2.1 Temperature Scales and Conversions
Key Takeaways
- Celsius and Fahrenheit are relative (interval) scales; Kelvin and Rankine are absolute scales starting at absolute zero
- Convert with °F = (°C × 9/5) + 32, °C = (°F − 32) × 5/9, K = °C + 273.15, and °R = °F + 459.67
- Radiometric and Stefan–Boltzmann calculations require absolute temperature (K or °R), never Celsius or Fahrenheit directly
- A 1 °C difference equals 1 K and 1.8 °F (or 1.8 °R)—convert ΔT by the degree-size ratio only, without adding 32 or 273
- Level II reports must state units clearly; NETA ΔT and ASHRAE envelope criteria often use °C or °F depending on the standard
2.1 Temperature Scales and Conversions
Quick Answer: Infrared thermographers work daily in °C and °F for reporting, but quantitative radiation laws use absolute scales—Kelvin (K) or Rankine (°R). Convert with °F = (°C × 9/5) + 32, °C = (°F − 32) × 5/9, K = °C + 273.15, and °R = °F + 459.67. Never plug Celsius or Fahrenheit directly into T⁴ power calculations.
Temperature is the physical quantity that thermography exists to measure, yet many Level I operators treat the camera’s number as a self-evident fact. Level II work demands more: you must convert scales fluently, know when a reading is on a relative versus absolute scale, and recognize how unit choice affects ΔT severity tables, baseline trending, and any calculation involving radiant power.
Relative Scales: Celsius and Fahrenheit
Celsius (°C) and Fahrenheit (°F) are relative (interval) scales. Their zeros are historical conventions, not the complete absence of thermal energy:
- Celsius sets 0 °C at the freezing point of water and 100 °C at the boiling point of water at standard atmospheric pressure.
- Fahrenheit sets 32 °F at water freeze and 212 °F at water boil under the same conditions, so the interval between freeze and boil is 180 °F degrees.
Because the zero points are arbitrary, a value of “0” on either scale does not mean “no heat.” A steel bus bar at 0 °C still radiates infrared energy. For trend work and severity classification, relative scales are fine—as long as every reading and every standard uses the same scale.
Conversion Formulas (Memorize Both Directions)
| Direction | Formula | Memory aid |
|---|---|---|
| °C → °F | °F = (°C × 9/5) + 32 | Multiply by 1.8, then add 32 |
| °F → °C | °C = (°F − 32) × 5/9 | Subtract 32, then multiply by 5/9 (~0.556) |
| °C → K | K = °C + 273.15 | Often rounded to +273 in field estimates |
| °F → °R | °R = °F + 459.67 | Often rounded to +460 in field estimates |
| K → °R | °R = K × 9/5 | Same ratio as °C/°F degree size |
| °R → K | K = °R × 5/9 | Inverse of above |
Interval rule (critical for ΔT work): A temperature difference of 1 °C equals a difference of 1 K, and a difference of 1 °F equals a difference of 1 °R. Converting a delta does not add or subtract 32 or 273:
- ΔT_°F = ΔT_°C × 9/5
- ΔT_°C = ΔT_°F × 5/9
Example: NETA-style Priority 3 often cites an >10–20 °C rise versus a similar component under load. That band is 19.8–36 °F, not “11–20 plus 32.”
Absolute Scales: Kelvin and Rankine
Kelvin (K) and Rankine (°R) are absolute thermodynamic scales. Zero on either scale is absolute zero—the theoretical point of minimum thermal energy (−273.15 °C or −459.67 °F). Absolute scales have no negative values in normal engineering practice.
| Scale | Zero point | Degree size matches | Common IR use |
|---|---|---|---|
| Kelvin (K) | Absolute zero | Same as °C | Radiometry, Planck, Stefan–Boltzmann |
| Rankine (°R) | Absolute zero | Same as °F | Some U.S. engineering texts |
| Celsius (°C) | Water freeze | Same as K | Reports, NETA ΔT, most cameras |
| Fahrenheit (°F) | (offset) water freeze | Same as °R | U.S. building / facilities reports |
Why absolute scales matter in infrared:
- Stefan–Boltzmann law states that total radiant exitance of a blackbody scales with T⁴ where T is absolute temperature. Using 80 °C instead of 353 K produces a wildly wrong power estimate.
- Wien’s displacement law relates peak wavelength to absolute temperature (λ_max · T ≈ constant).
- Object–background contrast in quantitative models depends on absolute temperatures of the target and surroundings, not on whether the display is set to °C or °F.
Level II cameras still display °C or °F because those are readable for humans. Internally, radiometric engines convert to absolute temperature for calculations. Your job is to know when the report, the spreadsheet, or a hand calculation must switch to K or °R.
