cable and connections for integrity. Somewhere in the Distromet doc it says that small drops may be lost if the cable is too long. If you have a shorter cable available, try the mister exercise with the shorter cable to rule out cable length as the culprit. 4) Collect some more data esp. in a longer lasting storm. It is possible that those early May storms occurred in a situation with non-saturated low levels and the small drops evaporated before reaching the ground. 5) If you get desperate send it the unit to me in Oregon and I will check it next to my JWD instrument. If there is a problem it can be shipped to Switzerland from here. I'm sure whatever is wrong it can be resolved. The site is excellent so therefore it is most likely something wrong with the instrument (or the atmosphere) :) Sincerely, Sandra