PrestonTom
-
Posts
4394 -
Joined
-
Last visited
Content Type
Forums
Events
Gallery
Posts posted by PrestonTom
-
-
Dr Who,
Thanks for your thoughts.
You are absolutely correct regarding the conundrum: On a continuum where the endpoints are either an anechoic chamber (no echoes or reverberation) or conversely to a reverberant room (where echoes are maximixed, yet the SPL is relatively uniform thought out the space): where do you want to be? That is a good question. There are certainly answers when it comes to concert hall acoustical design. But I am not trying to impose an additional concert hall on to what was originally recorded.
On your other thought on the time-alignment. Yes this is a tangent to the question, but there is an interesting relation to room setup issue. Basically, and simplistically, the bass bin effectively needs both to be in a corner yet at the same time about 6ft "in front" of the top section. This difference in propagation delay, would effectively re-align the drivers having different path lengths. I am ignoring the comparably slight difference in path length between the mid and the tweeter.
How to solve this. Well I can't have a corner "in front" of the top section. However, one could have the bass bin on the floor and position the top section in the corner but substantially elevated toward the ceiling (and angled downward).
My rough calculation, to create an extra 6ft (or about 6ms), would require the top section to be elevated by about 17 ft above the floor.
No, I only have an 8ft ceiling. So I can not do this trick. But consider this for a moment. It does not require any outboard equipment (no added distortion in the signal path), no extra amplifiers. What could be simpler! Does anyone have barn?
Good Luck,
Tom
-
Zealot,
You are posing a rather tough question actually. The choice of microphone is not as important since ultimately you will be making repeated measures throughout the space. If there were no boundaries, then the SPL would drop off at a rate of -6dB per doubling of distance. This in fact is part of the ANSI standard for measuring the low frequency cut off of an anechoic chamber. One can play this trick also to get a sense (spectrally) of the standing waves in a room.
The problem with using Freq domain approach (i.e., Fourier) is that it does not really tell you about echoes, only their results via standing waves and room modes. Inverting the FFT to go back to the time domain is tricky, since the algorithm assumes that the signal started a very long time ago and will contnue for a very long time. So the time information you get is the ongoing phase of the constituent cosines - if they were continuous. With a transient signal, the algorithm, in effect, asumes the transient is simply repeated indefinitely. There will be various
nobs" of the machine to apply various windows or filters; however, these must be used with some underlying knowledge of what is being done inernally. In this regard, many folks using an FFT & applying various windows are unintentionally filtering or smearing the signal.
Basically, with a many approaches (including FFT), you are either going to get spectral precision or temporal precision, but not both. In fact this headache is why waveform analysis using wavelets is so attractive, since you can get around some of this uncertainity. Alas, I am about to go off on a very different tangent .... I'll stop.
You really need the time-domain measures. Even a simple one like looking at the output on a scope (with the right test signal) will give some very solid clues about where the dominant reflections are coming from and also a ore macro vies of decay times etc. Although this can be a tedius process.
Good Luck,
-Tom
-
Westcott,
I appreciate your comments about the floor. But I am stuck on this one. The house is already built. The floors are hardwood. Currently there is an area carpet with a pad underneath. I will have to work with this (also the floor can not be raised since I only have an 8ft ceiling).
I also agree about the open floor plan being forgiving. I am banking on the various openings (to kitchen, hall, front entry) to alleviate some of the "building up" of the low frequencies. This also gives some nice asymmetries.
I am also sympathetic to your comment about the about the uniqueness of horns. I do, and will continue to, have my horns positioned so the listening chair is at a 45 deg with each horn. Your right in pointing out the advantage of direct to reverberant energy in this configuration. It also has the advantage of a flatter freq response since the listener is on-axis with the speaker and minimizing any off-axis frequency response drop off (I think this is sometimes forgotten about by many folks). The other, and more obscure, advantage to the 45 deg placement, is that the listening sweetspot is enlarged (left-to-right). I won't go into here, but for anyone interested there is a nice pictorial description in one of the Dope from Hope issues.
