No-Load Loss vs. Load Loss: The Line Item Most Transformer Buyers Get Wrong
A 1,000 kVA distribution transformer with 770 W of no-load loss sits on the grid, secondary open, delivering nothing. It still burns about 6,745 kWh a year.
No output. Just the core magnetizing and demagnetizing, cycle after cycle.
This is where purchasing decisions go sideways. The quote sheet lists two loss figures side by side. They look comparable. Over 25 years they are not.
No-load vs. load
No-load loss lives in the core. It shows up the moment the transformer is energized, load or no load. Hysteresis, from magnetic domains flipping once per cycle. Eddy currents, circulating in the laminations. Both depend on the core material-grade, thickness, flux density.
Load loss lives in the windings. No current, no load loss. Mostly I²R, plus some stray loss from leakage flux linking the tank and clamps.
Put simply, one is fixed, the other follows the load.
Load loss scales with current squared. At half load it is a quarter of rated. At 125% load it is 156%. That is also why peak efficiency usually sits somewhere around 60% to 80% load-no-load stays flat, load loss climbs, and the sum bottoms out before nameplate.
| No-Load Loss | Load Loss | |
| Common name | Core loss, iron loss | Copper loss, winding loss |
| Location | Core laminations | Windings, structural parts |
| When it occurs | Whenever energized | Only when current flows |
| Load dependence | None | Square of load current |
| Main mechanisms | Hysteresis + eddy currents | I²R + stray losses |
| Measurement condition | Secondary open | Secondary shorted |
| Annual behavior | 8,760 hours, constant | Varies with load profile |
| Design lever | Core material, flux density, thickness | Conductor cross-section, winding geometry |
What the efficiency tiers actually save
| Rating (kVA) | Tier 1 No-Load (W) | Tier 1 Load (W) | Tier 2 No-Load (W) | Tier 2 Load (W) | NLL Reduction | LL Reduction |
|---|---|---|---|---|---|---|
| 630 | 600 | 6,500 | 540 | 4,600 | -10.0% | -29.2% |
| 1,000 | 770 | 10,500 | 693 | 7,600 | -10.0% | -27.6% |
Two ratings here, 630 and 1,000 kVA, both following the same pattern: no-load down about 10%, load loss down 28% to 29%. Other ratings in the same family track close to this. Load loss has more room because it tracks copper content-more conductor, lower resistance. No-load is capped by the core material. To go lower you move up in grade or switch to amorphous.
Watt cost
No-load loss runs 8,760 hours a year. Load loss on a typical distribution transformer at 30% to 40% load factor works out to maybe 2,000 to 3,500 equivalent full-loss hours.
Utility capitalization factors usually land around $3 to $7 per watt of no-load loss, $1 to $2 per watt of load loss. At least 3:1.
So a 1,000 kVA unit going from 770 W to 693 W saves 77 W. At $5 per capitalized watt, that justifies $385 more on the purchase order. At $0.10/kWh it saves $67 a year, roughly $1,700 over 25 years undiscounted.
Two suppliers $200 apart, but the cheaper one has 50 W more no-load loss-the cheaper one is probably not cheaper.
Core material
Hi-B GOES cuts hysteresis loss 10% to 15% versus conventional GOES at the same flux density. Distribution transformers run a lot of hours, so that gap adds up.
Thickness matters too. Standard is 0.23 to 0.35 mm. Going to 0.20 or 0.18 mm cuts eddy loss, more so under harmonics. But thinner laminations mean lower stacking factor and higher processing cost.
Amorphous gets you 60% to 70% lower no-load loss than GOES. It also has lower saturation, it is brittle, and it costs more. We have seen it make sense for rural distribution where load factor is low and the transformer runs 24/7. We have also seen projects specify it where a good Hi-B grade would have done the job for less.
The trade-off
Core and copper budgets are finite. No-load and load loss trade against each other. Bigger core, lower flux density, lower no-load-but the window grows and you need more copper, which can push load loss up. Shrink the conductor and you save copper cost but load loss climbs.
Which way to go depends on load factor. Rural distribution and backup power spend most of their life lightly loaded. No-load loss dominates. Industrial facilities and data centers run near capacity, so I²R dominates. The ORNL loss evaluation guide has a figure-3.4-showing two efficiency curves for exactly this reason. Pick the wrong one and the money is gone.
Practical notes
Evaluate total owning cost, not sticker price. TOC = purchase price + A × no-load loss + B × load loss, where A and B are capitalization factors in $/W. A small no-load difference can flip the ranking of two bids.
Match the design to the load profile. A unit optimized for full load is wasted at 30% load factor, and a low-loss core is wasted at high load factor.
Check the reference conditions on loss data. No-load at rated voltage and frequency. Load loss at 75°C or 85°C. Numbers at different reference temperatures are not comparable.
And leave margin for harmonics. Rectifier loads and VFDs can push actual load loss 10% to 30% above the sinusoidal 50/60 Hz figure. Standard data can be optimistic.
Last thing
No-load loss and load loss are not two numbers you compare at face value. One tracks energized hours. The other tracks energy delivered.
Distribution transformers run at low load factors for decades, and the no-load part of the bill is usually bigger than people expect.
Convert both to lifetime cost before comparing quotes. That is the part that gets missed.
About TIG - Trust in Grid
TIG - Trust in Grid is a global transformer manufacturer with operations spanning complete transformer builds and transformer-related core material supply, serving markets across Asia, Europe, the Americas, the Middle East, and Africa.
Complete transformers: Oil-immersed and dry-type power and distribution transformers, designed and built to IEC, IEEE, GB, and other regional standards for grid connection requirements worldwide.
Core materials: Grain-oriented electrical steel (GOES) cores, amorphous alloy cores, and magnetic materials, with support for grade selection, thickness optimization, and supply chain reliability.
Global project support: From technical selection and loss evaluation through delivery and after-sales, with responsive one-stop service for international customers.
If you are working on a transformer project and need to evaluate core grade, thickness, loss targets, or supplier options, send us your specs-efficiency target, transformer type, volume, and timeline. We will give you a straight answer.
TIG - Trust in Grid
www.trustingrid.com
asia@trustingrid.com
References
1.Del Vecchio, R. M., Poulin, B., Feghali, P. T., Shah, D. M., & Ahuja, R. (2017). No-Load and Load Losses. In Transformer Design Principles (3rd ed.). CRC Press.
2.Kennedy, B. W. (2003). No-Load Versus Load Loss: Making Transformer Losses Part of the Purchasing Decision. IEEE Industry Applications Magazine.
3.Harlow, J. H. (Ed.). (2004). Electric Power Transformer Engineering. CRC Press.
4.Minimum Energy Performance Standards for Distribution Transformers. (2024). CTCN / Botswana.
5.Transformer Loss Evaluation. (n.d.). Oak Ridge National Laboratory / U.S. Department of Energy.
6.IEC 60076-1:2011. Power Transformers - Part 1: General. International Electrotechnical Commission.
7.Applications of Novel Heuristic Algorithms in Design Optimization of Energy-Efficient Distribution Transformer. (2023). IEEE Transactions on Power Delivery.
8.No-Load Loss and Load Loss Measurement. (n.d.). Independent Electricity System Operator (IESO), Ontario, Canada.
9.Grain-Oriented Electrical Steel: Grade Selection and Loss Characteristics. (2023). TIG - Trust in Grid Technical Notes.