Hydrogen and Gas Turbines: What a Blend Actually Saves and How to Verify a Hydrogen-Ready Claim

Hydrogen is the real answer to the question of where gas turbines go next. The number almost everyone quotes about it is the wrong one.

A 50 percent hydrogen blend by volume reduces CO2 emissions by roughly 22 percent, not 50. The blend percentage is volumetric and the decarbonisation is energetic, and those are different numbers because hydrogen carries about one third the energy of methane per unit volume.

This guide covers the arithmetic, what actually changes inside the combustor, why the blend limit is set by flashback margin rather than by the fuel, the five questions that separate a real hydrogen capability from a brochure claim, the NOx trade nobody mentions, and why dual fuel is the flexibility most owners can actually buy today.

What a Hydrogen Blend Actually Saves

A hydrogen blend is stated as a percentage by volume, and the CO2 reduction it delivers is always substantially lower than that percentage.

The reason is energy density. Hydrogen's lower heating value is approximately 242 kJ/mol against methane's approximately 802 kJ/mol, so hydrogen supplies roughly 30 percent of the energy that the same volume of methane does. Displacing half the fuel volume does not displace half the fuel energy, and CO2 tracks the energy.

Lower heating value, or LHV, is the energy released by combustion excluding the latent heat of water vapour in the exhaust. It is the basis on which turbine fuel and efficiency figures are normally quoted.

Those figures are consistent with published industry data. Global Energy Monitor reports that a 30 percent hydrogen blend by volume achieves a 12 percent reduction in CO2 emissions, and that a 75 percent blend reaches only a 50 percent decrease. Mitsubishi Power's own material states that a 50 percent hydrogen blend on the JAC class corresponds to a 22 percent CO2 reduction.

What this means for a business case

A board approving a "50 percent hydrogen-ready" turbine on the assumption of halved emissions has approved a 22 percent reduction. If your decarbonisation target is expressed as a percentage, convert the blend to an energy basis before the figure enters any model, offtake agreement or disclosure.

Does 100 percent hydrogen mean zero emissions

Zero CO2 at the point of combustion. The lifecycle figure depends entirely on how the hydrogen was produced, transported and stored. ISO 19870-1:2026 provides a methodology for determining greenhouse gas emissions at the hydrogen production stage. Colour labels such as green or blue are marketing shorthand and are not a contractual emissions metric.

Hydrogen versus the alternatives

Post-combustion carbon capture addresses CO2 at the stack rather than at the fuel, and applies to the full output rather than to a blend fraction. It is capital-intensive and parasitic on plant efficiency, but it does not depend on a fuel supply chain that does not yet exist.

Battery storage displaces fuel rather than decarbonising it, and is complementary rather than alternative. For sizing, chemistry and cost, see our guide to battery energy storage systems.

What Changes Inside the Combustor

Hydrogen burns roughly seven to ten times faster than methane, and that single property drives every engineering consequence that follows.

Hydrogen's laminar flame speed is approximately 7 to 10 times higher than methane's. Its adiabatic flame temperature is roughly 150 to 200 K higher. Its molecular diffusivity is far greater, producing a Lewis number below one, which makes the flame front less stable and locally uneven.

The practical result is that the flame sits closer to the burner, burns hotter, and is more willing to travel upstream into places it was never designed to be.

The combustor types and why it matters which you have

A diffusion flame combustor mixes fuel and air at the flame front. It tolerates variable and highly reactive fuels well, and it produces high NOx, which is why it is generally paired with water or steam injection.

A dry low NOx or dry low emissions combustor, abbreviated DLN or DLE, premixes fuel and air upstream of the flame to burn lean and cool, which suppresses NOx without water. Lean premixed combustion is the standard on modern high-performance machines.

DLN and DLE systems are far more sensitive to fuel composition than diffusion systems, because the premixed section is exactly where a fast flame can travel. Your blend limit is a property of your combustor, not of hydrogen.

The threshold below which nothing has to change

Siemens Energy's published position is that at low levels of hydrogen, up to 20 percent by volume, no changes to the fuel system or combustors are required. Above that level, material selection and fuel system sizing must be reconsidered, and combustion system design and operation require care to avoid flashback.

That 20 percent figure is the practical near-term answer for most installed fleets, and it corresponds to roughly a 7 percent CO2 reduction.

