One other thing that came out of the listening session with @HeadShake was imaging performance. I had thought the TX system had a slightly better performance in this aspect, and it was one of the first things Headshake commented on when we switched from the Tower Twin-8s to the TX system.
We discussed this for a while, and tried to understand why there would be a difference. We came to a plausible theory: The facetted bevels on the Tower Twin-8s have a lot of sharp edges. This was necessary because I wanted to veneer them. The bevels form a 35 degree angle to the front face, so these are gentle bevels, but still there is a sharp edge. Is it possible that these sharp edges could be producing a diffraction secondary sound source? Not as strong as a 90 degree edge of course, but still enough to slightly, ever so slightly, degrade the imaging performance?
The TX system has no sharp edge on the horizontal, it is gently radiused at 37 mm. The top edge is beveled at 45 degrees, so this would leave two sharp edges, but I suspect that a secondary sound source in the vertical is much less important than in the horizontal. The first sharp edge that the sound encounters on the horizontal is at the back of the cabinet.
I have noticed that several recent designs use tapered bevels that have a rounded edge. An example would be the Taipu Kero, which I believe is a Kimmo Saunisto design). Is this rounding more than just cosmetic? Does it soften the secondary sound sources and enhance the imaging performance?
Thoughts on this theory?
I also included the Power and DI charts for both speakers. Both speakers are +/- 1 dB from 100 Hz to 15 kHz on axis. But the DI curves are different. The TX system has a -4 dB dip between 1k - 2k, while the Tower is much better behaved.
j.
We discussed this for a while, and tried to understand why there would be a difference. We came to a plausible theory: The facetted bevels on the Tower Twin-8s have a lot of sharp edges. This was necessary because I wanted to veneer them. The bevels form a 35 degree angle to the front face, so these are gentle bevels, but still there is a sharp edge. Is it possible that these sharp edges could be producing a diffraction secondary sound source? Not as strong as a 90 degree edge of course, but still enough to slightly, ever so slightly, degrade the imaging performance?
The TX system has no sharp edge on the horizontal, it is gently radiused at 37 mm. The top edge is beveled at 45 degrees, so this would leave two sharp edges, but I suspect that a secondary sound source in the vertical is much less important than in the horizontal. The first sharp edge that the sound encounters on the horizontal is at the back of the cabinet.
I have noticed that several recent designs use tapered bevels that have a rounded edge. An example would be the Taipu Kero, which I believe is a Kimmo Saunisto design). Is this rounding more than just cosmetic? Does it soften the secondary sound sources and enhance the imaging performance?
Thoughts on this theory?
I also included the Power and DI charts for both speakers. Both speakers are +/- 1 dB from 100 Hz to 15 kHz on axis. But the DI curves are different. The TX system has a -4 dB dip between 1k - 2k, while the Tower is much better behaved.
j.
Attachments
Did you see that study facets vs roundovers. It's clear to me that roundovers are superior, the downsides is that it's just very difficult to veneer a large roundover. And besides, not everyone likes the look of a bubbly/rounded shape object.
The DI of the Twin Tower 8 is better. But the TX doesn't seem to be harmed by the directivity mismatch. Very interesting
Here's my speaker- no roundovers or facets or waveguides. Directivity mismatch around 4.5Khz due to hard edge diffraction.
But I don't think a drop @4Khz ever hurt anybody.
On the contrary, If you ever hear a screaming toddler in the shopping mall, whip out your phone can take an audio recording... GAWD it's that awful 4Khz that really pierces the skull...
The DI of the Twin Tower 8 is better. But the TX doesn't seem to be harmed by the directivity mismatch. Very interesting
Here's my speaker- no roundovers or facets or waveguides. Directivity mismatch around 4.5Khz due to hard edge diffraction.
But I don't think a drop @4Khz ever hurt anybody.
On the contrary, If you ever hear a screaming toddler in the shopping mall, whip out your phone can take an audio recording... GAWD it's that awful 4Khz that really pierces the skull...
The tapered bevels do not simulate as causing any real diffraction but it would take quite a lot of mesh resolution and frequency points to see everything at high frequencies.
The intersection points would be the most likely culprit of measured diffraction, the rounded facets smooth those out and present a more ideal transition, apart from difficultly in construction there is no down side to doing so.
I suspect the more likely cause is that the TX system is higher in directivity particularly at high frequencies, this will give more time between the direct sound and reflective surfaces. That really does have an effect on imaging, particularly being able to identify the sound as coming from a particular location.
My takeaway is that it is actually really hard to compare two things that are not the same if you are looking for a definitive conclusion.
It is hard to make sure that effective baffle width and other factors are not having a confounding effect and that you are actually seeing the difference from the edge treatment without something else superimposed on top.
Edit: The measurements stand for what they are, it is the interpretation of them to create some kind of generalised rule where things seem to break down.
Edit: The measurements stand for what they are, it is the interpretation of them to create some kind of generalised rule where things seem to break down.
How good would it be if sound did what we thought and not what it wants 😀
If it’s the right kind it’s quite easy, what type did you try it with?
