[dp70]

I think it's worth experimenting... but I'll bet placement of the microphone will be quite critical and if this works at all, it will probably cover only a small viewing area.


Also, there could be some weird, undesirable effects if the circuit picks up the movie soundtrack, so the microphone should probably be placed as close to the fan as possible. A high-pass filter on the output of the noise cancellation circuit would probably help too.

 

[rlsmith]

Interesting idea.


I believe that noise-cancelling technology can be applied to things other than earphones. For example, speaker phones use it to cancel out the sound from the speaker reducing or eliminating feedback.


This should work if done properly.

 

[dubmaster]

I inspected the earphone. It is quite difficult to reach the transducer and to pick up the signal from the tiny wires there. The phone costed 20 euros, i can also break it, but it would be wonderful to have sort of external device with mic in and line out.

I took a quick look on the web and found nothing.

With more time and patience i will experiment in the next days.


bye



------------------

dubmaster

 

[tommyboy2]

This was contemplated a while back by Cliff Watson. It is quite feasible, however, the sound must be "contained" more or less to effectively cancel it. The speakers on the projector may not do very much since the sound dispersion is rather wide. I think Cliff had thought about building projector ear muffs to put over the fans. The muffs would have a speaker aimed outward with slots to permit airflow on the sides and the mic inside. The problem was making it so that it could be used for something other than D-ILA given it is a niche market.


Nonetheless, someone could do this and Cliff may be willing to share his experiences from back when he was investigating the ear muffs. Actually, come to think of it I think I wanted him to name it a PCD - Projector Cloaking Device.


Tom

 

[dubmaster]

Hi, i just bought a sony headphone with noise canceling technology. It has small microphones and a dsp that mimics the noise in inverted phase. Testing it near the rather noisy DX3 gave astonishing results, the question is: if i take the signal from the earplugs, amplify it through a small speaker , could it make sense ?

I read that infocus is going to mount a dsp noise reduction on new dlp projectors.


In order to test i have to disassembly the earphone.


Just an idea..


bye




------------------

dubmaster

 

[dp70]

Could you use the projector's own speakers to amplify the output of the noise cancellation circuit? You could even use the projector's volume control to get the level correct!

 

[Man E]

Quote:

Originally posted by Mark Fontana: Could you use the projector's own speakers to amplify the output of the noise cancellation circuit? You could even use the projector's volume control to get the level correct!

Evidence of this forum's never ending pursuit of projector perfection, even from components that would otherwise be deemed useless for our application. Way to go Mark! (even if it doesn't work out)



------------------

The button is labeled "Play", not "Pay". STOP the MPAA!

Our Silent Angels

 

[dubmaster]

Mark, that's a great Idea. I already took my screwdriver and i'm looking at the circuits.


bye+



------------------

dubmaster

 

[Chris Carollo]

I was under the impression that noise-cancelling headphones work because the distance from the speaker to your ear is very tightly controlled. Since they know the distance from speaker to ear, they can cancel fairly high frequencies by timing their inverted-phase signal so that it hits your ear just as the actual-phase signal does.


With speakers, I'd imagine that you might be able to get it to work in a precise sweet spot, but that anywhere other than that you'd only be able to cancel fairly low frequencies (and it'd probably sound like an increase in high-frequency noise).


But hey, I might be wrong. And you could probably rig something up so that everyone gets little non-obtrusive headphones that only cancel projector noise (hooked up to a little mic by the projector?). Though perhaps that wouldn't be worth the work. http://www.avsforum.com/ubb/smile.gif

 

[Art Lloyd]



For the very little it is worth: I think one of the Japanese auto manufacturors has a working noise cancelation system for exhaust sounds in a car they sell only domestically. Also, didn't someone say that TAW was working on just such a system for the DILA model that they are going to carry? That may be only wishful thnking on my part but there is some kind of reference in the dusty pits of my mind. Art

 

[Zaccaria IV]

dubmaster,

non si vedono molti italiani nel forum. Vabbè, almeno non sono solo.

Saluti.

 

[Gary P. Gibbs]

Hi all. I am out of town until Friday so I won't be able to reply all that much. I work in the area of active noise control for NASA. We concentrate on interior noise in aircraft, and I am currently working on active boundary layer induced noise control.


Anyway I guess my generala comment is to encourage the investigation, but I can say that each active system is taylored to a particular problem, etc. It is by no means easy. People have been working in this area for 15 - 20 years, and there is only a few different type systems available. The active headsets have done very well, and there are some active tonal noise reduction systems for propeller noise in aircraft. These two systems are quite different and require a drastically different controller.


