noise problem 6

[gierszi]

Helo,

we have a problem with 2 booster fans which are the main source of noise. On the top of chimeney the level of noise is about 110 dB. 1km from the chimeney there is a control point and there is about 61 dB. We need to decrease the noise level to 55 dB at the day and to 45 at night. Background noise for this point is 42 dB.

We have four monts to solve this problem. What do you think, what will be the best sollution?

Thank for answears and sorry for my english.

regards,
Marcin Gierszal Poland

 

[GregLocock]

Hi Marcin, can you reduce the speed of your booster fans? Do you have a spectrum of the noise at the control point? Is most of the sound energy tonal or broadband?

Getting down to 45 at night is going to be rather difficult, since the background of 432 means you'll have to be at 42 yourself.

That's almost 20 dB reduction.

Even the 6 dB you need during the day is going to be a reasonably serious effort.




Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[SylvestreW]

Marcin,
Are the motors outside the building (on the roof) or inside? Is there a stack for the exhaust air?

If you could provide some pictures then you'll definitely get more assistance.

-
Syl.

http://www.aercoustics.com

 

[Rob45]

Marcin:
Based on your figures, my guess is your measurement at the top of the chimney is about 3 feet from the "source," and that since there seems to be rather little atmospheric attenuation, the noise is at or below 1000 Hz.
But this information is not of much use to you.
I can only restate what Greg said, that you need to reduce noise at the source by at least 19 dB.

IF you find the noise is highly tonal (e.g., fan blade passing frequency) then you might be able to achieve that degree of reduction with a Helmholtz tuner at the chimney outlet.

 

[gierszi]

Helo,

I reiceived more data about the problem and I going to put into forum. So we have the spectrum of noise source (two booster fans after FDG installation:

http://files.engineering.com/getfil...f725-4581-ad60-fc271df4763f&file=spectrum.jpg

Rob45: the measurements on the top of chimeney were done exactlly as you wrote. Could you say something more about Helmholtz tuner? I think that it is a type of active noise reduction.

SylwesteW: The motors are outside the building but we are sure that the main source of noise is a top of chimeney. Of course the noise is provided by booster fans.

http://files.engineering.com/getfil...56-4810-86fe-832fd01f4c5b&file=boosterfan.jpg

http://files.engineering.com/getfile.aspx?folder=a5250f63-908f-4d21-a5ed-b765859222e8&file=komin.jpg

http://files.engineering.com/getfil...-57b8-4987-b30d-0318058b964b&file=pomiary.jpg

GregLockok: the fans have const rotation speed. Adjusting is provided by changing blade rotation.

 

[davefitz]

As I recall from a few years ago, papers written by Putnam:

a)nightime noise is typically lowered by operating at a lower fan speed, which implies a load reduction at night and a 2 speed or VS fan

b)overall noise from the fan can be reduced by replacing the fan with a larger but lower speed unit, and blade geometry can be adjusted as well

c) outlet silencers can be both reactive / helmholz as well as absorptive type. Helmholz types are tuned for a specific frequency , likely some multiple of fan rpm x number of blades , easy example to view is on most auto's air intake to the air cleaner- that funny tumor on the side of the intake pipe is the helmholtz resonator- commonly used on air conditioning air ducts.Absorptive silencers are multiple parallel walls at the fan discharge containing absorptive material , such as fiberglass.

d) noise generated thru the wall of the fan housing or chimney could be reduced by application of material with signicant change of rho*c ( subs use a rubber type material whose brand name is rho-cee, as I once heard)

e)42DbA is pretty low, likely higher than current nightime background levels- a survey to show that current nightime levels are above 42 dBa may go a long way in allowing an increase in the permit levels.

 

[Rob45]

gierszi:
That is some tonal noise alright! 270 Hz and multiples thereof: I would recommend starting with either Helmholtz or quarter-wave tuners.

The Helmholtz tuner is a passive device that is generally inexpensive to construct and easy to prototype. It is the acoustic equivalent of a simple spring-mass damper, and a search of either literature or the internet should show you how to do this.

One often-neglected feature of the Helmholtz tuner, however, is the need for the product of the tuning chamber volume, in cubic inches (1 in.^3 = 16.4 cm^3), x the tuning frequency (in this case approx. 270 Hz) should equal or exceed 20,000. This is in order for the tuner to be effective.

You might also try 1/4-wave tuners at (or near) the chimney outlet: these are of course tubes closed at one end and 1/4 the wavelength of the frequency to be attenuated. For room temperature air, the 270 HZ tuner would be about 12.5 in. long, and should be near the duct (stack) diameter.

- R

 

[GregLocock]

Rob would you expect a short wide quinkie to work in this case? For instance if his duct is several feet in diameter then a pit 1 ft deep is going to look like a detail, not a tunable part.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Flareman]

