Thermal analysis is a family of techniques that share a heating programme and answer quite different questions, which is why a laboratory that buys the wrong one uses it twice a year. One measures heat flow and finds transitions and melting; one measures mass and finds water, solvent and decomposition; one measures dimension or modulus and finds softening under load. The question decides the instrument, and in every case the calibration, the atmosphere and the sample container do more to the answer than the brand does.
What thermal analysis instruments measure, technique by technique
Differential heat flow finds anything with an enthalpy: melting, crystallisation, a glass transition, a solid form change, and it is the technique a formulation question usually needs. Mass against temperature finds what leaves the sample and at what temperature, which separates surface water from bound water and shows where decomposition starts. Mechanical and dimensional methods apply a load or measure expansion and report softening and modulus against temperature, which is a materials question rather than a chemical one. Most laboratories need one of the first two and think they need all three.
Where a calorimeter instrument fits, and where an isothermal one is the honest answer
A scanning calorimeter ramps the temperature and reports heat flow against it, which finds transitions quickly and says little about slow processes. An isothermal instrument holds a temperature and measures heat over days, which is how a stability or a compatibility question is actually answered, because the reaction you care about at room temperature may be invisible in a scan. They are different purchases and the second is often the one a formulation group should have bought. Ask which of the two questions the laboratory asks most often before comparing specifications.
The pan, the atmosphere and the calibration do most of the work
A hermetic pan keeps volatiles in and moves a melt; an open pan lets water go and changes the baseline; a pierced lid is a compromise chosen deliberately. Nitrogen against air changes where oxidation appears by tens of degrees. And the temperature and enthalpy calibration, run with certified metals at the rate you will use, is what makes a number comparable to anybody else's. A result reported without the pan, the atmosphere and the heating rate is not reproducible, and it is worth insisting on all three in the method rather than in a footnote.
Buying one against sending the samples out
The instrument is not expensive by laboratory standards and the expertise is, which is the opposite of how these purchases are usually argued. Interpreting a thermogram is a skill that decays without use, so a laboratory running a handful of samples a year gets better answers from a contract laboratory that runs hundreds. Buy one when the sample count is steady, when the material cannot leave, or when the turnaround decides a development cycle, and budget the training and the annual calibration alongside the purchase.
A thermal conductivity tester and what it reports
Conductivity is a transport property rather than a thermal event, so it is measured on different instruments from the ones that find transitions: a steady state method drives a known heat flow across a sample of known geometry and reads the gradient, and a transient method puts a pulse or a line source into the sample and fits the response. The steady state route is slower and traceable more easily; the transient route is quick, needs less material and is more sensitive to contact between the probe and the sample. Either way the number depends on density, moisture and the contact as much as on the material, which is why a value quoted without the method and the sample condition beside it cannot be compared with another laboratory's. If the question is insulation or heat spreading performance rather than a transition temperature, this is the measurement to buy and a calorimeter will not answer it.
Questions people ask about thermal analysis equipment
Which technique for a glass transition?
Differential heat flow first, and a mechanical method when the transition is weak or the sample is filled, because the mechanical signal is far larger there. A modulated or a stepped heating programme helps on a faint transition, and the heating rate has to be reported either way because the value moves with it.
Can one instrument do heat flow and mass together?
Simultaneous instruments exist and they are a real convenience on a decomposition study, because one sample gives both curves on one temperature axis. The heat flow sensitivity is lower than a dedicated calorimeter's, so if transitions are the main question the dedicated instrument is better.
How often does it need calibrating?
Temperature and enthalpy against certified standards on a stated interval, and a check whenever the furnace, the sensor or the purge is disturbed. The interval belongs in the laboratory's own programme rather than the manual, and the check is cheap next to a set of results nobody can defend.