Cement Plant Transformer: How to Size the MV Distribution System for a Clinker Line, Why Mill Starting Current and Cement Dust Decide the Spec, and When Captive Power Changes the Design

The answer first: a cement plant transformer is sized from the grinding and pyro-process load list and from mill starting duty, not from a kVA-per-tonne-of-cement rule of thumb. The kiln runs 24 hours a day at a load factor above 0.8, so the load is not the problem; what decides the rating is that a handful of very large mill and fan motors start against that load, and that a growing share of it arrives through a variable-frequency drive (VFD), a slip-ring rotor starter or a synchronous motor. Build the load list area by area — crusher, raw mill, kiln and preheater, clinker cooler, cement mill, packing — apply demand and diversity, add design spare, check the voltage dip at the largest mill bus, derate for the non-sinusoidal current the drives actually draw, and only then derate for kiln-side ambient temperature and conductive cement dust before a standard rating is selected.

A cement plant transformer takes the voltage a utility or a captive power plant delivers to the works — commonly 33 kV, 66 kV, 110 kV or 132 kV, and 220 kV or 400 kV at large integrated sites — and turns it into the electricity that actually makes clinker and grinds cement: 11 kV or 6.6 kV for the big mill, fan and crusher motors; 11 kV or 6.6 kV down to 0.4 kV or 0.69 kV for motor control centres (MCCs), conveyors, packers, dust collectors and lighting; and 0.4 kV auxiliaries for instrumentation, compressed air, water systems and electrical-room HVAC. What separates cement duty from an ordinary industrial distribution load is not the total kVA. It is that the process cannot be stopped — a kiln that loses power is a week of lost production and a damaged refractory lining — that the largest motors start frequently and draw several times their rated current, and that the environment is simultaneously hot and full of alkaline, abrasive, moisture-absorbing dust. Those three facts drive the sizing, the harmonic specification and the enclosure, in that order.

Everything below is written for the person who has to buy or specify that equipment: a plant electrical engineer, an EPC contractor, or a project developer building a new grinding unit or clinker line. If you are sizing an ordinary industrial distribution load instead, the general method in our sizing and budget formula guide and distribution system sizing guide is the right starting point. If your site is a quarry feeding the plant, the load-cycle logic in our mine site transformer guide applies to that part of the works; and if you are adding a captive steel or scrap facility beside the plant, the furnace duty described in our electric arc furnace transformer guide is the relevant comparison. This article covers what changes when the load is a continuously running mill and kiln.

Key takeaways

• Start with kWh per tonne, not with kVA per tonne. Specific electrical energy consumption of a modern dry-process plant is commonly quoted in the range of about 90–110 kWh per tonne of cement, with cement grinding (roughly 25–40 kWh/t), raw grinding and the pyro-process fans making up most of it. Multiply by throughput and operating hours to get a yearly energy figure, then convert to a load list — the load list, not a rule of thumb, is the sizing input (industry experience / 行业经验值 for the split between process areas).
• The mill start is the design event. A mill motor starting direct-on-line draws roughly 6–8 times full-load current, a slip-ring motor with a rotor starter roughly 2–3 times, and a VFD-driven motor about 1.0–1.5 times. That single number decides whether the transformer ratio is 25% motor kW to kVA or 60%, and it is the cheapest place to save money on the whole substation. See inrush current for the related energisation event and voltage regulation for how the bus behaves under load.
• Continuous duty plus dust is a cooling and ingress problem. The kiln runs for 300–350 days a year, so transformer loading follows a flat, heavy profile rather than a duty cycle, and the unit must be sized, cooled and protected for that. Alkaline cement dust migrates into everything, coats bushings and insulators, and becomes conductive once it takes up moisture — so the specification is about IP class, creepage distance, cleaning access and cooling that still works when the cooling surfaces are dusty, not just about kVA. See cooling methods and temperature rise and bushing selection.
• Assume a distorted load and a weak supply. Mill and fan drives are a large, continuously operating non-linear load, and many cement works sit at the end of a long utility feeder or run on captive generation with limited fault level. Specify the harmonic duty (K-factor or a documented harmonic study, per our harmonics guide), the insulation level for the actual fault level, and voltage regulation for a grid that sags.
• Choose oil-immersed or dry-type by location, not by fashion. Outdoor mill substations and the main step-down normally use oil-immersed units; electrical rooms, MCC buildings and cable basements inside the plant favour cast-resin dry-type transformers because a mineral-oil fire in an occupied building is not acceptable. The comparison is set out in oil-immersed vs dry-type selection.

