 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
In Reply to: Re: From Wayne Parham - Re: Patentable claim posted by TomServo on July 04, 2002 at 16:48:59:
Hi Tom!You wrote:
>> Not having heard one, that would seem to be a remarkable call to make.
So, are you moving the discussion into the realm of the subjective, where we discuss what we think we hear with our ears? Or are we talking about the math involved to describe phase relationships and performance in the time domain. Because the former isn't something you could have used to impress the patent examiner.
>> A license is needed however as if a company knowingly allowed an
>> infringement and did nothing, that "looks bad" legally.You're right. It dilutes your claim to intellectual rights.
>> So far as "work", so far as I know everyone was happy with the
>> sound, don't forget they have several different strengths.I expect performance is acceptable. As I mentioned in other posts, it's one the few other implementations of large format conical horns besides my ten π's. Mine is a single-fold equiangular spiral and yours is straight, but they are both conical horns used for wide-bandwidth applications. But the claims of time alignment and of wide bandwidth loading are what I object to.
>> The name "unity horn" is the horn design which has multiple ranges
>> in one horn, Lambda sold kits to build one version of it which was
>> a 16 1/2 inch square mouth horn with a 300 Hz low cutoff.So you licensed the "Unity" trademark to Lambda and not the design? Having dissimilar branding may dilute your trademark. It would be like allowing someone else to call their adhesive bandage "Band-Aid." If it's the same, then you can license them the rights to manufacture without dilution. If not, you may have already diluted your rights to the name "Unity."
>> We sell several versions of the Unity including a 3 way horn.
But only the three-way implementation is covered by the patent, yes?
>> Two models can be mounted on a physical or acoustic boundary with
>> essentially no interference as well.I see. Smaller units to be used as an HF subsystem, I assume.
>> one can move the hf source even more foreword and find a spot
>> where the two outputs are temporarily in phase but the "Time" is
>> skewed and an impulse would still show the hf leading.That's the key here. "Temporarily" That's the point of all this. Moving the offset forward or back, either one is still a compromise. Your defense of the time alignment technique you've used only makes my case even more clear. There is no way to truly time align two different point sources at every frequency and at different positions with techniques as simple as position offset or the use of single-node filters. You can make them aligned at one frequency, and at one position and that is all.
I think Davies says it well, "It is one of the fundamental laws of linear systems that if their output depends only on previous input - that is, if they cannot see into the future - then the phase response is completely determined by the amplitude response."
>> Have to run, my kids and I are off to catch some pretty lights and
>> Transient response.Super! Enjoy the holiday!
Wayne Parham
Follow Ups:
> > Not having heard one, that would seem to be a remarkable call to make.So, are you moving the discussion into the realm of the subjective, where we discuss what we think we hear
with our ears?
Or are we talking about the math involved to describe phase relationships and performance in the time domain.
Because the
former isn't something you could have used to impress the patent examiner.No Wayne I was responding to "your" value judgment, passing judgment on a speaker and loudly proclaiming it on a
several public forums without ever even hearing it seemed like a bit much.
I see you have a forum for promoting your speakers that's nice, after looking through some of your posts I found another one which is puzzling
You assert that you are a little miffed at us that we (meaning me) would "copy" your approach and that the idea for
the Unity began as the result of a joke when you spent 20 years in the oven with your pi..
You must imagine the world revolves around you I guess and that no one has ever given any of this though before you.There is a loudspeaker world outside of yours where even they contemplate these issues, do a search and see how
companies like Meyer, EAW, Reinkus Heinz, Nexo and others deal with this stuff.
Bottom line and publicly, I am saying there is no way anything about the Unity evolved with any awareness of your approach and many of my ideas, come in a flash followed by a tail of details much like the impulse response of a line source (followed by a great deal of work to make it work). BFDLooking through your crossover paper I see you are in fact familiar with Spice etc.
You go through a long process showing all the circuit modeling but I did not see any indication that you actually
measured the acoustic results or are familiar with the time delays within the drivers.
Further, you go on to assert that people are "not sensitive to time" and so your arrangement of placing the hf AHEAD
in time is cool.You also say
Crossover phase also represents a difference in time between the two diaphragms. Crossover phase shows when the signa
l is generated by each motor, and the graph shown describes all reactive effects – including speaker driver characteristics.The graph plotted literally shows when each diaphragm makes a sound.
This is incorrect, one has to account for the intrinsic delay in each driver and other fixed delays only then do you see
the phase response of the driver.
I wonder in fact if you even have a way to measure time or real acoustic phase.
> > one can move the hf source even more foreword and find a spot
> > where the two outputs are temporarily in phase but the "Time" is
> > skewed and an impulse would still show the hf leading.That's the key here. "Temporarily" That's the point of all this. Moving the offset forward or back, either one is
still a
compromise. Your defense of the time alignment technique you've used only makes my case even more clear.
There is no
way to truly time align two different point sources at every frequency and at different positions with techniques
as simple as
position offset or the use of single-node filters. You can make them aligned at one frequency, and at one position
and that is
all.Again, it would seem that you do not have a good time measuring device at your disposal and given the choice of
making time and phase correct at crossover VS making the phase right but time off N 1/2 cycles (by moving the hf in
front of the lf) seems not like the right way to me but the wrong way.Having the phase right does make for a good amplitude response but making the time wrong insures poor impulse response. Granted in your design you have no way (other than an electronic delay like a BSS366) to arrive at the correct "time", the Unity is simply a way to do it with out any extra electronics.
You could aproximate the correct situation on one of your speakers or for that matter anyone with a multiway horn system (assuming you have the hf driver able to move front to back) and all you need is an oscillator.
Set the oscillator at your crossover point, where the sound is "equal" between the upper/lower drivers, adjust the
sound level to comfortable and then reverse the phase of the hf (just temporarily). This should put a deep null "on
axis", this null is much narrower and easier to hear than the broad addition hump in the "right phase".To compensate for the hf signal coming out of the crossover first and the LF driver having its own significant delay
and phase lag, move the hf unit to the rear of the lf driver by about 1/4 wavelength at crossover (at the half and half
subjective amplitude point the crossover phase shift is usually in the neighborhood of 90 degrees).
Move the driver back and fourth some to get the deepest notch on axis ( at the listening position).
If there is no notch, then reverse the phase again as this would mean that the previous forward offset was N
wavelengths instead of 1/2 wl's.
Once the distance physical distance is about 1/4 wl or a little more (depending on the LF driver's delay it may be
significantly more than 1/4 wl), then switch the hf driver to "the right phase" (additive) and both phase AND time
should be approximately correct.
Even without a TEF machine and proper crossover slopes, this can give decent results (but should be done outside for best) compared to out of time.
Since it would be so easy to try, I would suggest to anyone with multiple horn systems to try this.
I think Davies says it well, "It is one of the fundamental laws of linear systems that if their output depends only on
previous input - that is, if they cannot see into the future - then the phase response is completely determined by
the
amplitude response."That being true, it is still a mystery why you think that spreading out a complex signal in time such as your hf forward
alignment does would be superior to an approach that doesn't.
By the way, your assertion above regarding minimum phase systems (where the phase response is completely determined by the amplitude response) is only true for a single, linear, point source driver operating well below breakup.
As soon as two or more sources are involved (in differnet locations or delays) or are non linear or have changing directivity the system is not minimum phase.
A speaker system like you describe would measure a large change in acoustic phase in the transition between the upper and
lower sections even though the amplitude could be perfectly flat.
Unfortunately at least some of the MLS measurement systems have a potential flaw in that they calculate acoustic phase based on amplitude using a hilbert transform (the assumption that you refer to, that the system is linear and minimum phase) .
While the average user never notices, that the acoustic phase measurement is often very different than one taken
where the phase is actually measured (as in a TDS system).
