Fridge, Freezer and Mini-Split Solar & Power Station Calculator

Annual-label mode divides the entered kWh/year by 365. For measured total electricity, enter representative daily kWh times 365; that route includes defrost, ice making and other metered auxiliaries. Measured-cycle mode integrates $P_{off}+(P_{run}-P_{off})D$ over 24 hours and requires confirmation that these two states account for all daily energy. It is unsuitable for unaccounted defrost or auxiliary loads. Maximum non-start watts affect inverter sizing only.

Measured-cycle duty is adjusted by $D=D_{ref}(T_{amb}-T_{cab})/(T_{ref}-T_{cab})$ and capped at unity. This sensitivity is derived from constant cabinet conductance, cooling capacity and COP, not fitted to a refrigerator data set.

For the inverter mini-split, on-state electrical input is rated thermal capacity times the larger of demand fraction and minimum modulation, divided by operating-point COP. Below minimum modulation the duty is demand/minimum fraction. Outside the entered daily cooling window only appliance standby remains. Standby is already included in on-state watts. Compressor periods are fitted to an integer number of cycles within the window.

Average DC demand is average AC demand divided by inverter efficiency plus continuous DC idle draw. Available battery energy is nominal Wh times initial SOC above the reserve floor. No-sun endurance is available Wh divided by average DC W. Required nominal storage assumes starting full and discharging to reserve.

Running and maximum non-start apparent powers are calculated separately: $S_{run}=P_{run}/PF_{run}$ and $S_{peak}=P_{peak}/PF_{peak}$. Continuous inverter VA must cover the larger, plus the other steady load. If maximum-load PF is unknown, clear its availability checkbox: the inverter screen then stays incomplete. Starting VA comes from known starting current times voltage, or running VA times an entered current multiple. Continuous W, continuous VA, surge VA and duration are screened separately. Surge VA also provides a conservative upper bound on real starting watts; exceeding that bound does not prove an actual start fails.

PV nameplate sizing assumes all load energy passes through charging, conservatively dividing by charging efficiency, PV derate and peak sun hours. Recovery adds the requested backup energy divided by recovery days. Inspect the timeline for input clipping, battery-full curtailment and unmet demand.

For dispatch, accepted PV is piecewise linear between five-minute knots on the illustrative solar curve. Exact compressor switching times and cooling-window boundaries are merged with these knots. Each segment is split again wherever PV equals DC demand. Positive and negative energy flows are integrated separately; charging efficiency applies only to stored surplus. Full-charge and reserve hits are solved within each segment and recorded in the battery and cumulative-energy traces. Later surplus cannot erase earlier unmet demand. Only the power chart uses five-minute load and PV averages. The animated schematic samples the event results without recomputing dispatch.

One cooling appliance plus a steady unity-power-factor AC load. All quantities are user scenarios, not product presets. Default 475 kWh/year, 150 W running draw and a starting-current multiple of five are illustrative and must be replaced for equipment selection. Default maximum-load PF 0.8 is an explicit illustrative assumption. Clear the PF availability checkbox if no data or justified assumption is available. No soft-start or inverter-compressor surge reduction is assumed.

Annual-label cycles are energy-equivalent illustrations; they do not reconstruct a real compressor or defrost schedule. Label-mode temperature is not adjusted because the test climate and cabinet conductance are unknown. The measured-cycle temperature model assumes a single compartment and unchanged COP; combined fridge/freezer heat loads and high-ambient capacity loss require measurements. Mini-split COP is supplied at the chosen ambient and part load; it is not inferred from SEER.

Inverter efficiency is constant and excludes the separately entered idle draw. Battery capacity includes aging and thermal derating supplied by the user; no Peukert model or BMS dynamics are included. Keep AC outputs enabled for cycling loads; some power stations turn off at low draw.

Measured-cycle mode has no separate heater or auxiliary schedule. Its energy-completeness confirmation is mandatory; an automatic-defrost fridge with unaccounted heater energy must use metered daily-total energy via annual mode.

The PV curve is an idealized repeatable day centered on noon, approximated linearly between five-minute knots, with no cloud forecast. The energy dispatch assumes compatible simultaneous solar charging and AC output, adequate DC/BMS power capability, and no inverter trip. Low-battery and unmet-demand traces are energy accounting, not an appliance restart simulation.

Use for preliminary energy budgeting and capacity screening of AC compressor appliances within their manufacturer's operating limits. Results do not establish startup compatibility: verify waveform, voltage, frequency, motor current, surge duration, ambient derating, BMS discharge limits and equipment warranties. Do not substitute a resistive-load voltage-reduction boost rating for motor-start capacity.

The temperature correction is a first-order cabinet heat-leak estimate only. A duty of one flags potentially inadequate cooling. The mini-split model assumes the supplied COP applies while running at the modeled modulation point and does not model humidity, solar gains, transient room pull-down, cycling degradation or heating/defrost operation.

Average-load no-sun hours and storage sizing use the entire daily schedule. Exact endurance during a short outage depends on its start time. A single sunny day cannot prove indefinite service or recovery after several cloudy days. When the input limit clips PV, the panel count is a lower-bound nameplate estimate requiring further time-series design. Event dispatch resolves compressor switches, PV/load crossings and storage bounds within the piecewise-linear PV model. It does not simulate subsecond motor-start transients or battery power limits. The energy balance remains conditional on successful inverter operation.

The Solar Lab's measured totals support the scale of daily demand, but its running-watt and one-sixth-duty statements do not reproduce those totals; no universal divide-by-six rule is used. Virginia Tech's example validates electrical energy integration, not its generic refrigerator duty or voltage-times-current shortcut. Use true watts for energy and VA for inverter loading. GE documents a defrost heater operating while the compressor is off; compressor cycling alone does not capture its electricity. Measured-cycle mode must not be used with unaccounted defrost or auxiliary loads. Food temperature and food safety are outside this model. Installation work belongs to a qualified person.