Noise at flare stack 1

[C4Reactor]

Whenever there's hydrocarbon vapor release from our flare stack, the atomizing steam valve will be opened to ensure complete combustion therefore reduce black smoke formation. One persistent problem: with the steam discharge, come a defeaning noise (at 110 dBA monitored at grade level).
Happy to hear if anyone have similar experience and solution.
Regards

 

[TD2K]

Not surprising, the steam is usually required at reasonably high pressures (100 psig or so at the tips) and that is going to create noise when it flows to the atmosphere. On one hand, you want the turbulence as that brings in air and improves mixing in the flame, on the downside, it results in noise.

You can talk to the vendors to see if they can change the steam tip design to operate with lower pressure steam but that might increase steam requirements in order to get the same smokeless capacity and will require mechanical changes.

Typically, flaring is rare and no one is usually at the base of your flare stack, I'm not sure why I see noise as being an issue. Typically noise is more of a concern back where personnel can be or at boundary fences with neighbours.

The other option is to go to another flare design, totally enclosed ground flares are quite quiet, surprisingly so, but are quite a bit more expensive compared to a standard vertical flare stack. You also need to be able to supply at least 15 psig or so gas pressure to the ground flare and handle the resulting back pressure in the unit(s) going to the flare.

 

[cmtham]

In the ethylene plant I've worked in, the flare has the same noise description as yours. The noise as TD2K explained is due to the atomizing steam (we used 300 psig steam).

I'm not sure the local regulation at your side but mine specified continuously flaring for more than 15 minutes is reportable (irregardless of black smoke or not) and a 13m radius fenced area for heat radiation footprint. But, since a relief system is designed to relieve all liquid content in 15 minutes, unless the plant is proceeding to a shutdown and deinventory is required, the total "noise exposure period" should be quite short.

My opinion is that with all the regulations and codes in place, the noise for short period of time, is acceptable and safe.

 

[Flareman]

The noise at the flare during attempts to suppress smoke is certainly steam flow and turbulence related. However, there are things which can be done to reduce the noise level (110 dBA at grade sounds ridiculously high to me). If you have a full spectrum analysis of the noise it is sometimes possible to debug the probable major sources but that can't really be done from a simple dBA or dB number.

The most common problem is that operators usually open the steam valve to 100%, regardless of the specific requirement, on the basis that the prefered normal state is no-visible-flame. This oversteams the flame and generates much more noise than necessary. Sometimes this also has an audible pulse as well as being just plain noisy. Keep the flame clean but well visible and turn down the steam to achieve that case if possible.

If you have more than one steam feed to the flare, don't use the internal (center) pipe for smoke suppression. It is intended to assist burn-back suppression, and the steam jets are not intended to go to the sonic state, which they will do if you push up the pressure.

The noisy flares are usually those which are using an external ring of steam jets (like a crown around the top).
If you can't get the steam noise down to something you can tolerate, using the above techniques, you probably need to change the flare tip style completely. There is one on the market which is several dB quieter than most (at this point we're straying into a commercial area but I can make some suggestions)
Let me know if you have any questions.


David

 

[Mofya]

Anyone have any experience with low rumbling flare noise that seems louder at a perimeter (about a mile) outside the source. We have this problem (and the associated complaints!) during periods of high flaring. It appears there are lots of solutions for steam-related flare noises, but very few (if any) for inherent combustion noise. Any views??

 

[Flareman]

Yes!
You probably have a major flame instability due to oversteaming.
When you use steam to render the flame smokeless you are actually trying to put the flame out with the water (although you don't usually realize it). There is a tendency among many operators to set-and-forget a constant steam rate because it "hides" the flame and stops the plant manager's phone from ringing with complaints but it's a dangerous practice.
At very high steam rates, the flame becomes unstable and lifts away from the rim of the tip, even if it isn't extinguished. The base of the flame then has no-where to anchor so it moves up and down as the flow and the wind change. Each downward movement is actually an explosion with a physical noise and "rumble". It is often in the frequency range of 12 - 30 Hz and is largely below the audible range, so it's a bigger problem for peple who live a mile (or so) away because it rattles their windows.
If you have a flare tip which has an external steam ring, the position of the steam nozzles can sometimes contibute to the problem.
First - turn down the steam. I am fairly certain that you will find a smokeless condition which doesn't have the rumble noise associated with it. It may not be the visual picture you are hoping for but it should be satisfactory.