Good Luck,
-Tom
-
Mark,
I know of someone who actually did that. His reasoning, in part, was that it provided a rock solid image.
It is interesting to note that when doing that, the room acoustics problems are actually increased. Having two speakers, packs the room modes (standing waves) more closely together. This gives a more even distribution of SPL as a function of both frequency and location. With a single speaker, you lose some of that advantage.
Good Luck,
-Tom
-
Gil,
Thanks for your thoughts and I will certainly keep everyone posted (I think our solutions will be of interest to everyone).
As a starting point RT-60 decay times (I remember them as the "Sabine decays") are probably a good starting point. Although trying to keep them uniform across frequency will be tricky (trivial at high frequencies and requiring heroic efforts at low frequencies). This will be a blast getting back into this topic again.
A number of years ago I was the technical point of contact when our lab was renovating the anechoic chamber and reverberant rooms in our suite of laboratories. I am certainly familiar with Everest's Book. There was another more technical book by Kruttoff (?) on room acoustics. It was older but provided a solid technical background.
My goal is toward some workable solutions that are flexible, visually pleasing (at least not annoying), appropriate for others to borrow, and will not entirely alienate my girlfriend.
I will begin in earnest sometime in Jan (once my kitchen renovation is finished).
Good Luck,
-Tom
-
Folks,
I am bit embarrassed to confess this since I actually work in audio.
I will be finishing up a kitchen re-model, which of course has taken 3 times as long to complete then as I predicted. Which leaves me time for my next project: Serious (or semi-serious) sound treatments for my listening room.
I live in a raised ranch and use the living room as my listening room (K-Horns with 2-channel music & no eye toward home theater). The room is limited by 8 ft ceilings but measures 12 ft deep by 19 ft wide. The width spills open to a front entry stairwell, and a hallway. The depth spills open to a wide kitchen entrance with a cutout (window) between the rooms. There are enough complications so that a simple geometric model will not work. But the good news is that the space is not terribly small.
First question: When Artto and others first tackled their environments, what was the first step. As a scientist, I am inclined to do some subjective evaluation about what I like and don't like and then follow this up with some physical/acoustic measurements before starting
My dilemma is the measurement issue. Naively, one could use some long-term signals (flat noise or warble-tones) and measure long-term frequency response. This could tell me about standing waves. This frequency domain approach is attractive since it is straight forward and easily implemented ( I can borrow some of the equipment from work). However, the disadvantage of a frequency domain approach is that a highly reverberant room can also generate a sound field with a uniform SPL (room modes are densely spaced). Such a reverberant room can sound terrible (believe me I used to work with these).
What about the time domain? There are ways to measure this also but what about the treatment? One can diffuse, trap, and absorb much of the sound but then what are you left with? An anechoic chamber describes this extreme. Believe me you would not want this for your room. It is a rather dead environment (believe me again, I used to work in one). Again you would decrease the echoes and reverberation (at least at some parts of the spectrum), but the frequency response of the Speaker-Room system would not be flat.
The other aspect is sound deadening (making the room quiet & decreasing room-to-room transmission). This is difficult to achieve since the room already exists (this is NOT new construction) and this is also an "open" floor plan (which allows it to be a "larger" room).
Where does one start? I assume I will re-read my time domain measurement handbooks, borrow a microphone and spectrum analyzer and check out some DIY projects on fabricating sound absorbers, sound diffusers, and bass traps. And by the way, the treatments will require some aesthetic design criteria (you guys have girlfriends and wives so I don't need to explain this...).
Your thoughts please (and this should be an ongoing dialog),
-Tom
-
My own experience is living in a 12 x 22 ft room that opens onto a kitchen, hallway and front entry (making it seem larger)........placed them along the long wall so I am at a 45 deg angle from each. This has been fantastic.
I use these to listen to music (well-recorded music and played at realistic levels). Perhaps I am happy because I am not trying to play sound (or movies) at killer levels or listen to stuff that is recorded with an exaggerated bass.
i have exactly the same room / same experience]
Duke,
It sounds like you live in a raised ranch also. Actually my Winter DIY projects will be to measure/listen/install some room treatments. I am going to make it even better ....
