 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Natural Science Forum / Physics / Acoustics / November 2007Two Identical Utility Transformers - Resultant noise
Utility wants to add a second 10MVA transformer close to an existing one. Their consultant claims only a 3 dB increase. This may be true on average, but with nearly pure tones at 120 Hz, in phase, won't we get interference patterns with 6 dB peaks compared to a single unit. The pattern spacing would be pretty broad since the wavelength is about 9 feet. Am I correct? 10 x log (2 x P)^2 is an increase of 6. Mark Bagdon >Utility wants to add a second 10MVA transformer close to an existing >one. Their consultant claims only a 3 dB increase. This may be true >on average, but with nearly pure tones at 120 Hz, in phase, won't we >get interference patterns with 6 dB peaks compared to a single unit. >The pattern spacing would be pretty broad since the wavelength is >about 9 feet. Am I correct? 10 x log (2 x P)^2 is an increase of 6. Right. It could be the transformer is on another phase. A reversed phase would certainly help. greg > In article <1193965003.368087.280...@d55g2000hsg.googlegroups.com>, Mark B Greg: Greg: I already checked. They will be in phase. Mark <mbag...@novusengineering.com> wrote: > >Utility wants to add a second 10MVA transformer close to an existing > >one. Their consultant claims only a 3 dB increase. This may be true [quoted text clipped - 8 lines] > > greg >>Utility wants to add a second 10MVA transformer close to an existing >>one. Their consultant claims only a 3 dB increase. This may be true [quoted text clipped - 8 lines] > > greg I think reversing phase only moves the pattern... Asbjørn >>Utility wants to add a second 10MVA transformer close to an existing >>one. Their consultant claims only a 3 dB increase. This may be true [quoted text clipped - 8 lines] > > greg Wrong. It would certainly not help. Magnetostrictive hum occurs because the metal in the laminations changes size slightly in proportion to the magnetic field. IIRC, high fields _contract_ the metal. This effect does not care about the direction (polarity) of the field, so you get 120Hz from 60Hz AC. So whenever the voltage is high (positive OR negative), the metal contracts. So If you feed a second transformer with a voltage 180 degres out of phase with the first, you get THE SAME 120Hz hum, in phase with the first. Oh dear. Martin  SignatureM.A.Poyser Tel.: 07967 110890 Manchester, U.K. http://www.livejournal.com/userinfo.bml?user=fleetie > Utility wants to add a second 10MVA transformer close to an existing > one. Their consultant claims only a 3 dB increase. This may be true [quoted text clipped - 4 lines] > > Mark Bagdon Um, isn't this assumption only valid for coherent point sources? I would hardly consider a transformer a point source at 120 Hz, at least for the purposes of this calc and the (assumed) distances and dimensions involved. And I think this would also have to assume that the sound-generating surfaces of the myriad structural elements of one transformer were all vibrating in phase relative to the same elements on the other transformer, no? That seems unlikely to me. I'm skeptical, but curious. Savant > > Utility wants to add a second 10MVA transformer close to an existing > > one. Their consultant claims only a 3 dB increase. This may be true [quoted text clipped - 18 lines] > > Savant I agree. I suppose I should draw some more accurate diagrams to map it. The faces of the transformer are over 9 feet (one wave length) long and the sound seems to radiate from the whole surface as well as the cooling tubes. I suppose that that would destroy the coherance of the sound. We did find a surprising large peak on the 45 degree axis at some distance from the existing unit, which made me think that there might be some constructive interference going on between the two sides of the unit. > > Utility wants to add a second 10MVA transformer close to an existing > > one. Their consultant claims only a 3 dB increase. This may be true [quoted text clipped - 18 lines] > > Savant Savant: Also when I presented this question to Dan Driscoll, who what the acoustical analyst for the NYS Public Service Commission for many years, his immediate reaction was 6 dB peaks and a diffraction pattern. I have a call in to see of he has ever measured it. Mark Mark, I'm curious about this now, which could be good or bad... One IEEE paper I dug up lists a standard deviation for tonal noise at 120 Hz around the circumference of transformers (20+ MVA) of 4.1 - 4.4 dB (fans on - fans off, resp.) and states the following: "Comparisons of the circumferential traverse data of many transformers did not yield a single generally valid directivity pattern, and thus it does not seem feasible to identify a 'beneficial orientation' for transformers." {Source: "Field Study of Sound Radiation by Power Transformers" from IEEE Transactions on Power Apparatus and Systems, Vol. PAS-100, No. 7, July 1981 by Istv?n V?r (BBN), et al.} However, a different IEEE paper, published a year earlier, provides polar sound level plot predictions of tonal radiation based on surface accelerometer measurements at 250 points around the tank. Some of the plots are correlated with measured data. One that is not (unfortunately) shows a total delta of about 15-20 dB at 120 Hz around a 19.5 MVA transformer. {Source: "Prediction of Far Field Sound Radiation from Transformers" from IEEE Transactions on Power Apparatus and Systems, Vol. PAS-99, No. 1, Jan./Feb. 1980 by G.O. Usry (GE), P. Saha (GIT), et al.