Here's a drawing of the sensor and circuit:
The sensor is basically a variable capacitor. Since the dielectric constant of water and almost all fluids is different from tha of air, the sensor's capactance varies with the level of water in the sensor tube. The sensor is made from a 24" length of 7/32" ID, 3/8" OD hot/cold PEX (cross-linked polyethylene) water tubing. The stuff is milkly white in appearance and is becoming available in building supply stores. This material was selected since it absorbs very little water and I'm guessing it's dielectric constant isn't affected too much by water absorbtion. The copper conductors which form the plates of the capacitor are 1/4" wide solid copper "tape" with an adhesive on one side. The tape is sold at craft stores and is normally used for stained glass projects. Two pieces are applied to the tube 180 degrees apart. Over this goes a layer of 2 mil polyethylene tape which serves as an insulator. A 1/2" length of copper tape is left exposed at one end of the sensor and to each is soldered some wires. Next is a layer of aluminum foil wrapped and "scrunched" in place. The assembly is then inserted inside a piece of 1/2" CPVC pipe. The aluminum foil serves to reduce the affects of stray capactance and the PVC pipe sheathing acts makes the sensor a bit more rigid. The aluminum foil is connected to the circuit's ground along with one of the copper foil tape plates. This seems to help the performance of the circuit by reducing the effects of stray capacitance (or so I guess from playing around with the circuit...). The inside of the poly tubing was waxed with an automobile wax so that water will tend to run off the tube's inner surface. The transducer was plumbed to the container in which the liquid level is to be measured with 1/8" ID vinyl tubing. A simple oscillator circuit using the ever popular LM55 timer produces a signal whose frequency is inversely proporational to the level of water in the transducer. This signal is easy to interface to a microprocessor or computer for process control.
I plan to use the transducer with my RIMS
(Recirculation Infusion Mash System) to monitor and control the
level of fluid in the RIMS during sparging.
I calibrated the transducer and plugged the frequency and level data
into a spreadsheet to arrive at the following equation:
Level = 973.911 - (0.39475 * Freq)Here's a graph of the data and the derived curve. I think there's a good correlation!
Here's the data:
Meas. Calc. Freq. Level Level Error (Hz) (In.) (In.) (In.) 2464.0 1 1.24 -0.24 2462.0 2 2.03 -0.03 2459.5 3 3.02 -0.02 2457.0 4 4.01 -0.01 2455.0 5 4.80 0.20 2452.0 6 5.98 0.02 2449.5 7 6.97 0.03 2446.5 8 8.15 -0.15 2444.0 9 9.14 -0.14 2442.0 10 9.93 0.07 2439.5 11 10.92 0.08 2437.0 12 11.90 0.10 2434.0 13 13.09 -0.09 2432.0 14 13.88 0.12 2429.0 15 15.06 -0.06 2427.5 16 15.65 0.35 2425.0 17 16.64 0.36 2421.5 18 18.02 -0.02 2418.5 19 19.21 -0.21 2416.5 20 19.99 0.01 2413.0 21 21.38 -0.38