Good Luck,
-Tom
-
Folks,
I truly hope that a potential K-Horn user will not look at this discussion and conclude that anything less than a 25x40 ft room (or whatever) will be inadequate to enjoy a pair of K-Horns. Although overall room size is important, it is not the only issue (or solution) in room acoustics.
My own experience is living in a 12 x 22 ft room that opens onto a kitchen, hallway and front entry (making it seem larger). I added a false corner (half of a corner, actually) and l placed them along the long wall so I am at a 45 deg angle from each. This has been fantastic. It became even better, when I added a center. I enjoy these speakers and I would never try and use anything else and then try to supplement with a sub.
I use these to listen to music (well-recorded music and played at realistic levels). Perhaps I am happy because I am not trying to play sound (or movies) at killer levels or listen to stuff that is recorded with an exaggerated bass.
Corners are important, the other features of the room are also helpful but not as important. If it is an issue, then install a false corner. K-Horns are wonderful speakers (yes, I am biased). Please don't shy away from them because of the comments being made.
Good Luck,
-Tom
-
"You are not the only one confused about polarity! Not only is it different between networks but it's difficult even to say which way is correct"
Al, you posed this question about determining the "correct" polarity when hooking up the crossover/drivers.
You suggested many solutions. I believe you will do the best (as you mentioned) when you you look at the amplitude spectrum where the drivers overlap. Do this, then set the polarity to minimize the fluctuations across frequency. This, of course, will partially compensate for phase shifts introduced by the order of the filter (eg. invert if the filter has a 180 deg shift at the corner). Although, with a very high order filter (and little spectral overlap) then a consistent phase would be fine.
I thought this was the standard thinking. I guess I don't understand your objection.
Good luck,
-Tom
-
Chris,
Be thankful that the problem did not result in a fire. Speakers can always be fixed.....
I definitely agree with the freezing method. This should be done first. If that fails, telephone a dry cleaners, they have lots of tricks.
Good Luck,
-Tom
-
Tid,
I appreciate your dilemma; however, I there is another perspective.
Many of the folks are looking somewhat blindly at the cutoff frequencies for the various cabinets. Let me caution that the speaker response is not simply the output of the speaker, rather it is the output of "speaker-room " when viewed as a system. This is understood by all for the case of the K-Horn which does best in sealed corner. However, it does not stop there. It will also be true for the La Scala, a sub, etc. The output you will get will also be a function of room geometry, room treatment, location of cabinets and listening chair etc. So the difficulties one might have in placing a K-Horn will also occur when placing a La Scala etc.
One really needs to consider the cabinet as part of a system where the other half is room acoustics.
Incidentally, regarding the use of a sub, in my view one of the remarkable things about a K-Horn is that the deep bass is very clean sounding. I have yet to hear that from a sub. Sure, you can get a sub to put out plenty of low frequency energy, but I have yet to hear a sub generate the wonderfully accurate sound that a K-Horn produces. This is my bias.
Good Luck,
-Tom
-
Robert,
Glad that you are enjoying the Proton tuner. You are right about the NAD-Proton link (although I understood it was the other way around).
I have a NAD 4155 which I have been quite happy with. It is a strong tuner & great for pulling in campus stations (supplemented with a small Yagi on a rotator on the roof).
My collegue brought his Proton into work one day and low and behold, except for the nametag it is an absolute look-alike to my NAD.
Since they are not a well-known company, you can pick these up for a very good price on eBay.
Good Luck,
-Tom
-
Folks,
Just a bit of perspective here regarding low fequency sound.
Please remember at 100 Hz the wavelength is 10 ft, at 50 Hz it is 20 ft, & at 25 Hz it is 40 ft. These are big numbers. Add to it that these large wavelengths are relatively unaffected by things like carpet, drapes, & furniture.
