Calculation method · version 1.0.0 · 9 October 2026

How a thermogram becomes forints

Every example and the calculator run the same open method. Here you find the formulas, the parameters with sources and status flags, the assumptions, and why the uncertainty is ±30–50%.

Calculator Worked examples

All amounts are net HUF (excl. VAT)
The example facilities are fictional
An estimate, not a guarantee

01 / Limits

What we claim – and what we do not

What a thermogram gives us well

the location and extent of a defect (m², m),
its relative severity,
the approximate gap width (mm).

What it cannot give precisely

the real air flow (wind and stack pressure vary hour by hour),
the real conductivity of wet insulation,
the psi-value of a thermal bridge (no accessible EN ISO 14683 table, we use analogies),
the true price of a repair (needs quotes),
actual building use (opening hours, set-back).

A thermal camera measures surface temperature, not watts lost. A loss expressed in forints is therefore a model-based estimate: computed from the thermogram and on-site dimensions (m², m, gap width in mm) with standard physics and sourced parameters. Our figures are estimates, not guarantees; savings occur only if the fix is carried out and works. The survey itself saves no energy.

Please note: Engineering thermal-imaging survey – not an accredited or certified inspection. It does not replace an energy performance certificate, a statutory energy audit, mandatory electrical-safety or fire-safety inspections; those are carried out by professionals authorised under Hungarian law.

Full legal notice

Temperatures in our reports are thermal-camera surface measurements. Apparent temperature depends on surface emissivity, ambient and reflected temperature, distance, viewing angle and weather. Conditions at the time of measurement are recorded in the report, but a measured value is not the true internal temperature of the structure.

Figures for losses, costs and payback are model-based estimates. Actual energy use depends on operation, weather, energy prices and many other factors, so we do not guarantee any saving or payback period.

The service is not an accredited test, an official inspection, an energy performance certificate, an electrical safety inspection, a fire-safety inspection or a structural opinion. For those, a duly authorised professional must be commissioned. Before acting on a finding we recommend an on-site check by the relevant trade.

02 / Chain

The calculation chain: from a defect to forints

Every finding becomes a conductance H (W/K), then seasonal energy, gas and money.

Defectm², m, mm
HW/K
QkWh/season
Gasm³
MoneyFt, range
Paybackmonths, 5 yr

03 / Defect types

Four defect types, four formulas

Damaged / wet / missing insulation

H = (U_{def.} - U_{int.}) \cdot A

A [m²] from the thermogram. U(defect) = 2.5 W/m²K (1.5–3.5); U(intact) = 0.35 (older panel).

Linear thermal bridge

H = (ψ_{found} - ψ_{design}) \cdot L

L [m]; ψ = 0.35 W/mK (0.10–0.80). The as-designed 0.10 is already in the baseline, so it is subtracted.

Air gap (door, dock, panel joint, ridge, skylight)

H = 0.335 \cdot V

V [m³/h] from the leakage network (see below): length [m] × gap width [mm]. 0.335 Wh/(m³K) is ρ·cp of air (1.2 kg/m³ × 1.005 kJ/kgK / 3.6).

Open-door exchange (entrance, air curtain)

H = 0.335 \cdot Q_{open} \cdot f_{open} \cdot (r_{def.} - r_{base})

Q(open) is the buoyancy-driven exchange through the open door (formula below); f is the open fraction, r the residual transfer (vestibule, air curtain).

04 / Air gaps

Pressure, orifice equation and series connection

Stack effect

{Δp}_{k} = g \cdot h \cdot ρ₀T₀ \cdot (\frac{1}{T_{out}} - \frac{1}{T_{in}})

h = 80% of the hall height (distance between low and high openings); ρ₀T₀ = 353 kg·K/m³; T(out) is the season mean (4.4 °C).

Wind

{Δp}_{sz} = \frac{1}{2} \cdot ρ \cdot v^{2} \cdot ΔCp

Modest assumption: v = 2.5 m/s (season-effective at the opening, not the 10 m mean), ΔCp = 0.4 → 1.5 Pa.

Combined

Δp = \sqrt{{Δp}_{k}^{2} + {Δp}_{sz}^{2}}

We deliberately take the lower plausible pressure: about 3–4 Pa across the whole building (the lower half of the 2/4/8 Pa range).

Orifice equation

V = C_{d} \cdot A \cdot \sqrt{\frac{2 Δp}{ρ}} \cdot 3600

C_d = 0.6. Illustration: a 1 m long, 1 mm wide gap at 4 Pa gives 5.58 m³/h = 1.87 W/K.

Series leakage network

A_{ser} = \frac{1}{\sqrt{\frac{1}{{A_{low}}^{2}} + \frac{1}{{A_{high}}^{2}}}}

Air that enters must also leave: adding the flows of independent leaks would count the same air twice. We therefore separate a low level (doors, docks, lower joints) and a high level (ridge, skylights, roof joints); wall joints are split 50/50. A defect is charged only with the increase of flow it causes, with diminishing returns.

Open door

Q = C_{d} \cdot \frac{A}{3} \cdot \sqrt{\frac{g \cdot H \cdot ΔT}{T}}

C_d = 0.6; for baseline traffic and the air-curtain fault.

The baseline leakage (blower-door style q50) fills both levels: 6 m³/(h·m²) for an older steel shed.
The split between defects uses leave-one-out differences scaled so that they add up exactly to the joint increase. In the small-leak limit the flow is 0.354 of the free orifice flow, more conservative than a simple “halve it” rule.
Activity: a dock-seal gap only leaks while a trailer is docked with the door open (25% of heated hours, assumption); a sectional-door gap only while the door is closed (97%). The “normal” leakage of a door (EN 12426 class 1, 24 m³/(h·m²) at 50 Pa, the leakiest class) is subtracted from the gap found.

05 / Money

Seasonal energy, gas and forints

Season heat

Q = \frac{H \cdot HDD \cdot 24 \cdot f}{1000}

1 W/K × 1000 K·h = 1 kWh. The HDD matches the facility’s base temperature (Nyíregyháza, ERA5, 11 seasons). f = heated time fraction (night / weekend set-back, assumption).

Fuel

E_{inv.} = \frac{Q}{η} \cdot 1.1076

Hungarian gas is billed in MJ on a net-calorific basis: invoice kWh = MJ/3.2504, which is 10.8% more energy than physical NCV kWh. Efficiencies are on NCV.

Gas and cost

V_{gas} = \frac{E_{inv.}}{10.52}

cost = E_{inv.} \cdot p

34.2 MJ/m³ (sourced range 33–35; unverified, the invoice value rules). p = marginal, net, all-in gas price; the fixed capacity fee cannot be saved and is excluded.

Gas price (9 October 2026)

low: 32.1 Ft/kWh ≈ 338 Ft/m³
central: 40.4 Ft/kWh ≈ 425 Ft/m³
high: 56.9 Ft/kWh ≈ 599 Ft/m³

TTF quote + MVM Basic fees; the contract mark-up is an assumption. Low = Cal-27 + 5 EUR/MWh, central = front month + 15 EUR/MWh, high = MVM Basic.

Five years: two paths

(a) flat at today’s central price; (b) along the forward curve (40.4 / 35.8 / 27.1 / 27.1 / 27.1 Ft/kWh). A futures price is not a forecast; holding it after year 3 is an assumption. The two are the upper and lower bounds.

payback [months] = \frac{I + S}{annual saving} \cdot 12

{net}_{5} = \sum_{k = 1}^{5} s_{k} - I

I = fix, S = survey; s(k) = saving in season k. Priority: A ≤ 24 months, B ≤ 60 months, C otherwise. The “A items only” scenario shows what pays for itself.

Fix cost: per item low/central/high, indicative and quote-based (assumption) – built from the national minimum overhead labour rate (7,830 Ft/h) and typical materials. Items not in the research (thermal-bridge remediation per m, air-curtain replacement, roof-penetration sealing) are calc-local assumptions. CO₂: physical NCV energy × 0.202 kg/kWh (IPCC 2006 default).