For component-level innovation including blade cooling, coatings and aerodynamic improvements, see our overview of innovations for improved gas turbine productivity.

Flashback Margin: Where the Blend Limit Comes From

The hydrogen limit in a DLN system's fuel specification is set by flashback margin, which is why blend capability cannot be assumed across a product family or transferred between machines.

Flashback is the event in which the flame front propagates upstream from the combustion zone into the premixing section. Swirlers, fuel lances and premixing passages are not designed to withstand flame temperatures. They can sustain damage quickly, and in some events liberated hardware travels downstream through the power turbine, causing catastrophic damage.

A hydrogen blending ratio beyond roughly 30 percent significantly increases the risk of flashback and thermoacoustic oscillations. Thermoacoustic oscillation is a self-reinforcing coupling between heat release and pressure waves in the combustor, capable of destroying hardware through vibration alone.

Lean blowout is the opposite failure, where the flame extinguishes because the mixture is too lean or the flow too fast. Hydrogen widens the blowout margin while narrowing the flashback margin, which is why the two must be traded against each other rather than optimised separately.

How flashback is detected

Flashback damage has historically been identified through NOx emissions and exhaust temperature spreads, with the NOx increase typically proportional to the severity of the event. Exhaust temperature spread may increase or decrease depending on the state of the affected combustors, which makes NOx the more reliable indicator.

Why the Wobbe index is not enough

The Wobbe index relates a fuel's heating value to its relative density and is used to estimate heat input through a fixed pressure drop. Some OEMs use a modified Wobbe index that also accounts for fuel temperature. It is the standard fuel interchangeability measure.

It is also insufficient on its own. As Combined Cycle Journal sets out, two fuels with the same Wobbe index can have substantially different blowout, flashback, combustion dynamics and autoignition tendencies, so Wobbe index alone cannot be used to ensure safe, reliable operation.

Hydrogen is the clearest example. A blend can remain inside a Wobbe band while its reactivity, flashback margin and volumetric flow demand all change materially. Require the OEM's complete fuel composition envelope, not a Wobbe range.

What "Hydrogen-Ready" Means, and the Five Questions That Test It

Hydrogen-ready has no fixed definition, and the phrase is used to describe everything from a demonstrated full-engine capability to an engineering aspiration. Five questions separate them.

1. What percentage, and on what basis. Volume or energy. Almost always volume, and the distinction changes the decarbonisation figure by a factor of more than two.

2. Demonstrated where. A single-burner rig test, a full-engine test, or a grid-connected commercial installation. These are three very different levels of evidence and the market routinely presents the first as though it were the third.

3. On which machine. A result achieved on one combustor architecture does not transfer to another frame in the same product family. A product family advertised for hydrogen may require a particular combustor, fuel skid, controls package or retrofit before that capability applies to an installed unit.

4. Guaranteed or roadmapped. A commercial guarantee with liquidated damages is a different instrument from a development pathway with a target year. Treat the product-specific guarantee as the governing value.

5. At what NOx. A blend capability that requires abandoning DLN operation, or adding diluent injection, or installing SCR, carries costs that belong in the same decision.

Retrofit or new build

Retrofitting an installed unit for higher hydrogen generally means a new combustor system, revised fuel skid sizing, controls modification and materials review. Whether that constitutes reconstruction under the US New Source Performance Standards is a separate and consequential question, because crossing the 50 percent fixed capital cost threshold makes the unit an affected facility irrespective of any emissions change. See our guide to brownfield power generation decisions.

Where the retrofit outage exceeds what the operation can absorb, temporary capacity bridges it. See our guide to temporary and mobile power.

The OEM Position by Class

All four major gas turbine manufacturers now offer machines capable of at least 30 percent hydrogen blending across multiple frame sizes, with 100 percent capability either demonstrated or on a firm roadmap to 2030.

The aeroderivative point

Commercial 100 percent hydrogen deployment is currently led by aeroderivative machines rather than large frames. The LM6000VELOX order is the clearest example, and the smaller Kawasaki unit is the other. Aeroderivative combustor architecture, physical scale and modular replaceability all make full-hydrogen operation more tractable than on an advanced-class frame.

If your fleet is aeroderivative, you are closer to a real hydrogen option than most large-frame operators.

