Pump churning.

Thread Starter

recca02

Joined Apr 2, 2007
1,212
Hi everybody,
While learning about power plant operation today, when the pump for the condensate extraction was to be started, we noticed that a recirculation valve was open to 20%.

So,
we came to know that 20% recirculation of water in pump is done to avoid pump churning.

The Question:
But what exactly is this pump churning? When does it happen?
From what I Know, this seems to happen when discharge of pump is closed and perhaps the suction is closed(?). So for a high power pump (rated in MW) the small quantity of water entrapped in pump casing gets heated up. This rise in temperature can cause undesirable deformation in pump hence not desirable.
Am I correct?..Any ideas?

Edit: The pumps are of centrifugal type and maybe the phenomenon is sometimes simply referred to rise in temperature.
 

beenthere

Joined Apr 20, 2004
15,819
The pump may damage itself while running with low flow due to cavitation. The bubble formation can do mechanical damage to the impeller vanes.
 

Thread Starter

recca02

Joined Apr 2, 2007
1,212
Yes, cavitation in itself is a huge problem. Perhaps churning of fluid can lead to such a temperature where at suction there is not sufficient NPSH leading to formation of bubbles.
I think I'm now more or less convinced of the situations when churning can happen.
 

DC_Kid

Joined Feb 25, 2008
1,072
hmmm, wont a pump with in/out's closed still want to push what is has in the impeller casing... on a MW pump that could be a problem.
 

Thread Starter

recca02

Joined Apr 2, 2007
1,212
That is exactly where i have a little trouble analyzing. What happens after all the fluid has been pressurized and is now on the high pressure side? Regardless, I think the pump will continue to add energy to the fluid which now turns up as heat.
Does this condition also occur when suction valves are is open?

Edit: The problem is that I have found very little literature over this thing, though this is a huge problem when dealing with large size pumps, cavitation on the other hand is well explained in wikipedia and some books which deal a little bit with pumps. The recirculation valve is opened so that a little system resistance is introduced and the energy is lost in overcoming this resistance and establishing the recirculation flow.
Interestingly, even Turbines have churning problem when, High Pressure and Low Pressure(and perhaps Intermediate Pressure also) turbines are used in tandem.
Edit2: There is possibility of churning in HP turbine when load drops by about 20% from rated value, or the pressure in reheat line increases above certain pressure.
All this is what I have Inferred from bits of information here and there, it wud be nice if I could confirm these from somewhere.
 

DC_Kid

Joined Feb 25, 2008
1,072
That is exactly where i have a little trouble analyzing. What happens after all the fluid has been pressurized and is now on the high pressure side? Regardless, I think the pump will continue to add energy to the fluid which now turns up as heat.
Does this condition also occur when suction valves are is open?

Edit: The problem is that I have found very little literature over this thing, though this is a huge problem when dealing with large size pumps, cavitation on the other hand is well explained in wikipedia and some books which deal a little bit with pumps. The recirculation valve is opened so that a little system resistance is introduced and the energy is lost in overcoming this resistance and establishing the recirculation flow.
Interestingly, even Turbines have churning problem when, High Pressure and Low Pressure(and perhaps Intermediate Pressure also) turbines are used in tandem.
Edit2: There is possibility of churning in HP turbine when load drops by about 20% from rated value, or the pressure in reheat line increases above certain pressure.
All this is what I have Inferred from bits of information here and there, it wud be nice if I could confirm these from somewhere.
i have some studies in fluid dynamics (nuclear engineering studies). i'm figuring if the pump has huge torq turn on properties the impeller blades might get damaged if the fluid has no place to go. after that i suspect the frictional equations start kicking in as the fluid picks up speed, then when fluid gets to the cavitation point i suspect there will be both frictional and impact equations acting upon the impeller blades and housing..... the caviational issue will transfer forces into the drive shaft, which if at the right frequency can probably cause nasty harmonics in the drive and possibly break the drive mechanism(s).......
 