When dealing with acoustic phase (like where one has two sources adding together) only the real measurement of
acoustic phase is helpful.On still a different post you go on to tell someone else that only the mid drivers are horn loaded, all I can say is your
a lot of magnitude but no vector on this (that is U.B Wrong)
The unity speaker has a compression driver at the apex (on all the products so far), the response of the horn/
compression driver is only changed by the presence of the small mid entry passages below about 1500 Hz which is in
the neighborhood of the crossover. Above that the power response changes little "with or without" the holes and so
for the compression driver raw, the sensitivity is about 116 dB 1w 1m at 2 KHz. As the horn is very much a constant
directivty device, its high end response looks just like the Plane Wave Tube response for the compression driver
(which reflects the drivers actual acoustic power).
As the power on all compression drivers rolls off way before the hf limit (a typical 1" rolls off at about 2-2.5 KHZ
and a tad 2001 at about 4 KHz), here the real maximum sensitivity is set by the desired HF cutoff.
In this case, the compression driver into a the 60X60 degree horn is at about 104-105 dB 1W 1M at 18 kHz so that
is neighborhood of the maximum sensitivity which can be realized with flat response and passive circuitry.
That sensitivity is set by the coverage angle of the horn and compression driver, like any other CD horn.
Tom
Tom -You go throug a long diatribe that still misses the fundamental point. Phase is moving as frequency changes; The position of your drivers is fixed. Therefore, no "fix" of phase is possible.
The delay within the driver itself wasn't the point of the crossover document. Nor was the delay caused by a fixed offset, for that matter. The phase angles are described, showing the angles where planes of diffraction anomalies are present.
And - Yes - I am discouraged that we both use a conical horn as a wideband device. One of us asserts that it is providing acoustical loading over part of a driver's bandwidth, and the other asserts it is providing acoustical loading over the entire loudspeaker system's bandwidth. One of us says that speakers can't be made phase accurate because they aren't point sources and the other claims that it is relatively easy to do with position offsets and low order crossovers.
Tell you what. How about you explain two things:
1. How does the Unity act as an acoustic transformer, i.e. provide horn loading over the bandwidth generated by three audio subsystems?
2. How does the Unity manage to correct the movement in time over the span of frequencies within the overlap band of drivers in the crossover region? You have said, "generally a first or second order crossover gives the best results" in this design. The difference in time generated by these two networks is a factor of two. So which is it?Wayne Parham
Hi Wayne1. How does the Unity act as an acoustic transformer, i.e. provide horn loading over the bandwidth generated by
three audio subsystems?2. How does the Unity manage to correct the movement in time over the span of frequencies within the overlap band
of drivers in the crossover region? You have said, "generally a first or second order crossover gives the best results"
in this design. The difference in time generated by these two networks is a factor of two. So which is it?
Think about an exponential horn, it is a high pass filter who's low frequency cutoff is a function of the expansion rate
(and if acoustically small, its length).
A 30 Hz horn would have an area which would double every 24 inches or so a 300 Hz would expand 10 times faster
doubling about every 2.4 inches.When one drives a big conical horn from its apex, the driver couples into a part of the horn where the expansion is
very rapid and so, such a horn has poor lf response compared to an exponential of similar size.
If one slices up a conical horn and looks at each slice, what one finds is that the expansion rate is greatly variable and
that for example the same 16 1/2 inch mouth conical horn that would only allow flat response down to ~1Khz driven
from the apex, now with appropriate drivers, connected where the expansion rate is slower, can go down to 300 Hz.This probably sounds weird but get your self some paper or keyboard and plot the distance it takes to double in area
as a function of distance from the apex. What you see is that it starts as a high frequency horn and as it gets larger it
becomes a mid or even low frequency horn, all dependent on the rate of expansion.
Think about making the T&S parameters of the mid driver "right" for that throat area where the expansion is correct
for that frequency (making them larger in area and at lower internal SPL than a compression driver)At the low cutoff, one finds that there is still a main 1/4 wave resonance which is included in the design and as a result
except for defining the low cutoff, it is not evident in measurements. Driving the horn from a point forward of the
actual apex is also not an issue as the free volume rearward of the mids is in the model as well.
First, lets define time and how I measure it.
Set up a sine sweep that is some fixed rate, say from the below the low cutoff of the speaker under test to the above
the high cutoff of the speaker under test.
You place your microphone some distance away, say 1 meter.
The output of the microphone and the test signal go into a modulator (multiplier) which outputs a sum and difference
frequency.
Since it takes some time for the sound to get to the microphone and the sweep rate is constant, there is a difference
frequency produced, when the output of the sweep is viewed as an FFT, one see's an Energy VS time display which
shows not only the reflections in the room but also the shape (distribution in time) of the direct energy from the
speaker.Once the time is entered, the acoustic magnitude and phase can be measured by in effect making a tracking filter
analyzer which is the right number of HZ behind and this rejects reflections because by the time they get back to the
microphone, they are too far behind to be passed by the filter.
*I should say the TDS process is more complicated but it is sort of like a frequency agile radio which is tuned into the
direct arrival frequency.Since in a horn the acoustic paths are intermingled, the raw magnitude/phase measurement and time of each section is
done with all the drivers in place.
As you have identified, it is a pain to make a crossover with a real speaker as a load.
In my case, I have to have a crossover which also EQ's the power response of the compression driver and horn
and makes the "phase" happy in the region of interaction.I use a proprietary "spice like" filter design program, TEF and an HP network analyzer.
I take the upper and lower drivers response and phase and its impedance and phase measurements and use one as a
"series response" filter file and the other as a specified impedance load file.
I arrange a circuit topology which seems suitable and leave most of the components as open variables.
I assign the filter output to be "in series" as well and then have it curve fit to a flat response or what ever curve I asked for.
If the circuit topology permits it, this will find values that give the best fit shape of the magnitude and phase.
Just as in your spice simulator, one can define any topology high pass, low pass, all pass and combinations of all of them and then have it solve for what ever you asked for (magnitude and phase)Often there is more than one solution and then one has to decide which way to go, sometimes you have to manually change something to get it out of a local minimum in the math somewhere.
Sometimes it takes 5 minutes of solid
crunching to find the best fit, often it takes a day or two of fiddling to get something good.
In the beginning, this process took a week or more.It does make a lot of things easier, like the equivolent circuit models for drivers. One need only feed in the impedance curve data
setup the elements arangement with ball park values and then curve fit the impedance of the measured speaker letting the computer find exact component values.
The last compression driver model had 19 elements FWIW.One last thing, remember that the response curves we are matching up have a slope that is both acoustic AND
electrical and the part that matters is the region where they interact with each other which is to maybe -10 or 12 dB
or so on each side of center (-10 would be 90% from one driver and 10% from the other in power~ about 1 dB
error). By having the phase slope in the neighborhood or even right on, one can make the upper/lower driver
distances correct in the transition range.
All this combining takes place at an acoustic dimension which insures that the sum is still a point source in the horn,
there is no "array" related directivity at crossover etc.
When the distances are correct, one see's only one ETC arival and the acoustic phase has no clue as to there being a crossover or more than one driver.
I have a cube (which has a smaller 16.5 inch 2 way Unity horn) apart in the other room, when it is together, I'll send you a TEF measurement of the magnitude and acoustic phase of the horn part if you wish.Tom
Tom,
Even though I like reading your posts, don't waste your time trying to educate wayne. He already knows everything. You know it works, I know it works, anyone who's heard it knows it works, and most other people who might be interested in buying Unity horns know wayne is still living in the speaker building world of 1979.Hope to see in at the end of July,
John
John -Exactly when - "after 1979" - do you think the laws of physics changed to make several radiating planar diaphragms act as a single, unified "point source?"
And when - "after 1979" - do you believe that the bandwidth capability of horns grew so wide that a single horn suddenly became able to provide impedance matching from the woofer frequency range through the tweeter range?
Why stop there? By this logic, the horn must be capable of loading though the ultrasonics; Perhaps even into the RF band and beyond! Who needs optical technologies? - the Unity device can load all the way to angstrom scale!!!
I may not know everything, but I know an exaggerated loudspeaker claim when I see one. After all, speakers aren't the most complicated machines a person could build.
Wayne
Hello again Tom!Listen - I've read your work. Much of it is interesting.