So when you talk about the output of the speaker or sub, you really need to consider it as part of a system. That system is the "speaker-room" system. Even with an upper limit of 300-500 Hz, much of what you are listening to is simply room acoustics. There is no getting around this. Room treatments for low frequencies are difficult and expensive. Typically, foam will not do it. You really have to rely on room geometry, placement, and bass traps etc. Even these treatments are not simple nor entirely effective. In many cases they simply "move" the problems somewhere else (location or spectrally).
It really is a sobering thought, since trying to correct the problems will require some heoric efforts and a great deal of compromise.
Good Luck,
-Tom
-
DJK,
There was a caveat on my comment that soldering would not improve the sound. The caveat was: if the contact is clean and not corroded and the contact was secure. Then there would be no improvement.
Under those conditions how will the electrons "know" if they are going through a soldered connection or not? The possibility of screwing something up with a soldering gun is just too risky.
Good Luck,
-Tom
-
Careful, careful, careful
If the contacts are corroded or lose, then by all means clean and secure them. If this is done then soldering will NOT improve the sound.
Good luck,
-Tom
-
Tom,
Congratualtions!
No need to upgrade the equipment right away. Listen and enjoy.
However, I would encourage you to get a CD player. There is so much available on this format.
No need to spend a lot of money on a CD player. For less than $200, I would recommend a Dennon or Harman Kardon, or even a SONY (however not their bottom line models). You can spend more money , but the benefits will become elusive (the technolgy has gotten quite good, so newer is frequently better)
Once you have some well-recorded CDs, I would encourage you to spend some time on tweaking the placement of the speakers and listening chair. This can make a profound difference, more so than swapping amps etc.
Enjoy,
-Tom
-
Shawn,
Yes, these are interesting measures. Bench testing is to be commended. I am going to mention some things that I suspect you already know, but others may be confused about.
Using a square wave (and I am not sure how it is being generated) is a nice visual depiction for a time-domain analysis. The biggest thing that will mess it up (and make it look "less square") is a bandwidth limitation. You have nicely pointed out this limitation and where the harmonics are in a square wave.
However, what others may not be aware of is that this graphical depiction can also be "messed up" by two other problems. The first (and more trivial) is if the device produces a phase shift (it is still a linear device with plenty of bandwidth). The decompostion mentioned earlier needs to understood in that the harmonics comprising the square wave must be in cosine phase in order to generate the "square" edges on the waveform. However we are relatively (usually) insensitive to phase so this is not a terrible problem.
However there is a second problem that will also "mess up" the "squareness" of the waveform. That is the slew rate. This is simply the rise time or slope expressed in volts per time (microseconds). The bigger the number the better. Although I wil let others argue about the perceptibility of more modest slew rates.
The final comment gets into a tricky issue about amplifier measurement (and this was obviously beyond the scope of your measurements - so it is not a criticism of what you are accomplishing). The question is simply: Should the amplifier be measured as part of a system? By this I mean should one substitute an actual speaker to the output of the amp rather than using a power resistor. I am still referrring to measuring the electrical output of the amp (and not the SPL of a speaker in a room). But now the load is rather different. The impedance being driven is not constant across frequency and the demands on the current delievered by the amp can be rather different. Just think of an amp with limited current capability trying to drive low impedance electro-static speakers.
That raises a whole other set of issues.... However, realistically, no one can measure every combination of amp/speaker. I guess what I am trying to say is that this is an essential component of describing an amplifier; however, there are some other important issues (eg., transient current capability, damping factor, slew rate, distortion levels while driving a complex load, etc). Most of these are never adequately addressed & measured anyway.
Good Luck,
-Tom -
Bismarck,
The smartest thing you can do is put an in-line fuse on each speaker.
lstari gave some good advice about the Radio Shack ones. They are inexpensive and simple to install.
Good Luck,
Tom
-
Eric,
Sorry to hear about the bad news. However, just about everything can always be fixed. I certainly agree with DM & Gil about the most likely problems.
At the risk of sounding like a jerk, it is always a good idea to put an in-line fuse on your speaker cable. Thes are fairly cheap (20mm ones from Radio Shack are fine). What you have gone through is one of my worst fears when using older electronics or having someone bump into a volume knob etc.
