Finding the differences in a series of power supplies(lttlabs.com)
lttlabs.com
Finding the differences in a series of power supplies
https://www.lttlabs.com/articles/2026/05/05/testing-psu-series
7 comments
There's basically no point. Desktop PSU is a solved problem, most designs are all about cost engineering and the tiny sliver of higher power and ultra high efficiency options are not struggling with their current form factor.
In the data center, where power (and cooling) are the only significant OpEx, GaN point-of-load conversion is everywhere. Common as a rack distributed 48V to 12V bus or direct to processor Vdd (2% duty cycle is feasible with GaN thanks to fast on/off times). There was a while where GaN was used as part of the power factor correction for AC to DC in the server power supply, back when passing 400VAC or 800VAC bus around made sense. I think these days it's mostly back to DC buses, and AC-to-DC is all happening farther back, in part because of widespread solar deployments and trying to avoid DC->AC->DC double conversion losses when possible. So maybe GaN gets use on active secondary rectifier in the bus -> 48V now too.
In the data center, where power (and cooling) are the only significant OpEx, GaN point-of-load conversion is everywhere. Common as a rack distributed 48V to 12V bus or direct to processor Vdd (2% duty cycle is feasible with GaN thanks to fast on/off times). There was a while where GaN was used as part of the power factor correction for AC to DC in the server power supply, back when passing 400VAC or 800VAC bus around made sense. I think these days it's mostly back to DC buses, and AC-to-DC is all happening farther back, in part because of widespread solar deployments and trying to avoid DC->AC->DC double conversion losses when possible. So maybe GaN gets use on active secondary rectifier in the bus -> 48V now too.
Desktop PSUs aren't exactly space-constrained. For example, the Corsair SF1000 offers you 1000W in an SFX package (6.3cm x 12.5cm x 10cm). A regular ATX PSU has about 2.5x that volume available.
Efficiency isn't a big selling point either. They are all good enough that you're not going to notice any difference on your power bill, and they don't get hot enough to cause issues.
The average gamer cares more about things like cable modularity, fan noise, and whether it has RGB.
Efficiency isn't a big selling point either. They are all good enough that you're not going to notice any difference on your power bill, and they don't get hot enough to cause issues.
The average gamer cares more about things like cable modularity, fan noise, and whether it has RGB.
Some Asus ROG Strix PSUs market themselves as having GaN transistors, yes.
Of course, given that ATX power supply vendors produce a standard-sized product, and they're limited to the ~1600W that Americans can draw from their 110v sockets, the GaN products are neither smaller nor higher power (though they might run a bit cooler or more efficiently).
Of course, given that ATX power supply vendors produce a standard-sized product, and they're limited to the ~1600W that Americans can draw from their 110v sockets, the GaN products are neither smaller nor higher power (though they might run a bit cooler or more efficiently).
I was thinking about oldschool PC PSUs the other day. Specifically, my current portable https://github.com/vk2diy/hackbook-m4-mini is smaller than the PSU on my last Gentoo PC, and no doubt more efficient. Hate giving Apple money but I guess that's 40+ years of accrued engineering plus sacrificing serviceability for form factor and passive thermals.
Nice. It would have been interesting to dive into the exact differences in components. Are higher wattage PSUs in a series just using higher rated components or does the quality go up too?
Not really a fixed relationship between the two. Sometimes the entire blueprint of the PSU is from a different OEM. Corsair did this back in the day for a while where their HX750/HX850W PSUs were all made from a CWT design, their 1000W PSU was made from a different CWT design, and their 650/620W PSUs were a Seasonic design. I think this is less popular these days, but I think that's about as extreme as the difference gets.
At the component level, the focus is often on the sourcing and tolerances of the capacitors, which are used to clean up transients (very important) and power flow correction, among other things. I think the next most important components are the AC/DC conversion and the voltage transformers. Specifically for higher wattage PSUs vs lower wattage PSUs, the major difference is the amperage along the 12V rail.
A rough chain is: Outlet -> transient clean up circuit -> AC-to-DC conversion -> power factor correction -> PWM circuit (pseudo DC-to-AC) -> 3.3V/5V/12V transformers -> AC-to-DC conversion -> power delivery circuit (separate for 12V/5V/3.3V) -> power to components. The biggest difference between the wattages (if you keep the design fixed) would likely be in the power delivery circuit
At the component level, the focus is often on the sourcing and tolerances of the capacitors, which are used to clean up transients (very important) and power flow correction, among other things. I think the next most important components are the AC/DC conversion and the voltage transformers. Specifically for higher wattage PSUs vs lower wattage PSUs, the major difference is the amperage along the 12V rail.
A rough chain is: Outlet -> transient clean up circuit -> AC-to-DC conversion -> power factor correction -> PWM circuit (pseudo DC-to-AC) -> 3.3V/5V/12V transformers -> AC-to-DC conversion -> power delivery circuit (separate for 12V/5V/3.3V) -> power to components. The biggest difference between the wattages (if you keep the design fixed) would likely be in the power delivery circuit
Is anybody doing GaN power supplies for desktops? The premium market might be interesting.