06 / Pipes and electrical

Buried pipe and electrical hot spot

Buried pipe leak

E_{cond.} = w \cdot ℓ \cdot t / 1000

E_{make-up} = \frac{q \cdot t}{1000} \cdot 1.163 \cdot ΔT

w = 200 W/m (100–380; Ljubljana district-heating study: 200–380 W/m measured in wetted insulation, the low value is an assumption for small branches); t: 5,342 h heating season or 8,760 h year-round; heating the make-up water 1.163 kWh/(m³K) · ΔT (70 K: 81.4 kWh/m³). Cost: district-heat fee in Ft/GJ (Budapest FŐTÁV non-residential proxy; only the heat fee is avoidable) + make-up water Ft/m³ (assumption). The leak is downstream of the heat meter, so it is billed 1:1.

In a large district-heating network a closed-circuit water balance is only reliable above ~4 m³/h, so smaller leaks hide there; on a small private system the make-up water meter is the first check, and the camera is for locating the leak.

Electrical hot spot

P = n \cdot I^{2} \cdot R_{extra}

Example: 160 A, +0.5 mΩ → 12.8 W per phase. The energy is negligible (a few thousand Ft per year). The real stake is failure and fire, which we do not put a monetary figure on because we have no reliable source for downtime cost. The only sourced fact: of 736 fires investigated in Hungary in 2022, 73 (9.9%) had electrical energy as the cause (OKF sample, all building types, no breakdown for business premises). The local temperature rise (P · Rth, Rth = 1–4 K/W) is illustrative: why a few watts can show on a thermogram.

Not an electrical safety inspection.

07 / Baseline

The fault-free baseline and the “share of the bill”

To express a leak as a share of the bill we compute the fault-free baseline of every example facility: envelope (U·A), thermal-bridge supplement (8%), baseline infiltration, ventilation (with heat recovery), door traffic – with the same HDD, efficiency and price. “Share of the bill” = defect cost / (baseline + defect cost), central case. Sanity check (enforced by a test): small 1–3%, medium 5–12%, large 15–30%; the inputs were not tuned towards the target, and if they fell outside, the inputs would have to be revisited rather than forcing the result. The fault-free baseline is 56–101 kWh/m²·a of heat, a plausible order of magnitude for such buildings.

08 / Uncertainty

Why ±30–50%?

The loss of one finding is roughly the product of the following, roughly independent factors:

Factor	Band	Effect
Gas price (32.1–56.9 around 40.4)	−20 % / +41 %	direct
Weather (mild–cold season)	−17 % / +18 %	direct
Efficiency (0.72–0.88 around 0.80)	+11 % / −9 %	direct
U(defect) (1.5–3.5 around 2.5)	≈ ±50 % on ΔU	wet insulation
Gap size (from IR, ±), C_d, pressure	±30–50 %	air gaps

\sqrt{{0.20}^{2} + {0.15}^{2} + {0.10}^{2} + {0.40}^{2}} \approx 0.48

Root-sum-square gives ≈ 0.48, i.e. ±30–50% per item. The sensitivity analysis runs the 3×3×3×3 = 81 corner combinations (price × efficiency × weather × defect severity): the 10th–90th percentile of this pessimistic “all extremes at once” grid sits at about −37% / +72% around the central value on the large level, min/max at 0.5× / 2.5×. The central value is roughly the median, not the lowest figure. Several items together have a slightly lower relative spread, but the common system-level factors (gas price, weather, efficiency) do not average out.

The mandatory wording: “model-based estimate, range: see table”.

09 / Caution

Conservative choices – and what the method may underestimate

Deliberately cautious

lower pressure (modest wind, 80% of the height);
series leakage network;
activity factors (dock 25%);
subtraction of normal door leakage;
marginal (not average) gas price;
condensing boiler at 0.95 (not 1.0);
supermarket 20/12 °C HDD;
heated time fraction < 1;
district-heat loss only downstream of the meter, 1:1;
the forward curve as lower bound;
as-designed ψ subtracted from thermal bridges;
electrical risk is not monetised.

What it may underestimate

at busy docks several open doors can act as outflow openings;
flow computed at the mean pressure is slightly lower than the true time-average because cold hours weigh more (Jensen inequality);
deterioration of doors and seals over time.

10 / Parameters

Main parameters, sources and status flags

The values come from the calculation engine’s parameter set – the same one that drives the examples and the calculator.

Sourced: Taken from an opened source, with URL.
Derived: Computed from sourced values.
Standard / physics: Textbook physics or a standard factor; not fetched during the research.
Assumption: Our engineering judgement, no source behind it.
Unverified: Taken from a summary or brief, not confirmed.
Calc-local assumption: An assumption introduced by this calculation layer, not in the research parameters.

Parameter	Value	Status	Source
Gas price, marginal (low / central / high)TTF 2026-10-09 + MVM Basic fees; contract mark-up is an assumption	32.1 / 40.4 / 56.9 Ft/invoice kWh	Derived	oenergetice.cz mvmenergiakereskedo.hu nav.gov.hu
Forward curve (5 seasons)not a forecast; holding after year 3 is an assumption	40.4 / 35.8 / 27.1 / 27.1 / 27.1	Derived	oenergetice.cz
Gas: invoice kWh per m³34.2 MJ/m³ unverified; the invoice value rules	10.52	Derived	–
Invoice kWh / physical kWh	1.1076	Derived	mvmenergiakereskedo.hu
HDD: warehouse 16/12 · supermarket 20/12 · workshop 18/12 °CNyíregyháza, ERA5, 11 seasons; the balance temperature is an assumption	2,161 / 2,904 / 2,533 Kd	Derived	archive-api.open-meteo.com ec.europa.eu
Efficiency: radiant tube / warm air / condensingon net calorific value	0.80 / 0.85 / 0.95	Assumption	–
U(defect), wet / damaged insulation	2.5 (1.5–3.5) W/m²K	Sourced	cris.cobiss.net
U(intact), older panel	0.35 W/m²K	Assumption	ruukki.com
ψ industrial thermal bridge; as-designed ψ subtracted	0.35 (0.10–0.80) W/mK · 0.10	Assumption	phius.org
Orifice C_d · wind · ΔCp · stack distance	0.6 · 2.5 m/s · 0.4 · 0.8·H	Assumption	–
Baseline leakage q50: warehouse · retail · workshop	6 · 3 · 5 m³/(h·m²)	Assumption	–
Normal door leakage (EN 12426 class 1, 50 Pa)	24 m³/(h·m²)	Sourced	greenspec.co.uk
Dock active fraction · door closed fraction	0.25 · 0.97	Calc-local assumption	–
Wetted pipe section loss	200 (100–380) W/m	Sourced	cris.cobiss.net
District-heat fee (Budapest proxy)	9,965 (5,500–12,959) Ft/GJ	Unverified	fotav.budapestikozmuvek.hu
Make-up water	1,500 (800–2,500) Ft/m³	Assumption	–
CO₂ (physical NCV energy)	0.202 kg/kWh	Standard / physics	IPCC 2006 Guidelines vol. 2 ch. 2 tabl…
Fires 2022 (OKF sample)	73/736 (9.9 %)	Sourced	bacs.katasztrofavedelem.hu
Repair costs	see the table below	Assumption	–
Our survey pricenet + area surcharge; pilot 135,000 / 275,000 Ft	Quick diagnosis (2D) from HUF 190,000 · 3D package from HUF 390,000	Sourced	–

Repair costs (net HUF, indicative, quote-based)

