THERMAL LAB · ME-08 · METHODS
Method: side-by-side comparison
Most design decisions are choices between two options. A side-by-side comparison answers them directly, provided only one thing differs between the two sides. Getting that condition right is the whole method.
In this section
- ME-01 How we test
- ME-02 Method: thermal imaging
- ME-03 Method: contact temperature sensing
- ME-04 Method: power and energy measurement
- ME-05 Method: warm-up curves
- ME-06 Method: steady state and stability
- ME-07 Method: temperature uniformity
- ME-08 Method: side-by-side comparison
- ME-09 Method: cold conditions testing
- ME-10 Method: surface temperature limits
- ME-11 Method: repeatability and uncertainty
- ME-12 Standards we reference
- ME-13 Reading a thermal report
One variable
If two heating elements are compared in different stacks, at different voltages, on different days, the result says nothing about the elements. A fair comparison holds everything constant except the item in question.
| Held constant | How |
|---|---|
| Ambient conditions | Same room, same time where possible |
| Start temperature | Both samples rested together |
| Layer stack | Identical materials in identical order |
| Electrical input | Same measured power, or same voltage, stated which |
| Sensor positions | Matched and photographed |
| Duration | Same length of run |
Same power or same voltage
Two elements with different resistance draw different power at the same voltage. Both comparisons are valid and they answer different questions. At equal power, the comparison shows which element uses energy better. At equal voltage, it shows what each would do on the same battery. The report states which was used.
In one frame
Where size allows, both samples are imaged together. The camera, the background and the moment are then shared, and differences in the image are differences in the samples.
Swap and repeat
Position can bias a result. One side of a bench may be nearer a draft. We swap the samples and repeat. If the difference follows the sample, it is real. If it stays with the position, it is the bench.
Reporting
A comparison report shows both curves on the same axes, both images at the same scale, and a table of the figures that matter for the decision. It also says how large a difference would have to be to count, given the repeatability of the method.
What comparisons are for
- Choosing between heating elements, insulation or heat-reflective materials.
- Checking a new supplier's sample against the current one.
- Confirming that a cost reduction did not cost performance.
- Checking a new production lot against a reference.
A data sheet cannot do this. See what a data sheet does not tell you.