But I maintain the position that the Unity horn is not acting as an acoustic transformer for each of its drivers. I use similarly-shaped, large format conical horns as wideband devices too. But I realize that the horn is only loading the driver over a portion of its range, and certainly not loading the entire audio bandwidth.
I also maintain the position that you cannot make this configuration time-linear or "unity summed." I am sure that you have put a lot of work into time-aligning your drivers, but I insist that the time alignment method you chose is an approximation, just like everyone else who uses this sort of technique. Said another way, being an approximation, the device is still not acting as a point source.
One should still expect to use arrays and line source methods for proper coverage.
Wayne
Wayne responds:
-----------------
Hello again Tom!
Listen - I've read your work. Much of it is interesting.But I maintain the position that the Unity horn is not acting as an acoustic transformer for each of its drivers. I use similarly-shaped, large format conical horns as wideband devices too. But I realize that the horn is only loading the driver over a portion of its range, and certainly not loading the entire audio bandwidth.
-----------------Wayne, I have to ask if you actually read through and gave any thought to Tom's response in the post you are responding to here?(http://www.audioasylum.com/forums/HUG/messages/29609.html)
I wonder if you missed this portion of Tom's response, or chose to ignore it:
"When one drives a big conical horn from its apex, the driver couples into a part of the horn where the expansion is very rapid and so, such a horn has poor lf response compared to an exponential of similar size. If one slices up a conical horn and looks at each slice, what one finds is that the expansion rate is greatly variable and that for example the same 16 1/2 inch mouth conical horn that would only allow flat response down to ~1Khz driven from the apex, now with appropriate drivers, connected where the expansion rate is slower, can go down to 300 Hz." -TD
What part of this explanation goes against the theory, or doesn't make sense to you?
The key to the above explanation is in the realization that the expansion rate and mouth dimension which defines the loading. I am not sure if you are thinking in terms of only the angle and mouth area, but you need to examine each segment as part of a multi-way speaker. This could be further explained by breaking each segment of the Unity Summation Aperture into separate, discrete horns. In such a case, each segment would still operate over the limited bandwidth each segment is responsible for covering.
Now rather than just blindly stating that "it doesn't work," please explain what specifically would not allow proper loading.
Mark Seaton
Hello Mark!You wrote:
> > Wayne, I have to ask if you actually read through and gave any
> > thought to Tom's response in the post you are responding to here?( http://www.audioasylum.com/forums/HUG/messages/29609.html )
Yes I have. Tom has some interesting ideas. But I am not convinced of their implementation.
> > I wonder if you missed this portion of Tom's response, or chose to
> > ignore it:
> >
> > "When one drives a big conical horn from its apex, the driver
> > couples into a part of the horn where the expansion is very rapid
> > and so, such a horn has poor lf response compared to an
> > exponential of similar size. If one slices up a conical horn and
> > looks at each slice, what one finds is that the expansion rate is
> > greatly variable and that for example the same 16 1/2 inch mouth
> > conical horn that would only allow flat response down to ~1Khz
> > driven from the apex, now with appropriate drivers, connected
> > where the expansion rate is slower, can go down to 300 Hz."Yes, I've read it. It's an interesting idea, but my position is that the premise is wrong. Essentially what Tom is saying is that a conical horn with a 16.5" mouth provides response down to 1Khz if driven at its apex but it can provide response down to 300Hz if driven by an appropriate driver mounted somewhere nearer to the mouth.
He goes on to explain his ideas with the changing expansion rate analogy, but this can also be described as a throat discontinuity. Such a discontinuity is something that makes the horn less efficient, and it is my position that you can't go very far with this before you can't even consider the diaphragm to be horn loaded at all. The further the diaphragm is placed towards the mouth, the less the system is acting like the diaphragm is horn loaded at all.
> > What part of this explanation goes against the theory, or doesn't
> > make sense to you?To satisfy the conditions of the Webster equation, the horn must have constant pressure applied to its cross section. This means the diaphragm must be mounted in the apex, or the device isn't acting as a Salmon family horn anymore. Essentially, the conical horn must be driven at its apex or it isn't a conical horn. This is true also for catenoid, hyperbolic and exponential horns - all memebers of the Salmon family.
Now then, I'm not saying that a device that isn't built this way has no merit. It just doesn't fit the model. But this does mean that it has a throat discontinuity and that efficiency is reduced. I call them mal-formed horns, and they are appropriate for certain conditions - like perhaps yours. But I think it is important to call it for what it is, and this make plain the fact that the further out from the throat the drivers are placed, the less "horn loaded" they really are.
> > The key to the above explanation is in the realization that the
> > expansion rate and mouth dimension which defines the loading.
> > I am not sure if you are thinking in terms of only the angle and
> > mouth area, but you need to examine each segment as part of a
> > multi-way speaker. This could be further explained by breaking
> > each segment of the Unity Summation Aperture into separate,
> > discrete horns. In such a case, each segment would still operate
> > over the limited bandwidth each segment is responsible for covering.Actually, the diaphragms nearest the mouth are not really horn loaded at all. Those that are near the apex, but not at the apex, are less than optimal. And the size of the horn defines the range of frequencies where horn loading is accomplished the most - With a conical horn like this, you really can't expect more than about 10dB in a very narrow range of frequencies at its cutoff, and falling response after that. So the horn may be too large to be optimal for the compression device, even though it is at the apex. Above the horn's bandwidth, the device acts as a reflector or waveguide only.
> > Now rather than just blindly stating that "it doesn't work,"
> > please explain what specifically would not allow proper loading.I think I have done this above, yes? To provide horn loading, the device should properly satisfy the conditions of the wave equation.
Take care!
Wayne
You saidBut I maintain the position that the Unity horn is not acting as an acoustic transformer for each of its drivers.
I use similarly-shaped, large format conical horns as wideband devices too. But I realize that the horn is only loading
the driver over a portion of its range, and certainly not loading the entire audio bandwidth.If not acting as an acoustic transformer, what then explains the impedance curves which clearly show "radiation
resistance" and how then do the mid drivers reach an electroacoustic efficiency greater than the compression driver?
You are correct when you say a single driver will not load the horn over a wide frequency range and I have explained
several times about the horns low cutoff being a high pass filter who's frequency is a function of position in the horn.I am not familiar with your stuff enough to know what horn you use or how big it is, but if you have one to spare, try
adding mid drivers forward of the compression driver (a place to start is where the horn area is about 10 sq. ins).The point is that if one had a very large horn, that the lf cutoff is still dictated here by the expansion rate and that if
one wanted to use the horn at a lower frequency, one would have to place the drivers where the expansion rate was
appropriate for the frequency in question.
To date I have not made a horn which loads "the entire audio bandwidth" ( the 10 octave span from
20 Hz to 20 kHz ), the largest span is from about 200 Hz to 20 kHz but I am casually working on one which I hope
will cover 100 Hz to 20 kHz. Since the math says it should work, I don't see why it wouldn't but the issue is my
efforts are governed by what makes the most logical product.
you saidI also maintain the position that you cannot make this configuration time-linear or "unity summed." I am sure that you
have put a lot of work into time-aligning your drivers, but I insist that the time alignment method you chose is an
approximation, just like everyone else who uses this sort of technique.
Said another way, being an approximation, the device is still not acting as a point source.
You insist it won't work yet Energy vs time measurements show it does, you insist it can't be a point source yet polar
measurements show it is.
You insist it is an "approximation" while it was designed, built and fine tuned entirely based on actual measurements of
time and phase of the parts used.You assert that by using even a fastwoofer like the 2226 which only has an internal delay equal to .64 feet (I just measured one)
and by using a crossover where signals above Xo come out significantly AHEAD in time of those coming out of the low pass, that
by moving the hf source still further ahead in time, that "this" fixes the time problem. Maybe fixes isn't what you mean since you do say the ear is not sensitive to time.
You say that by measuring this time discrepancy and accounting for it until the measurements show ONE time, that this is wrong somehow.Picture the Indy 500, your suggesting that by the woofer being a full lap behind, that it is in the same phase and time
as the lead car at the finish. The woofer may be "phase" with the lead car but that is not the same as time, the woofer is still behind 1 lap.