The headset problem is by far the easiest because it is sort of a 1 - D problem. The air cavity formed between the headset and the ear is small so that for a huge frequency range the speaker, and error microphone, and ear drum appear co-located. Being co-located leads to great performance and relatively easy controller (electronics) design. How does this relate to the projector problem? For the projector you need to be able to recreate the noise field produced by the projector using the speakers. Depending upon the frequncy content of the noise and size and shape of the projector this can be quite difficult. You typically have to locate the speakers within 1/4 of a wavelength of the noise source at the highest frequency you want to control. Thus at 1000 Hz the wavelength is ~1 ft and thus the speakers should be within three inches of the noise sources. Thus if there are several areas on the projector radiating noise then you would have to have control sources spaced all over the projector. If you can recreate the noise field with the control speakers then there is a hope that you could get global noise reduction (quieter everywhere). If not you will get local control arround the error microphones. Thus you would want to mount the error microphones on your wifes ears like ear rings and have her hold very still http://www.avsforum.com/ubb/smile.gif http://www.avsforum.com/ubb/smile.gif


If you are lucky enough to be able to obtain global control then the error microphone location is not quite as sensitive, etc. The controller design will be totally dependent on the frequency content of the fans, the phase angles of everything, the geometric location of the noise sources, the control sources, and error microphones. It is etremely unlikely that the controller used in the headset that you have as a doner will work in this system.


I am sorry to rain on the parade, but the problems can be quite difficult. NASA, private companies, and universities have spent propably 100's of millions of dollars studying and developing these systems. I strongly encourage the work. We will all learn a lot from the findings, etc.


Good luck.





------------------

Gary


STOP DVI/HDCP!

DVI/HDCP! ~= HD-DIVX!!!

DO NOT SUPPORT JVC or anyone else who supports this!

 

[rick e]

Gary


That was a great summary of the current state of noise cancellation tech.

I'm not a techie but the noise cancellation discussion we had here last year came to similar conclusions.


As a G15 owner I too am very interested in noise control.


What do you think of sound silencers. This is a less elegant but is available now. These are somewhat like the mufflers used in motorcycles but are straight through without obstructions in the airflow. one company that I've found is Lindab at lindabusa.com.


They make very nice ducts ( Terry Gilliam would have loved these, remenber Brazil) and something called spirosilencer. These are sound attenuators for round duct systems and come in various diameters and lengths and even in elbows.


An example-

An internal diameter of 3", 12" in length could create a 19db loss at 1,000Hz.

One 24" long could experience a 33db drop at 1000HZ.


I say 'could' because there are specs in their literature that I don't understand. Air velocity is involved and the sound attenuation seems to be greatest at 2,000HZ throughout their line. What is the frequency spectrum of dila fans? Could one of the HVAC guys please look at this website and decifer the specs for us laymen?


What I think would work would be some type of plenum attached to the side of the dila Gxx. Then use their interesting duct systems or a silencing elbow to route the airflow to a discreet location above and to the side.


I hope to try this design in the next week or two. Part of the problem will be to get these products, there aren't any distributors in the western U.S. The factory said they might be able to sell to me but they might not want to deal with a small order.


Rick

 

[dubmaster]

Gary, many thanks for the reply, it was a very useful reality check and spared me a new pair of nc earphones, i can always wear the phones while watching movies


bye




------------------

dubmaster

 

[Gary P. Gibbs]

Originally posted by rick e:
Gary


What do you think of sound silencers. This is a less elegant but is available now. These are somewhat like the mufflers used in motorcycles but are straight through without obstructions in the airflow. one company that I've found is Lindab at lindabusa.com.


They make very nice ducts ( Terry Gilliam would have loved these, remenber Brazil) and something called spirosilencer. These are sound attenuators for round duct systems and come in various diameters and lengths and even in elbows.


An example-

An internal diameter of 3", 12" in length could create a 19db loss at 1,000Hz.

One 24" long could experience a 33db drop at 1000HZ.


I say 'could' because there are specs in their literature that I don't understand. Air velocity is involved and the sound attenuation seems to be greatest at 2,000HZ throughout their line. What is the frequency spectrum of dila fans? Could one of the HVAC guys please look at this website and decifer the specs for us laymen?


What I think would work would be some type of plenum attached to the side of the dila Gxx. Then use their interesting duct systems or a silencing elbow to route the airflow to a discreet location above and to the side.


I hope to try this design in the next week or two. Part of the problem will be to get these products, there aren't any distributors in the western U.S. The factory said they might be able to sell to me but they might not want to deal with a small order.


Rick


Passive noise control is not my forte, but I do have a reasonable working knowledge in the area. The first and foremost rule for passive noise control is called the "mass law". In layman's terms put mass between you and the noise source. For every doubling of mass in an enclosure you get about 6 dB of noise reduction.


The problem with projectors (heat producing equipment) is that you can not fully enclose the item with 3/4" MDF box. The way to handle these systems is to enclose them within boxes and then provide them with clean and sufficient airflow. You could supply inlet and outlet ducts which enter the box at locations that promote the normal airflow of the projector. These ducts would require a fan on either leg) and could draw and dump air in the room with the projector. In this configuration the ducts would be relatively short and require sound absorptive linings. As with every thing in acoustics it all depends on wavelength. The absorptive material will nominally be effective when it is ~ 1/4 wavelength thick and the ducts will have sufficient time to attenuate the noise if they are several wavelengths long thus at ~1000 Hz the ducts should be 2-3 feet minimum. The specs that you mention above (I tried there website but a link was broken to the spec sheet) 19 dB at 12" at 1 kHz is good, but if your projector puts out a good bit of energy below 1 kHz then the attenuation will be much less then 19 dB. Again the duct size and supplemental fans would be sized to promote normal airflow values.


