Fuel preparation — the circuit everyone forgets
Fuel is the second-largest line in the cost stack, and in the year to June 2026 the economics of it turned over completely. The argument most consultants reach for against petcoke is also the weakest one available — and it is weak for a reason this course keeps returning to.
On this page · 16 sections
- 01The one idea
- 02The parts
- 03First principles — why fineness has to track volatile matter
- 04What must be true
- 05The quantitative anatomy — and a basis error worth ₹4 a tonne
- 06What goes wrong here
- 07Scenarios from the field
- 08The numbers that matter
- 09Build the model — blended fuel cost, on one basis
- 10Upgrade paths
- 11The frontier, and who is running it
- 12What’s changing now
- 13Check yourself
- 14Mini case — the fuel strategy that was a year out of date
- 15Go deeper
- 16Carry forward
§ 01 The one idea
For a decade the Indian fuel question was simple: petcoke is cheaper per kilocalorie, so how much of it can we burn? That is no longer true. On a duty-paid landed basis in mid-2026, petcoke costs about ₹2.10 per thousand kilocalories against imported coal at ₹1.90 — roughly 10% more per unit of heat delivered.
Two things did it. Import duty on petcoke is 11% against 2.75% on coal, which as Argus puts it “largely offsets the calorific-value difference between the two fuels.” And the rupee fell about 9.5% year on year, which raises the cost of anything priced in dollars while leaving domestic coal priced in rupees.
Shree Cement went from 54% petcoke to 9% petcoke in a single year, replacing it with coal, which went from 26% to 74%. Nationally, Indian cement petcoke imports were 10.67 Mt in CY2025 and ran at 707,000 t in the first quarter of CY2026 against 2.47 Mt in the same quarter of 2025 — a 71% fall. The trade forecast for CY2026 is around 6 Mt.
Any material on Indian cement fuel written before mid-2025 has this backwards. If a client’s fuel strategy paper argues for raising petcoke share, check its date and its exchange rate assumption before anything else. A benchmark built on ₹83–85 to the dollar is obsolete.
§ 02 The parts
| Stage | What it does | What goes wrong |
|---|---|---|
| Coal mill VRM or ball mill | Grinds and dries in one pass, using hot gas from the kiln hood or preheater. Target residual moisture 1–1.5%. | Wet fuel derates the mill and cools the flame — each 1% of coal moisture lowers flame temperature by 10–14 °C. |
| Classifier | Sets fineness. This is where the R90 residue is decided. | Ground too coarse for the fuel’s volatile matter (§03), or too fine, which costs mill power for nothing. |
| Fine coal bin and dosing | Buffers and meters fuel to the burner and calciner. Typically 60% of fuel goes to the calciner, 40% to the main burner. | The highest-risk vessel in the plant. Fine fuel, warm, in air. See §06. |
| Main burner | Shapes the flame — momentum, swirl, primary air. Belongs to A7 and C6. | A burner cannot fix a fuel the mill prepared badly. |
| Alternative fuel line | Separate handling, storage, dosing and usually a separate feed point. Coarse, wet, heterogeneous material. | The binding constraint is almost never the kiln. It is the pre-processing ecosystem between the municipality and the weigh feeder (§12). |
§ 03 First principles — why fineness has to track volatile matter
A fuel particle entering a flame burns in two stages. First it devolatilises: heat drives off the gaseous fraction, which ignites almost immediately and burns fast. What is left is char, and char burns by oxygen diffusing to the particle surface and carbon dioxide diffusing away.
That second stage is the slow one, and it is diffusion-limited, so burnout time scales roughly with the square of particle diameter. Halve the particle and you quarter the time it needs. That single relationship is why fineness is not a matter of taste.
A high-volatile bituminous coal at 20% volatile matter has a fifth of its energy released in the fast stage, and a char particle that is already partly hollowed out and reactive. Petcoke has under 10% volatile matter, so almost all of its energy is in a dense, unreactive char — and the burnout clock starts later and runs longer.
