Probing the Unseen World

Ultra-fast response temperature probe allows a deeper exploration of heat and temperature

by Stephen Bannasch stephen@concord.org

• Introduction
• Thermal Conductivity From Our Skin's Perspective
• Broadband Radiometer and Thermal Radiation
• Hot Hands and Thermal Convection
• Pressure, Volume, and Temperature
• Endothermic Cooling
• The Thermodynamics of a Rubberband
• Transformation of Kinetic to Thermal Energy
• Making an Ultra Fast Response Temperature Probe
• Response Times of Temperature Probes
• For More Information

A shorter version of this article appeared in the Fall issue of our newsletter @Concord, http://www.concord.org/newsletter/2002-fall/probeware.html.

Introduction

When students use both sensor and model-based computer visualizations, will they gain a deeper understanding of the difficult concepts of heat and temperature? Can we extend that learning experience to the investigation of conductivity and thermal gradients in different materials? Can we go even further to include the concepts of radiation and convection, the subtleties of which are far more difficult to grasp? These are some of the questions driving our research with middle school students as we work with them to conduct hands-on experiments in temperature and heat flow.

Through our many years of work on models and probeware, we have encouraged students to use these tools for understanding the world around them. Now we are excited about a new development that will help them understand natural phenomena on a much deeper level.

For the Data and Models project, we have created an ultra-fast response temperature probe (on the right in Figure 1). For the TEEMSS project we have created a variation that is almost as fast (on the left in Figure 1). This new generation of highly sensitive probeware enables students to easily conduct investigations into heat and temperature phenomena that we only imagined possible before.

Figure 1

The simple experiments described in this article are either impossible or impractical with the temperature probes currently used in schools. A probe that responds this quickly will allow the kind of deep explorations that help students understand heat and temperature as indicators of physical and chemical processes. Research has shown that students who use data from probes to analyze experimental results gain a deeper understanding of science concepts. We must supply them with the most appropriate probes for the tasks.

Thermal Conductivity From Our Skin's Perspective

The first heat and temperature sensor most of us use is our skin. When we make a snowball in the winter or walk over hot beach sand in the summer our brain uses information relayed by nerve cells under our skin to decide whether something is "too hot", "too cold," or "just right." These qualitative judgments are based on the temperature of the nerve cells and how quickly that temperature is changing.

When you quickly let go of the handle of a pot that is too hot to hold, your brain has interpreted the quick rise in temperature sensed by the nerve cells under the skin as an alarm. In effect, your fingers sense the rate of heat flow into your fingers. A much greater alarm is elicited by an object that raises your skin temperature to 50 °C in one second than by an object that does the same in 30 seconds.

A classic heat and temperature "misconception" many students have is that metal is colder than wood. For example, take three similar small blocks of aluminum, wood, and rigid foam as shown below in Figure 2. Even though they are the same temperature the aluminum block will feel cool to the touch while the foam block feels warm.

Figure 2

The graph in Figure 3 measures the surface temperature of a finger (32 °C) as it is placed in contact with rigid foam, wood, and aluminum blocks at room temperature (24 °C). When touching the foam and wood blocks, skin temperature reaches a warmer equilibrium; however, when touching the aluminum block, skin temperature quickly dropped three degrees and was still cooling after 30 seconds. Heat from the finger was able to more quickly flow throughout the aluminum block because of aluminum's greater thermal conductivity when compared to wood or foam.

Figure 3

The ultra-fast response temperature sensor allows rapid investigations that help students tease apart the concepts of thermal conductivity and heat capacity in different materials. While the blocks are the same temperature, the graph in Figure 2 clearly shows skin temperature getting colder when touching the aluminum block. Yet, from our finger's point of view, the aluminum block is clearly colder because, after the initial warming of the aluminum surface, the heat from our finger continued to flow quickly into the aluminum block.

