PIPING NOISE PROBLEMS 2

[ryuchangmyong]

Dear All

Currently, we are now in a pre-commissioning test for Demi-water piping at Power Plant alongwith 8 bar operation pressure of Demi-water pump (Transfering water from Demi-Tank), this piping include restruction orifice at the middle of recirculation pipe from discharging to Demi-Tank, Globe valve and Cone type filter also installed at Dicharge line, However, especially too much noise and vibration occurs at Discharge. Currently I guess that the problems is due to thin pipe line even though required thickness in accordance with ASME B31.1 satisfied.
1) Suction line : A312-TP304, 10" SCH. 10S (Thk 4.19mm)
2) Discharge line : A312-TP304, 8" SCH. 10S (Thk 3.76mm)
3) Recirculation line : A312-TP304, 6" SCH. 10S (Thk 3.4mm)

At first, we guessed that Restruction size and Location occurs noise, so, we have tried and changed orifice hole size and installed Dual orifice, finally we tried to install with Multiple hole type in order to reduce Dynamic flow force through Orifice, However, nothing could we shooting this problems.
So, I strongly believe that thin pipe thickness occurs this noise because in suction line, cosidering NPSHa of 0.5 Bar, there is no expected Pump cavitation or surging, hammering. (See Attached Piping Photo)
Could you have any similar experience such as this kind of trouble.

 

[bimr]

You picture is missing. Maybe you can provide a sketch?

 

[georgeverghese]

Maybe the flow of demin water to user is low; so recirculation rate is high and recirculation line velocity is too high ? At max recirc flow (which should be total pumping rate when demand is zero), velocity should not be more than 3-4m/sec. Meanwhile, better engg practice is to return the recirc demin water back to tank and not to suction line. Noise can also be generated at the globe valve if it is pinching down a lot at low demand. Why do you have a globe valve on the discharge line?

 

[McDermott1711]

Is it possible that noise is related to some trapped air, somewhere in the line?
As bimr said, a sketch will be helpful.

 

[ryuchangmyong]

Dear ALL

Total flow rate is 170m^3/h(App. 30%) into Recirculation line from Discharge rate, and Globe v/v control this Rec. rate, Let me show you picture as attached. Recirculation orifice should be used in order to keep Discharge pressure, otherwise, Discharge pressure will be decreased.

 

[ryuchangmyong]

As for pipe minimum thickness, especially stainless Steel in case of thin thickness and restriction orifice installed, it should be considered with pipe noise for velocity of orifice flow.

 

[bimr]

You are showing the recirculation piping with a flow of 170 m/h. That provides a velocity of 8.5 ft/sec which is acceptable.

However, the 54 mm orifice plate is extremely small.

Do you have flow meters on the piping? The head loss across the 54 mm orifice plate should be about 60 m to pass the 170m/h flow.

You need to determine the flow in the pipe sections and where the pump is operating on the pump curve.

The noise is most likely generated at the orifice plate, valve, and pump. You should reevaluate the pump recirculation design. It does not look feasible. With a cost for power of $0.10 /kw hour, that recirculation system will cost around $40,000 per year to operate.

What type of plant is this? What is the pressure requirement for the demin water?

 

[ryuchangmyong]

Dear Bimr

I also agree with your orifice required hole diameter, However, What I'm talking is about thin stainless steel pipe may cause piping noise rather than restriction orifice.
We already have changed RO hole size and experience to install Dual type as well, multi type orifice.
However, nevertheless that noise could not be able to remove as to criterial.
So I strong believe that thin S.S pipe is the reason for this noise.

 

[georgeverghese]

When you are in precomm mode, demin water flow to user may be zero (?), and all flow may be going through recirc ( 170 / 0.3 = 570m3/hr), so a lot of the noise may be from the globe valve, and some from high velocity at 8.5 / 0.3 = 29fps = 9m/sec ?

 

[ryuchangmyong]

Dear sir,

As I noticed previous giving information, Recircultaion of flow rate(170m^3/h) is about 30% of total rate.

 

[BigInch]

Large pressure drop may cause flashing through the orifice.

 

[bimr]

I agree that the thin walled pipe may have more noise, but don't think that it is the main problem.

One fix for noise is to insulate the piping. Have you considered insulation?

 

[ryuchangmyong]

Dear Bimr

No piping insulation is doned for Demi-water piping.

 

[georgeverghese]

Okay, in that case pls check that the Q-h curve is not flat around the operating region.

 

[ryuchangmyong]

Dear ALL

This paper has been prepared by PES-ENG(

http://www.pes21.com)

for the guidance in decision of required thickness for piping design based on related CODE rule and industrial practice. Especially, this paper would like to indicate additional consideration in stage of minimum required thickness particularly, when Stainless Steel is to be selected rather than Carbon steel because relatively Stainless Steel thickness more thinner than that of Carbon steel. Those by, this may cause noise and vibration on piping line in case of restriction orifice installed at recirculation line due to pressure drop at back side of orifice, which means that front side cause back pressure where have to be keeping discharge pressure through Pump discharge pressure.
Therefore, the velocity will be increased as much as decreased pressure at rear side of orifice. Generally this velocity would be one of the reason cause of noise and vibration in piping line.
Specially, in case of thin thickness of stainless steel pipe would cause noise and vibration due to dynamic force of fluid velocity in case of which is passing through Elbow. Herewith, the dynamic force may be calculated as F={(P*A)^2+(eQV)^2}^0.5).
Even though, required minimum thickness should consider not only various external force but also additional thickness at the time of thickness selection in accordance with Code rule such as ASME B31.1 or EN13480-3, As well, where existing Fluid dynamic force such as Elbow flow would not be easy to apply to optimum thickness selection.

 

[casflo]

If the restriction orifice is close to the tank and has 60m of pressure drop, probably there is cavitation in the orifice. If the orifice has cavitation also has noise. The noise is transmitted outside the pipe and the pipe wall reduces the noise because it produces tranamission loss. This noise reduction increases with the pipe thickness.
In any case if you have cavitation you must eliminate it to eliminate the noise.
You say that have changed the orifice plate by two in series or by a multi-hole plate. With a total pressure drop of 60m and with no appreciable back pressure in the orifice, that is the orifice outlet pressure similar to the tank pressure, probably you need an orifice with more than 2 stages to eliminate the cavitation and the noise.
casflo

 

[gerhardl]

In addition to look thoroughly into the flow over the restriction orifice, which probably is (one of ?) the main contribution(s?)of noise, it could be sensible to look into other possible additional causes.

Question: from the picture it seems that this is one of several parallell constructions. If yes, is the problem and noise-level and operational data equal for all?

In my opinion the piping construction seems to be mostly self-bearing, lacking support near this (now problematic) construction. Perhaps extra support and extra fundaments could improve the noise level or at least alter the resonance characteristic of the system. Sharp 90 deg. bends, T's and short straight stretches do not improve the noise level.

The globevalve might also contribute to the noise, as well as the distance between the valve and orifice.

Replace the globevalve with suitable regulation valve and remove orifice restriction?

 

[bimr]

Regarding: "No piping insulation is done for Demi-water piping."

The point being made is that you may add insulation to the piping to reduce the noise.

 

[ryuchangmyong]

Dear gerhardl

This pump is parallel installed for 3 sets, noise is nearly same for each single set.

 

[saplanti]

You need to check the pressure at the inlet of the pumps. Noise may be generated by the high cavitation due to the low pressure.