I'd be interested to know whether you have success.


David

 

[121jaw]

We have almost no sound from our flare even at full relieving in case of a sudden plant trip. But steam pressure for atomising is quite low @ around 45 psia, to reduce the smoking we have snuffing steam all along the upper perimeter of the flare which is at around 35 psia
Recently we had a change of pilot gas to more of H2 rich gas due to recovery of LPG from the fue gas. We then experienced a continuous knocking sound from the flare, but not when flaring was high. This was later diagnosed to incompatibility of the old pilot to the H2 rich gas, we have overcome that problem by switching to LPG for pilots.

 

[Mofya]

Reduction of steam gives some success – limited however. The flare is still quite noisy at high (but relatively low compared to historical) rates. David you mentioned that the noisiest flares are the ones that use an external ring of steam jets and that the position of the nozzles makes a contribution to the problem??
Any more insight on that? And if you have any more suggestions on flare tip types, I would be very interested. The problem is getting up my nose!

 

[Flareman]

There are Flare tip designs which utilise steam as a motive fluid to inspirate air into the top of the tip and avoid most of the noise related issues completely. A couple of manufacturers make some design or other with this feature, which can reduce actual noise output by as much as 10 - 15 dB.

Some only take it so far and then revert back to the external ring because of physical limitation of the specific design and others also use mufflers around some of the steam jets to reduce that portion of the noise coming from some of the steam jets.

At this point I have to declare an interest because my employer makes the most effective noise reducer of these, which is used by many of the major oil companies in sensitive locations.
Contact me independendently via Flareman_xs@Netzero.net if you would like a short explanation of the specifics of the noise design or if you wish to pursue a commercial solution, and I can point you in the right direction.


David

 

[Flareman]

Picking up the noise theme from a general viewpoint, the noise from a flare may be thought of as coming from three different sources.

The combustion noise itself.
This is the result of the shockwaves generated when each molecule of cold gas and "N" molecules of oxygen are converted into "M" molecules of hot flue gas. The amount of the noise is related to the energy in the flame (heat release) and the dominant frequency depends on the reaction speed of the gas. In general, this noise tends to be centered around 100 - 250 Hz.

Turbulence noise.
Turbulence and mixing of inerts such as steam causes the flame to behave erratically, and this in turn cause the combustion noise to generated in pulses related to the erratic nature of the mixing. The amount of noise energy in this range depends on the turbulence, more turbulence = more noise. More erratic and more unstable flames generate lower frequencies than better mixed flames. Frequencies in the 20 - 60 Hz range are typical.

Mechanical noise.
All jets (gas and steam) create a noise pattern based on the mechanical expansions at the jet. The amount of noise from any jet is related to its mechanical stream power (basically the velocity pressure at the jet) and the peak frequency from the jet is inversely related to the diameter. 1/4" jets might have a peak around 1kHz.

Part of the problem is that we react differently to different frequencies and tolerate low frequencies better than high. In recognition of this the noise data are often given in dB(A) which tries to make an allowance for the physiological non-linearity.

Listen to the flare.

If the troublesome noise is a hiss, it's the steam injection nozzles themselves.
If it could be a musical note, it's probably combustion noise and you may not be able to do anything to change it.
If you have a rumble, it is turbulence.
If you can't hear it but you can feel it, it's flame instability.

Most of these are addressable in the tip design.


David

 

[Mofya]

We went for a wholesale reduction in steam flow by changing the characterisation block that calculates the flow. This has worked wonders so far. We have shut down and re-started the plant since (and have had high flaring events which previously would have atttracted complaints). No complaints - almost zero cost as well. Thanks

 

[DrNewton]

Most of the time a sudden pressure drop contributes in increasing noise level. Like In this case, steam pressure drops from 100 psig to atmospheric, that 100 psig potentail energy is converted into high velocity and of course noise.
If noise level is really a concern, I would suggest to install an inline silencer downstream of steam valve. That will reduce the noise level up to comfort level.