A second issue is what does that misalignment sound like, it may be favorable, for example there is a Pro Sound
subwoofer called a bassmaxx that has the higher frequency direct radiation arriving ahead of the horn radiation by
about 1/2 wl at mid band (corresponding to an HF lead distance of about 10 feet). Some people really like what that does to the sound of music, it adds an "attack" not present when the sound is reproduced strictly according to the input signal.
Lets not confuse that with accuracy however.
I am not sure if you really don't understand or are posturing in defense of "your approach" that you refer to but if you really don't get it, explain at what point the concept is inconsistent with the physics and I will try to explain it further.Also in another post you are vigorously slamming people about the notch you saw at 4 KHz on Nicks web site.
Each new design takes some time to evolve, at best I can only develop about 3 or 4 new speaker products a year
and even then I have been known to go back and change something.
For example our td-1, one of the products that
has been for sale the longest is having a new crossover shortly.When the new horn shape that Nick used was made, it also took a bit to get the kinks figured out.
Where the mid holes are, there is a small impedance discontinuity (due to the change in cross section), making the
holes as small as possible makes this as small as possible.
At about 4 KHz, that location in the horn is governing the directivity in the 4 KHz area and that causes a reflection.
This reflection causes the dispersion pattern to widen temporarily at 4 kHz which because the acoustic power spread
over a wider angle produces a "hole" (the notch you saw) exactly on axis.
This notch is not present off axis fwiw.
The reflection can be eliminated (or at least its effects once the sound exits the mouth) by placing 4 small foam
absorbers at the right location on the mouth and then the dispersion and amplitude response at 4 kHz are "fixed"
without other impact.
Even when not "fixed" the effect is minor and a narrower slice in frequency than one can discern.Again I would offer to send you some measurements if it really is an issue of understanding and I would urge you to
go to the speaker fest and hear them and think about it before you say much more publicly .While I do have some computer programs which are not publicly available, there is at least one program capable of
this kind of design. It is called AKABAK and can model just about anything one can think of.
You might want to look into it .
Tom
Hello again Tom!You wrote:
> > If not acting as an acoustic transformer, what then explains the
> > impedance curves which clearly show "radiation resistance" and
> > how then do the mid drivers reach an electroacoustic efficiency
> > greater than the compression driver?Filters modify impedance and act as reactance without matching impedance. I expect your horn will load the midrance band for a couple of octaves and that's all. That - and the fact that you use several midrange drivers - is why midrange output is higher. But the horn unloads a couple octaves above its flare frequency, and after that, the horn's flare is acting only as a reflector.
> > To date I have not made a horn which loads "the entire audio
> > bandwidth" ( the 10 octave span from 20 Hz to 20 kHz ), the
> > largest span is from about 200 Hz to 20 kHz but I am casually
> > working on one which I hope will cover 100 Hz to 20 kHz. Since
> > the math says it should work, I don't see why it wouldn't but the
> > issue is my efforts are governed by what makes the most logical
> > product.Your use of the phrase "cover 100Hz to 20Khz" is ambiguous. If you're talking about making a horn that provides acoustic impedance matching over a span of seven octaves, you're wasting your time. You can load the diaphragm(s) for two or three octaves, and not more.
> > You insist it won't work yet Energy vs time measurements show it
> > does, you insist it can't be a point source yet polar measurements
> > show it is.Did you ever consider that your measurements might be in error? Not to be rude, but I've even considered that they were "hand picked" and that you've only shown those that support your case. You have distanced yourself from the Lambda graphs, but I find them to be extremely telling.
> > You insist it is an "approximation" while it was designed, built
> > and fine tuned entirely based on actual measurements of time and
> > phase of the parts used.If you've done good work, then it is a close approximation. But to say it's perfectly accurate in both the time domain and the frequency domain simultaneously - as a true point source would be - this is not possible from the Unity device. It is not a point source. It is an array, and the scale of frequencies within the array is too course.
> > You assert that by using even a fastwoofer like the 2226 which
> > only has an internal delay equal to .64 feet (I just measured one)
> > and by using a crossover where signals above Xo come out
> > significantly AHEAD in time of those coming out of the low pass,
> > that by moving the hf source still further ahead in time,
> > that "this" fixes the time problem. Maybe fixes isn't what you
> > mean since you do say the ear is not sensitive to time.No, Tom, I have not. I do not promote alignment of drivers, by either moving them forward or back. I submit that neither will correct a speaker's time alignment. There are two planar sources separated by a distance, so even if they are completely 100% synchronized with one another - crossover, drivers, everything - there is still parallax between them and the listener. You cannot correct this problem in all positions and at all frequencies. You can only pick a position and frequency, and make an adequate compromise. The best solution is that which makes the best compromise for the task at hand.
> > I am not sure if you really don't understand or are posturing in
> > defense of "your approach" that you refer to but if you really
> > don't get it, explain at what point the concept is inconsistent
> > with the physics and I will try to explain it further.I understand completely. And "my approach" isn't germane to this discussion. I assert that a filter cannot be accurate in both the time domain and the frequency domain, simultaneously. Our loudspeakers are collections of filters - They are electronic filters and acoustic filters. So I have no approach but rather a position - and that position is that the Unity device is also bounded by these same physical laws too.
Tom, I'm sure that the Unity device is a good product. But you must understand - When you insist that you can make a horn provides acoustic loading from 100Hz to 20Khz, you are saying something that is not unlike what would have been said by snake-oil salemen of the 19th century. This is also true when you say you can make a filter - no, a collection of filters - simultaneously accurate in the time domain and the frequency domain.
I honestly expected more from you.
Wayne Parham
Wayne wrote:
---------------
> > If not acting as an acoustic transformer, what then explains the
> > impedance curves which clearly show "radiation resistance" and
> > how then do the mid drivers reach an electroacoustic efficiency
> > greater than the compression driver?Filters modify impedance and act as reactance without matching impedance. I expect your horn will load the midrance band for a couple of octaves and that's all. That - and the fact that you use several midrange drivers - is why midrange output is higher. But the horn unloads a couple octaves above its flare frequency, and after that, the horn's flare is acting only as a reflector.
---------------Wayne,
The impedance curves Tom is referring to are of the drivers mounted to the horn, with no crossover whatsoever. The increase in impedance through the passband of the driver directly demonstrates the horn loading of the system. Similar increases in passband impedance are observed for EACH segment of the horn.
---------------
Your use of the phrase "cover 100Hz to 20Khz" is ambiguous. If you're talking about making a horn that provides acoustic impedance matching over a span of seven octaves, you're wasting your time. You can load the diaphragm(s) for two or three octaves, and not more.
---------------All right, now we're getting somewhere! Your statement here of "You can load the diaphragm(s) for two or three octaves, and not more" is exactly what we are doing, for EACH set of drivers. In the 16.5" horn the midranges are loaded from about 300Hz well past 900Hz(three octaves) where the driver rolls off in the 1.2kHz range. The compression driver is operated for less than 4 octaves.
What we find is that if we extend the the mouth of the horn out further, as we move further from the apex past the midrange loading point, we will find a location where the expansion rate will co-incide with what is required to load the octaves below the midrange's low end cuttoff. This process continues out to however large we want to make the horn.
Do you not agree that drivers with suitable parameters loaded into the proper section/expansion rate of the horn will provide horn loading?
---------------
Did you ever consider that your measurements might be in error? Not to be rude, but I've even considered that they were "hand picked" and that you've only shown those that support your case. You have distanced yourself from the Lambda graphs, but I find them to be extremely telling.
---------------I've seen Tom continually question everything, especially when software predicts something which hasn't been done before. This is how Tom comes up with the unique solutions he does. The measurements are indeed valid, and have been verified by many well respected designers in the industry, even by other competing manufacturers. One significant quality our company is known for is that we meet or exceed all of our claims and specifications. Many out there massage published specs with "marketing markers" but we are in fact the ones bucking the trends here. If you would like to see some measurements, we can arrange that. There is no way to "hand pick" measurements which are repeatable by others under the same methods. These are not some newly thought up measurement method, but rather means which have been in use and proven for years.