If ducts can be run into other rooms that would not be used during movie watching, etc. then this would be an option.


I am also a bit offended by the noise of my LT100 and am considering a box.

I believe that there is free software available on the web to allow you

to use your sound card as a two channel analyzer. I will look into this and let everyone know. I think that the lapel microphones from rat shack are reasonably flat with frequency and are low cost. We could all spend a few bucks on mics and get the software and do a little noise testing to determine the noise frequency response of our projectors. This would tell us which frequency ranges dominate the spectra, etc. We could estimate ballpark overall noise levels (accurate if you calibrate your mic).


I am out of town next week at Johnson Space Center, but will be back at the end of the week and try to have some info on test equipment, etc.




------------------

Gary


STOP DVI/HDCP!

DVI/HDCP! ~= HD-DIVX!!!

DO NOT SUPPORT JVC or anyone else who supports this!

 

[dhanson]

IF the sound is mostly created by a high-velocity airflow out of the fan, couldn't you build a muffler with an expansion chamber, much like a rifle silencer or car muffler, and seal it right against the fan outlet? It would have to be completely non-restrictive, but I'd think it would reduce the noise coming out of the unit quite significantly.

 

[Gary P. Gibbs]

Yes this is possible if most of the noise is coming out of the vent ports. Some noise may be due to the fan causing the case to vibrate producing sound.




------------------

Gary


STOP DVI/HDCP!

DVI/HDCP! ~= HD-DIVX!!!

DO NOT SUPPORT JVC or anyone else who supports this!

 

[Iceman]

Dissipative silencers (as the ones rick is talking about) are slightly more complex in their noise reducing properties than Gary describes.


The silencers work by converting sound energy to heat by forcing the sound waves to travel through the fibrous material that lines the duct. The important factor is the ratio of absorbent thickness to the 'open' diameter of the silencer. There is quite a pronounced wavelength effect here and this is the reason for the attenuation increasing with frequency. For higher frequencies this type of silencer is quite effective if optimized and it is also the only type that offers wide band attenuation without dips for certain frequency ranges. The problem is that good low frequency attenuation requires very long silencers.


The typical fan spectrum (band limited pink noise type - air-rushing sound) usually has significant frequency content down to some 200 Hz. For these frequencies you would be happy to get perhaps 2-3 dB attenuation from a one foot long silencer. This will severely limit the usefulness of the dissipative silencer described as the total attenuation will be limited to perhaps 5-6 dB(A), depending on the noise spectral shape. A five foot long lined duct (silencer) will provide good attenuation (>10 dB)even at 200 Hz if the lining thickness to open diameter ratio is 2 or more. This probably makes you want to hide the installation somewhere, though.


The demand on the sound reduction from the hush-box walls is not very high and any resonably well constructed box made from 1/2" fibreboard or plywood should do the trick. Just make sure you do not get excessive vibration transfer from the projector to the box (some sort of elastic coupling - the more flexible, the better).


In addition to Gary's very good summary of active reduction methods, I would like to add that it will probably be quite difficult to get good results for a DIY active system for typical projectors unless used in ducts to and from a hush-box. There are several reports of 15-20 dB wide-band attenuation in ducts. I doubt the DIY system will even approach this, but it will be interesting to see what Gary comes up with.

 

[Gary P. Gibbs]

Iceman: I hope you did not think I was advocating the use of active systems on a DIY project http://www.avsforum.com/ubb/smile.gif


I totally agree with your general discussion and it correlates well to the point I was trying to make. I was trying to limit the amount of information that I was conveying. The end result is that duct attenuation systems are a function of wavelength and to handle low frequency you need a significantly longer duct.


The other items I mentioned (e.g. the spectrum analyzer and microphone) were really geared to hush box and duct type systems as I don't expect anyone to DIY an effective active system without getting there hands dirty in the details. What I hoped to express was that in order to design an effective passive (or active) system you should diagnose the noise problem. For example it would be nice to know what the overall sound level in dBA is and what 1/3 octave bands dominate the dBA level. If the dominant bands are above 1 kHz (for example) then a box with a two foot or so lined duct dumping back in the same room could suffice. If the noise is dominated by blade passage tones (and some airflow noise) down to 200 Hz I might recommend dumping the ducts in another room etc. because the lined duct length might be prohibitively long and expensive.


I think we are on the same wavelength http://www.avsforum.com/ubb/smile.gif I just did not go into enough detail on my passive discussion.


Thanks




------------------

Gary


STOP DVI/HDCP!

DVI/HDCP! ~= HD-DIVX!!!

DO NOT SUPPORT JVC or anyone else who supports this!

 

[Iceman]

Gary,

Good to hear from someone working with NVH and offering people some of his hard-earned insights. I understand your situation entirely: how much information is enough?


Other than that, active systems are generally difficult enough for professionals and even if the beauty of the concept always makes people interested, passive - or classic http://www.avsforum.com/ubb/wink.gif - techniques should still be considered the primary alternative.