Item	Low	Central	High	Status
Dock seal pad / curtain repair, per dock	150,000	300,000	600,000	Assumption
New complete dock seal / shelter	600,000	1,000,000	1,800,000	Assumption
Dock leveler pit seal	80,000	150,000	300,000	Assumption
Sectional door seal set (bottom, sides, top)	40,000	90,000	180,000	Assumption
Door adjustment / service	25,000	45,000	80,000	Assumption
Brush seal replacement per door	15,000	30,000	60,000	Assumption
Panel joint resealing per m (incl. lift access)	2,500	4,500	8,000	Assumption
Insulation patch / PU foam per m² (open, dry, fill, close)	12,000	22,000	40,000	Assumption
Roof skylight seal, each	60,000	120,000	250,000	Assumption
Pipe insulation repair above ground per m	3,000	6,000	12,000	Assumption
Buried pipe: locate, excavate, repair, reinstate (spot)	1,500,000	3,000,000	6,000,000	Assumption
Electrician hourly rate (industrial)	9,000	13,000	18,000	Assumption
Cabinet terminal re-torque + checks	25,000	45,000	90,000	Assumption
Replace terminal / contactor	60,000	120,000	250,000	Assumption
Thermal-bridge remediation (insulation strip + cladding) per m	8,000	15,000	30,000	Calc-local assumption
Air curtain replacement / overhaul	250,000	450,000	800,000	Calc-local assumption
Roof penetration sealing, each	25,000	50,000	100,000	Calc-local assumption

No Hungarian price list was retrievable; the estimates are built from the labour floor (7,830 Ft/h) and typical materials. To be replaced by supplier quotes.

The full parameter list (67 entries)