In another post you posed the question that if this indeed did work as claimed, you could create a horn which loaded into ultrasonic frequencies. As a matter of fact, you can! The limitation is in drive units available. The aperture can be extended on both ends to cover whatever frequency desired, so long as you have drive units with suitable parameters. I know Audax has made a few tweeters with Ti domes with rediculously high frequency response. If a proper compression chamber was designed, loading to very high frequency would certainly be realizable.
Mark Seaton
Hi Mark!You wrote:
> > In the 16.5" horn the midranges are loaded from about 300Hz well
> > past 900Hz(three octaves) where the driver rolls off in the 1.2kHz
> > range.A 1/4 wavelength of 1.2Khz is 2.8 inches. Are the distances between midrange diaphragms smaller than this? Is the distance between the midrange cluster and the compression driver shorter than this? Is the diameter of each individual diaphragms smaller than this? And not center-to-center, but furthest radiating surfaces - are they within 2.8 inches?
That is a mighty small area to try to keep stuff packed into, trying to establish a 1/4 wavelength maximum distance for the coupling that is claimed.
> > What we find is that if we extend the the mouth of the horn out
> > further, as we move further from the apex past the midrange loading
> > point, we will find a location where the expansion rate will co-incide
> > with what is required to load the octaves below the midrange's low
> > end cuttoff. This process continues out to however large we want
> > to make the horn.I know that you must realize the implications of what you are saying. You are saying that an infinitely large horn will provide loading of the entire audio range, in fact, above and below it.
What I'm saying is that this cannot be the case. The only scenareo where the horn meets the conditions of the Webster equation is when the diaphragm is loaded at the horn's apex. I'm not saying that the device has no merit if the diaphragms are installed somewhere else, but I'm saying it doesn't fit the model anymore and it has been shown to reduce efficiency.
The further out the drivers are placed towards the mouth, the more of a rapid transition you have from diaphragm to open space. The larger the rapid discontinuity is, the more the diaphragm acts as if it were driving into free space. This, in turn, reduces impedance matching and effectiveness of the horn as an impedance transformer.
Wayne
You wrote:> > If not acting as an acoustic transformer, what then explains the
> > impedance curves which clearly show "radiation resistance" and how
> > then do the mid drivers reach an electroacoustic efficiency
> > greater than the compression driver?Filters modify impedance and act as reactance without matching impedance.
True, but the "efficiency" I was referring to was based on the radiated acoustic power compared to the electrical input
power, the fact that part of the system is electrical, part electroacoustic and part acoustic seems secondary as it is the
system I was testing.I expect your horn will load the midrance
band for a couple of octaves and that's all. That - and the fact that you use several of them - is why midrange output is
higher.
But the horn unloads a couple octaves above its flare frequency, and afetr that, it is acting only as a reflector.I think your starting to get it, YES each driver or set of drivers can only load the horn for a couple of octaves.
But also that the frequency range of the horn is dictated by the size of the horn and that one can stack up horns of
differing sizes and frequency range to get a single horn that covers a very wide frequency band using multiple sets of
drivers.
"All" one has to do after that is arrange the drivers, acoustic paths and electrical filters such that when in impulse or
wide band signal is produced by the compression driver, that it is "joined" at the right phase and time with the mid
components and then so on down.
Since it possible to actually measure the drivers and filters separately or in combination, it is not impossible to get
pretty darn close to dead on. At that point the results are as good as the drivers permit.> > To date I have not made a horn which loads "the entire audio
> > bandwidth" ( the 10 octave span from 20 Hz to 20 kHz ), the
> > largest span is from about 200 Hz to 20 kHz but I am casually
> > working on one which I hope will cover 100 Hz to 20 kHz. Since
> > the math says it should work, I don't see why it wouldn't but the
> > issue is my efforts are governed by what makes the most logical
> > product.Your use of the phase "cover 100Hz to 20Khz" is ambiguous. If you're talking about making a horn that provides
acoustic impedance matching over a span of seven octaves, you're wasting your time.
You can load the diaphragm(s) for two or three octaves, and not more.I am talking about an efficient horn loaded system which uses one physical horn shape to cover that entire frequency
range. Polar plots SHOW that the radiation pattern of a Unity horn is that of of a single source, no lobes, no
interference.
> > You insist it won't work yet Energy vs time measurements show it
> > does, you insist it can't be a point source yet polar measurements
> > show it is.Did you ever consider that your measurements might be in error? Not to be rude, but I've even considered that they
were "hand picked" and that you've only shown those that support your case.
You have distanced yourself from the Lambda graphs, but I find them to be extremely telling.Surely you jest, dude, I have taken thousands of TEF measurements, I have had them at my disposal for nearly 20
years and I was even invited to present at a TEF workshop on "exotic measurements with the TEF" twice.
Ask anyone in Pro audio about our spec.'s and they will tell you our stuff meets its spec.'s.
For you to suggest that the countless measurements involved were all in error is, well, not likely.I will tell you what I told Nick about the "indoor" measurements he took, that for one if it was a TEF measurement, it
would show that there were reflections leaking in and that since I don't use MLS I couldn't tell much more.
I urged him to get an old TEF machine.
> > You insist it is an "approximation" while it was designed, built
> > and fine tuned entirely based on actual measurements of time and
> > phase of the parts used.If you've done good work, then it is a close approximation.
But to say it's accurate in both the time domain and the frequency domain simultaneously is not possible.
It is not a point source. It is an array, and the scale of frequencies within the array is too course.An array of sources that are small acoustically (less than 1/4 wl in size) can combine totally and coherently IF they are
less than 1/4 wl (1/3wl if some directivity is acceptable) apart, center to center.
In the range they operate in, the mid driver exit holes and horn dimension are less than 1/4 WL and load the horn as a
pressure (being too small acoustically to make a beam).
As the pressure is presented in a dimension less than 1/4 wl, acoustically, that is a point source.
> > You assert that by using even a fastwoofer like the 2226 which
> > only has an internal delay equal to .64 feet (I just measured one)
> > and by using a crossover where signals above Xo come out
> > significantly AHEAD in time of those coming out of the low pass,
> > that by moving the hf source still further ahead in time,
> > that "this" fixes the time problem. Maybe fixes isn't what you
> > mean since you do say the ear is not sensitive to time.No, Tom, I have not. I do not promote alignment of drivers, by either moving them forward or back. I submit that
neither will correct a speaker's time alignment.
There are two planar sources separated by a distance, so even if they are completely
100% synchronized with one another - crossover, drivers, everything - there is still parallax between them and the
listener. You cannot correct this problem in all positions and at all frequencies.
You can only pick a position and frequency, and make
an adequate compromise. The best solution is that which makes the best compromise for the task at hand.You say "planar sources" but the drivers we are using are acoustically small, a "point source" and in fact do combine
seamlessly IF they are combined at less than 1/4 wl at the largest point.
I don't care if the compression driver is not "in time" at 100 HZ or where any driver is in time when its way out of
band, the only time the driver to driver relationship matters is in the range where the exchange from one to another is
being made and when they are in the operating band.> > I am not sure if you really don't understand or are posturing in
> > defense of "your approach" that you refer to but if you really
> > don't get it, explain at what point the concept is inconsistent
> > with the physics and I will try to explain it further.I understand completely. And "my approach" isn't germane to this discussion. My approach is actually no approach
because I assert that a filter cannot be accurate in both the time domain and the frequency domain, simultaneously.
Our loudspeakers are collections of filters - They are electronic filters and acoustic filters. So I have no approach but
rather a position - and that position is that the Unity device is also bounded by these same physical laws too.Well I guess in Waynes world all knowledge is known leaving no room for a new approach.
You, in an earlier post, went on to recite how your work is based on a solid foundation and referred to a paper
written in part by Don Keele.
Next time you see Don, ask him about my designs, what I do, where I fit in the loudspeaker industry, I figure when he
hears the speaker, an explanation and then says "hey Tom that's a cool idea" that it means something too.
He "gets it" may he can explain it to you.
