pump flange loads victaulic coupling

[rugger80]

I have a piping contractor that is placing a pair of victaulic couplings between the anchor for the piping system and the main pump. I have concerns that large pipe loads will be transfered to the pump flange leading to a collection of problems such as early bearing failure due to shaft couplings forced out of alignment to possible pump flange failure due to fatigue. This is a 3500hp slurry pump operating in series to produce a gage pressure of 350PSI in the 30 inch pipe that is in question. If the victaulic couplings are not at their limit in linear movement, can someone give me a simple calculation to determine the loads transferred to the pump discharge flange?

 

[rconner]

I'm not sure you have provided sufficient information for anyone to conclusively answer your question; however, if there is no axial metal contact in either direction in a grooved and shouldered type joint, I'm not sure you can "transfer" any load axial load [this is a little hard for me to picture though in a practical sense, with assembly, valving/thrust foci, and temperature variations that might be involved etc. -- also, was just curious what style of 30" "Victaulic couplings rate 350 psi?]

 

[rugger80]

I found my information at vitaulic.com/docs/lit. The pressure thrust as they term it is equal to the pipe pressure applied to an area equal to the pipe OD. This thrust or force acts to seperate the pipe in a flexible Victaulic coupling until the groove in the pipe contacts the inner face of the coupling key. In this case there is the potential for 250,000 lbs. of force to be applied against the discharge flange of the pump. The pump is a very robust design but upgrades must be made to the base and foundation to avoid problems in the drive train.

If you are supplying or specifying pumps to be used in systems with Vitaulic couplings,I recommned going to Victaulic's website and reviewing their literature regarding design/installation requirements for their flexible couplings.

 

[rconner]

[Of course if there is no physical anchorage/metal contact of the pump to the piping, as it appears you may be sort of describing, while it appears you wouldn't necessarily be pulling or pushing much specifically on the discharge "flange" per se, it would appear there would however be a pressure induced force against the pump internals, and of course that somehow subsequently transferred on to the external anchorage of the pump by the pump case. I believe that pressure-created force at 350 psi could be something of the magnitude 350 psi[(3.14)(30)(30)/4]in2 ~ 250,000 lb, if 30" is the pressure diameter of the discharge. You might want to post your inquiry on the "Pump Engineering" forum to ask for effect of something like this on specific pump designs or anchorages (e.g. atleast if you don't get suitable answer here.)]

 

[rconner]

Glad you found what you needed. I was not able to get in to the link you supplied, but since your question apparently relates to (dead-end/bulkhead) pressure thrust, I know there is also a Table 17-6 on page 17-13 of the site

http://www.acipco.com/adip/products/Sect17.pdf

that might be helpful for future needs (at least if the piping is ductile iron etc.)

 

[NozzleTwister]

rconner is correct. The pressure thrust on the pump will be against the pump internals and that thrust will be there, with or without the Victaulic coupling. The opposing force will be against your external anchor. This thrust was restrained by the pipe wall when you were hard piped. This is an additional load on your anchor because of the coupling. Also, if you have a trunnion or base ell, you'll have moments also on the anchor.

Also, it worth pointing out that the coupling will allow fluid and pressure into a slightly larger area than just the pipe ID. The ID of this larger area will be the basis used to calculate the thrust against your anchor. The difference between this larger area and the area of the pipe (or pump nozzle) ID times pressure, will result in a force that will be on your pump flange that you didn't previously have.

Question: Why has the piping contractor been allowed to deviate from the installation drawings?

NozzleTwister
Houston, Texas

 

[rugger80]

There are two forces to consider here. One is the nozzle load or thrust which is the force due to the pressure head and change in momentum of the fluid through the pump. This load acts on the internal area of the pump and we have pumps operating with up to half a million pounds of nozzle load with acceptable service life.
The victaulic couplings introduce a second force or load that acts only on the pump flange. We have experienced pump failures ( broken discharge nozzle, bearing issues,etc.) when inadequately restrined flexible couplings have been used on the discharge side of the pump. The style Victaulic coupling in this application has .38 inch movement or .75 inch potential movement for the pair. This movement is restrained by the anchor and the pump, which brings up the concern over the magnitude of this force.

The Victaulics are being used at the request of the end user who cites maintenance issues as why they want the lateral flexibility the victaulics provide.

 

[rconner]

I guess due to unfamiliarity with exactly what you are describing I am having a little hard time understanding exactly the physical mechanism of how just (stretching?) movement in two joints only in line with the axis of a pump discharge would in itself break a pump discharge flange. Again, it would appear there would not really be any metal pull at all on the flange unless/until e.g. a thrust foci contained in piping direction away from the joints/pump somehow pulls all the space out and then makes contact in the keys/groove etc? On the other hand, I can easily see how such couplings/joints particularly if e.g. multiple sort of expanding joints were also used in some connecting piping at right angles to the horizontal pump discharge piping (and the "slack" were not initially/effectively removed from the connecting piping prior to pressurization by the installer?) and if there also happens to be an unblocked 90 bend not far away from the discharge might break a gray cast iron pump flange. The latter case of sort of hydraulically expanding transverse connecting piping would of course be like a huge hydraulic jack trying to break the cast iron flange off the discharge with bending moment (in sort of "wish/pully-bone" fashion). Many years ago I actually saw a 6" diameter flanged intersection broken myself in an event similar sounding to this latter case, when I saw the installer did not restrain across a sleeve-type coupling in some piping that was at right angles connected to the piping coming horizontal off a pump.