MadSci Network: Engineering
Query:

Re: Why does air trapped in a system of water filled pipes cause flow to 'Lock'

Date: Tue Nov 11 13:01:25 2003
Posted By: Gavin Ward, , Physics, AEA Technology
Area of science: Engineering
ID: 1065605369.Eg
Message:

Hi Richard,

The relevant parameters in this problem will be:
the quantity of air
the water pressure  
whether the system is open or closed to the outside.

Imagine a hose pipe with water running through it.  If you stop the water 
flow a bubble of air can enter the open end and run back up the hose 
pipe.  If you make an upward loop (inverted-u) or coil the hose so that 
the air bubble is trapped in the top of a loop (or a coil) it will trap 
any remaining water upstream.  The air bubble will always rise to the top.
If you turn the water flow on again, because the hose is open at one end 
it will push the water and air out of the pipe - if the flow is fast 
enough.  There will probably be a bit of spluttering as the air exits but 
eventually a continuous flow of water will resume.  

This was an OPEN system.


If the system is a CLOSED system of plumbing (as in a central heating 
system or a car's cooling system) and air is trapped in a loop it will 
block the water flow - an air lock.  This is particularly so under 
conditions of low flowrate or low pressure as is ususally the case in a 
domestic setting.  

So why does it occur?  The air bubble will always rise to the top of the 
loop and if it is a big bubble it will fill the diameter of the pipe at 
the top of the loop, it will effectively create a dam preventing the water 
from flowing over the dam (under normal domestic pressure conditions).  
And there the bubble will stay resisting the water flow if the force of 
the water flow is small compared to the force of the buoyancy keeping the 
air bubble at the top of the loop.  Just try submerging a block of 
polystyene foam in a swimming pool or a bath - it really doesn't want to 
go under!  So if the flow rate is low, and the bubble big, the buoyancy of 
the bubble will exceed the force of the flow and will resist it.   

In the situation we are examining, the trapped air bubble will be at (or 
close to) atmospheric pressure - that's about 1 Bar or almost 15 psi.
1 atmosphere pressure can hold up a column of about 780mm mercury or 34m 
of water! http://www.atmos.washington.edu/2003Q3/101/demonstrations/water_bar
ometer.h
tml

So how can the situation be resolved?  If you could decrease the size of 
the bubble to enable the water to flow over the dam that would work.  So 
you could try pressurising the system to increase the pressure:  this will 
compress the bubble in the top of the loop and if this makes the bubble 
small enough so that it no longer fills the diameter of the pipe the water 
will begin to flow again.  A smaller bubble is also less buoyant which 
will help.  (If the resultant flow is high enough the bubble might even be 
dispersed by disolving in the water, or broken up and swept away by 
the turbulent flow.)

Doubling the pressure will halve the volume of the bubble, quadruple the 
pressure and the volume will be a quarter (Boyle's Law). But will your 
pipes stand the pressure?  Maybe not, so this will only work if the bubble 
is fairly small.  What if the bubble is very big - such as in a long 
vertical run of pipe?   Increasing the pressure just won't be practical.  

Running the pump (if any) on your system acts in a similar way by 
increasing the pressure on one side of the bubble, but in a closed system 
it reduces the pressure on the other side so all it really does is push 
the bubble over a bit and make the bubble unsymmetrical at the top of the 
loop i.e. on one side the water level will rise and on the other side it 
will fall but the bubble size will be much the same.  However this might 
be just enough to raise the water level upstream sufficiently to enable 
the water to flow over the dam at the top of the loop and maybe the bubble 
will eventually be dispersed by turbulence in the flow.

However the real solution for a big bubble is like the hose pipe example.  
Open the system at a drain valve and pump fresh water though the system 
and the water will push the bubble out - if the flow is fast enough.

In a domestic setting a solution to an airlock in the hot water system is 
often to connect a hot tap to a mains pressure cold tap (using a length of 
hose).  The mains pressure is always higher, often much higher, and the 
hot system will be an open system as there will be an always-open 
expansion outlet - or just open the other hot taps on the system.

Irritating as an air lock may be in a domestic setting it's even more 
involved in an industrial setting where the quanitities of liquid may be 
enormous, dangerous (such as fuel), expensive or contaminating too - so 
you may not be able to flush it out down the drain, you may have to 
collect it - all of it.  So the plumbing is an important part of the 
design of an industrial installation or of a vehicle.  There are 
additional complications with vehicles, particularly fast vehicles such as 
aircraft where there are dynamic forces (G-forces) and the fuel can form 
its own bubbles by vapourising at low pressures (vapour locks) at altitude 
or due to characteristics of the system.   This limits, for example, 
operation of an aircraft on ordinary petrol (gasoline) which vapour locks 
relatively easily so different fuels are available for aircraft - but 
that's another story. 

I hope this helps somewhat.

Regards,
Gavin Ward

A "tap" in British English is the same as "faucet" in US English.



Current Queue | Current Queue for Engineering | Engineering archives

Try the links in the MadSci Library for more information on Engineering.



MadSci Home | Information | Search | Random Knowledge Generator | MadSci Archives | Mad Library | MAD Labs | MAD FAQs | Ask a ? | Join Us! | Help Support MadSci


MadSci Network, webadmin@www.madsci.org
© 1995-2003. All rights reserved.