Tom, I'm sure that the Unity device is a good product. But you must understand - When you insist that you can make
a horn provides acoustic loading from 100Hz to 20Khz, you are saying something that is not unlike what would have
been said by snake-oil salemen of the 19th century. This is also true when you say you can make a filter - no, a
collection of filters - simultaneously accurate in the time domain and the frequency domain. I honestly expected more
from you.Wayne Parham
Wayne I have to say I honestly expected more from you too.
I have been looking at a lot of old posts today , it seems that you have made dozens and dozens if not more posts
about the Unity horn in the last year or so, you had made this a personal crusade "to stomp this out".
In the end it is exactly the kind people I had wanted to Unitys to go to that you have harmed, how many that were
thinking about it didn't get it because of your tirades. Those tirades driven because you felt threatened or something
and based on your misunderstanding of how things work and lack of test equipment to find out for yourself..
Just as bad, you harmed Nick's business, he is a nice guy and one of relatively few places a DIY'r can get cool stuff,
these are the people that need support, not someone crusading against them, especially someone who is wrong to
boot..You insist that the Unity doesn't work but will not or cannot answer any of my questions related to getting you to
explain on what grounds you base your insistence. You even suggest that the measurments are wrong, when the fact
is, you are wrong, I would guess most of the folks reading this know it too.
I suspect you are feeling like you took a good long Whiz into a campfire and now are finding having to stand in the
"steam" unpleasant.
After seeing what you have been saying to people, I realize I have been too patient, there seems to be no substance
to you technical argument and continued jumping up and down saying it can't work is boring.
Enjoy the steam, breath deep.Tom
Hello again Tom!You wrote:
> > I have been looking at a lot of old posts today, it seems that
> > you have made dozens and dozens if not more posts about the Unity
> > horn in the last year or so, you had made this a personal crusade
> > "to stomp this out".Actually, there have only been two times I've given the Unity device a second thought. The first was a year ago, when Mark Seaton wrote remarks comparing the Unity with my Professional Series four π loudspeaker system. The second time was this week, when I was surprised to find that you had worked on patenting the thing.
I have written a lot of things in general about loudspeaker performance in the time domain; This is not necessarily about the Unity, and my opinions in this matter are first documented in the 70's.
Maybe you feel I've been a "big meanie." Maybe my approach has been rude, and I'm sorry for that. I don't honestly like "product bashing," and find it distasteful. So I don't like that this thread has taken this tone, and for my part in it, I'm sorry.
But I think the technical debate we've entered into has merit. I'd like to see it continue until it reaches a point of conclusion, if that's possible.
> > In the end it is exactly the kind people I had wanted to Unitys to
> > go to that you have harmed, how many that were thinking about it
> > didn't get it because of your tirades.You have yet to convince me that your claims have merit. My opinion stands, and with good technical grounds.
> > Those tirades driven because you felt threatened or something
> > and based on your misunderstanding of how things work and lack
> > of test equipment to find out for yourself.Your explaination of your device doesn't square with existing theory. You have the extra burden of proof. I do not misunderstand - I disbelieve.
> > Just as bad, you harmed Nick's business, he is a nice guy and one
> > of relatively few places a DIY'r can get cool stuff, these are the
> > people that need support, not someone crusading against them,
> > especially someone who is wrong to boot.And in an earlier post, you wrote:
> > You must imagine the world revolves around you I guess and that no
> > one has ever given any of this though before you.See, Tom, now you're showing your arrogance. You see in others what is most evident within yourself. It is you that appears to imagine the world revolving around you and "needing" your "revolutionary new explaination" of horn theory to replace our existing ones. You appear to actually believe that you have broken all the rules and hit new ground. So it is you that is predatory with your claims.
I'm not the only one that is building horn loudspeakers, and neither are you. You are not the only horn builder that caters to the DIY market.
But you and your associates are the only ones saying that you have a single horn flare that is capable of providing acoustic loading from 200Hz to 20Khz. You say that it is a perfect point source. Since the rest of us discuss our systems using traditional horn theory, that puts us at a distinct disadvantage. We speak in terms of mass loading and upper frequency cutoff, while you and your associates speak of ultrasonics from physically large horns. Quite a difference, and shocking since your device is based on a Salmon family horn, same as most of the rest of us. These are all fairly narrow bandwidth devices, limited to a few octaves or so. Certainly nothing like two decades. So when you claim that yours is capable of doing this, you are potentially hurting the business of everyone else.
...and you are wrong to boot.
> > You insist that the Unity doesn't work but will not or cannot
> > answer any of my questions related to getting you to explain on
> > what grounds you base your insistence.I have made my grounds very plain. I'l reiterate them here:
1. I do not believe that the Unity device acts as an acoustic transformer, i.e. provide horn loading over the two-decade bandwidth claimed.
2. I do not believe the Unity device manages to correct the movement in time over the span of frequencies delivered by its subsystems in this two-decade bandwidth, such that it acts as a point source. You have said, "generally a first or second order crossover gives the best results" in this design. This very statement means that you had not figured out how to make this thing work when you wrote it. Not only that, but the midrange drivers have wavelength-scale distances between them and across their diaphragms, and they enter a common chamber.> > I suspect you are feeling like you took a good long Whiz into a
> > campfire and now are finding having to stand in the "steam"
> > unpleasant.Touche! That's so funny!
But you appear to have become unbalanced and defensive. If your argument has merit, then it should be easier to explain than this. One would think you would have quantified it better by now.
> > An array of sources that are small acoustically (less than 1/4 wl
> > in size) can combine totally and coherently IF they are less than
> > 1/4 wl (1/3wl if some directivity is acceptable) apart, center to
> > center. In the range they operate in, the mid driver exit holes
> > and horn dimension are less than 1/4 WL and load the horn as a
> > pressure (being too small acoustically to make a beam).If that were true, I could accept this. But at high and middle audio frequencies, 1/4 wavelength is pretty small. How are you managing to get multiple midrange drivers packed within the space of less than 4 inches? A quarter wavelength at 1Khz is only 3.4 inches, and I expect you are running the midrange diaphragms even higher than this. If they are crossed over with a low order filter, then the regions of interest are pretty wide and wavelength scale gets very small.
So what are your dimensions and frequencies, exactly? Please quantify the scale instead of speaking in generalities and put this issue to rest.
We've discussed the time alignment issue quite a bit. So let's move to another issue that is perhaps even more telling. I've read your claims about wideband loading by putting diaphragms in the walls of your conical horn. It's always left me scratching my head. The rest of the world puts their diaphragms in the apex, and often with a compression chamber. This is always done to increase efficiency, and having constant cross section area is required to satisfy the conditions of the Webster equation. Failure to do this has been shown to reduce the effectiveness of the horn, and is what I sometimes refer to as a "mal-formed horn."
Most of your peers feel pretty good when we can get linear response from a horn over three octaves. Please don't simply show what you do, or discuss the phenomenon with the analogy of the "changing flare rate." Another way to describe this is a "large discontinuity" in the throat for the drivers mounted closer to the mouth. Some of my designs have this too, but I've always called it a mal-formed horn - a compromise - 'cause that's really what it is. In those designs, it's better than nothing, but it is used for economy of size and complexity, and not for performance.
So please expand on the mathematical model you've used to derive the wideband performance of your horn. Maybe you can discuss how it relates to the Webster equation.
I'm still laughing about the "steam." Nice shot, man!
Wayne
Wayne wrote:
> > > > That's the key here. "Temporarily" That's the point of all this. Moving the offset forward or back, either one is still a compromise. Your defense of the time alignment technique you've used only makes my case even more clear. There is no way to truly time align two different point sources at every frequency and at different positions with techniques as simple as position offset or the use of single-node filters. You can make them aligned at one frequency, and at one position and that is all. < < < <When the drivers are within 1/4 wavelength of each other you can. Keep in mind that the Unity has limited dispersion and a coaxial arrangement. Together with the carefully chosen spacing and an appropriately designed crossover, this makes off-axis lobing problems a non-issue in the Unity. The ONLY way in which a Unity does not act as a point source is due to the reflections you get from edges of the horn. This is a minor issue and, of course, not unique to the Unity. In fact reflection issues are probably less of an issue in the Unity than in discrete horn systems simply because you have fewer edges. I am sorry you are having trouble accepting that the Unity acts as a point source, but it is the truth.
