pressure pulsations in piping

[DYV1973]

LS,

If a piping system for the transport of oil is suffering from pressure pulsations, how can the effect further in the piping be calculated.
Pressure is approx 4 bar, and pressure peeks up to 25 bar (in miliseconds). What will the effect on the pressure be 50 metres further in the piping. Fluid velocity is 1 m/s and piping is carbon steel, and customer does not want to use pulsation dampeners.

Main issue for me is; what is the dampening effect from the length of the piping.

Thanks,
DYV

 

[C2it]

I think you have a potentially serious fatigue problem to deal with.

Damping will depend on diameter, wall thickness, material, fluid viscosity etc., but in any case, 21 bar is a big swing and must cause significant stress fluctuation.

 

[BigInch]

No. Your steady state velocity is 1m/s. The pressure wave is traveling much faster than that in the fluid, reaching the end of the pipe in 1/10th of a second or so, and even faster in the pipe material.

What's causing the spike?

I tend to think that's worse than a fatigue problem, meaning time to failure may be eminent, depending on material and wall thickness. At the least, it is also causing negative waves of the same magnitude meaning that cavitation, in addition to probable rapid strain hardening of the material. Unless you have a design pressure above the 25 bars, I wouldn't predict a very long life for this system, so your client will most likely change his mind soon.

Since the wave particle velocity is traveling at 1000 m/sec or so, local velocities in that wave are much higher. Pressure will decay from the 25 bar peak in relation to the friction caused by that local pressure wave particle velocity as it moves down the pipe.

From "BigInch's Extremely simple theory of everything."

 

[DYV1973]

@ BigInch
Pressure pulsations are caused by a piston pump. As customer does not want to take precautions, we can just warn him. At least I want to make a calculation for our delivered equipment. With normal flow, I can use Darcy-Weisbach to make a pressure drop calculation, but this will not work for a pressure wave of 1000 m/s. My aim is to know the approximate value of the pressure peak 50 metres furter downstream. Is there standard equations for it (with making some assumptions), as there might be a lot of factors having influence.

 

[BigInch]

Use DW head loss to get the pressure profile at 1 m/sec, then superimpose the spike on that head loss line as the spike travels down the length of the line. A short steel pipe is very rigid when used as an accumulator or dampener. This is more of a sonic pulse - echo situation, where you would probably get 4 or 5 echos in quick succession within 1/2 second or so, but each echo reflection about half the amplitude of the previous spike. With such high spikes, the low spikes are at 0, probably reaching vapor pressure of the oil, with cavitation, perhaps forming small vapor pockets, so accurate predictions become impossible.

How many piston strokes do you get per minute? More than 600 and basically you will have one big continuous spike. You didn't mention continuous spikes, so I don't know, but my guess is that you are probably running at a relatively low rpm in relation to wave speed.

From "BigInch's Extremely simple theory of everything."

 

[McCall920]

Bare minimum we would need fluid viscosity to figure the compliance through that pipe... Plus we'd need the dimensions of the pipe to factor volume for the compression/damping of the fluid. Frequency is going to be a key element here. Like BigInch said, at a high frequency, you're piping will effectively see a continuous peak of 25 bars, which is much preferable over a 21 bar delta at lower frequencies.

 

[MJCronin]

What about the piston pump manufacturer ?

Does he recommend pulsation dampners with his equipment ? Will he size and install them ?

Is this a used pump from another site ?

Evaluate some similar pumps from other manufacturers. Do all of thier Installation and Operation manuals recommmend dampners.

Present this case to the owner and ask him (in writting)why he does not want to use these devices.

I agree with the statements above.....with a 21 bar pressure swing, catastrophic failure will be soon.

 

[SeanB]

You are always going to have pressure spikes with a piston pump - just the nature of the way they work. For that reason the have pulsation dampeners on the discharge. If your client does not want want it, then get it in writing that this is being done against your recommendation.

 

[BigInch]

Yes, but we take care to design the pipeline to keep those within the transient allowables, dampeners or not, or increase the pipeline design pressure such that they are below max transient allowables. That is apparently not the case here.

From "BigInch's Extremely simple theory of everything."

 

[babelfish]

I'd like to say 'don't panic', but that would be bad advice.

For all intents and purposes this is a 'water hammer' situation and unless the pipe is designed for these surge flows then time to failure is closer than you think.

You say it is a piston pump, so I am assuming it is mechanically driven. Can you change the stroke length? A slower longer stroke is preferred. A faster stroke is likely to cause more problems as the liquid has to be accelerated and stopped more often. Meaning higher peak pressures.

If the pump is air driven reduce the pressure and get a bigger pump and/or a duplex or triplex pump.

Primarly you need to prevent a potential mishap. Of secondary imortance is the need for you to get written confirmation from the customer that they are taking on full legal culpability by operating against your recommendations. [You can do that politely, but that is what you are doing.]

"Are we speaking the same language?...engineering - it's the appliance of science."

 

[DYV1973]

Thanks for your replies.

The fluid viscosity is in the range of 12 mPas, with a density of approx 0,94 kg/l. Piping diameter is a DN50 steel pipe (exact material specs are not known). The frequencies of the pulsations are depending on the point of operation and are between 1 Hz and 10 Hz.

We have of course mentioned the need for pulsation dampeners, as we do want to avoid damage to the equipment that we have supplied. As the customer so far ignored our advise to use dampeners, I wanted to make some calculations to show that the pressure peaks are still present, even though it is 50 m further.

To avoid claims we require dampeners, but in the moment I would like to avoid a claim situation, and try to convince them to modify the system.

 

[BigInch]

frequencies of pulsations are 1 to 10 hz. Got news for you. That's full time overpressure. 1 to 10 hz might mean something in electrical pulses, but piping wise, it don't. There is NO accumulator, or whatever that will solve that problem.

What would you be doing, if you knew that you could not fail?

 

[hacksaw]

with those sort of over pressures what sort of kick-back valve are you using?