The industry encodes this in one of the oldest rules in cement operation: target a residue on 90 microns equal to about half the volatile matter percentage.
So bituminous coal at 18–22% VM wants R90 of about 9–11%, and petcoke at under 10% VM wants about 5%. Indian plants running 100% petcoke report going considerably finer than the rule implies — 1.6–1.8% residue on 90 micron — and achieving free lime of 0.5–0.8% with nil CO at the kiln inlet.
Provenance: the R90-equals-half-the-volatiles rule is stated in the cement operations literature and is long-standing, but it could not be traced to a specific handbook edition or standard. Describe it as an industry rule of thumb, not as a specification.
The commercial consequence is that a mill sized for coal is not a mill sized for petcoke, because it has to hit a residue less than half as coarse. That is a real constraint — and it is a capacity constraint, not the energy constraint everybody quotes. Which brings us to the most instructive number in this lesson.
§ 04 What must be true
Fuel preparation is judged almost entirely on the wrong axis. The circuit’s own power is a small number; what it controls is the stability and the ceiling of the largest cost line in the plant. These seven conditions are what the circuit has to deliver, and none of them is a grinding condition.
| What must be true | Why | Target | How you verify it |
|---|---|---|---|
| 1. The fuel is ground fine enough for what it is | Petcoke has almost no volatile matter, so it ignites late and burns slowly. The only compensation available is surface area. Grind it to a coal fineness and the flame lengthens, the burning zone moves, and free lime and refractory life both pay. | coal 10–15% petcoke 2–4% R 90 µm |
Ask for the residue and the fuel analysis together. A plant that switched to petcoke without changing its fineness target has made the kiln pay for a procurement decision, and the cost will be attributed to the kiln. |
| 2. The fuel is dry enough to flow and to burn | Moisture in fine fuel costs twice: heat to evaporate it in the kiln, and flow problems in the bin and the feeder. Petcoke arrives wetter than coal more often than not. | mill outlet <1.5% H₂O |
Mill outlet temperature and moisture, against the hot-gas source. If the fuel mill runs on its own hot-gas generator, that fuel is a cost nobody has booked to fuel preparation. |
| 3. The feed to the burner is steady to the minute, not to the shift | A kiln cannot average. A dosing swing arrives as a temperature swing, which arrives as a free-lime swing, which the operator corrects by adding fuel. The fuel rate a plant runs is set by its worst hour, and dosing stability is where the worst hour is manufactured. | — | One-minute fuel-rate trend against kiln torque and burning-zone temperature for two days. No plant, vendor or research body anywhere publishes a measured link between dosing stability and heat rate — it is universally asserted and never evidenced, so you must build it on the client’s own data. |
| 4. The circuit is inerted and the explosion risk is engineered, not procedural | Fine coal and petcoke in air is an explosible dust. This is the one condition in the course where the failure mode is not economic. | mill outlet O₂ held low |
CO and O₂ interlocks, inerting provision, explosion venting and the bin level and temperature regime. Check that it is interlocked rather than instructed. |
| 5. Alternative fuel is dosed by mass, not by volume | RDF bulk density varies by a factor of several between bales, so a volumetric feeder delivers a wildly varying heat input while reporting a steady rate. This is the single largest cause of a plant failing to hold the substitution rate its equipment is capable of. | gravimetric | Ask what the AF feeder weighs and how often. Indian majors recorded two to three percentage points of thermal substitution in FY26 from feed automation alone — the kilns were already capable; the feed was not. |
| 6. The chlorine and sulphur the fuel brings in have somewhere to leave | Alternative fuels carry chlorine. Chlorine circulates, condenses in the preheater and builds rings and blockages. The substitution ceiling at most plants is set here, not by the burner and not by the permit. | kiln feed Cl <0.015% |
Chlorine input balance across all fuels and raw materials, and whether a bypass exists. Every plant running above roughly 60% substitution that discloses how, discloses a bypass. |
| 7. The fuel supply is contracted, not merely available | A substitution target is an operating target only if someone has agreed to deliver a defined quantity at a defined calorific value and moisture. Otherwise it is a procurement aspiration with a kiln attached. | — | Ask to see the supply agreement and what it specifies on CV, moisture, chlorine and tonnage. A TSR target without a supply agreement is a target with no mechanism, and this is the binding constraint in India far more often than the equipment is. |
Six of these seven are engineering conditions a plant can act on this quarter. Row 7 is a contract, and it sits outside the plant gate. That is why fuel-preparation recommendations fail more often than any other category in this course — the capital gets approved, the equipment gets installed, and the substitution rate does not move because nobody secured the tonnage. Check row 7 before you cost rows 1 to 6.
