Why chamber temperature is not enough
The air in a chamber reaches the programmed value before the sample does. The sample heats or cools at a rate that depends on its mass, specific heat, conductivity and exposed area, and on heat transfer with the air (airflow, convection, radiation). The dwell time of a test must be counted from when the sample, not the chamber, reaches the temperature.
A useful model
If the sample is small or conductive enough to be considered uniform in temperature, its temperature approaches the chamber value along an exponential curve characterized by a time constant. That constant grows with mass and heat capacity and decreases with the exchange area and the heat transfer coefficient. After several time constants the difference is already small. If the sample is large or a poor conductor, there is an internal gradient and the center takes longer than the surface, so the model must be applied with care.
How it is determined in practice
- Instrument the sample with sensors (thermocouples or RTDs) at the point expected to be slowest, such as the center of a mass or a component with high thermal capacity, and another on the surface.
- Run a characterization test with the same profile, chamber load, position and airflow as the actual test.
- Record the temperature of the sample against time.
- Define the criterion: that the core temperature is within the tolerance band around the target value and that its variation over an interval is less than an agreed value.
- Establish the observed time, with a margin covering variation between samples and conditions.
Factors that change it
- The number of samples in the chamber, which reduces air circulation and adds thermal mass.
- Packaging, supports and fixtures, which act as insulators or thermal bridges.
- Energized equipment, which self-heats and does not tend toward the air temperature.
- The chamber ramp, which may be slower than the response of the sample.
What the engineer sees in practice
Two curves: that of the chamber, which rises and stabilizes, and that of the sample, which follows it with a delay. The stabilization time is the interval between the two, and it is noted in the procedure so that the test can be repeated with comparable results.