I am analysing a pump system. The nozzles loads are within allowables if friction is not considered, but if friction is considered the loads are more than allowables.
What is the general practice? shall the pump nozzle loads be calculated including friction or shall the friction be neglected because of it's transient nature.
Regards
lee

Use the [Search] feature (just above the Calendar to the right), and search for "friction".

I always recommend:

• a case without friction
• a case with the most reasonable low value of friction (mu)
• a case with the most probable value of friction (mu)
• a case with the most reasonable high value of friction (mu)

Then use the Load Case Option MAX to give you a summary of these conditions - which will give you the worst situation.

Remember, friction can only be used to hurt you, never help you.

Do you "live" in a friction less world?????? Think about your question and you will have your answer!

One solution is to add PTFE sliding plates in the supports close to the rotating equipment and thereby reduce the coefficient of friction.

Yes you can add slide plates to reduce the initial value of mu.

But what happens in five years after the support has been exposed to the environment:
- after it gets covered in dust or dirt
- after maintenance paints it
- after a bird builds a nest there

You need to consider the worst (realistic) scenario and account for that.

Friction and pump nozzles are often a problem. Note that B31.3, Para. 319.4.3, REQUIRES you to consider friction when calculating loads on terminal equipment. It's sometimes painful to find a satisfactory solution, but that pain is infinitely less than what you, as the analyst, will be subjected to if you ignore friction and the pump fails.

Sir,

Generally the sliding plate support is designed in such a way that even at the utmost displacement in all directions the sliding plate is not exposed . The upper and lower plates are made out of SS,and the dimensions are selected after checking the displacements.
Also i doubt whether painting will be done underneath the pipe shoe

The third condition, I have no answer , in that case also i doubt bird building a nest at almost grade elevation.

Please forgive me if i am wrong and thank you for clarifying

My point was that "mu" changes over time...

I totally agree with CraigB that we should go thro' this painful process of soving pump nozzles (also any rotating equipment nozzle) with friction. But is the same level of vigil required for static equipments as well?

People end up using lots of sliding plates even on piping connected with static equipments beacuse not all the time we have flexibility to change pipe routing per stress analyst's wish. I agree with Richard that sliding plates are not a reliable, real world solution. It is only a theoretical solution!

advice:

he uses 0,3 for steel - steel
0.15 for steel PTFE
0.1 for Steel - Lubrite

Hello every one,
As our teachers have mentioned above, I want just to add the following:
First run your analysis not considering the friction (0.3 or0.4) do wherever it has to be done ( spring support,loops or expansion joint,etc...) to reduce loads ( forces, momoents, stresses...) within the allowables, and then after satisfying the conditions set your friction (0.15 or 0.1) as additional safety factor.

"The friction will definetly change the pump loads, but should not be used as mothod of reduce pump loads."

Kind Regards,
Yani

With all the focus on Friction, I want to emphasize what Richard Ay mentioned in his first post to this thread.

A run without friction

My experience has been that while friction increases the forces on pump nozzles, many times it will reduce the moments on the pump nozzles by providing resistance to free thermal movement. This is especially true on the suction of a two or more pump system. The friction at the base ell supports actually reduce the moments on the suction nozzles.

It's not uncommon for Pump Nozzle Qualifications to pass for friction analysis and fail without friction. The reactions caused by friction disappear when the support movement stops.

The bottom line is equipment nozzle qualification should be done both with and without friction effects.

I agree with Abdul.The reason being: for static equipments,if we go in by meeting numbers,we in many cases will have to resort to slide plates, however, if we go in by the design by analysis route, where we may even go in for plastic analysis, this type of requirement( i.e. usage of slide plates) in many cases will not be required.

However with rotary equipments, there is no design by analysis route ( as least not known to me) and the issue is casing distortion and misalignment, hence very small numbers.Even though friction has a transient nature ,the momentary high loads during startup and shutdown also can create high loads which can result in casing distortion and shaft misalignment.However, I have a gut feeling ( I love to use the phrase for something which I cannot prove by mathematics or by experiments) that this momentary loading may not be contributing much to the alignment and distortion problem unless it is very high. However we have to follow the numbers mentioned in the standard as these equipments are highly strain sensitive. Personally I try to give more importance due to piping-pump misalignment related problems.

I have seen some consultancies follow the practise of not using friction ( in the analysis as well as non usage of pad ) at the first few supports close to pumps. They have an argument, coefficient of kinetic friction is a function of sliding velocity and drops down as the velocity increases. There will be some amount of vibration close to pumps and this will incresae the sliding velocity at supports thereby lowering the coefficient even further. I don't know how good this argument is.

In WRC 449,it is recommended that even if we are using low coefficient of friction at anti friction pads, the actual value of mu that should be used in analysis should be some percentage higher than the vendor recommended.

Best Regards

TFE slide bearings ah yes... the help for things that must slide... almost....

I was involved in the design construction and startup of a critical piping system that involved the use of these (TFE Slides) to allow 40 Kips of equipment to slide due to thermal growth. I was present during construction and start up to make sure things slid right along.

During start up I had calculated exactly at what temperature enough strain force would develop to cause the precisely made enormously heavy object to break free and start to slide in its progressive stick stop action.


The benchmark temperature came and went, and so with no slide action we poured 200F more strain into the system and still had no movement. So I had a come along applied to the weight and tensioned, then we exerted a 248# pull on the come along line and voila it slid. The published static mu and reality did not match up.

So the real world can often times not match up with your expectation when it regards friction, you have been fully informed of this its your choice how you chose to execute your design. I suggest to chose wisely when it comes to rotating equipment.

I came across thinking about this topic and remember this quote.

"I don´t believe ostrich climb trees to lay eggs"

Just to break a day with humor!

Regards,

The choice of including and excluding or using less friction also depends on the fluid service for which the equipment (Static & Rotating) is used. For example, water service of ambient temperature condition, where the leakage cause less harm, the routing can be simple and not to be complicated with much flexibility in piping. For those cases, the advantage of PTFE (mu = 0.1) can be taken.

As Mr.Richard Ay said that mu decreases with time - it is also the responsible of the operation team to check the supports such as spring hanger, PTFE pad specially provided for pump nozzle loads during the pump maintenance or shutdown time etc...

So from my point of view, the effect of PTFE can be taken for a less harm services and low temperature and pressure services....

I agree both friction and frictionless cases need to be analyzed.
I disagree with the practice of completely ignoring friction at the first support next to a pump.
WRC 449 suggests that piping on a can spring will deflect up to 10 mm horizontally without friction (approx. clearance between load column and spring housing).
Suggest using a factor of 1.5 on nozzle allowables for a non-running pump, and for friction case, as it is a transient condition, not permanent.