} My gut would tell me to rely on the field study more than the predictive study. But your measurements may say otherwise, at least for the specific transformer(s) you're studying. What can you tell about the measurements you made? How many were made around the unit? At what distance? What was the high-low delta at 120 Hz? Regards, Savant > Mark, > [quoted text clipped - 34 lines] > > Savant Sorry for delayed response. The sum of the 125 and 250 Hz bands (A- weighted) were 52.0 dB along the 45 deg line and 44.2 dB directly off the face at a distance of 107 feet. Quite a jump. The neighboring landowner pointed out the effect to us. Not the magnitude you mentioned but still significant. Can I get the IEEE documents on line? Noral - Dan does not have the BBN study but I have asked utility to see if they can get it. Mark > Sorry for delayed response. The sum of the 125 and 250 Hz bands (A- > weighted) were 52.0 dB along the 45 deg line and 44.2 dB directly off > the face at a distance of 107 feet. Quite a jump. The neighboring > landowner pointed out the effect to us. Not the magnitude you > mentioned but still significant. To attempt prediction of the pattern of a source of monotone noise (120Hz and some of its harmoics) that is larger than a wavelegth is folly. No design process I know of models such a source reliably because they will not know the relative vibration level of each face without extenive surface vibration measurements or an ultra careful mechanical design of the transormer case. Pertient field studies start with the type of data you have already provided. The next step is to make a simple model of the source, then to shield the surfaces you believe to be the sources causing the sound level in your complaintant position. The usual outcome is a barrier wall, often of "SoundBlox" which does not require encroachment to a live transformer and its structure. Noise control on the transformer case might proceed by mapping the vibration of all its surfaces, perhaps not an enjoyable task if the transformer is live. If louud surfaces are identified, they coud be shielded by an additional enclosure, if feasible. Large transformer oise control should start at the transformer factory where these parameters are more readily processed. AngeloCampanella > > Mark, > [quoted text clipped - 49 lines] > > - Show quoted text - If you're an IEEE member you can. Otherwise, you might be able to find them through a library. Perhaps a local (engineering) university? There's also an ASME paper of interest which presents more measured data: "Power Transformer Noise - Generation, Propagation and Control - A Review" by P. Saha (then with Blachford), it looks like the publication date is 1987, and I think it's from their Winter Annual Meeting (poor copy). Unfortunately, the measured data presented is from another author's study - a paper I haven't had any luck tracking down (so far). But the directivity graph of 120 Hz is even worse than the others I listed above - roughly 20 dB variation all the way around. This led to the author's (Saha's) conclusion: "This demonstrates that the propagation of noise from a transformer is poorly modeled as propagation from an omnidirectional coherent source." Also, I will note that I totally concur with Ang's assessment above. Savant > > > Mark, > [quoted text clipped - 72 lines] > > - Show quoted text - Thanks for all your input. The Planning Board met and decided to require a noise barrier for the new (actually old and relocated) transformer, not the existing. That should pretty much preserve the existing conditions - in the direction of the complainant, anyway. They have proposed a Sound Fighter wall which I believe is absorptive on one side. Mark Bagdon > Thanks for all your input. The Planning Board met and decided to > require a noise barrier for the new (actually old and relocated) > transformer, not the existing. That should pretty much preserve the > existing conditions - in the direction of the complainant, anyway. > They have proposed a Sound Fighter wall which I believe is absorptive > on one side. Correct. It's usuallybeen "SoundBlox, with the absorber openings on the inside surface to reduce reverberance and sound backscattering over the wall. Sheet metal or wood also works, the contra-issues being mainly cost and durability. As long as the area density is more than two pounds per square foot, this will be a good barrier to transformer nosie emissions as long as it is high enough. Make sure this wall is snug to the ground with no visible openings; add small scupper hole drains if necessary for rainwater. Angelo Campanella Mark, BBN did a major study for Edison Electric Institute back in the late 70's to provide guidance on predicting noise from transformers. You typically have two sources, the tonal noise of the transformer itself and the cooling fans. It is the tonal noise that concerns you and you are right that it can add to more than 3 dB but not necessarily a full 6 on average. The study is in two or three volumes and one volume concentrates on practical noise control and field predictions. It provides some guidance on combining both the tonal and overall noise of multiple transformers. I would suspect that Dan Driscoll would have a copy of it if he is convenient to you. > Utility wants to add a second 10MVA transformer close to an existing > one. Their consultant claims only a 3 dB increase. This may be true [quoted text clipped - 4 lines] > > Mark Bagdon |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2009 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.