gierszi
You don't give any clues about the source of this noise nor details of the specific process between the atmosphere at grade and atmosphere at the top of the stack.
What basic frequencies are generated by the blower?
What is the temperature of the gases in the stack?
Does the blower feed the boiler or dilute the flue gas?
How does this flow interact with the boiler?
How is the boiler fed with air?
Many tonal noise problems associated with combustion equipment and ducts result from the feedback of some other vibration into incoming air stream. This gives an oscillation to the incoming air flow. At the peaks of air flow, the flue temperature falls a little and this reduces draft. That reduces air inflow and the temperature surges again and the system goes into sympathetic resonance.(look up "singing flame")
As I read the spectrum, you are seeing a single peak at about 135 Hz and then integer multiples of 270 Hz.
The forcing vibration could come from the blower blade frequency or from a characteristic dimension of a some chamber or other resonant feature. Even if the 135 Hz represents a 1/4 wave and the stack temperature is (say) 560 degF, you would be looking for a characteristic length of about 11.5 ft (sound velocity/frequency) which seems out of range for the equipment shown in the photographs.
Certainly a Helmholz resonator would seem like a good candidate to tune out the specific tones. You may consider designing something with an adjustable volume like a water piston to give you variability after installation.
Another approach depends on the air inflow properties. If you have forced combustion air at low pressure (suppose 1" wg or less), the small sound pressure variations may be able to generate the initial disturbance. If that is the case, reconfiguring the combustion air for a higher pressure (say 6" or more) may be enough to overcome the oscillation. You need to take some pressure readings through the system and see where to impact (damp out) the feedback loop. If the boiler is natural draft, this approach is impractical.
If you do identify a physical dimension which couples with the frequency, try baffling the flow in that section or make the flow go in a swirl pattern rather than axially, to disturb the standing wave.
Unless you hit a lucky guess, you probably are not going to solve this unless you debug the theory of the noise process with some real numbers.

Good luck.
David

 

[Rob45]

Greg:
You're right, of course: a Helmholtz or Quincke tuner would be much more likely to do the job than a short wide "1/4-wave"

 

[Franko]

Sounds like you have two fans and one stack. Possible solution - could one fan rotate opposite direction to cancel sound waves? Many motors are reversible; one of the fans could possibly have blades reversed as well.

It may be that the frequency is equal to one of the axial mode "organ pipe" frequencies for closed/open conditions:

first mode = k x acoustic velocity / (4 x L)
where L = Length + ( 8 x diameter / 3 Pi)
and where k = 1, 2, 3, 4, 5 etc.

There are other transverse modes that may be resonant.

Active silencers can work below 500 Hz, that is if the pure tone travels as a plane wave, frequencies below the cut-off frequency:

f = .58 x Acoustic Velocity / diameter

search for silencers in:

http://material-handling.globalspec.com

It might be most cost effective to replace the fans.

http://www.multi-wing.com/

http://www.moorefans.com/notes4.htm

 

[gierszi]

Reply to Flareman:

- the basic frequency generated by blower is 273Hz
- the temperature is about 110oC
- the blowers (booster fans) are used to draught flue gases after the desulfurization process.

I've new inforamation. Before building the new chimney they were using to old chimneys. Height of those old chimneys is 250m so they are 100m higher than the new one. The noise level was much lower when the old chimney have been used. Perhaps it could be possible because of longer distance and attenuation.

 

[davefitz]

Its hard to believe the fan noise makes its way thru the new FGD flue gas desulphurization process. May be a site noise survey would show the main source is noise radiated out of the fan and flue casing upstream of the FGD system. If so, this would change the configuration of the most effective noise reduction solution .

 

[Flareman]

So! We know what the forcing element is ! Now you only have to find out where the resonance occurs.
Did you go to the combustion input end and investigate the pressures of the combustion air and fuel?

More thoughts.
Did the old chimneys have the same noise profile as now, but just with lesser dB? At 1 km (your photo), we wouldn't really expect to find any appreciable noise difference based on the height. Were the old chimneys brick built and the new one steel lined? That would affect the ability to absorb noise in the flue gas system. Also 2 replaced by one implies a changed diameter which may be an interactive feature.

David

 

[Rob45]

I don't think there'd be any opportunity for cancellation of the fan noise by having one of the two fans rotate in the opposite direction unless you had some clever way of phasing them.
And probably not then, either.

Also, based on the measurements taken at the stack outlet, I think you can say pretty conclusively that the noise is in fact coming from the stack outlet and is not coming from some other source.

 

[GregLocock]

You might phase them to get cancellation in one direction, but you'll get +6dB at right angles to that direction

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Tmoose]

Does the noise pulsate or "beat?"
If both fans discharge into one duct, and the fans are not running >>precisely<< the same rpm, I'd expect the sound to have a beat frequency.

Increasing the clearance between the blade tips and the "cutwater" in the fan housing can knock a few dB out of blade pass frequency noise. Options are reducing the fan OD, taper cutting the blade tips, simply shifting the fan pillow block bearings away from the cutwater, or fan housing surgery (cutwater-ectomy).

If the duct length happens to be just the right length for 270 Hz, adding a fairly short extension might clobber the resonance. Wavelength of 273 Hz in 120C air is ~ 4.8 feet.

http://www.sengpielaudio.com/calculator-speedsound.htm

"Turn on the quiet"
The former Digisonix made some pretty impressive active noise cancellation stuff.

http://www.appliedsignalprocessing.com/hvacapp.htm

 

[Strong]

Marcin,

I have done noise control engineering on several similar power plant stack sound problems. It is unrealistic to expect that this significant problem can be fixed in 4 months. My guess is the stack diameter is well over 5-ft, so sound waves do not propagate along the breeching ducts and stack as plane waves. If this is the case, then the automotive tail pipe acoustic devices that were suggested will not be appropriate. It was not clear to me that you confirmed that the fundamental frequency is at blade rate (number of blades times shaft speed). I worked on one project where the sound was flow induced resenance and not from the fan blades; so be sure!
The general noise control approach is to:
1) Reduce fan sound generation by evaluating system operating point (should be close to BEP), minimize inlet flow distortion (you have 90-degree elbow at inlet?), and open up blade-structure clearances
2) Identify and eliminate acoustic resonances in breeching ducts and stack, if present
3) Install silencer in breeching duct or stack (typically a custom-designed parallel baffle configuration with both absorptive materials and reactive chambers

A silencer is very effective, but is very expensive, causes a pressure loss, takes time to design-manufacture, and it is very expensive. Take a good look at items 1 and 2 above before specifying and purchasing a silencer.

Walt
w_f_strong@msn.com