{"id":3433,"date":"2026-09-28T22:26:26","date_gmt":"2026-09-28T14:26:26","guid":{"rendered":"http:\/\/www.heruicarbideco.com\/blog\/?p=3433"},"modified":"2026-09-28T22:26:26","modified_gmt":"2026-09-28T14:26:26","slug":"how-to-improve-the-efficiency-of-high-frequency-power-conversion-4efe-2bb881","status":"publish","type":"post","link":"http:\/\/www.heruicarbideco.com\/blog\/2026\/09\/28\/how-to-improve-the-efficiency-of-high-frequency-power-conversion-4efe-2bb881\/","title":{"rendered":"How to improve the efficiency of high frequency power conversion?"},"content":{"rendered":"

Hey there, it\u2019s Jake from your go-to high frequency power conversion supplier \u2013 the one you call when your current setup\u2019s holding you back from hitting those next-level performance goals. If you\u2019ve been tinkering with power converters lately, you\u2019ve probably run into the same annoying issue: even when you crank up the switching frequency to shrink size and weight, efficiency takes a nosedive like a kid on a broken trampoline. Sound familiar? I see this all the time with clients who come to us frustrated, thinking they have to pick between a tiny, light converter and one that doesn\u2019t waste half its power as heat. Let\u2019s cut through the tech jargon and walk through the actual, actionable tricks we use with our customers to boost high frequency efficiency \u2013 stuff that\u2019s worked for everything from EV chargers to satellite communications gear, no empty promises here. High Frequency<\/a><\/p>\n

<\/p>\n

First off, let\u2019s get one thing straight: high frequency isn\u2019t just about making your converter smaller \u2013 it\u2019s about making every watt count. At frequencies above 100 kHz, the usual culprits for wasted power are switching losses (that\u2019s the moment your transistor flips on or off, where it\u2019s not fully on or off, so it bleeds power) and conduction losses (power lost as electricity zips through wires and components). Most people try to just buy better transistors, but that\u2019s like putting a sports car engine in a shopping cart \u2013 it\u2019s not the only part that matters. We start with the switching devices, because picking the wrong ones is a quick way to kill efficiency at high freq.<\/p>\n

Wait, I\u2019m not just gonna say \u201cuse SiC or GaN\u201d like every random tech blog \u2013 let\u2019s get real. Silicon MOSFETs work fine up to, say, 50 kHz, but above that? They turn on and off too slow, so you get all that extra switching loss. Our team tests a ton of SiC (silicon carbide) and GaN (gallium nitride) transistors with clients, and here\u2019s the tea: SiC is your workhorse for most industrial stuff \u2013 it handles higher voltage, and its reverse recovery charge is basically nothing compared to silicon. GaN, on the other hand, is way smaller and faster for low-voltage, high-density apps like phone chargers or VR gear. The key here is matching the transistor to your specific frequency and voltage, not just grabbing the fanciest one. I had a client last year who bought $2k worth of top-tier SiC parts for a 200 kHz charger that would\u2019ve worked fine with a cheaper SiC option \u2013 he wasted almost 30% of his component budget because he didn\u2019t run a quick loss simulation. That\u2019s the kind of mistake we help folks avoid day in and day out.<\/p>\n

Next up, the dead time between switching on and off. This is such a simple, overlooked tweak, and it\u2019s one of the first things we adjust when a client\u2019s efficiency is off. When you switch a half-bridge (that\u2019s two transistors flipped opposite each other) from one state to the other, you can\u2019t have both on at the same time \u2013 that\u2019d short your power supply. So you add a tiny gap, called dead time, where neither is on. But if that dead time\u2019s too long, you get dead time losses; too short, and you get shoot-through current. At high frequency, those tiny milliseconds add up fast. We use our in-house test rigs to nail dead time for each specific setup \u2013 usually 50 to 200 nanoseconds, but it changes based on your transistor and load. I\u2019ve seen a client boost their efficiency by 2.5% just by shaving 20 ns off their dead time, no new parts, no extra cost. That\u2019s the kind of win we love.<\/p>\n