Where does the transformer sit in a cement plant's power chain?

A cement works is electrically simple in shape and brutal in detail: a small number of large loads supplied from a small number of large transformers, with almost no redundancy in the process itself. A single dry-process line typically has one main incoming substation, a ring or radial MV distribution to a handful of area substations, and unit substations at the mill and MCC buildings. The table below is the layering convention we see in most new plants, with realistic rating bands.

Two things are worth noticing in that table before you size anything. First, the plant's total connected load is dominated by maybe five or six motors, so the accuracy of your mill and fan motor data matters more than the accuracy of everything else combined. Second, the electrical rooms are laid out by process area, which means the transformer that feeds the kiln and the transformer that feeds the packing plant live in two completely different environments and should never share a common specification.

Why can't an ordinary distribution transformer do the job? Six ways cement duty differs

A standard distribution transformer will carry the amps. What it will not do is carry them in a cement plant for twenty-five years without being specified for six things that a distribution system never asks of it.

1. The load is continuous and nearly flat, so the loading guide governs the design

A cement kiln does not stop. Realistic annual availability for a well-run preheater/precalciner line is in the region of 300–350 operating days, and the kiln, cooler and mill circuits run near their design point for most of that time. A transformer on the mill or kiln bus therefore sees a load profile with a high load factor — very different from a distribution transformer that spends most of the day at 20–30% load and peaks for a few hours. That changes the calculation: the average winding rise, the effect of the hottest day of the year on top of that average, and the long-term ageing of the insulation all have to be checked against the loading guide rather than assumed away. The practical consequences are a smaller design margin between rated and calculated load than you would allow on a utility distribution unit, and a real interest in what the transformer does at 105% of nameplate for a month, not at 100% for a minute.

2. The largest motor starts against the load, and the motor-to-transformer ratio decides the kVA

This is the single most expensive decision in a cement substation. The mill drive, the kiln induced-draft fan and the largest pump motors are all big enough that their starting current is a meaningful fraction of the transformer's rated current. If a mill motor is started direct-on-line while the rest of the bus is already at 70% load, the busbar voltage collapses far enough to drop out contactors, stall the motor and trip the upstream protection. The conventional cure is to make the transformer bigger. The better cure is to change the way the motor starts — but that belongs to the process and drive engineer, and it has to be agreed before the transformer is ordered, because it is the difference between a 5 MVA and a 10 MVA unit.

3. A large share of the load arrives through a power-electronic converter

Mill drives, kiln and cooler fans, conveyors, roller presses and separator fans are increasingly VFD-fed, and VFD-fed means non-sinusoidal current with a harmonic spectrum rich in the 5th, 7th, 11th and 13th orders, plus switching-frequency components from the rectifier front end. A single large multilevel drive may be 12-pulse or 18-pulse and reasonably clean; a site full of 6-pulse drives on the same bus is not. The transformer sees the sum, and the sum appears as extra eddy-current and stray losses concentrated in the windings and structural parts. That is an eddy-current-loss specification problem (K-factor or a documented harmonic study), not simply a thermal-load problem. Details are in our transformer harmonics guide and K-factor selection.

