Magnus wrote:
My point was, the thermistor is never "happy." It always wants more
cooling. So it spins the fan up to full speed and is still too hot to
reach equlibrium.
Which only means that the thermistor setup is shifted and needs to
be adjusted to achieve the goal.
* * *
So, the question is, do we just replace fans when
they go bad and live with the poor thermal design, or do we try to
improve the thermal design? If we want to improve the thermal design,
the methods available to us are:
I think a sub-set of these are needed, but you also need to include
(v) change the balance-point for the thermistor stabilization.
Well, sure, you can raise the setpoint until the fan slows down and
regulates the temperature, but then the regulated temperature would
be too high. In other words, if the fan cannot keep things cool
enough by running at full speed, raising the set-point so the fan
runs slower is not the right answer even if it does result in some
approximation to isothermy. [Note that adjusting the setpoint *down*
will have no effect -- the fan already cannot cool the instrument
enough to reach equilibrium.]
I think the single biggest problem is too little air inlet, which
forces the fan to run in stupidly high speed without actually do
much, as the air input has too low cross-section and with several
small holes you need to create a large pressure difference in order
to achieve the air flow wanted. Also, those holes help to create
noise as they are not shaped to avoid turbulence. Best way is just
achieve large enough cross-section.
I have experimented with cardboard top covers that have various holes
in them. My conclusion is that even with more air inlet area, you
still want more cfm of air flow (i.e., a stronger fan). You also
need to adjust the airflow to the various sections of the interior to
keep everything sufficiently cool. As it is, there would be way too
much temperature variation from one part of the interior to another
even if you did reach a regulated equilibrium for the thermistor
location (wherever you put the thermistor).
Best regards,
Charles
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