Thread Starter

recca02

Joined Apr 2, 2007
1,212
The problem that happens in churning has perhaps got more to do with different rates of thermal expansion of impeller blades and casing and thermal stresses. there is very little clearance between them and a little difference in expansion is detrimental.
Cavitation OTOH is like a huge pressure difference created AFAIK when bubbles collapse under turbulence leaving a high pressure liquid rushing into a low pressure zone.
I will check up some books on this if possible and post back.
 

subtech

Joined Nov 21, 2006
123
Recca
I'm not quite sure, but if memory serves, it seems that there is a concern with the pump evacuating the condensate tank and pulling in air. Recirculation of some amount of water ensures that the pump is moving liquid 100% and no air that would allow foaming or other problems. I'm still reading a few articles I've found using google, if I find a coherent explanation, I'll be sure to post the link

Regards
 

RimfireJim

Joined Apr 7, 2008
22
I just joined the forum and saw this question, so my answer is a little late but hopefully useful.
The purpose of having the recirculation valve and line, or "churn" valve as it is called on some fire engines, is to provide some cooling for the pump in the event the discharge valve is fully closed. A centrifugal pump is non-positive displacement, meaning it can be turning even though the fluid is not pumping on through. The load on the motor goes down, as there is no mass transfer at that point, but a lot of energy is still being put into the system, and it has to go somewhere, so it goes to heat in the fluid. Eventually, things will get mighty toasty and bearings and seals will get cooked. Letting some of the fluid recirculate will dissipate some of the energy. Works up to a point, depending on how big the recirc line is and how long the pump runs that way, and other factors.
How do I know this? Four years working in a US Navy engineroom, a bachelor's in mechanical engineering, and 15 years as a volunteer firefighter, plus growing up around pumps in my dad's shop. One time on the ship, when I was still pretty new, I started up one of the fire system pumps and forgot to open the discharge valve. 100 hp electric motor, 6" line. The pressure gauge tap is before the valve, and the rest of the fire system is already at 150 psi, so you can't tell by looking at the gauge if the valve is open or not. I came back on my rounds a while later and that pump casing was so hot I couldn't even touch it. Fortunately, the bearings and shaft packing survived and I didn't get my heiney chewed out. (Nobody knew but me . . . and now, you.)
 

ASV2507

Joined Nov 26, 2008
1
Hi guys, I have just joined the forum after I got interested in discussion on Pump Churning.
I have two very interesting cases on the subject to share. In the first case, it was a cooling water pump, pumping sea water to the process units @ 2500 cu.metre/hour rate. I was informed that there was steam and hot water gushing out of the glands. Upon investigation, we found that the pump discharge valve seat had dropped when the pump was started a few hours back. Means, the valve spindle and the seat or gate got disconnected and the operator thought the valve was open (the spindle had come out) while actually it was closed. Since the pump was serving a system with two more pumps running, the failure to deliver could not be immediately noticed. Churning had caused so much rise in temperature of the trapped water inside the pump casing, that it came out of glands as steam and boiling water. And all this happened while the suction valve was fully open and enough water available in the basin. So I have no doubt that churning is caused even with suction valve full open.
In the second case, a pump in heavy oil service caught fire on its mechanical seal some 1 to 1-1/2 hours of operation with restricted flow. I could imagine from my experience with the cooling water pump how the temperature rise in the casing can cause the seal to blow. This is exactly what had happened and hot oil gushing out of the seal caught fire as soon as it came in contact with the atmospheric oxygen. Interestingly, the oil handled was supposed to be much below its flash point.
So guys, do not take churning lightly. It can play havoc if allowed to continue for long time.
 

BillO

Joined Nov 24, 2008
999
RimfireJim has given a great reply. With the output valve closed, all the energy being put into the pump ends up as heat. Several MW is a lot of heat. Once the fluid being pumped begins to boil or burn, and this will happen concussively as the churn guarantees even distribution of the available energy, the pressure rises catastrophically.

Also, as Jim mentioned, the rotational load on the pump is very low, as it is not doing much work (other than heating the fluid). While this is not a real problem for electric motor or IC engine powered pumps that tend to have built-in speed restrictions, it can be catastrophic for turbine powered pumps as the RPM can rapidly rise beyond the design limit causing the impeller to disintegrate.

All fun stuff.
 
Top