I agree with the concepts you pointed out but....You never pointed out what you consider to be "full spectrum"2- vs 3-Way Notes
As it is your thread, it is your responsibility to define the terms of the question you pose.
Lower Frequency Bound
There is no doubt about what can be heard/felt in the lower octaves. The lowest fundamental for a pipe organ is 8 Hz, with 16 Hz being a typical low limit for most large organs.
Upper Frequency Bound
Above say, 16 kHz. it depends on your inherent hearing capability, conditioned by your rock concert attendance history. The physical and acoustical attributes of beryllium are far superior to those of titanium, aluminum and magnesium; however, it is a toxic metal, difficult to form into tenable diaphragm geometries, and of course, expensive. Most will not hear the differences, even though they are clearly measurable, particular in regards to the frequency which marks the onset of diaphragm breakup modes. The dual annular diaphragm designs, made from Mylar film, provide a viable cost effective alternative. This particular design may be operated at a lower c/o frequency, as the dual compliances mitigate the rocking modes sustained by single surround, domed designs when they are operated in a comparable low frequency domain.
WHG
As it is your thread, it is your responsibility to define the terms of the question you pose.
Lower Frequency Bound
There is no doubt about what can be heard/felt in the lower octaves. The lowest fundamental for a pipe organ is 8 Hz, with 16 Hz being a typical low limit for most large organs.
Upper Frequency Bound
Above say, 16 kHz. it depends on your inherent hearing capability, conditioned by your rock concert attendance history. The physical and acoustical attributes of beryllium are far superior to those of titanium, aluminum and magnesium; however, it is a toxic metal, difficult to form into tenable diaphragm geometries, and of course, expensive. Most will not hear the differences, even though they are clearly measurable, particular in regards to the frequency which marks the onset of diaphragm breakup modes. The dual annular diaphragm designs, made from Mylar film, provide a viable cost effective alternative. This particular design may be operated at a lower c/o frequency, as the dual compliances mitigate the rocking modes sustained by single surround, domed designs when they are operated in a comparable low frequency domain.
WHG
Last edited:
Absolutly ...
… and that makes the system 3-Way. The special design problems unique to reproduction of the lower frequency domain (first decade) are not at issue when addressing those of the two upper frequency domains (next two decades). Recognition of this fact will lead you to cost effective system design. No matter whether your budget is large or small, you will be getting the most 'bang for your buck' by recognizing this reality. WHG
… and that makes the system 3-Way. The special design problems unique to reproduction of the lower frequency domain (first decade) are not at issue when addressing those of the two upper frequency domains (next two decades). Recognition of this fact will lead you to cost effective system design. No matter whether your budget is large or small, you will be getting the most 'bang for your buck' by recognizing this reality. WHG
Last edited:
I also asked the question in the beginning of the thread, in so many words, Two mains +stereo subs.....is that a 3 way or a 2.2 system. For what its worth.... I have chosen the two-way plus subs, not because the bare minimal 2way (no sub) doesn't cover my needs for what I want for "full spectrum"....rather, adding a sub makes it better, can't get around it. If the answer to the question is yes or no, it is no consequence to myself, for I did not create physics. I will be running a two way plus a sub (lol) regardless. The journey that lead me here was well worth it. The thorough lesson on driver selection, theory and application, priceless....all that information would not of came across as it did, had I just said, ok guess I'm building a 3 way, see ya later guys. The best 3 way (for me) is the 2 way concept exploited to the best of its abilities with a sub covering the bottom....My current system is very similar,(2 way) 4" covering 2.1khz to 130hz crossed to a 12".
Whats the best 2 way? A 3 way.....Whats the best 3 way.....a 4 way. Who says that?
Whats the best 2 way? A 3 way.....Whats the best 3 way.....a 4 way. Who says that?
Domain Confusion, Frequency vs Spatial
You are slicing both stereo (or more) spatial channels into three separate frequency bands. In the frequency domain that is a three-way system design. The fact that you are combining the lower (spatial) bands to drive a common subsystem does not make your system 2-way in the frequency domain. A lack of precision in the terms we use, leads to unnecessary confusion and misunderstanding. WHG
You are slicing both stereo (or more) spatial channels into three separate frequency bands. In the frequency domain that is a three-way system design. The fact that you are combining the lower (spatial) bands to drive a common subsystem does not make your system 2-way in the frequency domain. A lack of precision in the terms we use, leads to unnecessary confusion and misunderstanding. WHG
Well, if you put this way, then considering your desire for excellent HF, a ribbon super tweeter or whatever is the best these days and use the ~traditional decade 50-500-5000 or 30-300-3000 XOs or whatever decades works best with the super tweeter + separate sub system.
GM
Last edited:
Correct me if i'm wrong - you will be using hornloaded 1" driver, a 15" midwoofer and a sub ?
If the answer is yes, may i ask why ?
Since you are already using subwoofers, i think that it would be very wise to use 12" for midwoofer duty. The advantages are enormous. You can use your compression driver as low as your horn allows it, CtC distance is lower, cone breakup is higher, sensitivity is great and it will blend with your horn where ever you want. Eminence Deltalite II 2512 for instance:
Eminence Deltalite-II-2512 | HiFiCompass
Now, that is not complete set of measurements (off axis and higher V distortion missing but i've asked Yevgeniy to update it and he promised he will, later this autumn). It can do 100Hz-1000Hz perfectly and in a system with some large waveguide and a sub bellow ..... brrrr
In a 40 liter closed cabinet it will do -6dB under 70Hz - natural roll off. Volume so low makes it a small cabinet, so easy to get right with large roundovers and lots of struts inside for removing the wall resonances. Last but not least, it can be moved around the room for best sound quite easily.