Wayne wrote:
> > > > I think Davies says it well, "It is one of the fundamental laws of linear systems that if their output depends only on previous input - that is, if they cannot see into the future - then the phase response is completely determined by the amplitude response." < < < <This is true for linear "minimum phase" systems--phase is related to magnitude by a Hilbert transform, so if you know phase you can calculate magnitude and vice versa. This means that if a minimum phase system has a flat magnitude, it also doesn't have phase shift. I have heard this analogy between minimum phase systems and systems that don't "look into the future" before. It is a simplification that helps people intuitively understand what is required for a system to be minimum phase. However, it only works when talking about linear filters applied in series to a signal, not filters applied in parallel. Crossovers are, of course, filters that are applied in parallel and their output is summed.
First order crossovers are minimum phase--a flat on-axis magnitude response will also not exhibit phase shift. However, higher order crossovers are not minimum phase--they may have flat magnitudes, but they also have phase shift. Systems with flat magnitude and with phase shift are called all-pass systems. They are not minimum phase (phase is not unitquely determined by magnitude), but I wouldn't say that they "look into the future".
John
Hi John!You wrote:
> > I am sorry you are having trouble accepting that the Unity acts as a point source.
Not even a single driver acts as a point source; There is parallax across the diaphragm and with the listener. This device has many drivers in different positions, all entering a common horn. The crossover region between each of them is wide, so 1/2 wavelength is a large set of values having a wide range. About this business of having the distance between drivers limited to less than a 1/4 wavelength - that's smaller than the diameter of each diaphragm. A 1/4 wavelength of 1Khz is only 3.38".
There was performance data on the Lambda site for months which showed tell-tale frequency anomalies from the Unity horn. It was the sort of spiked dip that is evidence of diffraction interference caused by a phase reversal. Tom and Mark have tried to distance themselves from this by all means possible because it clearly shows what I say to be true.
Wayne Parham
"There was performance data on the Lambda site for months which showed tell-tale frequency anomalies from the Unity horn. It was the sort of spiked dip that is evidence of diffraction interference caused by a phase reversal. Tom and Mark have tried to distance themselves from this by all means possible because it clearly shows what I say to be true."Wayne,
This narrow dip you are referring to was at ~4kHz and was found to be a result of relfections off the outer edge of the horn. This has nothing to do with the integration of the drivers. The dip you are referring to does not change AT ALL when the midrange is disconnected and only the compression driver is tested, so clearly this is not a result of deconstructive interference with the midrange, as it occured in the Lambda Unity horn when the midrange was not operating. Take a look at the polar plots on the TD-1 and the frequency response of the Unity VFL loudspeaker. These problems you claim are not evident on these two other examples.
You also stated:
"About this business of having the distance between drivers limited to less than a 1/4 wavelength - that's smaller than the diameter of each diaphragm. A 1/4 wavelength of 1Khz is only 3.38"."The 8 entry holes for the midrange drivers are spaced at less than 3" appart. The beauty of this is that the driver diameter does not matter in this loading method so long as the R/L high frequency corner of the driver is high enough, and the drivers can be loaded into these holes. Seeing the solutions to the obstacles you are stuck on is exactly how Tom came up with the design!
Mark Seaton
Hi Mark,RE the 4kHz dip on the Lambda version, it was the holes, not the edge. When I did a quick mod on my Lambda Unities that Tom suggested it disappeared.
John
-------------
Hi Mark,
RE the 4kHz dip on the Lambda version, it was the holes, not the edge. When I did a quick mod on my Lambda Unities that Tom suggested it disappeared.John
-------------Correct John,
The throat geometry made for a problem with the holes that caused energy which reflected off of the outer portion of the flare, causing the problem to be seen only directly on axis. The dip is further minimized with some absorbent foam at the center of each final flare.
Mark
You two are a gas!You defend your position that the device sums true, but then argue about what caused the anomaly I've referred to.
Of course it's the geometry. When you've got two point sources that overlap, you're going to have positions that display this sort of anomaly. I understand that you've tried to make the flare angle narrow enough that the anomalous regions are outside the flare. But combining the sources in a common chamber ensures that you will have interaction, nonetheless.
Mark - You've said that "the 8 entry holes for the midrange drivers are spaced at less than 3" apart." But I'm sure you mean adjacent holes are less than 3" apart. So how far are the most distant holes?
John - You've said that you found the 4kHz dip on the Lambda version was the holes. This means that there was a diffraction anomaly, just as I've said.
So what argument exactly are you defending?
Wayne
--------------------------
You defend your position that the device sums true, but then argue about what caused the anomaly I've referred to.Of course it's the geometry. When you've got two point sources that overlap, you're going to have positions that display this sort of anomaly. I understand that you've tried to make the flare angle narrow enough that the anomalous regions are outside the flare. But combining the sources in a common chamber ensures that you will have interaction, nonetheless.
--------------------------Ok, then please be more specific in what you see a problem with. Your comments claimed an INTERACTION between the output of the drive units. Here you are talking about the effect of having entry holes to the flare. This effect, has no change whether the midranges are in place, connected, or just sitting there. If this was an acoustic interference problem, this would only be observed when the midranges where simultaneously driven and measured. Do you not agree?
--------------------------
Mark - You've said that "the 8 entry holes for the midrange drivers are spaced at less than 3" apart." But I'm sure you mean adjacent holes are less than 3" apart. So how far are the most distant holes?John - You've said that you found the 4kHz dip on the Lambda version was the holes. This means that there was a diffraction anomaly, just as I've said.
So what argument exactly are you defending?
--------------------------The holes are sized and spaced to minimize any discontiunities in the horn. With some simple refinement this anomaly becomes quite small and narrow in band, and even unseen in 1/3rd octave measurements. What was being defended was your accusation that the drive units do not sum as a single source, and your claim that this anomaly which is dependent ONLY on the compression driver in the horn, somehow demonstrated an improper summation of the acoustic energy of the compression driver and midrange. Again, the crossover is at 1200Hz, and the midrange output is attenuated at least 20dB by 4kHz when the acoustic output and crossover network are combined.
Mark Seaton
Hello Mark -You wrote:
> > then please be more specific in what you see a problem with. Your
> > comments claimed an INTERACTION between the output of the
> > drive units.That is what I see a problem with. There are several planar sources entering a common chamber with significant overlap between them. The frequencies of interest, i.e. the overlapping frequency range, is relatively high so that phase interaction is significant.
I know that you guys claim this isn't true. I've read that you say you have response charts to prove it. I've heard you say that many times. But the scale and dimensions of the components make this behavior something I would expect, and the Lambda response charts seem to confirm it.
Wayne
Wayne,This is absolutely infuriating. We try to correct your misconceptions and you answer with more nonsense. I don't know who you are trying to fool. To anyone who really knows acoustics and signal processing, it is immediately apparent that you're a fraud.
Now, back to the Unity. The narrow dip that I believe you are refering to on the Lambda site had NOTHING to do with driver offset--I repeat NOTHING. It was due to reflections off the midrange holes in the Lambda version of the Unity. The SoundPhysics versions do not exhibit this dip because they take efforts to avoid those reflections, efforts that were not implemented in the Lambda version. Nick (of Lambda Acoustics) explained to everyone how to make a quick mod to get rid of the dip, but he never updated his measurements.
You know Wayne, in a way I have to thank you. You're delusional, but you did succeed in getting Tom to write some interesting posts. I always learn something when he writes.
John
John -That was extremely rude.
For you to call me a fraud, and then to have admitted that what I've said about frequency anomalies in this device is true, is absolutely inconceivable.
Your words:
> > the 4kHz dip on the Lambda version, it was the holes, not the
> > edge. When I did a quick mod on my Lambda Unities that Tom suggested
> > it disappeared.Holes. Midrange. Multiple point sources. Diffraction. Anomaly.
You guys are really something.
Wayne
"You're delusional, but you did succeed in getting Tom to write some interesting posts. I always learn something when he writes."Agreed on both counts.
