David wrote:

You seem to have missed the next sentence, which said they are
"extremely quiet."  So quiet that I could reduce the noise of the
internal fan by 20dB and still not be able to tell whether the external
fan was running without looking at it. [1]  So no, as a practical matter
there is no additional noise.

Two different things.  The internal fan primarily cools the inside of
the counter, while the power supply heat is developed on the heatsink,
which is mainly cooled by convection outside the enclosure.  (Yes, the
heatsink has a couple of little holes in it where air from inside the
counter blows out -- but that is completely ineffectual at reducing the
heatsink temperature, and I doubt that's what the designers intended.
More likely, they were trying to minimize the amount of heat from the
heatsink that gets inside the box, by providing a path for air to flow
from the inside, past the heatsink, to the outside -- and at that, the
holes seem effective.)  So, if you reduce the internal fan airflow, the
internal circuitry will run hotter even though the power supply pass
devices won't (much).  Further, if you replace the heatsink-mounted
regulators with switchers that are located inside the box, you may very
well put MORE heat load on the internal fan even though the switchers
are much more efficient (because what loss there is would be dissipated
inside the box, rather than outside).

Best regards,

Charles

[1]  I actually tried an internal fan that reduced the noise by 22dB,
but it didn't have sufficient airflow to keep the interior properly cool.

In message CANy2iXq6ONvriDoFgNFkEBQjDKntP7t8KuE7BOUpxJWLCuxLsw@mail.gmail.com
, "William H. Fite" writes:

... at zero pressure differential, which is easy to do (Think: ceiling fan).

The hard part is to make fans silently move air when they also have
to build up pressure to get the air to move places.

This is where the gap between the wings and the frame must be minimized,
or the pressure will leak out through it.

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

On Fri, Jul 15, 2016 at 10:55 AM, Poul-Henning Kamp phk@phk.freebsd.dk
wrote:

Right.

You have to look at the curves on the data sheet that shows air flow vs.
static pressure (and be careful about the static pressure scale).  I found
that a 'quiet' fan would often be flowing one tenth as much air as the
original fan at the static pressure at which the original fan was rated.

In a given instrument, you may get away with the quieter fan, but how would
you tell other than putting a thermometer inside and making a before/after
comparison?

Orin.

The fact remains that there are some very quiet equipment fans besides the
little muffins that sound like drones lifting off.

On Friday, July 15, 2016, Poul-Henning Kamp phk@phk.freebsd.dk wrote:

--
Yes, Black lives do matter.

That is, in fact, precisely how you do it.

On Friday, July 15, 2016, Orin Eman orin.eman@gmail.com wrote:

--
Yes, Black lives do matter.

Actually, the lion's share of noise heard from a fan is NOT the moving
air, but the turbulence over the fan blades... This is a VERY large
topic that I don't want to bore the list with.

Clay Autery, KY5G
MONTAC Enterprises
(318) 518-1389

On 7/15/2016 8:17 AM, Dr. David Kirkby (Kirkby Microwave Ltd) wrote:

Yes, there are absolutely big differences between fans of same
physical footprint.

But most of the "so quiet we had to break the laws of nature" fans
aimed at the PC market don't do anything like that, and it would
be a big mistake to use such a fan to replace the HP5370 fan, where
a significant pressure is required.

For attaching to the cooling fins to get some air moving, they're
fine however.

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

Some thirty years ago fans for automotive radiator cooling were designed with 7 unevenly spaced blades to reduce the siren effect yet yield comparable air flow. Careful layout and blade sweep back along with an annular ring made it entirely feasible to have a statically and dynamically balanced fan as it came out of the mold. I always wondered why the folks who make these little fans can't figure out how to do that. Maybe getting comparable CFM from a much smaller fan violates some Reynolds number requirement for turbulence control.

From Tom Holmes, N8ZM

Hi

Back when these pieces of gear were being made “new”, HP Spokane had a
“fan silencer machine”. They used a strobe and a microphone to identify imbalanced
blades and then took chunks out of them. They claimed it made the fans significantly
less noisy….Watching it in action on a noisy factory floor … not so clear.

Bob