In order to help students develop more robust mental models we combined sensor and model-based visualizations in the software the students used. Students created simple virtual thermodynamic systems on the iPaq handheld computers using blocks of different thermal conductivities along with heaters and coolers. The model shows the temperature gradient within the blocks. You can also use the Heat Flow applet on our Web site to experiment with a similar modeling tool. A screen shot is below in Figure 4 showing heat conducting from a heater across a simplified model of a finger. See Monday's Lesson: Modeling Heat & Temperature on pages 6 and 7 of the Fall 2002 @Concord newsletter for an introduction to that tool.

Figure 4

Broadband Radiometer and Thermal Radiation

We also encourage students to investigate the thermal effects of radiation on the objects around them. The surface of the probe's sensing tip reflects both visible and near-IR radiation. The temperature rise when we hold the probe within six inches of a 100 watt incandescent lamp is less than 0.1 °C. The radiant energy from the lamp does not warm the probe as you might expect, because of the small surface area and the fact that the incoming radiation is mostly reflected rather than absorbed.

In order to experiment with the thermal effects of radiant energy we devised a simple instrument we call a "radiant lollipop" (Figure 5). It is, in effect, a broadband radiometer.

Figure 5

The small piece of rigid foam insulation is covered on one side by regular weight aluminum foil with the shiny side up. The other side is covered by aluminum foil painted flat black with high-temperature stove paint. The sensing tip of the probe is carefully slid behind the aluminum foil. Students find that it is quite difficult to raise the temperature of the shiny side of the lollipop. However, the black side is exquisitely sensitive to incoming radiation.

The equilibrium temperature reached by the black side is affected by two phenomena. These are the heat transfer due to conductivity with the surrounding air and the balance of emitted and absorbed radiant energy. When the black foil is at the same temperature as nearby emitting surfaces (such as the walls in a room) the radiant energy emitted by the black foil is closely balanced by the energy absorbed. This causes no net change in the temperature of the foil. However if the black foil is pointed at an object warmer than it such as the sun, a bright lamp, or even a warm hand, the radiation balance is shifted and for a time the incoming radiation will be greater than the radiation emitted. This causes the temperature of the foil to increase until the heat lost from both conduction to the air and increasing emittance balances the increased energy absorbed from radiation.

The radiant lollipop can also be used to indicate the lack of radiant energy. During a clear night the radiant energy coming from the night sky is very small. By taking the radiometer outside and first pointing it at the ground and then at the sky it indicates that the sky is cooler than the ground (Figure 6).

Figure 6

Hot Hands and Thermal Convection

The ultra-fast response probe can also be used for measuring the small, and often ephemeral, temperature differences associated with convective flow. Convective heat transfer is the movement of a gas or liquid propelled by buoyancy. A simple experiment, shown below in Figure 7, has a student firmly hold a copper tube in her hand and measure the change in air temperature above the tube. Within seconds of grasping the copper tube, her hand warmed the tube and the air inside. This lighter air then flowed up the tube and was detected by the probe.

Figure 7

The graph in Figure 8 shows the air temperature directly above the copper tube.

Figure 8

With this probe students can measure the thickness of a film of air falling down a cold window, the extent of a plume of hot air leaving a television, or even the effect of pouring cold air from a chilled bowl into another bowl.

Pressure, Volume, and Temperature

We can also plot the increase in air temperature caused by increasing pressure. We placed the sensor in a sealed plastic soda bottle and squeezed it hard reducing its volume and increasing the pressure. The graph in Figure 9 shows the air temperature rising 14 °C. As soon as the pressure was removed the temperature dropped back close to room temperature. In this experiment temperature variation is caused by changes in density, while in convection the density varies due to changes in temperature.

Figure 9

Endothermic Cooling

The fast response and low thermal mass of the probe also allow easy measurement of the endothermic cooling when a packet of sugar or salt is dissolved in water.

Figure 10

The plot in Figure 10 above shows the endothermic cooling associated with the dissolving of sugar crystals in water.