§ 05 The quantitative anatomy — and a basis error worth ₹4 a tonne
The petcoke grinding penalty, on the two denominators
CII’s benchmarking gives best-operating vertical roller mill power as 22.2 kWh per tonne of coal and 33.9 kWh per tonne of petcoke — Version 6.0, May 2023, more than 110 plants. Quoted as they stand, they say petcoke costs 53% more to grind, and a number like that is repeated constantly.
CII’s benchmarking website gives 21.2 kWh/t coal and 33.89 kWh/t petcoke with no edition and no plant count. The dated Version 6.0 PDF of May 2023 gives 22.2 and 33.9, and adds what the website omits: the full band, 22.2–43.9 for coal across 8 plants and 33.9–50.8 for petcoke across 9. The petcoke figures agree; the coal figures do not. This course uses the dated PDF. The practical rule is the one this difference teaches: a benchmark without an edition year and a sample size is not a benchmark, and the undated version of a number is the one most likely to be in circulation.
An earlier version of this lesson’s outline said the petcoke grinding penalty was “systematically under-modelled.” The opposite is true: on the denominator that matters it is systematically over-stated. The 60% headline is arithmetically correct per tonne of fuel and close to economically irrelevant per tonne of clinker.
This is F4 §06 arriving with money attached. Per tonne of what? The whole argument turns on it, and both numbers are published, and neither source says which denominator a reader should use.
If you are making a case against petcoke, lead with the price-per-kilocalorie inversion and with sulphur. The grinding-power argument is the weakest one available and a process engineer will dismantle it in the room.
Fuel prices and calorific values, mid-2026
| Fuel | Price | CV | ₹/’000 kcal | Basis — and it matters |
|---|---|---|---|---|
| Imported petcoke, US | $135/t CFR | 8,100 GCV | 1.58 | CFR, gross basis. Ambuja, June 2026. Peak was $160 in Dec 2025. |
| Imported petcoke, US | $150/t CFR | 7,500 NAR | 2.10 | Duty-paid, net basis. Argus, 2026. Sulphur 4–7%. |
| Imported coal, NAPP | $135/t CFR | 6,900 NAR | 1.90 | Duty-paid, net basis. Argus, 2026. |
| South African coal | $108/t FOB | 6,000 NAR | 1.70 | FOB — freight not included. Ambuja, June 2026. |
| Indonesian, HBA 2 | $96.92/t FOB | 5,300 GAR | 1.73 | Government reference price, 15 Aug 2026. 21% moisture. |
| Domestic coal, CIL G11 linkage | ₹1,394/t | 4,000–4,300 | 0.34 | Pithead run-of-mine only. Excludes royalty, DMF, sizing, transport, GST. Not a delivered cost. |
| Domestic coal, e-auction | ~₹2,105/t | 4,000–4,300 | 0.51 | Pithead, at the 51% average premium over notified price in April 2026. |
Read the basis column, not the numbers. Gross versus net calorific value moves the answer by about 8%; duty treatment by up to 11%; pithead versus delivered by a multiple. Indian cement companies do not state which basis their per-kilocalorie disclosures use. Every comparison in this table is only valid within its own row.
