Noise Measurements

[crmaris]

first one just baked
This is at 25-30C.

[Philipus II]

Are you sure the colours are right? because there's no dark green, so there's no sound level 25-30. or is it a semi-passive unit?

What's your new ambient level, I'm curious

[crmaris]

there is no dark green firstly because I am still on the making process of the chamber, so my ambient is above 30dBA and secondly because even with this ambient I can safely say that the PSU didn't hit a noise level between 25-30 dBA in these tests. The measurements below 25dBA are possible because the PSU features a semi-fanless mode, so I safely assume below 25dBA noise at this range.

I am pretty sure that even in the chamber most PSUs won't be able to register a below 30dBA noise with their fans spinning, but this is something that I will prove pretty soon once I find the time to finish my project.

[Philipus II]

But there's quite a lot of other green for the "less than 25 dba" area?

[crmaris]

"the measurements below 25dBA are possible because the PSU features a semi-fanless mode, so I safely assume below 25dBA noise at this range."

[Philipus II]

You might assume right

Perhaps just put a passive mode colour explanation there?

[crmaris]

nope the graph is only with db measurements. After all a not operating fan outputs zero noise (so <25 dBA).

[bobnova]

There's inductor whine and such sometimes, that's pretty obvious though.

[rafal_iB_PL]

Hence you cannot quantify a noise comprehensively into a single number. It is at least worth a comment, if noise comes from fan's bearing (esp. in case of ball bearing ones @low rpms), coil whine and such or is it just an airflow noise.

[crmaris]

guys the graph I give will have to do only with the PSU's fan.
Also coil whine has to do in a large degree with the components that the PSU feeds. It may not be present with simulated loads but in a real system to be quite audible. And the other way around is possible of course.

Noise is a very special area and you need tons of money and equipment to evaluate it right and I am just trying with alternative ways (which are carefully planned though) to provide the best possible (and accurate of course) results.