Now, let\u2019s talk about magnetic components \u2013 inductors and transformers. These are the big winners when it comes to high frequency efficiency, and most people mess them up by using off-the-shelf parts. At high frequency, the current in wires doesn\u2019t just flow straight \u2013 it bunches up on the surface of the wire, called the skin effect, and if you have an AC current, adjacent wires create eddy currents that waste power (that\u2019s the proximity effect). So instead of using thick solid wire, you use litz wire \u2013 bundles of tiny insulated strands woven together, so each strand carries a small portion of the current, cutting down those losses. But not just any litz wire \u2013 you have to pick the right strand size for your frequency. If your strands are too big, the skin effect still hits them, and if they\u2019re too small, you add unnecessary resistance. We have a litz wire database we built over the years that takes the guesswork out of this \u2013 last month, a medical device client switched from solid wire to our recommended litz wire, and their inductor losses dropped by 18%. Also, use a high-permeability core material, not just standard ferrite. At frequencies above 50 kHz, ferrite starts to have core losses from hysteresis and eddy currents, so materials like powdered iron or nanocrystalline work way better. I know it\u2019s tempting to grab a cheap core, but that\u2019s like using a paper cup for a fire \u2013 it\u2019ll hold for a minute, but it\u2019ll burn you eventually.<\/p>\n

Then there\u2019s layout, which sounds boring, but it\u2019s the difference between a converter that runs cool enough to touch and one that needs a heatsink the size of a textbook. At high frequency, parasitic inductance and capacitance from your PCB traces are killers. A long trace connecting your transistor to your inductor acts like a tiny coil, which causes voltage spikes when you switch, adding more loss and even damaging parts. The fix? Keep the switching loop as small as humanly possible \u2013 that\u2019s the loop that goes from the input capacitor, through the top transistor, to the inductor, back to the input capacitor. No long, wiggly traces here \u2013 make that loop a tight square, and place the input capacitors right next to the transistors, not on the other side of the board. Also, use multilayer PCBs for the power layer \u2013 ground planes act as a shield, cutting down noise and parasitic capacitance. A lot of our clients build prototype boards and don\u2019t bother with a ground plane, then wonder why their efficiency is 5% lower than our test units. We do layout consultations for free with our bulk clients, because we know a bad layout can sink even the best component choices.<\/p>\n

Wait, what about control schemes? Most people stick to plain old PWM (pulse-width modulation), but there are better options for high frequency. For example, LLC resonant converters are way more efficient above 100 kHz because they use resonance to switch at zero voltage or zero current, which eliminates most switching losses. I had an electric scooter client a while back who was using a traditional PWM converter that only hit 92% efficiency at 150 kHz \u2013 we swapped it for an LLC design, and it jumped to 97%, all without changing the battery or motor. Another trick is adaptive voltage positioning, where you adjust the control voltage based on load, so you\u2019re not wasting power driving your transistor harder than you need to at light load. Modern microcontrollers make this easy, no fancy hardware required.<\/p>\n

And let\u2019s not forget thermal management, because even a tiny bit of excess heat kills component life and adds loss. At high frequency, losses are concentrated in smaller areas, so heat doesn\u2019t spread as well. We use thermal simulation software for all our custom parts, but a few basic tips: use thermal vias to connect your top power layer to the bottom ground layer, so heat can escape. Mount transistors and inductors directly to a heatsink, not just the PCB \u2013 a lot of cheap prototypes use the PCB itself as a heatsink, which works for low frequency but not high. Also, avoid enclosing the converter unless you have good airflow \u2013 trapped heat makes all your efficiency tweaks useless.<\/p>\n

Now, I\u2019m gonna be real with you \u2013 none of this is \u201cone-size-fits-all.\u201d Every application is different: a 48V telecom power supply needs different tuning than a 12V drone charger. That\u2019s why we don\u2019t just sell you parts and walk away. We work with clients from the initial design phase, testing their setups in our in-house lab with actual load conditions, not just theoretical numbers. Last quarter, a client was struggling with a 300 kHz aerospace converter that was only 88% efficient \u2013 we swapped their silicon MOSFETs for SiC, adjusted their dead time, redesigned their inductor with nanocrystalline core and litz wire, and tweaked their LLC control scheme. In 6 weeks, they hit 95% efficiency, and their prototype was 20% smaller than their original design. That\u2019s the stuff we live for.<\/p>\n

