Geddes on Waveguides

MaVo said:
Hi JLH, i wonder if the distance between both horns will lead to increased interference, lobing, combfilter effects. Gedlees speakers have a smaller distance between both sources. I dont know guidelines to figure out if this is a problem in this case, so if someone knows better, please clarify.


These were all good posts and good work. I do think that the larger waveguide and horn might have polar response problems because of their size. Lobbing cannot be avoided, but it can be minimized and spacing is the critical thing. All these tradeoffs - sorting them out is indeed a real task.

What I would ask of JHL - why don't you try my approach first and see how you like that. It far easier to impliment that what you are proposing and you just might find that it changes your mind about the dynamics of a direct radiator. If you still aren't satisfied then you can add in your Tractrix horn, but at least you will have a well established basis for a comparison.
 
Disabled Account
Joined 2007
gedlee said:



Insults are seldom an effective way of making friends. And don't claim that it was a joke. Good jokes don't make fun of people.

No insult intended, just a comment from the sidelines.

We need to lighten up a bit.





Humility: The defining characteristic of an unpretentious and modest person, someone who does not think that he or she is better or more important than others. Synonym: humble
 
Re: different mouth distance

angeloitacare said:
i have drawn a tractrix lower midrange horn with 27" size at the mouth, and a oblate spheroid midrange horn with 20" mouth, and 10" deep. When the two horns are time aligned, the vertical mouth distance one to the other is about 13". How might this discrepance affect the perceived sound and integration of the two channels?
I'd think it's always best wrt to a consistent time behaviour vs. radiation pattern to align the mouths on the baffle, not the diapraghms... which means, when the HF part comes out to be ahead of the bass as it is here, the HF must be delayed... hard to do in a passive/analog XO. This is another point where the approach in Dr. Geddes' Summa is well balanced, he has the bass before the HF in time and that is easily taken care of in the XO. And both parts are integrated onto a single baffle which, besides of more benign diffraction issues in general, is quite probably making off-axis time behaviour very consistent to on-axis, at least in the sideways dimension.

Therefore, another +++ from my side, too...

- Klaus
 
I'd think it's always best wrt to a consistent time behaviour vs. radiation pattern to align the mouths on the baffle, not the diapraghms... which means, when the HF part comes out to be ahead of the bass as it is here, the HF must be delayed... hard to do in a passive/analog XO. This is another point where the approach in Dr. Geddes' Summa is well balanced, he has the bass before the HF in time and that is easily taken care of in the XO. And both parts are integrated onto a single baffle which, besides of more benign diffraction issues in general, is quite probably making off-axis time behaviour very consistent to on-axis, at least in the sideways dimension.
hi Klaus

there is no baffle. While with time aligned diaphragms the distance on axis to the listener is the same, off axis, the reverbant sound takes different paths, when the mouth is not aligned. Crossover delay is another chapter... I am not interested in comparisons between the two designs - summa - and lower midrange horn configuration in a multihorn system , they are two worlds, but the best method of horn alignment each to the other.

offaxis.jpg
 
gedlee said:



...
What I would ask of JHL - why don't you try my approach first and see how you like that. It far easier to impliment that what you are proposing and you just might find that it changes your mind about the dynamics of a direct radiator. If you still aren't satisfied then you can add in your Tractrix horn, but at least you will have a well established basis for a comparison.
I second this proposal. Most perceived limitatiodynamics heard in direct radiators are the more more complicated wave generated from various parts of the diaphram which are evident when looking at off-axis response. This normally does not exist if the xo point is below the beaming point in lower frequency drivers.
 
Angelo,

I find in real practice you can have the two horns/wave guides much closer than what your picture shows. It is a subjective call as to how close they can be. For the most part, the mid-range wave guide’s sound reflection off of the back lip of the mid-bass horn just becomes indirect sound. However, you still don’t want too much reflection.

l_02edde372e175ccaddc35b96f258845f.jpg





Klaus,

You make some very good points. My problem is I’m an Analog man and very much dislike digital devices and artificial time delays. In my past installations, I have had very good success with time aligning the diaphragms in a vertical arc. The convergence at the listening position becomes very good with this method. Of course, I have never done it with an OS wave guide before. I’ve only done this with tractrix horns. I’m unable to tilt the horns in the image, so I just drew lines to show how they should be tilted along the vertical arc.