§ 06 What goes wrong here
| The failure | The mechanism, and what to do |
|---|---|
| Comparing fuels on mixed bases. A CFR gross-basis petcoke price against a duty-paid net-basis coal price. | The two conventions can differ by a fifth. Build one table, one basis, all fuels — and put the basis in the column header where it cannot be lost. |
| Sulphur cycles, misread as combustion problems. Petcoke runs 4–7% sulphur against coal at 0.5–1.0%. | A plant on 100% petcoke at 6% sulphur reported dusty kiln conditions and ring formation toward the kiln inlet with every burner and oxygen parameter normal. That is the signature of sulphur accumulation, not of a mis-set flame. The control variable is the sulphate-to-alkali molar ratio, target 0.8–1.2, not the burner. |
| Chlorine from alternative fuel, underestimated. | Coal carries 0.01–0.3% chlorine. PVC plastics carry 20%. Animal meal 0.6–1.6%. And above about 0.015% chloride in the raw meal, recirculation is bad enough that preheater blockages become inevitable. This is the real constraint on RDF rate, and it is a materials-acceptance question, not a kiln question. |
| Quoting a refractory-life figure by fuel. | No published figure exists for burning-zone brick life on coal versus petcoke versus high-RDF. What is published: over 60% of refractory wear cases are caused by salt infiltration, and the chlorine loading contrast above. Present the mechanism; do not invent the months. |
| Treating the coal mill as ordinary rotating equipment. | Fine fuel dust ignites at 500–625 °C, a smouldering layer can self-ignite as low as 160 °C, and the minimum explosive concentration is 50–100 g/m³ — which a fine coal bin exceeds continuously. Oxygen must be held below about 13%, typically by drawing ~3% oxygen gas from the preheater. |
| Assuming the mill can make the fineness. | Petcoke needs roughly half the residue coal does (§03), and a mill sized for coal will be throughput-limited before it is power-limited. No published derating percentage could be sourced — get it from the plant’s own trial data, and be suspicious of a fuel-switch business case that does not contain one. |
§ 07 Scenarios from the field
Ramco Jayanthipuram slowed the coal mill table to reduce the water spray needed for bed stability: 1.5 kcal/kg of clinker, no investment.
JK Cement Nimbahera cut the coal mill auxiliary auto-stop delay from 15 minutes to 5: 1.03 lakh units a year, ₹5.65 lakh, ₹0 spent.
Neither is a project. Both are settings that were correct once and were never revisited — the ten minutes of auxiliary running was presumably a commissioning-era caution about restart. This is the marginal-gains argument in its purest form, and it is why F6 §07 tells you to ask for the register: items like these are never in a capital plan, because there is no capital.
Petcoke 54% → 9%, coal 26% → 74%, in a single year. And the consequence the company disclosed alongside it: the clinker conversion factor deteriorated from 1.58 to 1.50, attributed to lower coal quality — roughly a 5% efficiency loss.
Shree also landed at ₹1.95 per ’000 kcal, the highest of the six disclosed producers, despite running 74% coal. That does not reconcile with cheap domestic coal, so either most of that coal is imported or its delivered domestic cost is far above pithead. Either way the lesson is the same: a fuel switch made on price per kilocalorie can be undone by quality, and the quality effect shows up two departments away as a conversion factor.
