proudly sponsored by
|
archiving of SOWACS proudly sponsored by |
|
On 2 Feb 99, at 17:58, owner-sowacs@aqua.ccwr.ac.za wrote: > Currently I am using a TDR system supplied by Campbell Scientific and > comprises CS615 sensors and a CR10X logger. I need to measure the soil > water contents on different silvicultural treatments. Now these plots are > really large and I will need to have cables up to 200m in length between > the logger and sensor. Does the cable length affect the accuracy of the > reading and what is the maximum cable length that can be used with these > sensors? Second, I suppose like the neutron probe the best way to > calibrate the sensors is gravimetrically? Anyone else using the same > equipment? Here is one official Campbell response: The CS615 generates a frequency output. When measured with a CR10X using the period measurement, the effects of longs cables on the measurement of that frequency are insignificant for any practical cable length. This is because the thresholds for detecting the waveform are small compared to the peak to peak waveform (+/-2.5 V) and attenuation of the signal would be small, except with very high capacitance cables (which we don't supply). There may be more significant limiting factor caused by voltage drops in the power reaching the sensor for very long cable lengths, but again this is up at the level many hundreds of metres. However, we now impose a limit of 100 m on the cable lengths we sell with the CS615 for another reason, which was inferred in another response to this group. This is because we know that the longer the cable run, the more prone the sensor is to damage caused by lightning strikes to ground. All probes made in the last 2 years or so, have internal transient protection in the sensor head. However, because the probe is isolated from ground, the transient protection decreases in effectiveness with the distance from the ground reference point, which is normally at the logger. We have discussed with some users the possibility of them using shorter cables and fitting transient protection and ground connection in junction boxes, at points along much longer cable runs, i.e. between the logger and remote sensor, but this is not a technique we have tried or tested so we officially do not support it. As to the calibration. The CS615 can be used in some soils without any recalibration, as like TDR it measures the dielectric water content. However, as readers of the mailing list will be aware there a some notable soil types which do not conform to the standard TDR calibrations, plus in addition the CS615 is more sensitive than a full TDR trace-analysis system, to the effects of salinity. Therefore if absolute accuracy is required you are advised to at least do a single point calibration in typical soil and more points to get greater accuracy. The CS615 does measure volumetric water content so you have to make sure you take this into account. Simply drying technique a known volume of soil and measuring the weight change is the normal technique used. Andrew Sandford P.S. I am not going to get into the debate of whether this probe operates in the time versus frequency domain. The sensor is fundementally an oscillator whose frequency varies according to the transit time of fast edges along the rods of the probe. The oscillator runs at around 50 MHz so the edges are faster than simple low frequency oscillator probes, but not quite as fast as is used in a conventional TDR system. The oscillator output is scaled down in the head to the low kHz range for transmission back to the logger, so loss in the cable runs back to the logger are not the same as one has to consider in the length of long cable runs between a true TDR unit and the TDR waveguides (probes). -- Andrew Sandford Phone: +44 (0) 1509 601141 Campbell Scientific Ltd. Fax: +44 (0) 1509 601091 80 Hathern Road, Shepshed, email: andrew@campbellsci.co.uk Leics, U.K., LE12 9RP http://www.campbellsci.co.uk/