Which machines dominate new build

Global Energy Monitor reports 175 GW of gas turbine capacity under construction worldwide, with GE Vernova leading at almost 55 GW. Roughly 82 GW, or 47 percent, of turbines in plants under construction are capable of blending 50 percent hydrogen. GE Vernova's 9HA alone accounts for 20 GW of that total. Three manufacturers supply nearly two thirds of capacity under construction and serve over 75 percent of projects.

NOx: The Trade Behind the Decarbonisation

Hydrogen raises flame temperature, and thermal NOx formation rises steeply with flame temperature, so a hydrogen blend increases NOx emissions unless additional measures are taken.

Thermal NOx is nitrogen oxide formed by the reaction of atmospheric nitrogen and oxygen at high combustion temperatures. It is the dominant NOx mechanism in gas turbines and it is exponentially sensitive to peak flame temperature, which hydrogen raises by roughly 150 to 200 K.

The three mitigations and what each costs

Keeping DLN operation within the qualified blend limit is the cheapest answer, because a properly qualified burner can maintain low NOx at its released blend limit. This is why the guarantee, not the marketing claim, is the number that matters.

Diluent injection using demineralised water, steam or nitrogen cools the flame and reduces NOx. It also reduces efficiency and, in the case of water and steam, adds a water treatment and consumption burden.

Selective catalytic reduction treats NOx after combustion by injecting ammonia or urea across a catalyst. It adds capital, adds exhaust backpressure that slightly reduces output, and introduces a reagent and catalyst replacement cycle.

The regulatory consequence

In the United States, 40 CFR Part 60 Subpart KKKKa, finalized in January 2026, applies to stationary combustion turbines constructed, modified or reconstructed after 13 December 2024 and sets NOx standards of performance reflecting the best system of emissions reduction for each subcategory. For large turbines above 850 MMBtu/hr operating above a 45 percent twelve-month capacity factor, the EPA determined that the best system is combustion controls plus post-combustion SCR.

Subpart KKKK governs turbines whose construction, modification or reconstruction commenced after 18 February 2005. Establish which subpart your unit falls under before evaluating a hydrogen scope, because a retrofit can change the answer.

The net position is uncomfortable and worth stating plainly: a hydrogen blend that reduces CO2 by 22 percent may increase NOx, and the fix for the NOx may reduce efficiency, which partially offsets the CO2 gain. Model all three together.

Fuel Supply: The Turbines Are Ready, the Hydrogen Is Not

The binding constraint on hydrogen in gas turbines is not turbine technology. It is hydrogen production, transport and storage infrastructure that does not yet exist at the scale a fleet would require.

Every full-scale hydrogen demonstration to date has occurred at a site where supply was specifically arranged for the project. Whyalla has a dedicated hydrogen plant. HYFLEXPOWER was an EU-funded research facility. These prove the machines work. They do not represent fleet availability.

The case for commercial blending remaining below 30 percent through 2030 is holding, though under pressure. An owner specifying hydrogen readiness in 2026 is buying an option on a fuel, not a fuel.

What that means for a decision today

Specify hydrogen capability where the incremental cost is low, where a credible local supply pathway exists, or where an offtake or regulatory requirement makes the option valuable. Do not build a business case on hydrogen fuel cost, because there is no established delivered price to build it on.

Ammonia is sometimes proposed as a hydrogen carrier for turbines. It is a genuine research direction and it is early, with its own combustion and NOx challenges. Treat it as a watch item, not a plan.

Emissions reporting

Hydrogen combustion produces no CO2 at the stack, so blending reduces reported Scope 1 emissions in proportion to the energy share displaced, not the volume share. Upstream emissions depend on production route and fall into Scope 3 under most frameworks. Confirm the accounting basis your framework requires before claiming a reduction, and use ISO 19870-1:2026 methodology rather than a colour label.

Infrastructure, Materials and Safety

Hydrogen requires roughly three times the volumetric flow of natural gas for the same energy input, which resizes the entire fuel supply system before a single combustor component changes.

That volumetric factor drives larger piping, larger valves, higher compressor duty and a redesigned fuel gas skid. Ansaldo Energia's service organisation has been explicit that the fuel skid and controls require special attention to avoid leakage and maintain safe turbine control.

Materials

Hydrogen atoms are extremely small and readily permeate the internal structure of metals, causing hydrogen embrittlement, a loss of ductility and fracture resistance. Material selection for piping, valves, seals and fuel system components must be reviewed against hydrogen service, not carried over from natural gas.