§ 08 The numbers that matter
| Metric | Reference | Basis and source | Your plant |
|---|---|---|---|
| Blended kiln fuel cost | ₹1.53–1.95 | Per ’000 kcal. Six producers, Q1 FY27. JK Cement 1.53 · JK Lakshmi 1.65 · Ambuja 1.66 · ACC 1.67 · JSW 1.80 · Shree 1.95. | — |
| Import duty | 11% / 2.75% | Petcoke / coal. The reason the economics inverted. | — |
| Petcoke sulphur | 4–7% | Against coal at 0.5–1.0%. Roughly 3–5× the sulphur per unit of heat. | — |
| Sulphate/alkali molar ratio | 0.8–1.2 | The control variable for sulphur build-up. | — |
| Raw meal chloride ceiling | ~0.015% | Above this, preheater blockage from recirculation becomes inevitable. The binding constraint on RDF rate. | — |
| R90 target | ≈ half the VM | Coal 18–22% VM → R90 9–11%. Petcoke <10% VM → ~5%. Indian 100%-petcoke practice: 1.6–1.8%. | — |
| Coal mill outlet temperature | 70–75 / 90 °C | Direct firing / indirect firing. No petcoke-specific limit is published. | — |
| Residual moisture | 1–1.5% | Each 1% of coal moisture costs 10–14 °C of flame temperature. | — |
| Oxygen for inertisation | <13% | Below the explosion threshold for bituminous dust. Typically ~3% O₂ gas drawn from the preheater. | — |
| India TSR | ~6% | GCCA India, Nov 2025 — “some plants exceeding 20%”. Europe 40–58% (sources disagree; CEMBUREAU 58% for 2022, WBCSD 40–50%). | — |
| Disclosed company TSR | 7–11% | JK Cement 11.29% at June 2026, targeting 35% by FY30. Ambuja 7.1% Q1 FY27. Most producers do not disclose. | — |
§ 09 Build the model — blended fuel cost, on one basis
What the model holds fixed. Import duty at 11% on petcoke and 2.75% on coal; CII grinding power at 33.9 kWh per tonne of petcoke and 22.2 per tonne of coal (Version 6.0, May 2023). Domestic coal is entered as a delivered price because no credible all-in delivered figure is published — pithead run-of-mine for CIL G11 is ₹1,394 on linkage and about ₹2,105 at the April 2026 e-auction premium, and everything between that and the plant gate is the client’s own number. The grinding line is computed per tonne of clinker, for the reason in §05.
§ 10 Upgrade paths
| Tier | Intervention | What it preserves, and the honest assessment |
|---|---|---|
| 0 | Rebuild the fuel comparison on one basis | Free, and it is the first thing to do. Half the fuel strategies in circulation compare CFR gross against duty-paid net, and the error is worth a fifth of the fuel bill. |
| 0 | Revisit the mill’s timers and setpoints | JK Nimbahera’s auxiliary auto-stop delay: 15 minutes to 5, ₹5.65 lakh a year, nothing spent. Ramco’s table speed: 1.5 kcal/kg, nothing spent. Ask when each setting was last changed and why. |
| 1 | Fineness trial against the VM rule | If the plant is grinding to 6% R90 on a 20% VM coal, it is grinding finer than the rule requires and paying mill power for it. A controlled trial settles it. Preserves everything. |
| 1 | Automate the alternative-fuel feed | Indian majors recorded 2–3 percentage points of TSR in FY26 from feed automation alone. The kiln was always capable; the feed was manual and therefore conservative. The best-evidenced AF intervention available. |
| 2 | Sulphur and chloride monitoring | Hot-meal sampling and a sulphate/alkali balance. Turns ring formation from a mystery into a controlled variable, and it is the prerequisite for any petcoke or RDF rate increase. |
| 3 | RDF handling and feeding system | The main retrofit in this area, and now partly compulsory (§12). The one disclosed Indian cost is ₹15 crore for 100 tpd with a 7–8 year payback — in 2017. Escalate before quoting; budget ₹22–28 crore. A 500 tpd pre-processing plant was ₹25 crore on the same 2017 basis. No 2024–26 disclosed Indian capex could be found. |
| 4 | Chlorine bypass | Only when RDF rate is genuinely constrained by the 0.015% raw-meal chloride ceiling and the source cannot be cleaned up. Expensive in capital and in heat — a bypass throws away hot meal. Establish the chloride balance first; the answer is often in the RDF specification, not the kiln. |
§ 11 The frontier, and who is running it
The fuel-preparation frontier has moved entirely to the alternative-fuel side. No new solid-fuel mill order, commissioning or upgrade is published anywhere for 2024 to 2026 — the trade press has nothing, and every coal or petcoke vertical mill reference that surfaces dates from 2018 or earlier. Meanwhile the substitution numbers at named European plants have run away from the Indian average by more than a factor of ten. Both halves of that sentence matter.
