3 replies to this topic

[ASH25320]

What is the effect of foaming in solutions on the performance of pumps? Can we expect some erosion on impeller shrouds due to implosion of bubbles?. In the Benfield system of CO2 absorption there is always possibility of foaming of solution due to impurities, which is suppressed by adding antifoaming agent to the suction of pumps. Basically question is of academic interest as what really happens to pump in case of foaming? Thanks in Advance...

[latexman]

Assumption: you are referring to centrifugal pumps.

A foamy solution is a multiphase solution with gas/vapor and liquid which are stabilized by the foam at their interface. It will be more compressible than an identical fluid that is not foamy. I think a foamy solution would affect the capacity, head, and efficiency of a pump in direct relation to how it's effective viscosity and density changes once it gets foamy. Furthermore, since it is stabilized by the foam, I would expect less impact from implosion of bubbles (cavitation).

[JEBradley]

Thanks for making this post. I am (I confess) more a scientist than an engineer in spite of my education and job title as such I'd like to see more posts aimed at encouraging a bit of engineering philosophy

I'm not trying to offer an absolute answer - just add my thoughts.

I believe bubbles caused by foaming would not damage the impeller in the same way that cavitation does - and may not damage it at all. Cavitation is 'cavities' (by which I mean a void) imploding whereas foam bubbles are pressurised 'gas-filled bubbles' (couldn't think of a better word) - how could they implode??

There's also a question here of what is a foaming system. I have heard it used to mean :-

• A system that has the capability of foaming (but to all extents and purposes it's just a liquid)
• A Body of foam - like washing-up lather
• Somewhere in-between. A recognisable body of liquid with a head on top - the pint of beer model

In the first instance pumping would not be effected if the pump has been designed correctly for the system as the tendency for foam formation would be automatically suppressed.
For the second case - The impeller would just cut through the foam with any momentum transfer being taken in by the foams inherent elasticity.
The last case is a real teaser. Like Latexman says - the pump efficiency would decrease along with head and capacity. But in addition the action of the impeller would cause the system to foam further therby losing further head and capacity. I think at the other end of the pipe you'd just get a stream of foam with a trickle of liquid.

Anti-foaming agents would stop this - but that would spoil the question wouldn't it

[djack77494]

Whether a system is foaming or not, there is still an interface between the liquid phase and a second dilute phase - be it vapor or a foamy mix of vapor with some liquid. Centrifugal pumps like to see just the liquid phase, so you must provide adequate NPSHa and vortex breaking so that vapors do not enter the pump suction. Whether the dilute phase is a condensable vapor or a non-condensable gas, it is not meant for the pump. I suggest that you're making too big a distinction between a "clean" liquid-vapor interface and sloppier interface between liquid and (very much like a vapor) foam.

If you want to know more about how sucking foam into a pump suction would differ from conventional cavitation, then search for topics dealing with "Entrainment" of gases into pump suctions.
HTH,
Doug.