The standards that apply, by correct designation

Complying with NFPA 2 is not a complete safety case. The hazard and operability study and the emergency plan must cover leaks, venting, purging, loss of ventilation, failed ignition, blend excursions, compressor trips and interaction with adjacent natural gas equipment.

Insurance

Property and machinery breakdown underwriters price on fuel, protection systems, area classification and maintenance regime. Hydrogen service changes all four. Engage your broker during design rather than after commissioning, because a hydrogen scope can change both premium and deductible.

Dual Fuel: The Flexibility You Can Buy Today

If the requirement driving your interest in hydrogen is fuel supply security rather than decarbonisation, dual fuel capability delivers it now at a known cost.

Dual fuel means a turbine configured to run on gas or liquid fuel and to transfer between them, typically with on-site distillate storage. It protects against pipeline interruption, gas price spikes and interruptible supply contracts.

Dual fuel capability typically adds in the region of $150 to $250 per kW to a project. It also complicates emissions compliance, because liquid fuel operation carries a different emissions profile and usually a separate permit limit, and it adds fuel storage, treatment and periodic transfer testing to the operating scope.

Prismecs operates eight TM2500 dual-fuel units totalling 260 MW at Birr, Switzerland, specified as dual fuel to guarantee winter reliability on a reserve plant.

The honest comparison: dual fuel solves supply security today at a defined cost. Hydrogen addresses carbon at an undefined cost on a fuel that is not yet available. Many owners discussing hydrogen actually need the first one.

Writing It Into a Specification and Verifying It

Specify hydrogen capability as a guaranteed percentage by volume at a named NOx limit, demonstrated at a named evidence level, on your specific machine, or the clause is unenforceable.

What to require in the bid package

ASME PTC 22, Performance Test Code on Gas Turbines, governs output and heat rate testing. Specify that the test is run on the contracted fuel, with correction methodology and measurement uncertainty agreed before mobilisation. API 616, Gas Turbines for the Petroleum, Chemical and Gas Industry Services, applies where the machine serves process duty.

Handling composition variability in operation

Pipeline hydrogen content is beginning to vary in some networks, and a change inside a gas pipeline tariff is not automatically inside your turbine's fuel specification. Require your supplier's certificate of analysis and rate-of-change data, compare both against the OEM fuel specification, and set alarm and trip limits accordingly.

Blend excursion is an operating condition your procedures must cover. For how procedures, alarm response and operator competency should be structured, see our guide to plant operations and maintenance scope.

Warranty and service agreement

Running outside the qualified fuel envelope will generally void combustion hardware coverage, and a hydrogen retrofit can reset or terminate an existing long term service agreement. Establish this before award. See our guide to choosing a power plant O&M provider for how service agreements handle scope change.

Why the Decision Is Being Forced Now

Gas turbine capacity is sold out for years, which means the machine you specify today is the machine you will run for decades, and its fuel flexibility is decided at order rather than later.

GE Vernova booked 18 GW of turbine orders in Q4 2025 alone and ended 2025 with a backlog stretching to 2029, reaching 116 GW of combined equipment backlog and slot reservations by the end of Q2 2026. Siemens Energy reports a record backlog. Lead times across the three major manufacturers now stretch up to eight years, and turbines ordered today are generally not delivered until 2028 to 2030.

Manufacturing capacity, not demand, is the binding constraint. GE Vernova is ramping from roughly 48 heavy-duty turbines a year toward 70 to 80 by 2026, and Mitsubishi plans to double capacity over two years.

Combined cycle configurations hold roughly 70 percent of the market in 2026. For heat rate, capacity factor and configuration selection, see our guide to power generation systems compared.

The consequence is that fuel flexibility has moved from a nice-to-have to a specification decision with a twenty-year consequence, made under time pressure in a seller's market.

How the Hydrogen Decision Changes by Sector

The physics is constant. What changes is whether the option is worth paying for.

Power utilities and IPPs

Dispatch economics and emissions compliance dominate, and the Subpart KKKKa capacity factor threshold interacts directly with any uprate. Where a jurisdiction has a declared hydrogen blending mandate or a low-carbon offtake, the option has quantifiable value. Where it does not, specify the highest blend capability available at no incremental cost and stop there.