l_328c848baf1eac6faae7d5a27296ad70.jpg



Dr. Geddes,

Thank you for your reply. You make a good argument for the direct radiator. However, I’m still not sure if there is one made that has the punch I’m used to. In addition, since I’ve always used horns all my amplifiers are of the low power tube type. My most powerful amplifier is 6.5 Wrms. Maybe a Karlson coupler would be an interesting compromise?

karl_frt.gif


Some of the few direct radiators I know I could live with are unobtainable at a reasonable price. The Western Electric 728B, the original Altec 515A, and a few Klangfilm woofers are just a few that come to mind. Earl, do you have a suggestion for a good direct radiator?

Thanks, JLH
 
Rather ironic, maybe deliberately, that the example given earlier was Frank Lloyd Wright. A man ahead of his time, highly egotistical, or maybe opinionated about his own value and whilst designing some of the best made many mistakes and was openly castigated for them by architectural people and owners.

This didn't diminish his achievements and ideas one little bit though. His points. maybe about 30 odd, on what to do or look out for when designing a residence should still be read by every architect and homeowner/handyman out there.

Back on topic to a degree. These posts and forums are invaluable when they stick to discussing thoughts, experiments and ideas but the persistent carping on in two or three of them is starting to make them a hard read. I believe that Graham Maynard mentioned somewhere that people could be, or are being, frightened off from the hobby if their knowledge level, like mine, is on the low side. Just a few thoughts for what its worth. Please all keep up the good work as we do learn more each day.
jamikl
 
JLH said:
Angelo,

Dr. Geddes,

Thank you for your reply. You make a good argument for the direct radiator. However, I’m still not sure if there is one made that has the punch I’m used to. In addition, since I’ve always used horns all my amplifiers are of the low power tube type. My most powerful amplifier is 6.5 Wrms. Maybe a Karlson coupler would be an interesting compromise?

karl_frt.gif


Some of the few direct radiators I know I could live with are unobtainable at a reasonable price. The Western Electric 728B, the original Altec 515A, and a few Klangfilm woofers are just a few that come to mind. Earl, do you have a suggestion for a good direct radiator?

Thanks, JLH


You are putting way too much imporatnce in the driver - its not that critical, its the system design that matters.

There are lots of good direct radiators, I just happen to like the B&C drivers becaues they have everything that I look for at a reasonable cost.
 
gedlee said:

In general I don't see much advantage to a changing cross-section in the waveguide, certainly not from circular to square. Circular to elliptical I could understand - if done right!!! - but not to square.

Thank You so much! :)

at last this is exactly the kind of answer I always expect from You Dr Geddes :)

the discussion can go on without the invaluable help from Mr Hansen ;)

"from circular to rectangular" (but NOT square) was as I mentioned above the Japanese guys first try (experimental, as they declared) in "Grand Scepter GS-1"
their second try – in "Scepter 2002" (image appended) – was exactly something You pointed to – "from circular to elliptical"
(Japanese comment from www.audio-heritage.jp put through "yahoo babelfish translation"):

Super hyperbolic horn with the technology which was developed the case of GS-1 development, concerning the change of the cross-sectional area adopts the super elliptic function which includes flat function and form function, hyperbolic function, concerning the change of shape by the fact that these simultaneous equations are solved, has become the horn which has the smooth extent which does not have sudden form change.

so it looks they had the right ideas or at least right intuitions?

the question is: did they actually do it right then? Or did anybody ever?

and what is most interesting – what could be the advantage of such design?
as You know the Japanese guys claim was "improved impulse response"
what can be the mechanism behind such an improvement?

best regards,
graaf
 

Attachments

  • jpg.jpg
    jpg.jpg
    18.4 KB · Views: 605

Attachments

  • gs_1_126.jpg
    gs_1_126.jpg
    93.6 KB · Views: 658
hi John

thanks for your answer. The polar response matching of two horns at crossover point might in fact help integration . I've not given importance to the vertical distance between the two horns at my draw, as this was not the focus of the question. There will be the tweeter between them, so there is a little more distance , and i thought to place it on axis, vertically, too. Or do you think, it would be better, to bring the two horns as close together as in your example, and the tweeter on side of them ? When you say 112hz of the 27" horn at half size, you want to say at half space, floor positioning, right ? I am getting 187hz free space, and 160hz half space. This , if the horn loads down to Fc. Have you tried, where the best crossover point would be with a 35" midbass horn , and the size of the OB in combination with it ? Lets say, we use a Fane Studio 8M in a 27" horn, which loads the driver to 160hz. Below it transmits down to ~ 90hz as a direct radiator. What difference would be perceived compared to the same driver, used in a 35" horn, which loads down to 125hz at half space ? would it be only efficiency, or might there be a different tonal behaviour ?

