As a supplier of amorphous alloy transformers, I get asked this question all the time by customers, engineers, and even facility managers who’ve heard these transformers are better for energy efficiency but want the full picture—including where the small losses still happen, even with such advanced core material. Let’s break down the power losses in an amorphous alloy transformer, explain why they’re different from traditional silicon steel transformers, and why that matters when you’re evaluating whether to upgrade your equipment. Amorphous Alloy Transformer
First, I need to start with a quick refresher that’s rooted in real-world conversations, not textbook jargon. Amorphous alloy is a metal alloy made by cooling molten iron, boron, and silicon at like a million degrees per second—so it forms a non-crystalline, glass-like structure. That structure drastically cuts one of the biggest transformer losses: core loss, which is the energy wasted as heat when the core is magnetized and demagnetized by alternating current. But that doesn’t mean there are zero losses. No electrical device is 100% efficient, and amorphous alloy transformers have their own set of power losses, some unique to their design, others similar to traditional units but smaller in scale.
Let’s start with the two main loss categories every transformer has, because that’s the foundation everyone learns first: no-load losses and load losses. For amorphous units, no-load losses are where they shine, but they still have measurable losses here, plus load losses that matter when you’re running the transformer at full capacity.
No-load losses, also called core losses, are the power drawn by the transformer when it’s connected to the grid but not carrying any actual load—like when a shop is closed, but the transformer is still sitting there humming. This is where amorphous alloy’s non-crystalline core wins big. Traditional silicon steel cores have a crystalline structure, so when AC flips polarity 50 or 60 times a second, the domains in the steel rearrange, rubbing against each other and generating heat. That’s called hysteresis loss, and amorphous alloy has almost none of that because there’s no crystal structure to restrict domain movement. There’s also eddy current loss in the core, which is when small circulating currents form in the metal core because it’s a conductor. Amorphous alloy is a thinner material than silicon steel, so eddy currents are smaller too.
But wait—there’s still no-load loss in amorphous alloy transformers. Why? Because even with the best core material, you have to account for manufacturing variations, right? For example, when we stamp and assemble the amorphous alloy cores to fit the transformer, we have to cut the thin metal sheets, and cutting creates small localized stresses in the material. Those stresses temporarily distort the amorphous structure, adding a tiny bit of hysteresis loss in those stressed areas. We do everything we can to minimize this—using laser cutting instead of mechanical shearing, annealing the core after assembly to relieve stress—but there’s still a minor no-load loss that’s almost always less than a third of what a comparable silicon steel transformer would have. I’ve seen data from field tests where a 500 kVA amorphous transformer has a no-load loss of around 200 W, while a silicon steel unit of the same size is around 700 W. That’s a huge difference, but the 200 W isn’t zero, and that’s part of the losses we talk about.
Now moving on to load losses—these are the losses that happen when the transformer is actually carrying power to your equipment, so they’re tied to the current flowing through the windings, the core, and the structural parts of the transformer. This is where amorphous alloy transformers have some key differences from traditional ones, and where some of the less talked-about losses come in.
First, winding losses. These are the I²R losses, right? The heat generated when current flows through the resistance of the copper or aluminum wire in the primary and secondary windings. Amorphous alloy transformers don’t change the fundamental resistance of the wire, so I²R losses are similar in magnitude to a silicon steel transformer of the same size, but there’s a catch here. Because amorphous alloy cores have lower no-load losses, manufacturers sometimes design the transformer with a smaller core window, meaning the windings are a bit tighter or shorter, which can slightly reduce winding length and thus resistance—so these losses are actually a touch lower, but not by as much as core losses. Wait, but there’s another component to load losses: stray losses. Stray losses are when magnetic fields from the windings induce currents in metal parts that aren’t the core or windings—like the transformer tank, the clamps that hold the core together, or the bus bars that carry power in and out.
Here’s where amorphous alloy core design brings a specific stray loss consideration. Amorphous alloy cores saturate at a slightly lower magnetic flux density than silicon steel. That means if you design the transformer to operate at the same flux level as a silicon steel unit, the magnetic field around the amorphous core is a bit stronger at high loads, which can increase stray losses in the surrounding metal parts. That’s a loss we have to mitigate during design: we use non-magnetic clamps for the core, optimize the shape of the tank to redirect magnetic fields, and sometimes switch to aluminum bus bars instead of copper to reduce induced currents. If we don’t do that, stray losses can be higher than expected, eating into some of the efficiency gains from the core. I remember a client who tested our first-generation amorphous transformers and thought we’d made a mistake because stray losses were higher than their calculations—until we adjusted the core gap and clamp design, and they dropped right back to levels comparable to silicon steel, just with way lower no-load losses. That’s a real-world example of how these losses aren’t just theoretical; they require careful design to manage.
Another load loss-related area: eddy current losses in windings themselves. Windings are made of solid wire, so when AC flows through them, it creates eddy currents within the wire that are more concentrated near the outer surface—this is called skin effect, and it’s worse at higher loads or higher frequencies. Amorphous alloy transformers operate at standard grid frequencies (50 or 60 Hz, depending on the region), so skin effect is similar to traditional units, but since we often design windings with slightly larger cross-section wire to offset any small resistance increases from smaller core windows, skin effect losses might be a tiny bit lower. Still, this is a minor loss, and it’s something we model in our design software to make sure it doesn’t add up to more than it should.