Key	Low	Central	High	Value	Status
`physics.rhoCpWhM3K`Wh/(m3 K)				0.335	Standard / physics
`physics.rhoAirKgM3`				1.200	Standard / physics
`physics.gravity`				9.810	Standard / physics
`physics.rhoT0KgKPerM3`				353.000	Standard / physics
`physics.cdOrifice`				0.600	Assumption
`physics.cdLargeOpening`				0.600	Assumption
`physics.leakExponentN`				0.650	Assumption
`physics.waterKwhPerM3K`kWh/(m3 K)				1.163	Standard / physics
`leakage.windSpeedMs`	1.50	2.50	4.00		Assumption
`leakage.windDeltaCp`				0.400	Assumption
`leakage.stackSeparationFraction`				0.800	Assumption
`leakage.baselineQ50.industrialShedOld`				6.000	Assumption
`leakage.baselineQ50.retailBuilding`				3.000	Assumption
`leakage.baselineQ50.workshopShed`				5.000	Assumption
`leakage.doorClassEN12426.class1_m3_per_hm2_at_50Pa`				24.000	Sourced
`leakage.doorClassEN12426.measuredGoodSectionalAt50Pa`				20.300	Derived
`defects.uDefectWPerM2K`	1.50	2.50	3.50		Sourced
`defects.uIntactPanelOld`				0.350	Assumption
`defects.psiIndustrialWPerMK`	0.10	0.35	0.80		Assumption
`defects.psiPanelJointWPerMK`	0.05	0.15	0.30		Assumption
`defects.pipeWettedLossWPerM`	100.00	200.00	380.00		Sourced
`defects.psiBridgeAsDesigned`				0.100	Calc-local assumption
`defects.dockActiveFraction`				0.250	Calc-local assumption
`defects.doorClosedFraction`				0.970	Calc-local assumption
`gas.mjPerM3Ncv`	33.00	34.20	35.00		Unverified
`gas.mjPerInvoiceKwh`				3.250	Sourced
`gas.ncvKwhPerM3`				9.500	Derived
`gas.ncvToInvoiceKwh`				1.108	Derived
`gas.invoiceKwhPerM3`				10.520	Derived
`gas.priceFtPerInvoiceKwh`	32.10	40.40	56.90		Derived
`gas.forwardPathFtPerInvoiceKwh`				40.4 / 35.8 / 27.1 / 27.1 / 27.1	Derived
`gas.co2KgPerKwhNcv`				0.202	Standard / physics
`electricity.allInFtPerKwh`	104.00	132.00	153.00		Assumption
`districtHeat.ftPerGJ`	5,500.00	9,965.00	12,959.00		Unverified
`districtHeat.gjToKwh`				277.778	Standard / physics
`water.makeupFtPerM3`	800.00	1,500.00	2,500.00		Assumption
`climate.bases.warehouse_16_12.hdd`	1,788.00	2,161.00	2,546.00		Derived
`climate.bases.shop_20_15.hdd`	2,613.00	3,126.00	3,493.00		Derived
`climate.bases.supermarket_20_12.hdd`	2,456.00	2,904.00	3,334.00		Derived
`climate.bases.workshop_18_12.hdd`	2,122.00	2,532.50	2,940.00		Calc-local assumption
`climate.heatingHoursLe15C`				5,342.000	Derived
`climate.heatingHoursLe12C`				4,520.000	Derived
`climate.hoursPerYear`				8,760.000	Standard / physics
`climate.designOutdoorC`				−15.000	Unverified
`efficiency.radiantTube`	0.72	0.80	0.88		Assumption
`efficiency.warmAir`	0.78	0.85	0.92		Assumption
`efficiency.condensingBoiler`	0.90	0.95	1.00		Assumption
`efficiency.districtHeatLoss`		1.00			Calc-local assumption
`repairCosts.dock_seal_repair`	150,000.00	300,000.00	600,000.00		Assumption
`repairCosts.dock_seal_new`	600,000.00	1,000,000.00	1,800,000.00		Assumption
`repairCosts.dock_leveler_pit_seal`	80,000.00	150,000.00	300,000.00		Assumption
`repairCosts.sectional_seal_set`	40,000.00	90,000.00	180,000.00		Assumption
`repairCosts.door_service`	25,000.00	45,000.00	80,000.00		Assumption
`repairCosts.brush_seal`	15,000.00	30,000.00	60,000.00		Assumption
`repairCosts.panel_joint_reseal_m`	2,500.00	4,500.00	8,000.00		Assumption
`repairCosts.insulation_patch_m2`	12,000.00	22,000.00	40,000.00		Assumption
`repairCosts.skylight_seal_each`	60,000.00	120,000.00	250,000.00		Assumption
`repairCosts.pipe_insulation_above_m`	3,000.00	6,000.00	12,000.00		Assumption
`repairCosts.buried_pipe_spot_repair`	1,500,000.00	3,000,000.00	6,000,000.00		Assumption
`repairCosts.electrician_hour`	9,000.00	13,000.00	18,000.00		Assumption
`repairCosts.cabinet_retorque`	25,000.00	45,000.00	90,000.00		Assumption
`repairCosts.terminal_replace`	60,000.00	120,000.00	250,000.00		Assumption
`repairCosts.thermal_bridge_fix_m`	8,000.00	15,000.00	30,000.00		Calc-local assumption
`repairCosts.air_curtain_replace`	250,000.00	450,000.00	800,000.00		Calc-local assumption
`repairCosts.penetration_seal_each`	25,000.00	50,000.00	100,000.00		Calc-local assumption
`electrical.terminalRthKPerW`	1.00	2.00	4.00		Calc-local assumption
`uncertainty.typicalBand`	0.30		0.50		Assumption

Sources

Parameters (CSV)every parameter with value, status and source Worked examples (CSV)the output the pages show

11 / Gaps

Known gaps, before you treat the numbers as exact

A real gas invoice and offer (the mark-up and capacity-fee class).
The local non-residential district-heat tariff (Nyírtávhő, Tiszaújváros).
2–3 Hungarian quotes each for dock seals, door and panel-joint repair, PU foam patching, excavation.
EN ISO 14683 Annex C and the EN 12426 table.
The design outdoor temperature (−15 °C, unverified).
A source for downtime cost.
Local wind statistics.

After the first 1–2 real measurements (blower door, door test) q50, C_d and wind can be calibrated. Until then the numbers are estimates, together with their ranges.

Try the method with your own numbers

The calculator runs this same model in your browser.

Calculator Get a quote