Thanks for bringing Mr. Danley to this site!
Magnetar
Mike -I don't appreciate your tone, and find your opportunistic position is disgusting.
You know, the point you always drove into the ground with me and my designs, was my choice to push the upper limit of my conical horns. You always discounted my decision to allow them to be used past the point where the horn had unloaded.
But now, here, you side yourself in a debate with a group who has said that they believe their similarly shaped horns can be used through three subsystems - woofer, midrange and tweeter.
I don't appreciate your comments; It is clearly evident that they are strictly personal and not relevent to the discussion at hand.
Wayne
Your speakers are far and away different then what Danley is doing. You of course have never measured or heard a Unity horn system but puff up your child like chest and make unfounded proclamations and yourself look like a fool.Yeah it's personal -- personally I think this is about as funny as it gets.
Magnetar
Hey Mike -Did you continue to follow this thread? I was wondering why you didn't interject your technical comments, questions - Anything except your few isolated insults? Maybe you hoped no one else would notice.
Ah, but by-gones be by-gones. How are your sales going? Can you even give stuff away? Maybe you've decided to join me and Danley by selling large-format rectangular horns with conical flares for wideband use?
Then again, even if you did then two weeks later you'd change your mind again.
Wayne
Bates -You just can't keep from making it personal, can you? Well, OK then. How about telling us just how much penis envy goes into those monstrosities you build and drag around to every show you can find? You change them every two weeks, in what appears to be an attempt to be all things to all people.
At least I'm consistent. I build a product that works well, and is successful. I don't change my mind about what I think works best every two weeks. It's a tried-and-true design, and I stick by it. Naturally, I'm proud.
That has nothing to do with this discussion. I was surprised to find a patent on the things that were claimed. I am surprised to see someone in the industry claim to have a horn design that provides acoustic loading through two decades, from 200Hz to 20Khz. And I'm surpised to see someone saying they think they have a design that is simultaneously accurate in both the time domain and the frequency domain. Physicists say these things can't be done.
But, Mike, you just keep on changing sides. Maybe your moniker should be "the Chicago Chamelion."
Wayne
I don't pick sides. I build whatever i feel like building and don't really give a shit what other people think about it. If I want to change something I do, it's an addiction I have. Nothing more, nothing less. I drug a BIG system to Lima and will never do it again. Trying to integrate a big system in a shitty room isn't my idea of FUN. Oh, that's one "show" which wasn't really a show, it was a gathering of hobbiest. I have a much smaller speaker now I can load in my van and take it with me, that's what I'll take if I go to any new gatherings. Maybe you should go to Indiana with your big speakers and let folks compare them to the Titans or the Unities- or for that matter my little 2 ways. You can spout all you want about the technical aspects (debate -lol)of loudspeakers but it's my position the majority of the people that read this board find you about as interesting as a broken Burle Ives record.
Magnetar
Mike -What exactly did your last post have to do with this discussion? For that matter, what did any of your comments have to do with this discussion?
Nothing.
This thread was started with John posting news of the Unity patent. I was surprised because I didn't honestly think it was patentable. I wouldn't expect the claims to be accepted, any more than I expect over-unity energy machines to be. So the focus of the dialog became one where Tom explained what his device is really capable of doing, and what should be expected from it. The discussion took the form of my challenging claims and his defending them. In the process, we discovered exactly what is really meant by those claims - Separating the behaviour that is actually expected from the device from that which was only semantic.
So where do you want to fit in all of this?
This forum is for the exchange of ideas, for debate and for sharing experience about sound systems. It is NOT a place for making insulting personal remarks.
Wayne
Wayne,As far as I can tell this "debate" has been about you trying to stroke your own ego by inventing conspiracies (eg. Tom copying your products and steering clients away from them) and belittling the invention of someone who is clearly your superior. You have done so by making claims that make no sense and you have supported those claims with nothing more than the weight of your personal authority. There is absolutely nothing intellectual about this endeavour so don't pretend there is...and if Mike wants to call you on your BS then in my view he is justified. Please stop compensating for your inadequacies at someone else's expense.
John
John -Where's that FEA model? If you can't show the horn provides acoustic loading from 200Hz to 20Khz, then shut the f**k up.
Wayne
Hi Wayne,Pressure getting to you? That's a pretty strong statement coming from someone who has done nothing but deliver conspiracy theories and faulty analysis all week--and loads of it. As I said before, I'll try to get to it some time this week. I do have a life after all. I wouldn't think you'd be so anctious to be proven wrong again.
John
John -Do you really think you're going to "prove" that a horn flare can provide acoustical loading through two decades?
And by the way, how do you propose to solve for a non-symetrical flare? The midrange diaphragm enters into a large discontinuity which does not have the required equal pressure across its cross section.
Wayne
nt
How about the famous peer reviewed "crossover document."
Magnetar
Go ahead Mike -Give us your technical evaluation.
If you can.
Wayne
Wayne,You're belittling someone else's technical abilities? Now that's the pot calling the kettle black...
John
Hey John -How about you join us in the debate. Please show us all how that midrange diaphragm in the Unity can have its leading edge driven behind the trailing edge, as far as phase is concerned. I guess it really is a proprietary design, because it warps the diaphragm by an amount that increases with frequency. The Unity will need to do these things to be in phase across its bandwidth, you know.
Also, solve the wave equation for this horn, and show how it loads from 200Hz to 20Khz.
Thanks,
Wayne Parham
Wayne,You are a lunatic...
> > > > How about you join us in the debate. Please show us all how that midrange diaphragm in the Unity can have its leading edge driven behind the trailing edge, as far as phase is concerned. I guess it really is a proprietary design, because it warps the diaphragm by an amount that increases with frequency. The Unity will need to do these things to be in phase across its bandwidth, you know. < < < <
What the hell are you talking about (leading edge behind trailing edge)???? Just give it up Wayne. You might have been able to get away with your fraud on this list before, but not while I'm around. I don't mind your lunatic ravings so much, what bothers me is you deceiving DIY'ers who don't know enough to recognize your nonsense for what it is.
> > > > Also, solve the wave equation for this horn, and show how it loads from 200Hz to 20Khz. < < < <
That's a pretty ironic suggestion given that I just today made a web site about solving the wave equation using finite element models. You know Wayne, I already solved the wave equation for the Unity over a year ago. I'll see if I can find some time to post it to my web site this week.
Breath deep, Wayne.
John
Hi John!You wrote:
> > What the hell are you talking about (leading edge behind trailing edge)?
Notice that each midrange driver has an edge closer to the listener and an edge further from the listener. That's "what the hell I am talking about."
> > You might have been able to get away with your fraud on this list
> > before, but not while I'm around.I can see that you're really a quick study, so please continue to enlighten us. But first, look up the phrases, "leading edge" and "trailing edge." Also study up on phenomenon at wavelength scale, specifically those of quarter wave and half wave.
> > I don't mind your lunatic ravings so much, what bothers me is you
> > deceiving DIY'ers who don't know enough to recognize your nonsense
> > for what it is.Well, I have yet to see what sort of mechanism is twisting the diaphragms on those midrange units so they'll be in phase.
> > That's a pretty ironic suggestion given that I just today made a
> > web site about solving the wave equation using finite element
> > models.I saw that. It looks like good stuff.
> > I already solved the wave equation for the Unity over a year ago.
> > I'll see if I can find some time to post it to my web site this week.Super! It will be interesting to see. Don't forget that the radiators aren't a point source, so don't approximate them as such. For those frequencies where their dimensions exceed 1/4 wavelength, you'll have a bit of a problem. You'll need to solve for them using an array of radiators as your closest approximation.
Please post your models!
Wayne
Ah Wayne, your lunacy knows no bounds. You do realize that the midranges enter the horn through small holes located in the middle of the drivers?Get some rest Wayne, John
Hi John!Now we're getting somewhere! And the distance between these holes and the apex, where the compression driver is located? And the overlap region in the crossover? And the anomoly that you found in your Unity horn?
Anxiously awaiting your FEA model,
Wayne
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