4. Cement dust is alkaline, abrasive and hygroscopic

Cement and clinker dust is not inert dirt. It is alkaline, it is mildly abrasive, and it absorbs moisture from the air; the deposit that forms on a bushing in a dusty plant is therefore a semi-conductive path, not an insulating one. In practice that means: creepage distance matters as much as flashover voltage, the bushing and insulator profile must be specified for a polluted environment rather than a clean one, cleaning access must be designed in from the start, and the enclosure has to be genuinely sealed rather than merely louvred. The classic failure mode — external flashover across a dust-contaminated bushing on the first humid day after a dry spell — is a specification error, not an operating one.

5. The kiln area is hot, and the hot ambient is not the ambient of the datasheet

Standard transformer ratings assume a maximum ambient of 40°C with a defined daily and yearly average (for oil-immersed units, maximum 40°C, daily average 30°C, yearly average 20°C, per IEC 60076-2). The kiln and clinker-cooler area can exceed that, and radiant heat from hot surfaces, plus an electrical room whose ventilation is fighting process heat, makes it worse. Every degree of real ambient above the design value comes off the loadability of the unit, and the effect is not linear: it hits the winding hot spot and the cooling-air temperature simultaneously. The same logic governs altitude, discussed in high altitude and temperature derating.

6. The supply is often weak, captive or both

Several cement works sit far from a strong utility bus, at the end of a long overhead feeder, or run on captive generation with a limited fault level and its own governor behaviour. Weak supply does three things: it makes motor-starting dips worse, it makes the fault level — and therefore the acceptable transformer impedance — a matter of system study rather than habit, and it makes tap-changing voltage regulation on the main step-down transformer genuinely useful instead of decorative. Where a plant adds a waste-heat recovery (WHR) turbine, the step-up transformer that ties it into the plant ring has to handle paralleling, load sharing and island operation, which is a different specification again.

How do you size a cement plant transformer? Six steps from kWh per tonne to nameplate

The method is the same as any load-list sizing exercise, but the inputs and the checks are cement-specific. Do them in order; each step can change the answer of the one before it.

Step 1 — Convert process data into a load list, area by area

Start from tonnes and kWh, not from kVA. Take the plant's design throughput (t/h for clinker and for cement, and the yearly operating hours), then apply the specific electrical energy consumption of each process area — crusher and raw grinding, the pyro-process including preheater, kiln and cooler fans, cement grinding, and the utility/packing area. Published plant benchmarking puts total specific electrical consumption for a modern dry-process works in the region of about 90–110 kWh per tonne of cement, with cement grinding usually the single largest consumer. Convert each area's energy figure into an installed motor list with rated kW, efficiency, power factor, starting method and duty, because the transformer must be sized for current, and the current depends on efficiency and power factor, not only on kW. Where a figure cannot be sourced from the plant's own design data, mark it as an assumption and carry it through to the sensitivity check rather than silently embedding it.

Step 2 — Apply demand and diversity factors per area

Not every motor runs at once, and not every motor runs at nameplate. Apply a demand factor per load group (how much of the connected kW is actually drawn at the design point) and a diversity factor across groups (which groups peak together). Cement plants have unusually high factors because the process is continuous: the cement-mill area typically runs at 0.85–0.95 demand, the pyro-process area similarly, while workshop, packing and dispatch drop more sharply. The main area that genuinely varies is the crusher and quarry circuit, which is why it is often given its own transformer. Document every factor you apply — the single most common cause of a wrongly sized cement substation is a diversity assumption buried in a spreadsheet with no owner.

Step 3 — Convert to kVA, add design spare, and pick a standard rating

Sum to kVA, add the design spare the client's basis of design demands (commonly 15–25% for a plant that may be debottlenecked later), and then round to a standard rating. Do not stop at this step: the next three steps routinely move the answer by a rating step in either direction, so treat this kVA as a provisional figure, not a decision.