| Status | What | Plant, company, country | The number |
|---|---|---|---|
| OPERATING | The highest sustained substitution anywhere | Solnhofen — Solnhofer Portland-Zementwerke, Bavaria, Germany. Presented by the operator, 2024. | TSR 60–65% → 90–95%, enabled by a bypass system. No capex disclosed and no bypass rate disclosed. The step is attributed to the bypass, which is row 6 of §04 doing exactly what it says. |
| OPERATING | High substitution on an SRF line | Rohož̌ník — Danucem Slovensko, Slovakia. Belt dryer commissioned March 2023, hot disc, calciner feeder. | TSR above 80%, SRF at 12 t/h. Presented by the operator, 2024. |
| OPERATING | A substitution step with a date on both ends | Souselas — Cimpor, Portugal. New kiln burner with satellite burner, belt drier. | TSR 43% (2023) → 65%, one third of the alternative fuel as RDF. One of the few cases with a dated before and a dated after from the plant itself. |
| OPERATING | Capex and outcome in the same sentence | Derrylin — Mannok, Northern Ireland. Kiln upgrade, August 2025. | €2.5m for 30% SRF substitution, with a phase two target of 65–70%. This is the cleanest capex-to-outcome pairing in the whole area, which is itself a comment on how rarely the two are published together. |
| OPERATING | A chlorine bypass with money and a result attached | Vidal Ramos — Votorantim Cimentos, Santa Catarina, Brazil. First chlorine bypass in Brazil, December 2022. | US$10m. Co-processing capacity 45,000 → 73,000 t/yr (+62%); petcoke 65,000 → 53,000 t/yr; CO₂ down 12,200 t/yr. The bypass rate as a percentage of kiln gas is still not disclosed — three of the four variables you would want are here, which is one more than usual. |
| OPERATING | Asia, for scale | Citeureup — Indocement, Indonesia. Hot disc commissioned 2023. | Kiln 8 reached 53% TSR in December 2025; company-wide 29% in 2025. Note the gap between the kiln figure and the company figure — 24 percentage points, and both are true. |
| BENCHMARK | The national frontier | Germany — VDZ environmental data, 2023. | About 74% alternative fuel share nationally, one third of it biomass. CEMBUREAU members averaged 58% in 2022, up from 53%. The EU average is around 52%; Japan is above 50%; the United States was 16% in 2023, with 60 of 87 US plants below 20% and 39 of them below 5%. |
| BENCHMARK | India, and the width of the gap | CII Version 6.0, May 2023, more than 110 plants; GCCA India, November 2025. | CII: industry average TSR 7%, best single plant 30%, top-ten band 14–30%, best plant up from 21% in 2014. GCCA India: average 6%. The best-performing Indian plant is anonymous, and no Indian cement plant publishes its own plant-level TSR. Only group figures exist. |
| FAILS CHECKING | “100% alternative fuel” | Holly Hill — Holcim, South Carolina, USA, March 2023. Małogoszcz — Holcim, Poland, May 2026. | Holly Hill ran more than 40 continuous hours at 100% TSR on 30–31 March 2023; the monthly average was 89.7%, itself a plant record. Małogoszcz was a test lasting several days, with no fuel mix and no annual rate disclosed. The monthly 89.7% is the citable number. A 40-hour demonstration is a demonstration. |
| FAILS CHECKING | A vendor ceiling read as a plant rate | A Heidelberg Materials plant in Italy, Entsorga Pelican system. | Conference material attributes 90% TSR to the plant. The vendor’s own release names no plant and no country, and says the system will let the plant increase substitution “by up to 90%” — a design ceiling in vendor language. The two sources also disagree on handling capacity. Do not state that plant runs at 90%. |
| FAILS CHECKING | A group average worn as a plant figure | Heidelberg Materials, “about 30% TSR”. | An explicit global group average — 50% SRF and RDF, 15% tyres, 15% biomass, plus solvents and waste oil. No individual plant in the same dataset has a disclosed TSR. The same trap catches Ambuja’s 9%: a group number, useful as a group number only. |
