proudly sponsored by
|
archiving of SOWACS proudly sponsored by |
|
At 01:49 PM 11/14/96 -0800, you wrote: > >> From majordom@aqua.ccwr.ac.za Sat Nov 9 10:33:25 1996 >> X-Sender: dynavis#pop.hip.cam.org@192.168.0.1 >> X-Mailer: Windows Eudora Light Version 1.5.4 (32) >> Mime-Version: 1.0 >> Date: Sat, 09 Nov 1996 12:01:55 -0500 >> To: SOWACS@aqua.ccwr.ac.za >> From: Romain Gagnon <RGagnon@smartrain.com> >> Subject: Re: TDR and soil salinity >> Sender: owner-sowacs@aqua.ccwr.ac.za >> Precedence: bulk >> X-Lines: 66 >> Content-Type: text/plain; charset="us-ascii" >> Content-Length: 2777 >> >> At 12:49 96-11-08 -0800, you wrote: >> > >> > Romain Gagnon wrote the following >> > >> >+Why would you want to add heat-shrink tube to TDR or FDR probes ? >> > >> >+It will not prevent the signal from being attenuated by the saline soil even >> >+though there is no direct contact between the probe and the soil. Also it >> >+would lower the sensitivity therefore the precision of the sensor even after >> >+calibration. >> > >> >It has been our experience that the TDR can be used in much more conductive >> >soils by insulating the rods of the probe with fiberglass resin, or any other >> >insulator. In cases where no visible reflection is seen from the probe end, >> >the insulation enables a strong reflection to be recorded from the probe end. >> >You are correct that the sensitivity of the system is decreased, but good >> results are still obtainable. This loss in sensitivity can be compensated for >> >to some degree by using a longer probe length. (shorter un-insulated probe >> >lengths would be the traditional approach to using tdr in more conductive >> >media. Naturally sensitivity is also lost by shortening the probes. The >> obvious problem is that the system has to be >> >calibrated for each conductive soil it is to be used in. Also insulating the >> >probes is difficult and it is unknown how long such coatings will last in the >> >field. In cases where we have measured the amplitude of the reflected TDR >> signal, amplitude remains constant even as conductivity increases. Ofcourse in >> >the case of an uninsulated probe amplitude of signal decreases with increasing >> >conductivity and this fact has been used widely to calculate soil >> >bulk conductivity. It seems to me that if an insulated probes signal amplitude >> >does not decrease with increasing conductivity that in effect signal >> attenuation >> >has been prevented? >> > >> >Sam Hokett >> >Desert Research Institute >> >Las Vegas, Nv >> >Sam@snsc.dri.edu >> > >> > >> > >> If you insulate TDR probes, the reflected signal will obviously be less >> attenuated. However, as you say, the sensitivity will decrease accordingly. >> Now, if you make the probe longer to compensate for sensitivity loss, then >> your signal should be as much attenuated as the initial shorter and >> non-insulated probe...at least in theory. >> >> The fact that there is no contact has an impact on low frequencies but not >> on high frequencies. If you do observe a difference, then the frequency >> spectrum of the reflected pulse is different from the initial one. >> Therefore, the calibration required is more complex than a mere linear one >> based on probe length. Does that match your observations? > >------------------------------------------------------------------------------- >You may well be correct that the high frequency signal is still being attenuated and that only the lower frequencies are being preserved by the coating. >But what ever the >case the insulated probes can still be used to accurately determine water >content at a salinity range where TDR couldnt be used at all previously. >A simple correction is applied to the account for the effect of the coating >on the measurement. The coating has a larger effect in wetter soils than in >drier soils and naturally thinner coatings have less effect than thicker ones. >The relationship of Ka corrected for coating effect >and water content is the >same as with uncoated probes. I should mention though that there is a >conductivity limit >beyond which even insulated probes can not be used. >At least this has been my observations. > >Sam Hokett >Desert Research Institute > > This is very interesting. Your observations indicate that the lower part of the frequency spectrum of your TDR pulse or step is playing an important role. Using a coating for an FDR would not make a difference since it uses a single tone signal. At Smart Rain's, we measure in the 100-325 MHz range. Under 100 MHz, the ganularity of the soil will influence apparent permittivity. Therefore, the standard deviation error will be higher when calibrating for any soil (clay, silt or sand). This may explain why our approach is slighly more precise than TDR. Romain Gagnon Eng Smart Rain Corporation inc 1505, Place de l'Hotel de Ville, suite 102 St-Bruno, Quebec, Canada --- J3V-5Y6 E-mail: Info@SmartRain.com Phone: 1-514-441-4289 FAX: 1-514-441-2147 WEB: http://www.SmartRain.com