Step 4 — Check the starting voltage dip at the worst motor

Take the largest motor on the bus, the starting method actually chosen, the transformer impedance, the cable run, and the load already on the bus; calculate the voltage at the motor terminals during starting. Common design limits are a maximum 10–15% dip at the MCC busbar for an infrequent start, and often 5–10% where the plant's own standard is stricter or where contactor drop-out is a concern (industry experience / 行业经验值 — the limit must come from the client's basis of design). If the dip is too deep, your options in order of cost are: change the starting method, add a dedicated transformer for that motor, reduce the impedance of the transformer, or increase its rating. Verify the short-circuit current the new transformer will deliver against the switchgear rating at the same time, using the method in short-circuit current calculation.

Step 5 — Apply harmonic duty and verify the distorted-load capability

Establish the harmonic spectrum at the bus — from the drive supplier's data, from a study, or from a measurement on an existing line — and specify the transformer for it. With a documented spectrum, the capability is checked against the derating method; without one, the pragmatic approach used across the industry is to specify a K-factor (K-4, K-9 or K-13 depending on how much of the load is drive-fed) or to require a harmonic-loss-tested design. At the same time, confirm what the site has to comply with at the point of common coupling: IEEE 519 for sites in North America, IEC 61000-3-6 for MV and HV installations, and the local grid code elsewhere. If the plant cannot comply with filters or a higher-pulse-count drive, that is a design change, not a transformer change.

Step 6 — Apply environment derating, then finalise cooling, noise and losses

Now derate for what the site actually is: maximum ambient at the kiln and cooler area rather than at the site gate, altitude if the plant is in the highlands, and any indoor temperature rise in an electrical room. Then choose the cooling class (ONAN for most outdoor units, ONAF where you need the extra 20–30% of capacity on the same core and tank, dry-type with AN/AF for indoor units), the sound level against the plant's noise specification, and the loss grade against the applicable efficiency regulation. Only now do you write the rating.

A worked example (illustrative)

The numbers below are invented to show the method, not to describe a real plant. The point is where the answer moves.

Look at the final line and note what drove it: not the average running load of 3.7 MW, but the combination of a starting event, a distortion specification and a hot, dusty location. That is why a cement transformer cannot be bought from a kVA calculation alone.

Why is the mill motor start the sizing event, and what are your options?

The transformer only has to survive the start; the motor and the process have to live with the starting method chosen. So the conversation belongs with the drive engineer, but the electrical consequences land on the transformer specification. The table below sets out the trade in terms a transformer buyer can act on.

The engineering conclusion is blunt: spending money on the starting method usually saves more transformer kVA than spending it on transformer capacity. A VFD on the mill main drive typically costs far less than the extra 3–4 MVA of transformer, switchgear and cable it makes unnecessary — and it also removes the process disruption caused by a failed start under load. But it shifts the problem to harmonics, which is the subject of the next section, and it means the transformer specification must now include the drive's harmonic signature. Whichever route you take, check the transformer's own energisation transient as well, described in inrush current.

How do you deal with the harmonics that mill and fan drives put into the winding?

Cement grinding is one of the largest installed bases of high-power motor drives in heavy industry, and the harmonic problem is well understood: the drive current is non-sinusoidal, the harmonic components produce additional eddy-current losses in the transformer windings and stray losses in the tank, core clamps and structural steel, and those losses heat parts the transformer's own temperature sensors may not be measuring.

What the harmonics do to a cement-plant transformer

• Additional eddy-current loss in the winding, concentrated near the winding ends and proportional to the square of the harmonic frequency and current, which raises the hot-spot temperature well above what a purely sinusoidal load at the same RMS current would produce.
• Increased stray loss in structural parts, which can produce local overheating in the tank wall, core clamps and busbar supports — invisible to a top-oil thermometer.
• De-rated capability in the sense defined by the loading guides: the transformer can carry the same RMS current but not the same fundamental current, so a 5 MVA unit may only be good for around 4.5 MVA of real load (illustrative).
• Higher dielectric stress and resonance risk where the drive's switching frequency or a capacitor bank interacts with the transformer and cable capacitance — a commissioning-stage problem that is far cheaper to prevent by specification.