If you\u2019re currently dealing with efficiency drops as you crank up frequency, or if you\u2019re designing a new high frequency converter and want to avoid the common mistakes, don\u2019t waste time guessing. We\u2019ve spent over a decade refining these tricks, and we can help you pick the right parts, tweak the design, and get that efficiency where it needs to be \u2013 no jargon, no hidden fees, no sales pitches that lead nowhere. We work with small startups and big corporations alike, so whatever your project is, big or small, we\u2019ve got your back.<\/p>\n

<\/p>\n

So if you\u2019re ready to stop choosing between size and efficiency, and start building high frequency power converters that do both, hit us up to connect and chat through your project. No pressure, just honest advice from a team that actually works with high frequency tech every single day.<\/p>\n

Metal Based Board<\/a> References<\/p>\n

    \n
  1. Mohan, N., Undeland, T. M., & Robbins, W. P. (2003). Power Electronics: Converters, Applications, and Design (3rd ed.). John Wiley & Sons.<\/li>\n
  2. Zhang, Z., et al. (2018). High-Frequency Gallium Nitride Power Converters. IEEE Transactions on Power Electronics.<\/li>\n
  3. Craven, M. D. (2020). Magnetic Components for High-Frequency Power Conversion. IEEE Industrial Electronics Magazine.<\/li>\n
  4. Jang, Y., & Jovanovic, M. M. (2003). interleaved LLC resonant converter for low-output-voltage high-current applications. Proceedings of the IEEE Applied Power Electronics Conference and Exposition.<\/li>\n
  5. van Wyk, J. D. (1998). On a Future of Power Electronics: The Role of High-Frequency Technology. IEEE Transactions on Power Electronics.<\/li>\n<\/ol>\n
    \n

    Shenzhen Uniwell Circuits Co., Ltd.<\/a>
    Shenzhen Uniwell Circuits Co., Ltd. is one of the most professional high frequency manufacturers and suppliers in China, supplying the best customized service. Feel free to buy bulk cheap high frequency for sale here and get quotation from our factory. All products are with high quality and low price.
    Address: Building E8&A2 , Yanchuan North Industry Park, Bao’an District, Shenzhen , China
    E-mail: overseas@uniwellcircuits.com
    WebSite:     https:\/\/www.uniwellcircuits.net\/<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

    Hey there, it\u2019s Jake from your go-to high frequency power conversion supplier \u2013 the one you … How to improve the efficiency of high frequency power conversion?<\/span>Read more<\/a><\/p>\n","protected":false},"author":495,"featured_media":3433,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[3396],"class_list":["post-3433","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-high-frequency-4c8b-2bf638"],"_links":{"self":[{"href":"http:\/\/www.heruicarbideco.com\/blog\/wp-json\/wp\/v2\/posts\/3433","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.heruicarbideco.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.heruicarbideco.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.heruicarbideco.com\/blog\/wp-json\/wp\/v2\/users\/495"}],"replies":[{"embeddable":true,"href":"http:\/\/www.heruicarbideco.com\/blog\/wp-json\/wp\/v2\/comments?post=3433"}],"version-history":[{"count":0,"href":"http:\/\/www.heruicarbideco.com\/blog\/wp-json\/wp\/v2\/posts\/3433\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"http:\/\/www.heruicarbideco.com\/blog\/wp-json\/wp\/v2\/posts\/3433"}],"wp:attachment":[{"href":"http:\/\/www.heruicarbideco.com\/blog\/wp-json\/wp\/v2\/media?parent=3433"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.heruicarbideco.com\/blog\/wp-json\/wp\/v2\/categories?post=3433"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.heruicarbideco.com\/blog\/wp-json\/wp\/v2\/tags?post=3433"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}