There are some naysayers who will recite the usual cable voodoo about the sound degradation that supposedly occurs when the electrons are forced to transverse 20 mm length of thin copper. But that is nonsense. The actual surface contact between mechanical connections between components or within a component are comparably small. The electrons do not know and I won't even mention how small the gauge is on a voice coil winding.
Again, I am sorry to hear about the bad news. Statistically, it is most likely a diaphragm in the tweeter that was blown. That can be replaced and is not too costly. However, I certainly recommend to you and to others to install an in-line fuse on your speaker cable. Accidents do happpen.
Good Luck,
-Tom
-
Yes, that does bring back memories of the Ohm F (walsh) drivers.
As I remember, there was a big to-do over the sound field. Supposedly, the sweetspot was just about the entire room. They sounded pretty good when they were loud. At this time the Carver-designed Phase Linear amps were quite popular. That was the combination that I remember
Ahh memories,
-Tom
-
Jim,
Congratulations!.
There is no hurry on this stuff. You already have a decent amp and the real trick in matching an amp to a high efficiency speaker is to find one that is quiet and clean at the low power levels. The speakers do not present a low impedance, so you need not worry about an amp that will deliver high current levels.
At this point the biggest impact on your system will come from spending some time on locating the speakers/listening chair. This is not sexy but it is absolutely necessary.
Next some room treatment is in order. This can range from simply moving/adding/removing drapes, carpets, pads & furniture to more exotic treatments such as foam and bass traps. Before you perform the latter (since it can be expensive) you need to get a clear idea about what it is that you like and dislike about the sound.
Refreshing the crossover is certainly in order also.
The isssues of amps & pre-amps are secondary to the set up & room tweaking.
Good Luck,
-Tom -
You are correct DrWho.
When the signal is not split (as is sometimes the case in audio but not the other applications) you can still cancel the noise by subtracting the noise (picked up on the inverted, other wire). Signal-to-noise ratio is still improved.
But the key elements still remain: noise cancellation via a phase inversion followed by a summation between the two wires.
Good Luck,
-Tom
-
Gil,
That was a thoughtful, and thought provoking, review. BTW a couple of the "pluses and minuses" are inverted - but the context clarifies your intent.
It sparked something in my memory banks about a derived center channel. The summation version for a center is usually used to "anchor" the center of the image when the speakers (L & R) have a large separation. So this takes care of the "hole in the middle" problem. Not as frequently mentioned is that it will also enlarge the listening "sweet spot".
What you sparked in my memory was when I first came across the Bell Lab's etc symposium. One of the frequently mentioned observations was that the third "channel" now gave a sense of depth to the sterophonic image. This has always been intriguing to me. A number of folks get a sense of spaciousnes when there is a bit of de-correlation produced by reverberation, but the reports from the symposium (and these were individuals who could knowlegable and could articulate some of the percepts) was that there was now a sense of depth to the audio-image. The intrigue comes from the fact that this is difficult to explain. The stereophonic image, including the phantom center etc, is well-described / predicted by "stereophony" (as conceptualized and described by Ben Bauer). However, the sense of depth remains elusive in its mechanism.
Bottom line: I know it when I hear it, but I just can't explain it.
Good Luck,
-Tom
-
dboxmeyer,
This is all very simple. The problem is that your wife does not like having 5 speakers in the room that are all in a black finish.
Consider replacing them with just two K-Horns.
The logic is as follows. They have nice wood finish and fit nicely into the corners of the room. There are only two of them (you really don't need the other three), so you have already made a good-faith-effort to consolidate.
Now if you can convince your wife with this logic (practice in front of the mirror first), let us know.
Good luck,
-Tom
Hey Al.
in Technical/Restorations
Posted
Artson,
That is a very clever way to access the different taps.
will be upgrading to an ALK universal this coming year. I know I will want to go back and forth testing the different squawker settings. You have just saved me a bunch of time.
Good Luck,
-Tom