What you can specify instead of oversizing blindly

One caution that is specific to cement plants: the harmonic duty is not the same on every bus. The cement-mill bus may be 80% drive-fed, the kiln bus dominated by one large fan drive, and the packing and workshop buses nearly sinusoidal. Specifying a K-13 unit everywhere is a way of paying for a problem you do not have, while specifying one everywhere on the mill bus and assuming the rest is clean is a way of buying an early failure. Do the study per bus.

What do cement dust and kiln-side heat demand from the transformer and its enclosure?

In most industries the enclosure is an afterthought. In a cement plant it is a reliability decision, because the two dominant ageing mechanisms — insulation contamination and thermal ageing — are both driven by the environment the transformer sits in.

Dust: the mechanism, and what to specify against it

Cement dust is fine, alkaline and hygroscopic. Three failure paths follow directly from that description. First, surface contamination: dust settles on bushings, cable boxes, insulators and the tank top, and once it absorbs moisture it forms a partially conductive film along which flashover can track. Second, cooling degradation: dust coats radiator fins and dry-type winding ducts, so the cooling surface the thermal design assumed is no longer there — a transformer can be within its temperature limit on paper and over its hot-spot limit in service for this reason alone. Third, abrasion and corrosion of paint systems and of the galvanising that protects the tank, the conservator and the enclosure.

The specification response is concrete, and it is cheap compared with the consequences:

• Ingress protection matched to the location. Indoor electrical rooms in clean areas can use IP21 or IP23; rooms adjacent to mills, packers and filters need IP54 or better; outdoor units in the dustiest areas often get IP55 or IP65 enclosures and sealed cable boxes. For North American projects the equivalent discussion is in NEMA enclosure types (12 for dust-tight indoor, 4 or 4X for outdoor wash-down areas).
• Creepage distance and insulator profile for a polluted environment. Specify an increased specific creepage rather than accepting a standard profile; the cost difference on a bushing is trivial next to an outage. The selection logic is in our bushing guide.
• Cleaning access designed in. Maintainable means the radiator bank, the bushing assembly and the cable box can be reached, washed and inspected without scaffolding. Ask for this at the drawing-approval stage, not after installation.
• Filtered or pressurised ventilation for dry-type units. A dry-type transformer in a dusty room must have its cooling air filtered and its windings cleanable; unfiltered forced air is a method of moving dust directly into the winding ducts.

Heat: the derating you must not forget

Standard ratings are declared at a defined ambient, and a cement plant routinely violates it in the kiln and cooler areas. The practical handling is to establish the site's real design ambient from measured summer data per area rather than from a single site value, and then to derate or to upgrade cooling. Consider three levers in this order: move the transformer to a location with lower ambient (an electrical room with filtered forced ventilation, or a shaded position away from radiant heat); increase the cooling class (ONAN to ONAF, or AN to AF on a dry-type unit); or accept the derating and increase the rating. Note also that forced cooling changes the loss structure and the noise level, and that on a dry-type unit AF cooling can be noticeably louder than AN, which matters if the substation sits next to an operator cabin. See cooling methods and temperature rise and noise level.

Choosing between oil-immersed and dry-type, in a cement plant context

The full trade between the two families — including impedance, harmonic behaviour, short-circuit withstand and life-cycle cost — is covered in oil-immersed vs dry-type transformers. Whichever family you choose, verify the actual losses against the applicable efficiency regulation, because a continuously loaded cement transformer is exactly the case where the regulated minimum is worth exceeding: see energy efficiency standards comparison and loss grade comparison.

Should the plant rely on the grid, a captive power plant, or both?

Power cost and power reliability are both material to cement economics, which is why so many works have captive generation, a waste-heat recovery turbine, or both, operating in parallel with the grid. That choice changes the transformer scope considerably.