Last edited:
Good point.
When it became clear to me that audio was being driven by marketing and not by reality I decided that I would try my best to educate people. Professionals don't need this education and as you imply, most stay away from the amateurs especially on sights like this.
Among the most knowledgeable people in audio there is wide consensus about the topic. Among the less knowledgeable there is a very large dogma of false believes. The pros mostly just laugh and move on.
So the real answer is that I desire to spread my knowledge in hopes that some sanity can be brought to the table.
Sad, but true. But there is a limit to the amount of disrespect that anyone will take.
This is getting borderline insulting - to suggest that I make statements of fact without having any support for them is entirely false.
And your example is entirely wrong, at least as you seem to view it. I do check low level signals and I claim, and you can check this out, that these are where amplifiers show their differences. I even developed a test to find distortion products that are usually hidden by noise. You can find that description here at DIY.
I test acoustics at mid levels, not high, not too low ither. And I have talked extensively why loudspeakers will not be prone to errors at low levels except in extreme cases (and I even discussed this at one point and described the rare instance where I saw it.
Continue your disrespect and I will simply cease to respond to you comments.
This is entirely correct. Floyd Toole once asked me how I can hang around these kind of forums - he won't.
"One minor hiccup" - seriously, you think that this is uncommon?
Thanks Bill
It used to take me a week or more to have enough data to do a crossover. But then I developed software that allowed me to use a base set of polars and then manipulate the crossover in software. This cut down the process to a few days. With active it is even shorter since interactions of components is far less it can occur in minutes (once one has the complete set of polar data, which always takes a lot of time.
I don't follow "SHT", but a good waveguide, like mine, can go well into the upper ranges with absolutely no problems. The fact that all the round devices that you have heard you don't like, despite the claims (which ARE often false,) does not mean that it can't be done.
I'd agree with this. There can be some small differences, but in effect it is the waveguide that dominates the results. Poorer waveguides will tend to mate differently to different CDs. But I have found that a great waveguide can take just about any driver and sound just about the same.
My intention is not to insult anyone. Please explain then what did you mean by this:
Is it possible that i have misunderstood you ?
Last edited:
This is true, but waveguides are never made for the LFs (at least I never do for just this reason) they are HF devices and there they make a huge difference. They really only work well after the wavelengths become short enough to actually see the waveguide.
Its not because they are round, it's because they are poorly designed.
Just to be clear it is absolutely the case that no crossover can change the polar pattern of any driver. It does and can change the summation of the drivers, but that is all. The polar responses are completely linear so superposition is always possible. That's what allows me to do one set of polars and simulate any possible crossover from that set.
I can only imagine the capabilities of the software that can be available. But when I first started doing the crossovers (decades ago) nothing could do what I needed so I wrote all my own stuff. At the time I considered this to be an incredible competitive advantage and mostly kept this to myself. Today my code would be a joke as we found out some time back when we tried to bring this to the mainstream. Someone tried to rewrite the code in a more contemporary language and it just kind of collapsed due to the difficulty of doing it. I was sorry that this happened, but well, time moves on.
I was doing what Klippel is selling today, and had a patent along these lines, about ten years before Klippel. His system is impressive, but so is the cost. He has a staff of engineers, I worked alone.
And interesting sideline experience of mine:
Some years back (many years back) I was interested in expanding my business. I went to NYC for a few days to meet with a potential investor who had bought my speakers. His comment was that it was clear to him that I was passionate about my technology, but not so much about my product. In other words, I was a great scientist, but a poor businessman. He respected my position but he declined to invest. In hindsight, he was right of course and I never again thought of expanding since it was very clear that the business just didn't interest me. The science did.
CDs all have different characteristics and one MUST EQ them to a common curve. WHen this is done then it is very difficult to tell any differences.
I'll quote again the study that I did where I had two Summa's one with TADs and one with B&Cs, each optimized to a common goal. In a blind test of some 25 attendees, four systems (two identical designs with different drivers and two other different designs,) there was no statistical difference between the two driver sets while there was a huge difference between system designs.
I hate normalized polars!!!
Hi Bill
For all practical purposes the kinds of distortion that PWK was talking about have been well under control for a while. Today, I think that what most people hear as "muddy" is not nonlinear distortion but diffraction, which I showed could be heard as a similar flaw to nonlinearity. It's just that one is in the source and the other in the receiver.
This isn't true. In a completely linear system there is no modulation distortion and as the nonlinearities fall so does the modulation. This happens to the point where we can no longer detect them and this is pretty much the state that we are in today.
But yes, cone mass is irrelevant in this regard.
True, but obsolete. Back when driver nonlinearity was NOT well controlled this was the only cure. But drivers today are far more linear than there were in the past and so this rationale no longer holds water.
For home use this is not an issue as I can't get anywhere close to the upper power capabilities in my room as that would likely cause me to go deaf.
For pro use this would be a disaster.
It's hard to define "crossover frequency" when one uses different slopes and different gains. In essence the crossover moved about somewhat depending on specifics, but was always somewhere between 700 and 850.
Those plots are antiques. Look at the more modern ones on my site.
Depends on who you are talking to.
My mind's made up, I've concluded the optimum in any given physical size for home use.
The rest are just debates about small insignificant issues, or unsupported with objective data.
My work is to be the less laughtable and the more concise as possible
The technical communication to neophytes is tricky...
No, for the medium range frequencies, the radiation of a compression driver with a high directivity horn create a larger lobe of direct sound than a direct radiator and when you raise the compression driver size and its horn size it lowers this lobe frequency range.Being ironic?
You are just preferring the near field sound, that's all.