proudly sponsored by
|
archiving of SOWACS proudly sponsored by |
|
BlFtЁ Bruce-- Thanks for your interest--I have revised some of my comments to make them more accurate, complete and specific. (You may want to edit your responses accordingly.) Please also advise people that my enthusiasm may be tempered by experience--I am new both to Civil and Environmental Engineering and to using TDR and the like for soil measurements! >Sowacs > >Today's discussion was sparked by Dave Barnett who has recently >joined sowacs - welcome Dave! > >.. >The point is that sowacs was started partially for me not to >reinvent the wheel, but also for people like you to collaborate >with others in development. >The problem is, is that there seem to be several if not many >centres internationally that are independantly going the TDR >route. They are also doing it as commercial ventures (sometimes >indirectly or in the long term), and as a result, the >collaboration has been scant. (Perhaps some people have got into >contact privately through the list - but that is good as well.) > >Almost every "new" sensor that is anounced these days - and >there have been a lot recently - use a "novel" patented >technology. >This is always either TDR, FDR or heat dissipation. What is novel >about it is that the electronics have been developed to simplify >its use and packed as a unit with the sensor. >I wish, however that the producers would be open and in the specs >somewhere, say that it is in fact a "TDR-based" or "Heat >dissipation-based" system. This way, the 'technos' amongst us >will have something to latch onto straight away rather than having >to play detective to figure out its principle of operation! > Dave>--In academia, the model has to be fully explained and results repeatable and verifiable to be accepted--Proprietary technology doesn't lend itself to that. >This, I think, is an indication that the race for sensor >technology is about to be finalised - and once there is a winner >(or winners), mass production and consumption will reduce the >price so that the required numbers of sensors become affordable. > > >>Dave> My first project here at UMass was to duplicate a "4 Electrode Soil >Salinity Sensor" described in a paper from 1979. Because the ICs used were no longer available and because I did not believe the circuit could possiblyhave worked properly to begin with, I designed a new one with a digital display and an output to a computer A/D board. In the original circuit, the IC analog switches did not provide a DC discharge path, so the comparator would only have seen maximum values. Also, soil conductivity is more dependent on moisture content than salinity per se. (Stick Ohmmeter probes into a bag of dry rock salt and see how much conductivity you get--then add water!) Therefore, my circuit only promised to measure resistivity, not salinity. > > >>Dave>Now I am trying to come up to speed on this technology and assist one of the Environmental Engineering faculty in getting into TDR soil moisture content measurements in an economical way. > >HAVE YOU SEEN THE TDR CLEARINGHOUSE? Link off my sowacs page - >but they are into the original uses of TDR - ie cable testing >etc. > >>Dave>Yes. And we also have a quote from Campbell Scientific. Campbell seems to offer good equipment, but they are in the business of selling data loggers. To use their TDR system out of the box requires the use of their data logger and their software. According to one of their applications engineers, interfacing CS615 probes directly with a computer I/O board would require writing our own software. We have discussed this with the grad student on the project and he feels he can write the code or enlist any help he might need; therefore, for the cost of a probe and an I/O board, we may be able to get into TDR very economically. > >I don't think this is entirely true.. The CS615 probe can be >measured two ways if I remember correctly. Pulse or period >measurement. Most capable loggers should be able to measure one >of these? It can then later be transformed into a soil water >content after downloading to computer. (Perhaps I'm just used to >using a good logger like that :-) > > >>Dave>..I was glad to find info and active discussions about the relative merits of various technologies in SoWaCS. Having an RF background, I also think I can design a better uStripline probe than Campbell's parallel rods which must not stay parallel (causing impedance problems) when pushed through soils.) > >This was an argument we never cleared up on sowacs - whether the >CS615 probe waveguides have to stay exactly parallel in order not >to introduce errors. Can you shed some light on this? >What is a uStripline - something like radio antennae wire? > >>Dave>..TDR measures how long it takes for reflections caused by discontinuities in a transmission line to return to the source. Thesereflections are caused by impedance mismatches and can occur in at least four ways: a break (open) in the transmission line conductor, a short between conductors, a change in the distance between conductors and a change in the effective dielectric constant between (and around) conductors. Using a Tektronix 1502 TDR (US$8,500), an experienced operator should be able to determine the specific nature as well as the distance to the defect in a cable. According to their brochures, the Campbell CS615-L (US$195) is optimized for measuring changes in [effective] dielectric constant. Since the CS615-L probes are so economical and consist of an unterminated balanced line only 30 cm long (2 nano seconds for a pulse to bounce off the end and return!), I have to believe they have made some broad (and hopefully, valid) assumptions in their design. I am waiting for a demonstration of this equipment. Yes, think of microstripline (uStripline) as a coax cable (such as used on your cellphone antenna cable) which has been carefully slit lengthwise and uncurled so the shield and inner insulation lie flat with the center conductor on top. Practical uStripline transmission line begins with doubly clad printed circuit board material. The cladding on the bottom side is left intact and acts as a ground plane. The top side is etched to a leave a conducting trace. The width of this trace is determined by several factors: Desired impedance, type of dielectric and thickness of the dielectric. These factors are interrelated, which is why I added "effective" to Campbell's use of the term dielectric constant above. 50 Ohms is the most common impedance in RF applications. On .062 in. thick G-10/FR-4 PC board material (the common greenish epoxy filled fiberglass), a .10 in. wide trace would have an impedance of about 50 Ohms. (The narrower the trace, the higher the impedance.) However, above 150 MHz and with square-wave pulses, G-10 exhibits high dielectric losses and teflon filled substrates may be more desirable. uStripline transmission line construction is widely used in RF and high speed computer applications because they are easy to fabricate and offer predictable, low-loss performance. I hope I haven't lost you, because here is the important part: Under certain circumstances, uStripline can interact with its environment in predictable ways. Performance of uStripline is primarily determined by the width of the top trace, thickness of the underlying dielectric and type of dielectric. Normally, this allows uStrip to be a good neighbor, neither absorbing nor radiating spurious signals. However, by adding a thin, sealed layer of dielectric over the trace, soil moisture content could be determined in a way unaffected by displacement of nominally parallel rods. (By leaving the trace and ground plane uninsulated, the probe could also be excited with a low frequency sine wave to determine net Ohmic resistivity.) Terminating the uStrip with a known resistance could allow the probe's use over a wide range of frequencies, allowing for the idiosyncrasies of various soil and moisture content conditions. >Cheers>Regards > >-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-. >Bruce Metelerkamp SOIL WATER RESEARCH OFFICER >Institute for Commercial Forestry Research, >University of Natal, PO Box 100281 >Scottsville, ZA3209 >Rep. of South Africa Voice:27 331 62314 >E-mail: bruce@icfr.unp.ac.za FAX:27 331 68905 > brooz@pobox.com > Bruce.Metelerkamp@pobox.com >URL http://www.icfrnet.unp.ac.za/~metele > ../SoWaCS.html > ../RR.html >Host of SoWaCS (Soil Water Content Sensor) Discussion List. >SEND info sowacs OR subscribe sowacs TO >majordomo@aqua.ccwr.ac.za >