• Grid-only plant. The main step-down transformer carries the full contract demand, and the specification must include the utility's requirements at the point of common coupling: harmonic limits, power-factor and voltage limits, protection co-ordination, and often the fault-level contribution the utility will allow the plant to draw. A weak or long feeder argues for OLTC and for a lower impedance than habit would suggest.
• Captive generation in parallel. The step-up transformer that connects a captive turbine to the plant ring must handle synchronising and paralleling transients, load sharing between the generator and the grid, and reverse power flow. Where the captive plant is the only source during a grid outage, the transformer must be rated for the island condition as well — which is usually the condition that decides the transformer's ability to start the mill motors, because the generator's fault level and inertia are far smaller than the grid's.
• Waste-heat recovery (WHR). A WHR boiler and turbine on the kiln exhaust typically generates a few megawatts — enough to make a real contribution to plant consumption and, on some sites, to export. The step-up transformer for that generator follows the same rules as any generator transformer plus the kiln-specific one: its loading pattern tracks the kiln's, so when the kiln stops, the generator stops, and the plant must be able to take the load back from the grid without a voltage or frequency excursion that trips the process.
• Captive generation only, in remote locations. Where the works is genuinely islanded, the transformer specification has to be developed together with the generator specification, because the motor-starting study that sizes the transformer is the same study that sizes the generator, and doing them separately is how sites end up with a transformer that is fine and a generator that cannot start the mill.

What belongs in the technical specification and on the nameplate?

A cement-plant transformer specification is short if the process information is complete and long if it is not. The table below is the parameter set we would expect to see in an RFQ for a mill or kiln-area unit; anything left blank will be filled in by the supplier's own assumptions.

Two specification habits from our own engineering practice are worth repeating here, because they save arguments later: state the secondary voltage as the value required at the motor terminals at design load (not the nominal secondary of the transformer), and state the motor data — kW, efficiency, power factor, starting method, starting current multiple and inertia — as an attachment to the transformer RFQ even though it is motor data. A transformer supplier cannot check a starting dip without it, and the alternative is to size on a ratio that may be much larger than necessary. See also customisation options for what can reasonably be specified and what should be left standard.

Which tests prove the unit is right?

Tests are the only evidence a buyer gets before the transformer is energised, so they should be chosen for the duty rather than copied from a template.

• Routine tests on every unit: winding resistance, ratio and polarity/vector group, impedance (short-circuit) voltage and load loss at reference temperature, no-load loss and no-load current, applied and induced dielectric tests, and, for oil-immersed units, oil dielectric and dissolved-gas-screening tests before filling. These are the tests that catch manufacturing errors immediately.
• Type tests: temperature-rise test on the design, dielectric type tests, and lightning-impulse tests where the specification requires them. For a repeat duty — the same design going to another line — confirm the validity of the previous type-test report and the design number it refers to rather than paying for a repeat.
• Special tests worth asking for in this application: a sound-level measurement against a defined standard and measurement distance; a harmonic-loss or K-factor verification where the harmonic duty is significant; a short-circuit withstand report for a transformer that will see repeated motor starts and fault contributions; and a frequency-response analysis (FRA) as a fingerprint, which costs little at the factory and becomes invaluable during commissioning and later condition assessment.
• Site acceptance: insulation resistance and polarisation index, winding resistance, ratio and vector-group confirmation, oil tests after filling and settling, tap-changer operation through the full range, protection relay and interlock checks, and a thermal image under load once the process is running. See commissioning tests checklist and pre-shipment inspection checklist.

What does a cement-plant transformer cost, and what drives the price?

Rather than quote a price that would be wrong for your voltage class, copper specification and site, it is more useful to know what moves it.

