The first impression we had when opening this power supply was that we were in front of a very low-end (“generic”) unit that was put inside a nice housing as the printed circuit board was too small for the size of the housing...
This power supply uses a mix between new and obsolete designs, showing us that the manufacturer instead of creating a new design from scratch, adapted an old design.
The main difference between this power supply and newer (and better) models is how power is distributed. This power supply was created when most of the power drawn by the computer was concentrated on the +5 V line and not on the +12 V line like it is today. We can say this because it uses a rectifier with lower specs for the +12 V line and the rectifier with higher specs for the +5 V line.
When we tried increasing one amp (from 16A to 17A) at +12 V, noise jumped to 190 mV and skyrocket to 680 mV when we tried pulling 18 A from it – and, remember, according to the power supply label +12 V could deliver 18 A.
The conclusion is that according to our methodology Thermaltake Purepower 430 W NP isn’t a 430 W power supply, but a 350 W model! We also could only pull 16 A from its +12 V output, while the label says the limit is 18 A.
This is an old ATX power supply where the manufacturer added a 24-pin motherboard connector, SATA power cables and a PCI Express auxiliary power cable to make it compatible with computers available today. Simply updating the cables doesn’t make this power supply an updated product. This is so true that this power supply is listed as ATX12V 1.03 by Thermaltake, and not as ATX12V 2.x, despite the presence of the 24-pin motherboard connector and the 6-pin PCI Express auxiliary power cable for video cards.
The main problem with this power supply is that it can’t deliver its labeled power. It is, in fact, a 350 W power supply.
You could buy it as if it were a 350 W unit, but when we pulled 355 W from this power supply noise level was touching the maximum admissible limit and efficiency was at 69.6%. With other load patterns the maximum efficiency we saw was 76.9%.
Our conclusion is pretty simple: don’t buy this power supply.
Bookmarks