Technical Notes These notes briefly define platinum and thermistor temperature sensors and signal-conditioning methods used with each. They also identify some common sources of measurement errors and performance problems. Where error magnitudes are noted, they are based on products manufactured by Logan Enterprises, Inc., and do not necessarily apply to those of other manufacturers.
Platinum Based Products
The most widely used platinum sensor has a resistance of 100 ohms at 0°C and an average sensitivity below 0.4 ohms per °C. It is necessary to reduce the effect of lead resistance to achieve accurate temperature measurement. For this reason most platinum RTD's are manufactured with three or four external leads, rather than the two needed for simple resistance measurements. The extra leads are used in various ways to perform the function.
Signal-Conditioning Four Wire Measurement- figure 1 This is the most effective method to measure a low resistance device. Two leads are attached directly to each end of the sensor and excitation current is passed through the sensor and one pair of leads. The voltage developed across the sensor is detected by a high impedance circuit using the second pair of leads. The result is an accurate measurement of the sensor's resistance. Since it is not necessary to determine the resistance of any of the leads, the circuit needs no compensation. Lead resistance is limited only by the voltage available to provide excitation current, allowing long runs from sensor to signal-conditioner with light guage leads. This approach is used in all precision bench-top meters and in some industrial instrumentation.
Three Wire Measurement-figure 2 This is the most common method used in industrial signal-conditioning devices. As in the four wire system, excitation current is passed through the sensor and a pair of leads. However, the voltage detection method differs somewhat. In this case, the voltage developed across the sensor and one excitation lead, Eout1, is detected. The voltage developed across the second excitation lead, Eout2, is also detected and is subtracted from Eout1. If the resistance of both excitation leads is equal, the circuit will provide an output proportional to the sensor's resistance. Any mis-match in resistance of the two excitation leads will result in a measurement error of about 0.1°C for each 0.04 ohm difference. For this reason, both external excitation leads should be of copper of identical wire gauge. | | Four Wire Compensating Loop This method, used in early devices, functions in similar fashion to three wire compensation. A 'loop' of wire is installed in the RTD along with the sensor's leads. Four leads run to the instrumentation. The instrument measures the resistance of the sensor plus its leads and subtracts the resistance of the 'loop' for compensation. While some of this equipment may still be in use, it has generally been replaced with three or four wire designs.
|
Moisture Damage and Prevention
In many applications- storage tank, refrigeration, and soil temperature monitoring are examples- water intrusion into the probe can be a severe problem. Although the connection area in the rear of most of our probes is imbedded in epoxy, the primary function of the epoxy is to anchor the leads, and it does not provide a good seal against moisture. If the probe is immersed or if its temperature is lowered enough to allow moisture to condense within it, its performance will quickly degrade. The short-term effects may be subtle- erratic or drifting outputs, or alarms in systems which monitor circuit-to-ground impedances. In the long-term, oxidation and other chemical reactions will occur and the probe, regardless of sensor style, will fail unpredictably. The following probe styles and options can prevent these failures from occurring.
Waterproof Probes The 4119 and 4120 are heavy-duty probes, designed originally for outdoor applications. They feature a neoprene jacketed cable and a vulcanized neoprene junction or tip. Both are rated for operation between -40 and 100°C, although the sensing end of a 4119 with a long sheath can withstand high temperatures. Both can be used for temperature monitoring in water tanks, lakes and streams, or soil. Because the sensor is imbedded in metal within the tip, they can be immersed to several hundred feet or can be buried without damage.
The 4159 and 4160 feature reduced junction and cable sizes. Although not as rugged as the 4119 and 4120 probes, the probes are easier to handle, and they provide installation options not available for the larger probes. The K Option If a probe doesn't need to be immersed or buried, this option offers greater temperature range and design flexibility than do the waterproof probes. It is designed to operate to cryogenic temperatures, and the probe is hermetically sealed. Although we don't rate the probe to high temperatures- the suggested operating range is -200 to 400°C, and the seal area is limited to 200°C maximum- our experience indicates that the design provides an extremely stable device over this temperature range. The option is available on most 4100 style probes and may be used with thermistors. | |
Vibration Wirewound sensors will withstand reasonable levels- to 30 g's or more- of low frequency vibration. This is sufficient to allow the use of standard probe configurations in most applications. Where higher levels of shock or vibration occur, it may be necessary to provide additional protection or to use a more rugged sensor.
The VI Option
In this option, we combine a method of reducing movement of the probe's internal leads with, in the larger diameter probes, a fiberglass sleeve over the internal assembly. This increases resistance to shock and vibration while retaining the high temperature capability of the probe.The use of this option increases the thermal response time and reduces the pressure capability of the 3/16 inch and 1/4 inch diameter probes. It has no detectable effect on the performance of the 1/8 inch diameter probe.
Where very high levels of vibration or shock are expected, the best choice may be to use a platinum film sensor with the VI option. This sensor has a much lower mass than the wirewound sensor, and the sensing element is a glass-encapsulated film rather than a fine wire. It does have certain disadvantages- we don't recommend its use above 400°C, it's available only in the .00385 ( DIN ) alpha, and it is not as accurate as a wirewound sensor at elevated temperatures.
Thermistor Based Products
The thermistors most commonly used for temperature measurement are sintered mixtures of metallic oxides with large negative temperature coefficients of resistance- typically on the order of -4%/°C. Because of their relatively high resistances and high sensitivity, they do not require lead resistance compensation. Their small size- typical diameters of 0.12 inch down to 0.005 inch- make them suitable for miniaturization. They are used in applications ranging from cryogenic temperatures to 400°C. The most precise of these are more accurate than many platinum sensors in the temperature range 0 to 100°C, and they are generally less expensive than platinum sensors of equivalent accuracy. Individual device specifications are usually stated over fairly limited temperature spans, with typical accuracies for precision thermistors of ± 0.1°C to ± 0.2°C between 0 and 100°C. Accuracy degrades rapidly outside the specified range. Many indicators, controllers, and hand-held thermometers are available for operation with thermistors. Because of non-linear characteristics and a lack of standardization, instrumentation is usually designed for fixed ranges and often for a specific manufacturer's thermistors. |