• Copper versus aluminium winding is the largest single material swing, and the difference is not only in first cost: it affects losses, size and impedance. See copper vs aluminium winding.
• Loss grade is a design choice you pay for once and recover monthly. On a unit loaded continuously for 20 years, the capitalised cost of load loss is usually many times the incremental first cost of a lower-loss design. The arithmetic is in total cost of ownership.
• Impedance, insulation level, harmonic duty and K-factor rating all add design and material cost, and all four are being bought because of something real in your system study. Buying them speculatively is the waste; omitting them on the mill bus is the risk.
• Cooling class, IP rating, corrosion system and special fittings add cost in proportion to how far they depart from a standard design.
• Tests are a small share of the price and the cheapest insurance in the whole package, particularly impulse, sound level and FRA.
• Delivery and logistics to a remote plant, plus crane and offloading constraints, can outweigh several specification choices. Current lead-time logic is in delivery lead time, and the price structure from ex-works to port in FOB quotation breakdown and distribution transformer price breakdown.

What maintenance keeps a cement-plant transformer alive for 25 years?

A cement plant's maintenance calendar has one dominant feature: the annual or biennial kiln shutdown, when the line is cold and the electrical system can be taken out of service. Everything that requires an outage should be planned into that window; everything else should be continuous and non-intrusive.

Continuous and periodic condition monitoring

• DGA (dissolved gas analysis) on oil-immersed units — the standard early-warning method; see DGA guide. Trending matters more than a single reading.
• Oil quality: dielectric strength, moisture, acidity and interfacial tension on a schedule, not only when a problem appears.
• Thermal imaging under load, on bushings, cable terminations, tap-changer contacts and the tank, ideally at the same load condition each time so images are comparable.
• Temperature and load logging, so you know whether the unit is running where the design assumed — and whether the ambient you specified is the ambient you got. A unit consistently at 90% load in a 45°C room is a unit you should plan to replace before it fails.
• Dust and cleaning regime: bushing and radiator cleaning frequency should be set from the observed contamination rate, and the work should be scheduled, resourced and recorded. Most cement-plant transformer failures we see traced to contamination were maintenance-schedule failures rather than design failures.
• Tap-changer maintenance on units with OLTC, including contact wear checks and oil changes in the diverter compartment; a plant that tap-changes daily needs a real maintenance plan.
• Protection function testing and verification that alarms still reach the control room, plus a spares holding for the items with long lead times: bushings, gaskets, temperature indicators, relays and oil.

Five specification mistakes that cost cement projects money

• Sizing from a kVA-per-tonne ratio. It is quick, it is indefensible in a system study, and it usually produces a unit that is wrong in both directions at once — oversized for the running load and undersized for the starting event.
• Ignoring the starting method until after the transformer is ordered. The motor-starting decision is worth several MVA. Making it late means paying for capacity instead of for a drive.
• Specifying the transformer for the site's ambient instead of the substation's ambient. The kiln and cooler areas are hotter, and an electrical room with poor ventilation is hotter still.
• Treating dust as a housekeeping issue. It is a specification issue: IP class, creepage, cleaning access and filtered cooling all have to be in the drawing package.
• Choosing the enclosure after the layout. Whether the unit is oil-immersed outside or dry-type inside determines the fire separation, the room ventilation, the cable routing and the maintenance access. Deciding it late means redesigning the substation building.

None of these require exotic engineering to avoid. They require asking the process and drive questions before the transformer RFQ is issued — which, in a plant that is continuously loaded and hard to stop, is the cheapest engineering hour in the project.

Put your own numbers into the calculation

If you already have the plant's throughput, the specific electrical consumption by area, the motor list and the starting method, the arithmetic in steps 3 to 6 above can be done in a few minutes with our power system calculators, which also build the equipment list for the substation; if you are still choosing a configuration, the block configurator walks through the same voltage, duty and environment decisions as a series of choices. When the load list is fixed and you want a quotation with the transformer, switchgear and cable sized together, send us the motor list and we will return the equipment list and a priced offer — see the selection guide and industrial equipment solutions for how that scope is normally divided.