There are also other, less obvious losses that aren’t as commonly discussed but add up, especially for large transformers. Let’s talk about dielectric losses—those are the losses in the insulating materials around the windings, like the oil in oil-immersed transformers or the resin in dry-type amorphous units. Dielectric loss is when the insulation material has a small amount of electrical conductivity, so some current leaks through it and generates heat. Amorphous alloy transformers don’t use any special insulation, so dielectric losses are roughly the same as in traditional transformers, but they’re a fixed cost—more important for transformers that operate at high voltages, where the insulation has to be thicker. For example, a 138 kV amorphous transformer might have dielectric losses of a few hundred watts, which is negligible compared to core losses but still part of the overall loss profile.
Then there’s mechanical loss—wait, no, more accurately, the loss from cooling systems. Most large transformers use fans or pumps to circulate oil and keep the core and windings from overheating. The power used to run those fans and pumps is part of the total losses of the transformer system. Amorphous alloy transformers run cooler than traditional ones, right? Because their core losses are lower, so they need less cooling capacity. That means smaller fans, smaller pumps, and lower parasitic loss from the cooling system. I’ve had customers tell me they can downsize their cooling system by 20-30% when switching to amorphous, which is an added benefit that reduces the overall energy waste of the setup, even though it’s not a loss in the core or windings themselves.
Wait, I should also mention no-load current loss—this is different from core loss, right? No-load current is the small amount of current that flows through the primary winding when the transformer is not carrying a load, just enough to magnetize the core. For amorphous alloy, this current is tiny—around 0.1-0.5% of the rated current, compared to 1-3% for silicon steel transformers. That’s a big difference, but again, not zero. This current flows through the winding resistance, so it creates a small amount of I²R loss in the primary winding even at no load. That’s a separate component from core loss, and it’s why no-load loss is a tiny bit higher than just core loss, but again, it’s far smaller than in traditional units.
Let me also put this in perspective for someone who’s actually using a transformer, not just an engineer in a lab. A lot of my customers are industrial facilities with transformers that run 24/7, 365 days a year. The no-load loss is constant, even if the facility isn’t running at full production. So if a traditional transformer has a no-load loss of 700 W and an amorphous unit has 200 W, that’s 500 W saved 8760 hours a year—about 4,380 kWh per year, which adds up to around $500-$600 a year at average industrial electricity rates. Multiply that by the 10-15 year lifespan of the transformer, and that’s $5,000-$9,000 in savings just from no-load losses. The load losses might be a bit higher in amorphous units if they’re poorly designed, but when we do it right, they’re almost identical, so you’re getting all the savings without any extra cost.
I should also address a common myth I hear: some people say amorphous alloy transformers have no load losses, but that’s not true. The core material is better, but manufacturing, design, and minor induced losses mean there’s always some energy lost as heat. The key is that those losses are dramatically lower, not eliminated. And as a supplier, I tell customers that understanding these losses helps us design the right unit for their application. For example, a commercial building that has the transformer running at light load most of the time (like a office building that’s empty at night) will benefit most from low no-load losses, while a manufacturing plant that runs at full load 24/7 will focus on keeping load losses low, which we can do with our optimized designs.
Wait, let’s make sure I’m not missing any key losses. What about harmonic losses? More and more facilities have variable frequency drives, LED lighting, and other non-linear loads that create harmonics in the current. Harmonics increase both core losses and winding losses, because the higher frequency components mean more eddy currents. Amorphous alloy cores actually handle harmonics pretty well, but not perfectly—harmonics increase the hysteresis and eddy current losses in the core, so amorphous transformers will have higher losses with harmonic loads, just like traditional ones, but since their baseline core loss is so low, the total harmonic-induced losses are still lower. That’s another point to mention, because it’s relevant for modern facilities.
Let me wrap this up by tying it back to what matters for you, the reader, whether you’re evaluating a new transformer or looking to upgrade an existing one. The power losses in an amorphous alloy transformer are split into no-load (core) losses and load losses, with a few smaller, design-specific losses like stray losses and no-load current losses. While none are zero, they’re drastically lower than those of standard silicon steel transformers. As an amorphous alloy transformer supplier, my team and I spend hundreds of hours optimizing each design to minimize these losses—using precision cutting, stress relief annealing, non-magnetic core clamps, and tailored core flux levels—so you get the maximum energy savings without sacrificing reliability or performance. If you’re ready to calculate how much you could save by switching, or if you have questions about which amorphous transformer is right for your facility, we’re here to help with a no-obligation consultation and quote.
Transformer References:
- McHenry, H. et al. Amorphous Metal Transformers: A Review of Design, Performance, and Applications. IEEE Transactions on Power Delivery, vol. 25, no. 4, 2010, pp. 2754-2762.
- Energy Efficiency & Renewable Energy Office. Amorphous Alloy Core Transformers: Technology and Market Status. U.S. Department of Energy, 2018.
- Pillai, S. et al. Stray Loss Mitigation in Amorphous Alloy Distribution Transformers. International Journal of Electrical Power & Energy Systems, vol. 95, 2018, pp. 567-574.
- German Energy Agency. Technical Guide: Amorphous Core Transformers for Industrial Applications. 2020.
Henan GNEE Electric Co., Ltd.
Henan GNEE Electric Co., Ltd. is well-known as one of the leading amorphous alloy transformer manufacturers and suppliers in China. If you’re going to buy customized amorphous alloy transformer made in China, welcome to get pricelist from our factory. Quality products and low price are available.
Address: 25TH FLOOR HUAFU COMMERCIAL CENTER ANYANG HENAN CHINA.
E-mail: sales@gneesteels.com
WebSite: https://www.chinasiliconsteel.com/
