Hardware Insights Forum

So, this was to be my last review here, but unfortunately, I just don't believe my results. So, I have opted not to publish them. I'll just say that it all seems far too good to be true. So, as a compromise, I'll just post a few pics of it and its internals here.

It looks to me like they used the resistance of the filter coils to measure the +12V currents.

Obviously that 15A rated input draw is impossible and the +5V output can't be capable of 24A if it and +3.3V can deliver no more than 100W between them. And that upside-down SMD resistor is also a stunning QC fail.

I'd also prefer if they used single thick wires rather than bundles of thinner wires to the modular connector board, e.g. 1 × 12AWG instead of 4 × 18AWG, although that's perhaps nit-picking on my part. I can't say I feel totally comfortable with that naked EMI filter board in fairly close proximity to a few of the secondary-side wires, for that matter, and that threaded stud attached to the rear panel is clearly intended for the earth connection, so why didn't they attach the earth wire there (instead of holding it under the PCB screw and running the risk, however slight, that someone attempting to repair the unit will fail to put it back)? If it's any consolation, at least they used a larger MOV than in many other units.

c_hegge wrote:So, this was to be my last review here,

Is it okay if I ask why (is Behemot taking over?)?

Yes. By the way, I've also been given the role of quality control here. (I've done it on an article c_hegge has drafted on dealing with dust, and would have done it on the intended review of the unit in this thread if it wasn't for the questionable results.)

I may eventually write my own content, although I presently have other priorities.

Wester547 wrote:Is it okay if I ask why (is Behemot taking over?)?

Pretty much for the same reason as Longrunner. I just have too many other important things going on right now and don't have the time (or desire, really) to write reviews anymore.

What was the problem with your measurements BTW? You haven't told me on BadCaps…

I hinted at it with the comments on one of the pictures. First of all, not even severe crossloading made the voltages budge. It was absolutely 0% variation on all 3 rails the whole time. Not even Delta can do that on a $400 PSU, let alone a no-name $80 consumer PSU.

The efficiency was also quite a bit worse than what 80plus got - it failed to meet the required 85% at 20% load, even though it was being run on 230V, not 110V (and so it should have been a bit more efficient).

Between the unusual results and the fact that I don't have particularly high end equipment, I question the accuracy enough not to publish it.