This article is an excerpt. Download the complete Technical Note (TNTECM-012) with all figures, equations, and data tables in PDF format.
Download Full PDF
1. Introduction
In thermoelectric cooling (TEC) modules, the maximum achievable temperature difference DTmax is not a fixed constant — it depends on the absolute hot-side temperature Th at which the module operates. Engineers frequently observe that a TEC rated for DTmax = 68 C at Th = 27 C will reach roughly 76 C at Th = 50 C and over 96 C at Th = 100 C. The hardware has not changed; the physics simply favors warmer operation.
This article derives the relationship from first principles, then explains in plain language why higher Th produces a larger DTmax.
2. The Thermoelectric Energy Balance Equation
The net cooling capacity Qc at the cold junction is governed by three competing physical phenomena:
•
Peltier cooling — active heat pumping proportional to current I and cold-side absolute temperature Tc: Qpeltier = alpha * Tc * I. This term is the root cause of the Th-dependence.
•
Joule heating — parasitic resistive dissipation (I^2 * R) inside the pellets; one half flows back into the cold side.
•
Fourier conduction — parasitic thermal back-conduction K * DT from hot to cold side through the pellets.
The cold-side energy balance is therefore:
Qc = alpha * Tc * I - 0.5 * I^2 * R - K * DT (Equation 1)
where alpha is the Seebeck coefficient (V/K), Tc is the cold-side absolute temperature (K), I is the operating current (A), R is the electrical resistance (Ohms), K is the thermal conductance (W/K), and DT = Th - Tc.
3. Deriving DTmax — Optimum Drive Current
DTmax occurs when Qc = 0 — the operating point where all of the module's pumping capacity is consumed by its own internal losses. Setting Qc = 0 in Equation 1 and solving for DTmax as a function of I gives:
DTmax = (alpha * Tc * I - 0.5 * I^2 * R) / K
Optimizing the current by differentiating and setting d(DTmax)/dI = 0 gives the optimum current:
Iopt = alpha * Tc / R (Equation 5)
Note: Iopt scales linearly with Tc, so the optimum drive current itself rises with Th.
Substituting Iopt back and simplifying:
DTmax = 0.5 * (alpha^2 / (R * K)) * Tc^2 (Equation 6)
4. The Figure of Merit Z and the Cooling-Side Master Equation
The grouped material parameter alpha^2 / (R * K) is the thermoelectric Figure of Merit Z, with units of K^-1:
Z = alpha^2 / (R * K) (Equation 7)
Equation 6 collapses to the master form for the cooling side:
DTmax = 0.5 * Z * Tc^2 (Equation 8)
This is the central result of the derivation. It states that the maximum temperature difference scales linearly with Z and with the square of the cold-side absolute temperature. For Bi2Te3 at room temperature, Z is approximately 2.5 * 10^-3 K^-1.
5. Why Higher Th Helps — the Physics in One Sentence
Equation 8 is in terms of Tc, but engineers control Th. Because DT = Th - Tc, raising Th raises Tc as well — even at the maximum-DT operating point. And the master equation says cooling power scales with the square of Tc. So:
higher Th -> higher Tc -> more Peltier heat per amp -> larger DTmax
6. An Intuitive Analogy — Ants Carrying Heat Buckets
Picture each electron as a tiny ant carrying a bucket of heat from the cold side up a flight of stairs to the hot side. The bucket capacity is set by physics: each ant can carry an amount of heat proportional to the absolute temperature where it picks up the load (qe = alpha * Tc).
•
On a cold day (low Th, and so a low Tc), each ant's bucket is small. The ant struggles up the stairs and barely moves any heat per trip; the module's pumping margin is modest, and DTmax stays around 68 C.
•
On a hot day (high Th, and so a high Tc), each ant carries a much larger bucket. Same number of ants (same drive current I), but more heat per trip overcomes more parasitic loss before the cooling capacity is exhausted. DTmax climbs to roughly 92 C — even though the hardware is identical.
7. Expressing DTmax in Terms of Th
Equation 8 is in terms of Tc, but for engineering use we need it in terms of Th (which the heat sink fixes). Substituting Tc = Th - DTmax and expanding into a standard quadratic gives the physically meaningful root:
DTmax = Th + 1/Z - sqrt(2*Th/Z + 1/Z^2) (Equation 11)
Use Equation 11 for any quantitative work with real Bi2Te3 materials.
8. Numerical Example
Using Z = 2.5 * 10^-3 K^-1 (typical for Bi2Te3), Equation 11 gives the following values:
• Th = -20 C: DTmax = 51.05 C
• Th = 0 C: DTmax = 57.91 C
• Th = 25 C: DTmax = 66.87 C
•
Th = 27 C: DTmax = 67.60 C (Reference)
• Th = 50 C: DTmax = 76.22 C
• Th = 75 C: DTmax = 85.94 C
• Th = 100 C: DTmax = 96.01 C
• Th = 120 C: DTmax = 104.30 C
9. Practical Implications for System Designers
1
Specify DTmax at the actual operating Th, not the datasheet reference. Datasheet DTmax values are quoted at Th = 27 C by convention, but most real systems run hotter. Multiply by your datasheet value and use that for design.
2
Use Equation 11. The simplified parabolic form overshoots the exact value substantially for any modern TE material at typical Th.
3
Match Iopt to Th. If the system runs over a wide ambient range, set the controller current limit to track Iopt(Th) using a Th sense input. This squeezes out the last 5-15% of DTmax that a fixed current limit leaves on the table.
4
Keep Tc above the dew point if condensation is a concern. In humid environments, the cold side at DTmax can drop below the dew point and accumulate water. Run the module sub-DTmax, or seal/desiccate the cold-side enclosure.
10. Summary
The maximum temperature difference of a thermoelectric cooler is governed by the cooling-side master equation DTmax = 0.5 * Z * Tc^2. Because Tc rises with Th at the DTmax operating point, the achievable temperature span itself rises with Th. Each electron carries Peltier heat alpha*Tc per coulomb, so the 'bucket per ant' grows with absolute temperature; that is the whole story behind the Th-dependence.
Reference: Analog Technologies, Inc., Technical Note TNTECM-012.