Data centers

The binding constraint is time to power, not carbon. Hydrogen readiness is frequently specified for procurement and disclosure reasons rather than operational ones. Be explicit about which it is, because a hydrogen scope that delays energisation defeats the purpose of the project.

Oil, gas and petrochemicals

Some of these sites produce hydrogen as a process by-product, which makes them the one segment with genuine local supply. Refinery fuel gas already contains hydrogen in many cases, so the fuel envelope question is live today rather than hypothetical. Hazardous area classification under IEC 60079 and NFPA 70 Article 500 already governs the site.

Industrial cogeneration

Smaller industrial machines reach higher published hydrogen percentages more readily than large frames, and the Kawasaki 1.8 MW-class unit at 100 percent is the clearest example. Where a site has a coincident thermal load and a hydrogen source, the case is stronger here than anywhere.

Remote and island grids

Hydrogen supply logistics are the deciding factor and usually rule it out. Dual fuel capability addresses the real problem, which is supply interruption.

Below 20 MW

Packaged machines in this class frequently offer higher hydrogen percentages than large frames, but the infrastructure cost of hydrogen supply does not scale down proportionally. The threshold question is whether a hydrogen source exists within economic transport distance.

Outside the United States

The NOx framework differs. The EU Industrial Emissions Directive and its BAT conclusions, national permit limits and local grid codes replace the 40 CFR framework. The combustion physics, the standards for hydrogen piping and safety, and the verification questions are unchanged.

What Prismecs Deploys

Prismecs installs, commissions, operates and maintains aeroderivative gas turbine fleets, which is the machine class currently leading commercial hydrogen deployment.

Verified scope includes three LM6000PC units installed and commissioned, adding 150 MW of fast-start reserve; eight TM2500 dual-fuel units totalling 260 MW at Birr, Switzerland, delivered as a fast-track reserve plant with a new 220 kV interconnection and specified dual fuel for winter reliability; four TM2500 units at Duqm, Oman totalling 110 MW with O&M teams, CMMS and parts support; an LM2500XPRESS plant at Miaoli, Taiwan delivered in ten months; and an LM6000 fleet decommissioned in Norway, transported, then reassembled and recommissioned at a new site.

Prismecs is OEM-agnostic, which means the fuel flexibility recommendation is not tied to one manufacturer's roadmap, and it works on the owner's side through owner's engineering on projects where independence from the supplier is the requirement.

Apply this article's questions to any supplier, including us. Ask for the guaranteed blend percentage with its energy equivalent, the evidence level and the machine it applies to, the guaranteed NOx at that blend, and the ASME PTC 22 test basis.

To request a fuel flexibility assessment, send your turbine model and commissioning year, current fuel specification, expected capacity factor, applicable emissions permit and any hydrogen supply pathway to sales@prismecs.com or call +1 (888) 774-7632. We return a blend capability assessment, an emissions and efficiency trade analysis, and a specification checklist.

Frequently Asked Questions

How much CO2 does a hydrogen blend actually save?

Far less than the blend percentage suggests, because blends are stated by volume and CO2 tracks energy. Hydrogen's lower heating value is roughly 242 kJ/mol against methane's 802 kJ/mol. A 30 percent blend by volume cuts CO2 by about 12 percent, a 50 percent blend by about 22 percent, and a 75 percent blend by about 48 percent. Convert to an energy basis before the figure enters any model.

What does hydrogen-ready actually mean?

It has no fixed definition. Test it with five questions: what percentage and on what basis, demonstrated at what level (rig test, full-engine test or commercial operation), on which specific machine and combustor, guaranteed commercially or only roadmapped, and at what NOx. A product family advertised for hydrogen may require a particular combustor, fuel skid, controls package or retrofit before the capability applies to an installed unit.

What hydrogen percentage can a gas turbine burn without modification?

Siemens Energy's published position is that up to 20 percent hydrogen by volume requires no changes to the fuel system or combustors. Above that level, material selection and fuel system sizing must be reconsidered and combustion design requires care to avoid flashback. Twenty percent by volume corresponds to roughly a 7 percent CO2 reduction, which is the practical near-term position for most installed fleets.

Why does hydrogen cause flashback?

Hydrogen's laminar flame speed is approximately 7 to 10 times higher than methane's, and its adiabatic flame temperature is 150 to 200 K higher. In a lean premixed dry low NOx combustor, a faster flame can propagate upstream into the premixing section, which is not designed for flame temperatures. Hardware can be damaged quickly and liberated debris can travel downstream through the power turbine.

