- Joined
- Mar 1, 2013
Previously, I wrote about how PFC doesn't really benefit home users and its cost and efficiency penalties. Back then, I mostly worked on power supplies for stationary equipment where weight and size are relatively small factors in the overall design of the product. An extra pound is of little importance in a machine that weighs 40lbs and there's plenty of space to enlarge the PSU by a few extra cubic inches. (Very high power loads like EV chargers do benefit from PFC in a residential setting but are outside the scope of this discussion.)
But now, I working on a portable device (compact gaming/media center/LAN party PC) where size and weight are what set it apart from existing high performance PCs on the market. (It's actually a project requested by a friend.) Basically, it's a PC about the size of a Bluray player, and that includes a quad core CPU and a standard desktop GPU. In such an application, not only space is at a premium but also heat dissipation. As it is intended for home use, PFC offers little benefit but plenty of disadvantages. The parts needed to implement it cost something, take up valuable space, add weight (not that an extra pound would make much difference), increase heat dissipation, decrease efficiency (could be mostly mitigated with digital PFC, but at a higher cost), and decrease reliability. Of those, size and heat are especially significant factors in the particular application I'm targeting. Then there are applications like portable device PSUs and retrofit LED lighting where size, heat, weight, and cost are even more of a big deal. I have taken apart a laptop PSU which failed because the PFC stage failed in a spectacular manner from overheating, and a surprising amount of its internal volume was spent on PFC.
Now the problem lies with clueless consumers who assume PFC is good to have because the PSU manufacturers advertise it. It doesn't help that PSUs without PFC tend to be cheap, poor quality generics, thus associating the lack of PFC with poor quality. And then there are those who perpetuate the myth that a PSU without PFC is akin to a naughty girl practicing omorashi in the water at a pool party, when PFC causes even more EMI. (There's a reason the FCC doesn't care about harmonics below 150kHz.) PLC devices like Homeplug are very sensitive to EMI on the line yet PFC rectifiers bother them a lot more than non PFC rectifiers. Finally, a number of efficiency standards (e.g. 80 Plus) ironically require PFC.
What I think would be a good solution is to revise the efficiency standards to exempt strictly home use devices from the PFC requirement as long as they pass stricter tests for efficiency, reliability, and/or EMI. For example, where a PFC PSU needs to get 89% efficiency at a given condition a non PFC PSU must get at least 92% efficiency. In other words, make PFC optional but with strings attached so OEMs don't just pocket the savings from cutting out PFC, but rather deliver a superior product to the customer.
But now, I working on a portable device (compact gaming/media center/LAN party PC) where size and weight are what set it apart from existing high performance PCs on the market. (It's actually a project requested by a friend.) Basically, it's a PC about the size of a Bluray player, and that includes a quad core CPU and a standard desktop GPU. In such an application, not only space is at a premium but also heat dissipation. As it is intended for home use, PFC offers little benefit but plenty of disadvantages. The parts needed to implement it cost something, take up valuable space, add weight (not that an extra pound would make much difference), increase heat dissipation, decrease efficiency (could be mostly mitigated with digital PFC, but at a higher cost), and decrease reliability. Of those, size and heat are especially significant factors in the particular application I'm targeting. Then there are applications like portable device PSUs and retrofit LED lighting where size, heat, weight, and cost are even more of a big deal. I have taken apart a laptop PSU which failed because the PFC stage failed in a spectacular manner from overheating, and a surprising amount of its internal volume was spent on PFC.
Now the problem lies with clueless consumers who assume PFC is good to have because the PSU manufacturers advertise it. It doesn't help that PSUs without PFC tend to be cheap, poor quality generics, thus associating the lack of PFC with poor quality. And then there are those who perpetuate the myth that a PSU without PFC is akin to a naughty girl practicing omorashi in the water at a pool party, when PFC causes even more EMI. (There's a reason the FCC doesn't care about harmonics below 150kHz.) PLC devices like Homeplug are very sensitive to EMI on the line yet PFC rectifiers bother them a lot more than non PFC rectifiers. Finally, a number of efficiency standards (e.g. 80 Plus) ironically require PFC.
What I think would be a good solution is to revise the efficiency standards to exempt strictly home use devices from the PFC requirement as long as they pass stricter tests for efficiency, reliability, and/or EMI. For example, where a PFC PSU needs to get 89% efficiency at a given condition a non PFC PSU must get at least 92% efficiency. In other words, make PFC optional but with strings attached so OEMs don't just pocket the savings from cutting out PFC, but rather deliver a superior product to the customer.