No named plant anywhere publishes its coal mill kWh/t. The only figures that exist are CII’s anonymised Indian bands: 22.2–43.9 kWh/t for coal across 8 plants, 33.9–50.8 for petcoke across 9, Version 6.0, May 2023. No plant publishes a petcoke grinding power penalty — the CII pair is the only quantified comparison in existence, and it is per tonne of fuel, which is the wrong denominator (§05). No published capacity-derating percentage for converting a coal mill to petcoke. No plant-level petcoke fineness disclosure — the 2–4% R 90 µm figure is practitioner consensus, and CII’s observed band runs 2–16%. Fuel-feeder vendors publish no dosing accuracy percentage at all in current literature, and no plant publishes a measured one. No named plant publishes an AI or advanced-control result on the fuel circuit or on alternative-fuel dosing; the figures usually offered — Tokuyama’s 3% specific heat, Anhui Conch’s 1% kiln fuel — are kiln-side, not fuel preparation.
Two frontiers, running in opposite directions. The equipment that grinds fuel has stopped developing, which means a client’s coal mill is very unlikely to be a capital opportunity. The equipment that receives and doses alternative fuel is developing fast and has named plants at 65%, 80% and 90% substitution with dates attached.
The Indian position is not an engineering position. The Indian average sits at 6–7% against a German national figure of about 74%, and the gap is not explained by kilns — the best Indian plant is at 30% on the same equipment everyone else has. It is explained by what row 7 of §04 asks for: segregated, contracted, specified fuel. The three published numbers in circulation for India — 7% (CII, 2023), 6% (GCCA, 2025) and a regional “Asia 3%” — do not agree with each other, so quote the range and attribute each one.
On price: no RDF price for India could be found in any source, for any date. The only public figure anywhere is UK RDF export at £90–140/t in March 2026 — and that index closed on 1 April 2026, so there is now no free public RDF price series for Europe either. A business case built on an assumed RDF price is building on the one input nobody will confirm.
§ 12 What’s changing now
Current as of August 2026 · refresh every six monthsGazette notification S.O. 388(E), published 28 January 2026, in force 1 April 2026. Rule 11 requires industrial units using solid fuel and located within a specified distance of an RDF plant to replace part of that fuel with combustible fractions from solid waste. The distance is 400 km for RDF Grades I–III going to cement kilns, and 100 km for the segregated combustible fraction above 1,500 kcal/kg.
The schedule is 6% from the effective date, 10% after three years, and 15% after six. RDF suppliers must maintain a minimum 3,000 kcal/kg when supplying cement plants, and report availability monthly through a central portal.
What it means: check the baseline you are given. The PIB release, at least one major law firm’s client note and a policy think-tank all summarise this as “5% to 15%”. The gazette text says 6%, and so does the legal press reading the gazette. It is a one-point difference that will be quoted at you in a meeting, and it is a good illustration of why you read the notification rather than the summary. The other number to establish on day one: is this plant inside 400 km of an RDF plant? That is what determines whether any of this binds.Q1 CY2026 imports were 707,000 t against 2.47 Mt a year earlier, a 71% fall, with the full year forecast around 6 Mt from 10.67 Mt in CY2025. Cement remains exempt from the 2018 Supreme Court petcoke import ban — the constraint is price, not permission.
