- 52 months ago
Got a PM on reddit recently about listing PSU tiers and PSUs with 30C rated components.
I've seen comments that some power supplies only have capacitors rated for 30C. Stop. Please stop. It's like a bad rumor that's taken on legs and the story gets more outlandish the longer it goes on. Next thing you know we'll have people saying that if the room exceeds 30C your PSU will take out all the components in your build. Oh wait, that's already happened.
If we're going to talk about specs on a PSU, show proof and talk technical details. Cite reliable sources. If you're going to say it isn't designed for gaming builds, show proof. (Marketing blurbs saying a PSU is designed for price-conscious users does not prove it can't power a gaming build.) Real and reliable bench tests don't lie.
As for my background, I studied and competed in electronics servicing. Even won at the national level. I loved it - I studied college EE textbooks in high school, studied EE in college until switching to computer engineering (effectively a CS major with EE minor). Electrical engineering has always been my first technical love, and it's a domain I continue to try and grow in. I've built plenty of circuits from class A amplifiers to custom PC expansion cards to (non-switching) power supplies, etc. Designed and built plenty of other circuits, and fixed many more.
Plenty of people talk about capacitors like it's some linear component - like if the capacitance value is double it's twice as good. But it doesn't always work that way. The values are chosen and tuned, and what size they need to be, what voltage they need to be, and what tolerance rating they need to have depends on how it's used and what kind of circuit it's in. Sometimes you need narrow tolerance caps. Other times, having one that is rated for -20/+85% (like a lot of electrolytics) is perfectly fine and the circuit is designed to handle that variation and remain in spec. That's the EEs job. There is not just capacitance, but also have leakage currents, inherent resistance, temperature coefficients, etc. It's not as simple as brand X vs Y. But people without the background don't know this, and assume that cap X vs cap Y is so much better, or that cap Z is bad, etc.
So when it comes to claims about components, particularly power supplies and voltage regulation, my bar for what constitutes valid evaluation and feedback is considerably higher. My tolerance for bogus facts and erroneous claims is extremely low. If you see posts on review sites talking about capacitor "performance", consider it with a high degree of skepticism. No EE that I know of talks about caps in some absolute term of "performance". There's rated specs and adherence to that, but not some linear scale "performance".