A Portable Water-Flow Calorimeter
This unique system permits on-site calorimetric evaluation of a wide variety of devices ranging in power from a few watts to hundreds of watts. The calorimetry is based upon the fact that 1 calorie of heat (4.186 joules) is required to raise 1 gram of water 1° C. By circulating water at a known flow rate through a heat exchanger wrapped around the device under test (DUT) and measuring the temperature rise that occurs in the water, the heat being dissipated by the DUT can be determined.
A constant delivery pump (FMI) circulates water around a loop that contains a Peltier heater/cooler, an inlet temperature probe, the heat exchanger, an outlet temperature probe, and an open reservoir. The Peltier heater/cooler, under control of the system computer, provides constant temperature water. The probes, DUT, and heat exchanger are well insulated to reduce heat losses to negligible levels.
Electrical power to the DUT is monitored and reported to the system computer by a Clarke-Hess 2330 Power Analyzer. This versatile instrument can accurately measure the power in arbitrary waveforms from DC to 400 kHz.
In order to measure the heat energy it must be transferred from the DUT to the circulating water. In general, a custom heat exchanger is fabricated by simply winding 1/4" Cu tubing around the device. In some cases the tubing can be applied directly to the exterior surface of the device as shown in the following pictures.
In other cases, particularly where the DUT must operate at a temperature substantially higher than the water, the tubing can be fashioned into a rigid shape that encloses the DUT completely. An example of such a heat exchanger is shown in the photo to the left. With the DUT completely surrounded by heat exchanger, an arbitrary amount of insulation can be placed between the DUT and the heat exchanger without adversely affecting the heat recovery. Of course, the thermal response time is affected by such insulation but this does not affect the accuracy of the calorimetric measurements.
After the DUT is enclosed in an appropriate heat exchanger, thick cotton insulation can be easily applied, even to irregular shapes, to provide an excellent thermal barrier. These photos show examples of the application of cotton insulation to completely enclose the DUT and Cu tubing heat exchanger.
The following image shows the appearance of the computer screen during a typical run. As the run progresses, current data are displayed digitally in the upper portion of the screen and selected parameters are plotted vs time in the lower portion. The traces are color-matched to the digital displays. For example, the electrical input power, Pin, is plotted in purple and the heat output power, Pout, is plotted in green. The vertical scales for each trace are typically different and are given by the numbers in parentheses that follow the digital display for each parameter. For example, both Pin and Pout are plotted on a vertical scale that runs from 9 watts at the top of the plot to -1 watt at the bottom. Thus zero watts occurs at the first division up from the bottom (marked by a solid grey line).
On this particular run, the input power was off for the first 2.8 hours to allow thermal equilibrium to be reached. You can see the Pout trace rise up and closely approach P = 0 (within about 0.1 watt). Then, approximately 5 watts of electrical power was applied to the DUT. The Pin trace responds instantly but the Pout trace responds slowly because of the thermal mass of the DUT. Within an hour, the Pout trace has responded fully and you can see that it matches the Pin trace almost perfectly (within 0.1 watt). At about 4.5 hours, the electrical input power was turned off but the calorimetry was kept running to collect the heat energy stored in the DUT. Note the Ein and Eout traces, which show the energy totals (obtained by integrating the power signals over time). As the Pout trace returns to zero, indicating that the DUT has cooled back to its starting temperature, the Eout trace rises to match the Ein trace, indicating a good energy balance for the run.
The DUT for this run was an ordinary resistor. However, the electrical input power was somewhat unusual. As shown in this scope trace, shorts bursts of 40 kHz sine waves occurring at a rate of about 60 per second were delivered to the resistor. During each burst, the power level was about 50 watts but the average power was only 5 watts. In fact, this run was conducted to demonstrate the accuracy of the Clarke-Hess 2330 power analyzer under highly non-uniform conditions.
The entire system is visible in the photograph below. From the right you can see the laptop computer, the Clarke-Hess 2330 power analyzer, the water system, and a small calibration DUT wrapped in insulation.
The following photo shows the entire system ready for transport (the water system is in the case on the left). These cases meet the size and weight restrictions for airline baggage. A typical field trip with this system can be completed in 2-3 days. Only a few hours are required to fabricate the heat exchanger and apply the insulation (an ample supply of which serves a dual purpose as packing material inside each case). The rest of the time can be spent making measurement runs, which usually last 4-8 hours each. If necessary, the system can be left running unattended overnight.