Earl

i don't agree that the driver used is of not so much importance, while only the system design is. I think both are important, drivers, horns and enclosures, specially, bouth, adapted as a unit working together. Your experiment between a B & C driver, and a TAD driver, is not indicative or representative in the sense, that, even if the price difference is very big, others that made comparisons between the same drivers , stated that the B & C is a top performing driver as well, even if cheap, and has a quality naturally on par with the TAD's. Not of all other drivers can be said the same thing. The comparative of these two cannot be generalized. There are very big quality and sound differences from one speaker to the other. And even when both used appropriately, differences appear.I had already a few different compression drivers to compare with, and find substantial differences. Why does Duke appreciate a TAD 1102 so much, but not , let's say, a cheap Eminence driver ? Why are some drivers in high regard, others not ? certainly not only because of the appropriately used horns and cabinets.

Angelo
 
graaf said:


...
so it looks they had the right ideas or at least right intuitions?

the question is: did they actually do it right then? Or did anybody ever?

and what is most interesting ?what could be the advantage of such design?
as You know the Japanese guys claim was "improved impulse response"
what can be the mechanism behind such an improvement?

best regards,
graaf
I am wondering what improvement they mean? Most irregular impulse shape would probably be due to the throat and it's matching to the driver as Dr. Geddes has mentioned. Looking at the impulse in his papers and the AI site, these two look a bit different which makes me wonder if the AI site has problems with matching the throat with the driver or not. But it would be interesting to see what the Japanese impulse looks like.
 
jamikl said:
Rather ironic, maybe deliberately, that the example given earlier was Frank Lloyd Wright. A man ahead of his time, highly egotistical, or maybe opinionated about his own value and whilst designing some of the best made many mistakes and was openly castigated for them by architectural people and owners.

Do keep in mind that Frank's and Earl's design methods are worlds apart. Wright was not well educated in science or engineering. His designs, though brilliant, were mostly based on intuition and his own aesthetic sense. When these conflicted with what the engineers were telling him, more often than not he ignored the engineering analysis. More often than not, that didn't work out well. Earl, on the other hand is a competent and accomplished scientist and engineer, and many of the criticisms being directed at his work are based on a less technically sound understanding....
 
MEH said:


Do keep in mind that Frank's and Earl's design methods are worlds apart. Wright was not well educated in science or engineering. His designs, though brilliant, were mostly based on intuition and his own aesthetic sense. When these conflicted with what the engineers were telling him, more often than not he ignored the engineering analysis. More often than not, that didn't work out well. Earl, on the other hand is a competent and accomplished scientist and engineer, and many of the criticisms being directed at his work are based on a less technically sound understanding....
I think ignoring one or the other are two extremes that should consider second thoughts. There are many phenonemas that cannot be fully expressed in general mathematics equations, that is why FEA methods are used to close the gap between measured data and linear equations. Evolution of technology always starts from intuition, then methods to reliably predict results are developed and continuously improved.
 
MEH said:


Do keep in mind that Frank's and Earl's design methods are worlds apart. Wright was not well educated in science or engineering. His designs, though brilliant, were mostly based on intuition and his own aesthetic sense. When these conflicted with what the engineers were telling him, more often than not he ignored the engineering analysis. More often than not, that didn't work out well. Earl, on the other hand is a competent and accomplished scientist and engineer, and many of the criticisms being directed at his work are based on a less technically sound understanding....

Actually Wright was a trained engineer and architect, but he did throw out a lot of classical assumptions of design and fabrication. The fact that he basically invented steel reinforced concrete is no small feat.

Being compared to him, egotistical or not, is an honor. I have always admired FLW and his designs thrill me to this day. I have visted virtually everything that he has done including his own homes and studios. The man was a genious. But what I like best is that he was an unknown until he was about 50. He did his best work in the last 30 years of his life, well after most people retire. That is a true testiment to the mans fortitude.