What it means: a plant that invested in petcoke grinding capability over the last decade now owns an asset whose economic case has reversed. That capability is not wasted — the price relationship will move again, and duty is a policy variable. The right question is whether the plant retains the optionality to switch, which depends on mill fineness capability and on sulphur handling, not on the current price.India generates 62.4 Mt of municipal solid waste a year, heading for 165 Mt by 2030 — but Indian MSW runs 30–50% moisture and 30–40% ash, rising to 40–50% moisture in the monsoon. In Karnataka in 2021, of 3,476 t/day of dry waste, only 215 t/day became RDF and only 118 t/day reached a cement plant. 75% was landfilled unprocessed.
What it means: the field ratio quoted in the sector is that a truck of coal equals nine trucks of waste. The kiln is almost never the constraint. Pre-processing capacity, moisture and chlorine are — and tipping fees paid by the tonne reward volume over calorific quality, which is a contract-design problem rather than an engineering one. A TSR target set without a supply agreement is a target with no mechanism.§ 13 Check yourself
§ 14 Mini case — the fuel strategy that was a year out of date
A 2.8 Mt/yr plant, running 25% petcoke, commissioned a fuel strategy review. The draft recommendation was to raise petcoke to 55%, on the argument that petcoke was cheaper per kilocalorie and the plant’s mill could make the fineness.
What the draft compared. Petcoke at $135/t CFR on a gross calorific value of 8,100 — ₹1.58 per ’000 kcal — against imported coal duty-paid on a net calorific value of 6,900, ₹1.90. A 17% advantage to petcoke, and a clear recommendation.
What the comparison actually was. Petcoke on CFR gross; coal on duty-paid net. Put petcoke on the same basis as the coal — net calorific value of 7,500, 11% duty — and it is ₹2.10, not ₹1.58. Coal is 10% cheaper per kilocalorie, not 17% dearer. The recommendation inverted.
The size of it. At 730 kcal/kg of clinker and a 0.68 clinker factor, a plant making 2.8 Mt of cement burns about 1.39 × 10¹² kcal a year. Moving 30 points of that from ₹1.90 fuel to ₹2.10 fuel is ₹0.20 per ’000 kcal on 30% of the heat — about ₹8.3 crore a year, before touching sulphur, refractory or the mill.
What made it defensible-looking. Both prices were real, both were sourced, and both were current. Nothing in either number announced its basis. The reviewer had built the table from two credible documents that happened to use different conventions, and no line in the model said so.
A composite scenario. The prices, calorific values, duty rates and exchange rate are sourced to June 2026; the plant is not a real one.
The transferable move. Build the fuel table yourself, in one basis, with the basis in the column header. Never assemble a fuel comparison from figures collected out of different documents — the conventions are invisible, they are never stated, and the error runs in whichever direction your sources happened to differ.
§ 15 Go deeper
§ 16 Carry forward
- “Petcoke isn’t the cheap fuel any more. Duty-paid, per kilocalorie, it’s about ten percent above imported coal — eleven percent duty against two and three-quarters, plus the rupee. Shree went from fifty-four percent petcoke to nine in a year.”
- “Before we compare those two fuels — gross or net calorific value, and is the duty in? Those two conventions are worth about a fifth of the number.”
- “The sixty-percent petcoke grinding penalty is per tonne of fuel. Per tonne of clinker it’s eight-tenths of a kilowatt-hour, about five rupees, against a nine-hundred-rupee fuel bill. If we’re arguing against petcoke, that’s not the argument.”
- “Grind to about half the volatile matter as residue on ninety microns. Char burnout goes as diameter squared, and petcoke is almost all char.”
- “Your RDF ceiling is chloride, not the kiln. Above roughly point-oh-one-five percent in the raw meal, preheater blockages become inevitable — and PVC is twenty percent chlorine against coal’s point-three.”