Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Entrained Air Issue?

Status
Not open for further replies.

eekman56

Industrial
Feb 6, 2007
7
Hello. I have a diesel operated 6" trash pump. I'm pumping from a undergound chamber which is 17' deep, 23' long and 8' wide. A 36" line dumps into the chamber @ 6' from the top. My pump suction is located at the opposite end. Vacuum gauge reading when operating is 15". On the discharge side, I have a constant reading of 35psi. The pump is operated via a float switch so the suction never runs dry, never creates a vortex. Whenever the pump turns off via the float switch, and goes to come back on, we are getting air trapped at the top of the volute and the unit won't prime for some time. The unit is equipped with a venturi type priming system which incorporates a float in the priming chamber. We put a small ball valve in the top of the volute and cracked the valve to allow for air release. We also ran a small tube from this back into the water source to ensure the priming system still works when it needs to. This seems to work most of the time, but isn't failsafe. Without it, we have torn engine couplings. We even tried an electric drive unit with the same results. I guess my question is am I dealing with entrained air and if so what are my options. Any guidence would be appreciated.
 
Replies continue below

Recommended for you

I am assuming that when the pump is down, water is held in the case and suction line by the closed discharge check valve and is under a vacuum.
If this is the case I would suspect small air leaks through the shaft seals or various gaskets.
 
The priming chamber has a non-return valve which hold water in the chamber. The sight gauge on the seal oil is clear meaning there isn't any contamination. The couplings on the suction hose is Bauer which is a ball and socket fitting. We took the precaution of checking the O-rings and even wrapped them in the handy-man's secret weapon - Saran Wrap to make sure we weren't drawing air in.
 
Couple of thoughts:

Are you certain that the installation vents the stuffing box completely and the seal is designed for the service? Once fluid begins to move, any air that is back in that area will become trapped by the heavier water being slung to the outside. The seal then may leak.

Is the ambient temperature higher than the water in the sump? If the water warms up in the suction line while idle, less gas is soluble causing air to come out of solution, even if it weren't air entrained from the plunging inflow. You might be able to look at the Flowserve pump manual, which I recall as having a diagram for the proper relationship between in and outflows in a sump.
 
1. The water is probably aerated because the in supply to your sump is free-falling from the supply pipe - this can drive huge amounts of air into the water, however the system seems to be able to tolerate this as you say the pump runs without any problems but would account for the air in the system at each shut-down.

I assume you are talking about an air-driven venturi system. Is this a commercially built pumpset or put together for the job

2. The problem of self-priming after shut down sounds as if it connected to your priming system not the aerated water, float switches etc.
2.1 I would check that the venturi system is not partially blocked and that air supply line is not blocked.
2.2 Check that the nozzle and venturi are not worn and that the nozzle / venturi are positioned correctly in relation to each other
2.3 Check to make sure that the float system is operating properly and not stuck in the "shut" position ie, negating the air-handling capability of the venturi.
2.4 is the unit fitted with a non-return valve on the discharge?

What is a "long time" for repriming - 30 seconds - 10 minutes?

I do not understand why you are tearing couplings. What is the mode of failure as I fail to see why repriming or not repriming has anything to do with coupling failures.



 
I'm not sure why the couplings are tearing but it has to do with the application because of the failure happening to any pump we put in service. The pumps are Gorman Rupp Model PA6D 6" portable diesel trash pumps. We rent these for a number of different applications. The seal is in an oil cavitaty which allows it to run dry and keeps it lubricated. The venturi is operated by an engine mounted air compressor. Both the oil in the seal and the venturi are operating within specifications. I am positive that the fluid cascading into the structure is causing a massive amount of air entering the system but we are at a loss as to what to do to remedy the situation. I am sending a tech out today to remove the non-return valve at the top of the priming chamber to see if starting with an empty suction hose helps the situation.
 
To remedy the problem of air entrainment modify the inlet pipe to the sump to discharge below the nomimal low water level.
 
Great idea Artsi. We will give it a try. What about the couplings? Do you think that the air is creating an unbalancing of the pump which results in the engine coupling coming under stress?
 
I've even seen a "scrubber" on the suction of a water pump to allow air to break out. Then a level control to a valve on the top to vent the air/backpressure for NPSH
 
I can't see the air causing the problem of coupling failure unless of course the coupling is underrated for the application.
As Gorman Rupp pumps are designed to run all day "snoring" (pumping then not pumping then repriming- pumping not pumping repriming)the driver and coupling should also be selected to accommodate this mode of operation.

I assume you are shutting the diesel engine down at each cycle, for interest, what is the frequency and timing (run time - stop time) of start / stop cycles.
 
Artsi,
The unit runs for about 15 minutes and is down for @ 40 minutes. The unit is designed for snoring applications however we have found that whenever we are pumping into pressure, ie forcemain, etc., the volute tends to get airlocked at the beginning but once its released, the air doesnt come back. Unless we vortex.
 
I think I might have the answer to your problem - it is a problem come across about 60 years ago when Henry Sykes Pumps UK were designing what became their Univac and Velovac vacuum assisted self-priming contractor pumps.

On some occasions the pumps would "half-prime" on a re-prime sequence - to overcome the problem they invented and patented the "Univac" pipe - this consisted of a tapping point into the pump case behind the impeller and close to the stuffing box - from there it lead to the priming tank via the "Univac pipe" in which the float level control apparatus was positioned - the point of entry into the tank was below the nominal water level maintained in the tank when the pump was operating- a later refinement was fitting a non-return valve in the univac pipe with the direction of flow from the back of the pump housing to the priming tank.

This was to overcame the problem of air entraped in the volute, this is a catch 22 problem -- the air can't be pushed from the casing until the pump is fully primed and pumping and it can't be fully primed and pumping until the air is removed.


I did have a copy of the patent - but think it has vanished and I can not find it on the web so hope this description is clear. It is probably worth considering as I can see now other explanation.

As for the coupling failures - still in the dark.
 
Sounds like a great idea. In fact I had read an article in pump and systems magazine where someone was having a similar problem and their solution was to inject water into the suction side which would help "push" the air through the volute. We will give it a try. Thanks all of you for your help. I'm sure I'll be counting on you again in the future.
 
A quick fix I have used many times for pumps drawing from a lower level, is to fabricate a loop on the suction side of the pump. The loop is high enough so that when the pump is stopped the casing is always flooded. On start -up of the pump it is self primed.
Other alternatives are to fit a check valve on the suction or a foot valve at the inlet pipe (not recommended for slurry)
 
I have seen self priming pumps burn the belts off when running dry for extended periods of time. As the pump is running dry you are converting the mechanical energy into heat in the casting this liquid can start to build steam pressure in the casting. This is why Gorman Rupp started installing pressure relief valves on the casings a few years ago. At the higher pressures at no flow you could be overloading the coupling grid.
 
I fail to see how an increase in pressure in a centrifugal pump can cause any problems to a belt drive. In a pump like Gorman Rupp with its radial flow impeller as the flow reduces and head increases, power reduces. It follows then at no flow the couplng is not overloaded.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor