The temperature tool is our oldest production logging sonde. Several types of sensors have been used in modern temperature tools; the three most common are briefly discussed below.RTDs
Most premium tools contain a platinum Resistance Temperature Detector (RTD*). The RTD exploits and optimizes the tendency of metallic conductors to exhibit a change in resistance with temperature. An RTD is essentially a temperature-sensitive resistor with a positive temperature coefficient, which means that the resistance of the metal increases with temperature. RTDs are made from a number of different metals, but the platinum RTD has captured most of the industrial market, and is now used in most modern temperature tools. A platinum RTD will perform over a very wide temperature range, and is inherently relatively linear. The major drawback is a rather small change in resistance per degree of temperature change, mandating special signal processing (amplification) in the "front end" of the tool. The GO / MLS Differential Temperature Tool (20 counts per degree) used a nickel RTD rather than platinum presumably because the resistance change per degree is somewhat larger; here is the Resistance versus Temperature Table (pdf file) for this oddball RTD.Thermistors
Platinum RTDs are available in the newer "thin film" or the older, and increasingly less common, "wire wound" constructions (wire wound platinum RTDs are not typically available in greater than 500 ohms). The reduced mass of thin film elements results in a faster response time than wire wound RTDs, making the thin film construction desirable. RTDs are made in different resistances, with the resistance usually specified at 0°C (32°F), known as the R0 value. 500 ohm wire wound or thin film and 1000 ohm thin film R0 platinum RTDs have been commonly used in temperature logging tools over the years since the higher values mean a greater change in resistance per degree (the most common industrial RTD is 100 ohms, but that value is not really suitable for downhole tools). The grade (purity) of platinum affects the RTD's temperature coefficient (a), which is the average slope of the RTD from 0°C to 100°C. DIN grade platinum RTDs produce a temperature coefficient of 0.00385 W/W/°C (±0.000012); this is the old European standard, and has more recently become the default standard in the USA. Reference-grade 99.999% pure platinum produces a temperature coefficient of 0.00392 W/W/°C; this was the industrial standard in the USA in the past but is not now as common as 385 slope platinum RTDs. 0.00375 W/W/°C slope platinum RTDs are also readily available. To describe platinum RTD accuracy, the International Standards IEC Publication 751 for DIN RTDs specifies two classes: Class A (high accuracy with an ice-point tolerance of ±0.06 W) and Class B (standard accuracy with an ice-point tolerance of ±0.12 W).
AnaLog Services, Inc. stocks RTD probes as well as replacement elements or capsules in a number of configurations and resistance values, both in thin film and wire wound constructions for repair or modification of customer's tools. See RdF's So, what is a Platinum RTD? for more information, and Minco's RTD Calc to generate tables for most common RTDs.
*The acronym "RTD" is variously used to represent Resistance Temperature Detector, Resistance Temperature Device, Resistive Temperature Detector, and Resistive Temperature Device.
Some tools contain thermistor sensors, metal oxide semiconductor devices. Both positive (PTC) and negative (NTC) temperature coefficient devices are manufactured. Thermistors provide a large change in resistance per degree, but they are highly nonlinear devices. There are a number of linearizing schemes, but in logging tools, where a dynamic range of 300 degrees Fahrenheit or more may be needed, thermistors are problematic. AnaLog Services, Inc. was asked to rework some old thermistor tools for use in shallow wells requiring a relatively narrow temperature range. We developed an Excel spreadsheet template based on the Steinhart & Hart equation, and using the classic three point single resistor linearization formula. Old thermistor based temperature tools, especially the old GO unijunction models, can thus be resurrected for some applications. But a better solution to resurrect old temperature tools is our remanufacturing process which includes the installation of a platinum RTD and new electronics, including replacement printed circuit boards (PCBs).Diodes
Diodes have been used as temperature sensors in a few tools; Comprobe and Bell manufactured diode based tools. Diodes produce a relatively high change in voltage per degree, but there are probably limitations with respect to higher temperatures. Burr-Brown's Application Bulletin AB-036, Diode-Based Temperature Measurement contains an interesting and useful discussion of diode based thermometers.
Finally this cautionary tale from the Wireline Reflector Mailing List:
"For years I have replaced the old sensors in temperature tool thermowells with new RTDs either for repair or upgrade of tools. I usually have to heat the thermowell with a propane torch to soften the epoxy many manufacturers used to pot the sensors in the little tubes. This is followed with a cleaning with appropriate sized drill bits, then rinsing using a little capillary tube and a spray can of perc, but I digress.
"Yesterday, I was doing one of these little jobs on an old GO thermistor tool. The thermistor was really glued in there, so I proceeded with the heat. I have had these things pop and carry on as the epoxy softens, but never has one actually fired the sensor out of the thermowell like a bullet. This thing made a deafening sound and shot the guts out of the thermowell with tremendous force. After it was all over, and we regained our composure (and found the thermowell on the opposite end of the shop), the entire shop stank of hot epoxy.
"Anyway, I mention this so that if any of you technicians ever do an RTD replacement, you will know to be cautious. We were just lucky neither of us were in the line of fire. We will be more careful doing this procedure in the future now that we know how much energy can be released. In case I have not been candid enough here, this little misadventure scared the shit out of both of us, and my ears are still ringing!"