Sources / 资料来源

• Source: IEC 60076-1 — Power transformers, Part 1: General (rating basis, temperature limits, test regime)
• Source: IEC 60076-2 — Power transformers, Part 2: Temperature rise for liquid-immersed transformers (maximum ambient 40°C, daily average 30°C, yearly average 20°C)
• Source: IEC 60076-3 — Power transformers, Part 3: Insulation levels, dielectric tests and external clearances in air
• Source: IEC 60076-5 — Power transformers, Part 5: Ability to withstand short circuit
• Source: IEC 60076-7 — Power transformers, Part 7: Loading guide for oil-immersed power transformers
• Source: IEC 60076-10 — Power transformers, Part 10: Determination of sound levels
• Source: IEC 60076-11 — Power transformers, Part 11: Dry-type transformers
• Source: IEC 60076-18 — Power transformers, Part 18: Measurement of frequency response (FRA fingerprinting)
• Source: IEC 60214-1 — Tap-changers, Part 1: Performance requirements and test methods
• Source: IEC 60034-1 — Rotating electrical machines, Part 1: Rating and performance
• Source: IEC 60034-12 — Rotating electrical machines, Part 12: Starting performance of single-speed three-phase cage induction motors
• Source: IEC 60909 series — Short-circuit currents in three-phase a.c. systems
• Source: IEC 61000-3-6 — Electromagnetic compatibility (EMC), Part 3-6: Assessment of emission limits for the connection of distorting installations to MV, HV and EHV power systems
• Source: IEEE Std 519 — IEEE Recommended Practice and Requirements for Harmonic Control in Electric Power Systems (voltage and current distortion limits at the point of common coupling)
• Source: IEEE Std C57.110 — Recommended Practice for Establishing Liquid-Immersed and Dry-Type Power Transformer Capability When Supplying Nonsinusoidal Load Currents
• Source: IEEE Std C57.91 — IEEE Guide for Loading Mineral-Oil-Immersed Transformers and Step-Voltage Regulators
• Source: NEMA MG 1 — Motors and Generators (motor enclosure, starting performance and application definitions referenced in North American specifications)
• Source: ISO 12944 series — Corrosion protection of steel structures by protective paint systems (corrosivity categories)
• Source: ISO 9223 — Corrosion of metals and alloys: corrosivity of atmospheres, classification, determination and estimation
• Source: ISO 1461 — Hot dip galvanized coatings on fabricated iron and steel articles
• Source: EU Commission Regulation (EU) No 548/2014 — Ecodesign requirements for small, medium and large power transformers
• Source: US Department of Energy 10 CFR Part 431 — Energy conservation standards for distribution transformers
• Source: GB/T 6451 — Technical parameters and requirements for three-phase oil-immersed power transformers; and GB 20052 — Minimum allowable values of energy efficiency and energy efficiency grades for power transformers
• Source: EN 197-1 — Cement: composition, specifications and conformity criteria for common cements (product context for plant capacity definitions)
• Source: IEA Technology Roadmap — Low-Carbon Transition in the Cement Industry, and IEA cement-sector energy and emissions analysis (process energy intensity and specific thermal/electric consumption ranges)
• Source: Global Cement and Concrete Association (GCCA) — Roadmap to Net Zero Concrete and published cement industry energy-performance material
• Source: World Business Council for Sustainable Development, Cement Sustainability Initiative (CSI) — cement sector energy and CO2 performance reporting framework
• Source: CEMBUREAU — the European Cement Association, published production and energy statistics
• Source: USGS Mineral Commodity Summaries — cement (annual world and country production statistics)
• Source: publicly reported cement plant electrical benchmarking studies — specific electrical energy consumption of modern dry-process plants in the range of approximately 90–110 kWh per tonne of cement, with cement grinding the largest single consumer
• Source: QDTB Engineering Team — industry experience / 行业经验值 for the split of specific electrical consumption between process areas, typical rating bands per substation layer, demand and diversity factors, motor-starting voltage-dip limits, motor-to-transformer ratio practice, K-factor and harmonic-duty specification practice, enclosure, IP and coating practice for dusty hot sites, the first-cost versus loss-capitalisation conclusion, and the maintenance items normally scheduled into the annual kiln shutdown