Why is the Wobbe index not enough to qualify a hydrogen blend?

The Wobbe index relates heating value to relative density and estimates heat input through a fixed pressure drop, but two fuels with the same Wobbe index can have substantially different blowout, flashback, combustion dynamics and autoignition tendencies. Hydrogen is the clearest case: a blend can sit inside a Wobbe band while reactivity, flashback margin and volumetric flow demand all change. Require the complete composition envelope.

Does hydrogen increase NOx emissions?

Yes, unless additional measures are taken. Hydrogen raises adiabatic flame temperature by roughly 150 to 200 K, and thermal NOx formation is exponentially sensitive to peak flame temperature. The mitigations are staying within the burner's qualified blend limit, diluent injection using water, steam or nitrogen which costs efficiency, or post-combustion selective catalytic reduction which adds capital and a reagent cycle.

Which OEMs offer hydrogen-capable gas turbines?

All four major manufacturers offer at least 30 percent blending across multiple frame sizes. GE Vernova publishes 50 percent on the HA class with a pathway to 100 percent, and has a commercial 100 percent aeroderivative order for four LM6000VELOX units at Whyalla. Siemens Energy ran an SGT-400 on 100 percent hydrogen in HYFLEXPOWER. Mitsubishi Power demonstrated the JAC grid-connected at 30 percent. Kawasaki launched a 1.8 MW-class 100 percent unit.

Are aeroderivative turbines better suited to hydrogen than frames?

Commercially, yes, at present. The clearest 100 percent hydrogen commitments are aeroderivative, including GE Vernova's LM6000VELOX order for the 200 MW Whyalla plant and full-scale aeroderivative hydrogen testing begun in Texas in January 2026. Combustor architecture, physical scale and modular replaceability make full-hydrogen operation more tractable than on an advanced-class frame. Aeroderivative operators are closer to a real option.

Is hydrogen fuel actually available?

Not at fleet scale. Every full-scale demonstration to date has run at a site where supply was specifically arranged, such as the dedicated hydrogen plant at Whyalla or the EU-funded HYFLEXPOWER facility. The turbines are ready; the production, transport and storage infrastructure is not. The case for commercial blending staying below 30 percent through 2030 is holding, though under pressure.

What standards govern hydrogen at a power plant?

NFPA 2, the Hydrogen Technologies Code, covers generation, installation, storage, piping, use and handling. ASME B31.12 covers hydrogen piping and pipelines including materials requirements. ISO 21789 covers gas turbine installation safety. IEC 60079 and NFPA 70 Article 500 govern hazardous area classification. ISO 19870-1:2026 provides hydrogen greenhouse gas accounting methodology. Compliance with NFPA 2 alone is not a complete safety case.

How much larger does the fuel system need to be for hydrogen?

Roughly three times the volumetric flow is required for the same energy input, because hydrogen's volumetric energy density is about one third that of methane. That drives larger piping, larger valves, higher compressor duty and a redesigned fuel gas skid. Materials must also be reviewed for hydrogen embrittlement, since hydrogen readily permeates the internal structure of metals and reduces ductility.

Should I specify hydrogen readiness or dual fuel?

Dual fuel if the real requirement is supply security. It solves pipeline interruption and price volatility today, typically adding in the region of $150 to $250 per kW, at a known cost with a known fuel. Hydrogen addresses carbon at an undefined cost on a fuel not yet commercially available. Many owners discussing hydrogen actually need dual fuel, and the two are not mutually exclusive.

Does a hydrogen retrofit trigger new source requirements in the US?

It can. Under 40 CFR 60.15, reconstruction occurs when the fixed capital cost of replacement components exceeds 50 percent of the cost of a comparable entirely new facility, and reconstruction makes the unit an affected facility irrespective of any change in emission rate. A combustor, fuel skid and controls replacement can cross that threshold, pulling the unit into Subpart KKKKa. Run the calculation at scope definition.

Which test code proves performance on a hydrogen blend?

ASME PTC 22, Performance Test Code on Gas Turbines, governs output and heat rate testing. Specify that the test is run on the contracted blend rather than on natural gas alone, with reference conditions, correction methodology and measurement uncertainty agreed before mobilisation. A guarantee stated only on natural gas tells you nothing about performance on the fuel you intend to burn.