[Amps] tube cooling

[Ian White, G3SEK]

Gary Schafer wrote:
>
>You missed the point. Again: "You can have as much turbulence sucking
> air as you can by blowing it."
>This is the key: "It depends on the speed of the air and
> the surface that the air is passing over as to how turbulent it is."
>
>If the air is not going through a duct, and a fan is in free space then 
>yes there is a difference between inlet and outlet air flow. It is more 
>concentrated at the output than at the input. There is more velocity 
>directly in front of the fan than there is on the inlet side of the fan.
>
>Cooling a tube you are forcing or pulling air through a duct. (the tube 
>chimney)
>
>Blow air through the cooling fins of a tube with a chimney or suck the 
>air through and you will get the same amount of cooling of the tube.
>
We've been arguing at slightly cross-purposes. I agree that the in the 
finned anode cooler and the chimney, it doesn't matter whether you suck 
or blow. That is because the air flow is so tightly confined, only one 
flow pattern is possible and the pressure difference is the only thing 
that matters.

But that  is the *only* part of the entire amplifier where the airflow 
is tightly confined in a duct. Everywhere else in the amplifier the 
layout is much more open, so there are many possible flow patterns. 
These patterns will be very different between "suck" and "blow".

In a fairly large volume like the power supply compartment or the output 
tuned circuit compartment, the air flow will always take the path of 
least resistance. That path is generally through the most open spaces - 
in other words, the air flow will mostly *avoid* the components that you 
want to cool! Blind ends, corners and locations like the middle of the 
electrolytic stack will tend to remain stagnant and not well cooled.

The only way to get cooling air into those dead spaces is to *blow* it 
in, using the turbulence created by a fan located close upstream. 
Sucking harder at some point downstream won't help much... it will only 
pull more air along the path of least resistance, leaving the stagnant 
areas pretty much the same as before.

The auxiliary fan in the Alpha-type airflow layout is so effective 
because it swirls cool air all around the transformer and the power 
supply compartment.
But as John points out, the auxiliary fan *must* be big enough to 
deliver the same flow rate as the main blower, or else it will choke the 
airflow. Then the total flow rate through the whole amplifier will 
decrease - the power supply will be cooled better, but the tubes will be 
cooled worse.

(Actually the aux fan must be able to supply a slightly higher flow rate 
than the main blower, because some of that flow will be lost through the 
vent holes in the power supply compartment  - tnx for pointing that out, 
Sam.)

Anyhow, the Alpha owners now have their answer from the horse's mouth!


-- 
73 from Ian G3SEK         'In Practice' columnist for RadCom (RSGB)

http://www.ifwtech.co.uk/g3sek