Figure 11

The plot in Figure 11 above shows the endothermic cooling associated with the dissolving of salt crystals in water.

The Thermodynamics of a Rubberband

Figure 12

The graph above in Figure 12 displays the temperature of a rubber band as it is stretched and then relaxed. The probe was taped to a rubber band, at 20 s the rubber band was stretched to approximately four times its relaxed length. At 40 s tension on the rubber band was released and the rubber band relaxed back to its resting length.

Transformation of Kinetic to Thermal Energy

Figure 13

This experiment involved smashing a board with a sharp hammer blow positioned so that it will squash flat a ball of clay around the temperature probe as shown in Figure 13 above. After flattening, the clay rose approximately one degree as shown in Figure 14 below. I speculate that the cause of the 2.5 °C peak at the instant of the blow was caused by the friction of the layers of clay sliding past each other.  Others have speculated that the transient peak was caused by the physical shock to the thermocouple.

Figure 14

Making an Ultra Fast Response Temperature Probe

The probe developed for the Data and Models project uses Type E thermocouple wire made of Nickle-Chromium and Copper-Nickle alloys 0.005" in diameter. The two ends of the thermocouple wire are spark-welded together to create an extremely low thermal mass temperature sensor. In still air the sensor will return to 90% of equilibrium temperature in about 10 seconds. This response is more than an order of magnitude faster than normal temperature probes. The calibrated range is -50 to 110 °C with a resolution of 0.1 °C. The uncalibrated range of the sensor extends from -200 to 900 °C. The cold junction of the thermocouple is algorithmically compensated with an analog output IC temperature sensor. At 20 °C Type E thermocouples have a response of 60 micro-volts per °C. We measure the response of the thermocouple and the IC temperature sensor directly with the Linear Technology LTC2402 2-channel 24-bit analog-to-digital converter.

The TEEMSS project developed a very similar fast response temperature probe however the thermocouple wire was 0.010" in diameter. This made the probe more robust. In addition because of production limitations instead of spark welding the ends we soldered the twisted ends with tin solder. Both of these changes affected the response time of the probe. The TEEMSS probe will return to 90% of still air equilibrium in about 20 seconds.

Response Times of Temperature Probes

Both of these temperature probes have a much faster response than temperature probes now commonly in use. In the table and graph below I compare our fast response probes to the sampling of temperature probes available from Vernier and Pasco in 2004. In 2004 Vernier redesiged their Surface Temperature Sensor and replaced the slower responding earlier model with a small exposed glass-encased thermistor sensor. In Nov 2004 I tested the newer model of this probe in which the small thermistor is uncoupled to any other thermal mass.  This probe has the best thermal response yet tested for commercially available probes.

The table below shows the time to return to 90% of still air equilibrium temperature for each temperature probe. Total temperature difference was approximately 10 °C. The probes were initially warmed by holding them between two fingers. Figure 15 shows a graph of the responses for each temperature probe.

Figure 15

For More Information

For more details of these and other interesting experiments including information on how we made this probe, as well as our research with middle school students visit the Data & Models project Web site. Data & Models project Web site: http://www.concord.org/data-models/

To find out about the interface hardware and the open source software CCProbe that allows probe visualization on PalmOS, PocketPC, MacOS, Windows, and Linux go to the CC Probeware Web site: http://www.concord.org/ccprobeware/

The Spring 2001 issue of @CONCORD, we described our thermal conductivity block system for investigating heat flow: http://www.concord.org/newsletter/2001spring/evidence.html

To read about our research in the middle school classroom, see Barbara Buckley's article, Investigating Heat Flow at the Hot & Cold Club: looking closely at students' mental models , on our Web site: http://www.concord.org/data-models/

A shorter version of this article appeared in the Fall issue of our newsletter @Concord, http://www.concord.org/newsletter/2002-fall/probeware.html.