Worked Examples Relevant to IR Work
Example 1 — Motor winding hotspot (report units)
A Level I report lists a motor connection at 185 °F. The client’s maintenance standard is written in Celsius. Convert:
°C = (185 − 32) × 5/9 = 153 × 5/9 = 85 °C (exactly).
If a similar phase connection runs at 70 °C, ΔT = 15 °C = 15 × 1.8 = 27 °F. Severity tables must use one consistent scale for both the reading and the criterion.
Example 2 — Absolute temperature for radiation comparison
Two surfaces measure 40 °C and 80 °C. Naively, 80 looks “twice” 40 on a relative scale—but radiation scales with absolute temperature to the fourth power.
- T₁ = 40 + 273.15 = 313.15 K
- T₂ = 80 + 273.15 = 353.15 K
- Ratio of blackbody exitance ≈ (353.15 / 313.15)⁴ ≈ (1.128)⁴ ≈ 1.62
The hotter surface radiates about 62% more power as a blackbody, not “twice as much.” Level II quantitative thinking starts with this conversion habit.
Example 3 — Ambient and reflected ambient on mixed-unit jobs
You measure reflected apparent temperature (RAT) with crumpled aluminum foil as 72 °F. Your camera’s atmospheric-temperature field is set in Celsius. Convert before entry:
°C = (72 − 32) × 5/9 = 22.2 °C.
Entering 72 into a °C field would force the camera’s atmospheric and reflected corrections into nonsense territory and can shift indicated target temperatures by several degrees—especially on low-emissivity metals.
Example 4 — Building envelope ΔT (ASHRAE-style condition)
A common rule of thumb for meaningful building-envelope IR is roughly 10 °C (18 °F) interior-to-exterior air temperature difference. If outdoor air is 35 °F and you need at least an 18 °F ΔT, indoor air should be about 53 °F or higher for a heating-season survey—or you wait for colder weather. Confusing the 10 °C criterion with “10 °F” understates the required gradient by almost half and can produce false “no problem” surveys.
Choosing Units on Level II Deliverables
| Context | Preferred unit | Notes |
|---|---|---|
| Electrical NETA-style ΔT tables (many U.S. practices) | °C (sometimes dual) | Confirm client standard; never mix in one table |
| U.S. facilities / HVAC client reports | Often °F | Dual-unit (°C / °F) improves clarity |
| Scientific radiometry, research, camera SDK math | K | Required for T⁴ and Planck functions |
| International manufacturing / ISO-oriented plants | °C | Default SI reporting |
| Steam systems (gauge vs absolute pressure context) | °C or °F for temp | Do not confuse temperature scale with pressure “psia/psig” |
Professional habit: State the unit on every annotated image, every table column header, and every severity call. “Hotspot = 95” is not a quantitative finding.
Common Conversion Traps on the Exam and in the Field
- Adding 32 to a ΔT — Differences convert by the 9/5 factor only.
- Using °C in T⁴ — Always convert to kelvin (or rankine) first.
- Camera unit mismatch — Emissivity is unitless, but RAT, atmospheric temperature, and ambient fields inherit the camera’s selected scale.
- Rounding absolute zero — +273 or +460 is acceptable for rough field estimates; use 273.15 / 459.67 for report-grade hand calcs.
- Negative Celsius outdoors — Valid on relative scales; convert carefully (e.g., −10 °C = 14 °F = 263.15 K).
Practice Drill (Mental Fluency)
Work these until they are automatic:
| Given | Find |
|---|---|
| 100 °C | 212 °F, 373.15 K |
| 0 °F | −17.8 °C, 255.4 K |
| 25 °C ambient | 77 °F |
| ΔT = 35 °C (Priority 1 style threshold in some NETA tables) | ΔT = 63 °F |
| 300 K scene | 26.85 °C ≈ 80.3 °F |
Level II competence is not memorizing trivia—it is refusing to let unit confusion corrupt severity classification, camera parameters, or radiation math.
Summary for Level II
Temperature scales are the language of thermometry. Relative scales (°C, °F) dominate client communication and ΔT criteria. Absolute scales (K, °R) dominate physics. Convert differences with the degree-size ratio alone; convert absolute values with the full offset formulas; and always declare units on quantitative infrared work.
A thermographer measures a bus connection at 90 °C. What is the equivalent temperature in Fahrenheit?
Why must Stefan–Boltzmann radiant-power calculations use Kelvin (or Rankine) rather than Celsius?
A similar-component temperature rise is reported as ΔT = 18 °C. What is the equivalent rise in Fahrenheit degrees?
Which pair correctly describes absolute versus relative temperature scales used in thermography?