← Back to portfolio

Thermal energy storage · PCM · Lab experiment · KTH MJ2386

Latent Heat Mobile Thermal Energy Storage – RT60 HC PCM Discharge Lab

Experimental characterisation of a Mobile Thermal Energy Storage (M-TES) prototype using RT60 HC phase-change material and RENOLIN THERM 300X heat-transfer fluid. Discharge measured over 8 hours via 10 PT100 sensors; three calculation methods compared against theoretical capacity — Method 1 achieved 99.9% agreement.

M-TES PCM thermal lab visual

Experimental setup

The M-TES unit was characterised during discharge using two thermostatic oil baths regulating the HTF temperature, a coil heat exchanger (HX), an HTF reservoir and pump with frequency controller. Ten PT100 sensors (Chn 201–210) monitored temperatures along the M-TES unit from inlet (201) to outlet (210); a type-T thermocouple tracked the HTF reservoir buffer temperature (Chn 115). Data was logged via a Keithley acquisition system at 10-second intervals.

Cooling was provided by a cold water faucet when buffer tank temperature fell below 100°C, enabling controlled heat discharge. Flow was measured by an Omega flow meter.

PCM and HTF physical parameters

  • PCM — RT60 HC: phase change 60°C (solidification 62→51°C); density 0.75 kg/L (liquid) / 0.85 kg/L (solid); Cp = 2 kJ/(kg·K); latent heat 264.5 kJ/kg; total volume 44 L → mass 33 kg
  • HTF — RENOLIN THERM 300X: density 875 kg/m³ at 15°C; Cp = 2 kJ/(kg·K) at 50°C (temperature-dependent); flow rate 2 L/min = 3.33×10³ m³/s
  • Tank: stainless steel shell, 150 kg, Cp assumed 0.5 kJ/(kg·K)

Theoretical storage capacity

Three heat contributions were calculated for the discharge window (71.1°C → 40.1°C):

Qlatent = 8,728.5 kJDominant term: L × mPCM = 264.5 × 33 Qsensible solid = 719.4 kJCp × m × ΔT below solidification range Qsensible liquid = 601.3 kJCp × m × ΔT above solidification range

Total theoretical storage capacity: 10,049.2 kJ — latent heat accounts for 86.9% of the total, confirming the dominance of phase-change over sensible contributions in this operating window.

Sensor temperature behaviour

Key observations from the 10-sensor temperature-vs-time plot over 8 hours:

  • Inlet (Chn 201) drops first and most sharply — HTF entering at the lower oil-bath temperature drives the discharge front from the inlet.
  • Outlet (Chn 202) decreases steadily throughout, unlike all other sensors, reflecting integrated heat extraction across the unit.
  • Sensor pairs 203–204, 205–206, 207–208, 209–210 follow a stepped pattern: each pair drops sequentially with approximately 40-minute intervals, indicating a thermal front propagating through the PCM from inlet to outlet.
  • A sudden drop in the oil reservoir Chn 115 temperature was identified as a likely measurement error.
  • At 8 hours, all sensors converge to 40°C — confirming full thermal equilibration with the oil bath temperature and complete discharge.

Experimental heat transfer — three methods

Heat transferred from PCM to HTF was calculated by integrating the inlet–outlet temperature difference over the 10-second timestep sequence:

QHTF = Σ Cp,HTF × ΔT × V̇ × ρ × Δt,   where ΔT = TChn202 − TChn201

  • Method 1 (temperature-dependent ρ and Cp): Density and Cp varied with temperature using datasheet curves. Cp,HTF decreased from 2.033 to 1.945 kJ/(kg·K); ρ increased from 839.6 to 854.9 kg/m³ during the experiment. QHTF = 10,035.6 kJ
  • Method 2 (constant ρ = 825 kg/m³, Cp = 2 kJ/(kg·K)): QHTF = 9,814.6 kJ
  • Method 3 (including tank heat absorption): Qtank = 150 × 0.5 × 30 = 2,250 kJ. Total: 12,285.6 kJ (Method 1 + tank) or 12,064.6 kJ (Method 2 + tank)

Theory vs experiment comparison

Theoretical: 10,049.2 kJPCM sensible + latent heat sum Method 1: 10,035.6 kJ0.1% below theoretical — excellent agreement Method 2: 9,814.6 kJ2.3% below — constant-property simplification

Method 1 (variable properties) achieves 99.9% agreement with theory when comparing PCM-only contributions. This close match validates the theoretical model and confirms the sensor arrangement and integration approach are accurate.

The Method 3 values (12,064–12,286 kJ) exceed theoretical predictions because the stainless-steel tank absorbs 2,250 kJ of heat not included in the PCM-focused theoretical model. This is the largest identified discrepancy source, not measurement error. Minor additional contributors include heat losses to the environment and 10-second data-logging resolution.

Limitations and future work

  • Theoretical model does not account for tank heat absorption — including it would close the Method 3 gap.
  • Density and Cp functions were digitised from datasheet plots rather than tabular data; minor digitisation errors may affect Method 1 accuracy.
  • Single test run — repeated charge/discharge cycling would quantify PCM stability and degradation.
  • Only discharge characterised; a charge experiment under equivalent conditions would complete the round-trip efficiency picture.
  • No heat loss correction applied — adding insulation sensors or using a heat-loss model would improve the energy balance closure.

Relevance

Why this matters

This lab demonstrates the hands-on measurement side of thermal energy storage: setting up a real PCM discharge test, deriving the heat balance from sensor data using two property-handling approaches, and reconciling theory against experiment at 0.1% agreement. The workflow — theoretical capacity calculation, integration over measured time-series, sensitivity to property assumptions, and identification of unmodelled heat sinks — directly parallels the analysis required for TES prototype evaluation and heat exchanger commissioning in industrial settings.

The RT60 HC latent heat result (8,728.5 kJ vs 9,441 kJ sensible only) quantifies why PCM storage is so attractive for compact mobile TES applications: latent heat provides 87% of the total energy stored in a 31°C window, enabling significantly higher energy density than sensible-heat-only systems.

↓ Download lab report (PDF, 1.8 MB)