Daily thought from mainframe / fluid message exchange:
if taking this vaneering issue to first principles its easy to decide what to do. Make priority order which should come first for current project, sound quality or looks? Loudspeakers are full of compromises, trade-offs, and if not guided by the root level priority order then its gonna be a compromised build
Example, if looks is more important than sound quality then no question, hard edges etc. vaneering, and what not to make eye happy. But if sound quality is first priority why even consider vaneering, it doesn't relate to sound in anyway? Alright, consider both important, how to determine which one should it be? Figure out if vaneering is more important than slight reduction in diffraction. Now that you have based an answer to this question, and all questions regarding the build, from first principle which compromise to take, there is no question it was right or wrong and you can live with it and keep going There is no confusion about it, it was conscious decision and you know what you are doing. For project that has first priority in sound quality there is no question about vaneering, its obvious trade-off to take, leave it out. No trade-offs in sound quality leads to best sound quality, its a philosphical thing.
One could also circumvent the whole issue by using a waveguide, no question about vaneering ever gets asked. But its again sound quality over looks trade-off, for some at least. Or perhaps consider using hard wood for the hard-to-vaneer sections so there is no need to.
I've got three root level bins to push trade-offs to: sound quality, looks and cost. Cost includes everything from size to money to time spent, family relations and all Looks contains also all kinds from visual appeal to ego stuff, like brands and makes and models, or preconceptions to avoid something or favor something without exactly knowing why. Taking trade-offs is now rather easy, when there is two things to take compromise on just look which bin they belong to. To maximize sound quality push all trade-offs to the other bins, to cost like size and money, looks is just secondary ever even considered. If there is a choice when both options are trade-off in sound quality its time to take the concept back and rethink it.
And if you expand about this, you'll see most if not all commercial speakers are bound to the looks and cost bins Anyway, just having fun with the hobby is the main goal of course. Being conscious what one is doing, and why, leads to results one can live with.
if taking this vaneering issue to first principles its easy to decide what to do. Make priority order which should come first for current project, sound quality or looks? Loudspeakers are full of compromises, trade-offs, and if not guided by the root level priority order then its gonna be a compromised build
Example, if looks is more important than sound quality then no question, hard edges etc. vaneering, and what not to make eye happy. But if sound quality is first priority why even consider vaneering, it doesn't relate to sound in anyway? Alright, consider both important, how to determine which one should it be? Figure out if vaneering is more important than slight reduction in diffraction. Now that you have based an answer to this question, and all questions regarding the build, from first principle which compromise to take, there is no question it was right or wrong and you can live with it and keep going
There is no confusion about it, it was conscious decision and you know what you are doing. For project that has first priority in sound quality there is no question about vaneering, its obvious trade-off to take, leave it out. No trade-offs in sound quality leads to best sound quality, its a philosphical thing.One could also circumvent the whole issue by using a waveguide, no question about vaneering ever gets asked. But its again sound quality over looks trade-off, for some at least. Or perhaps consider using hard wood for the hard-to-vaneer sections so there is no need to.
I've got three root level bins to push trade-offs to: sound quality, looks and cost. Cost includes everything from size to money to time spent, family relations and all
Looks contains also all kinds from visual appeal to ego stuff, like brands and makes and models, or preconceptions to avoid something or favor something without exactly knowing why. Taking trade-offs is now rather easy, when there is two things to take compromise on just look which bin they belong to. To maximize sound quality push all trade-offs to the other bins, to cost like size and money, looks is just secondary ever even considered. If there is a choice when both options are trade-off in sound quality its time to take the concept back and rethink it.And if you expand about this, you'll see most if not all commercial speakers are bound to the looks and cost bins
Anyway, just having fun with the hobby is the main goal of course. Being conscious what one is doing, and why, leads to results one can live with.
Last edited:
Whilst I agree with you about the looks/sound trade off in this case I don't see there being one. If you use paperback veneer sheets with heatset glue you can literally use an iron to follow the curve and the bigger the roundover the easier it is I would say that it is actually harder to do it the way Jim did on the towers and get a neat edge and match the grain. I have some thicker veneer sheets in Santos Rosewood and Tineo and the Rosewood is difficult to get right on a flat surface without the grain splitting.
I would say that it is actually harder to do it the way Jim did on the towers and get a neat edge and match the grain. I have some thicker veneer sheets in Santos Rosewood and Tineo and the Rosewood is difficult to get right on a flat surface without the grain splitting.
Nice, a win-win Yes, getting deeper into a question can reveal its wasn't a problem at all and was just lack of knowledge. Another example, I had question of active or passive crossover and went for active as it seems to enable stuff that can make for better sound quality while trade-off being mostly complexity, cost. Now it looks like this was false question as well, why not use both!
Yes, getting deeper into a question can reveal its wasn't a problem at all and was just lack of knowledge. Another example, I had question of active or passive crossover and went for active as it seems to enable stuff that can make for better sound quality while trade-off being mostly complexity, cost. Now it looks like this was false question as well, why not use both!
Last edited:
Jim, I want to agree with fluid here. The comparative tests which have been discussed before, those done by Alexander Heißmann and the ones that mabat did when he introduced the enclosure feature to Ath, indicated that your compromise of a facetted solution is very competitive and effective. There might be various other factors and the most likely would be that both of you where hearing effects related to the different sound signature that come from reflections.
I like that theory, Fluid. Both systems create a convincing 3D image, but it is just slightly better with the TX system. However, when I pull the towers out into the room by an additional foot, the imaging improves a bit, and "a bit" is all it takes. When I do this, I am getting a bit more direct sound and a bit less early reflections. This all makes sense in the context of your theory. Unfortunately, this location is not friendly to the upper bass.
So true. We can often subjectively assess that something is not quite right, but attributing that to a specific cause is very difficult.
My wife lets me do whatever I want with speakers. They can be any size, any shape, they can be positioned in the most optimum location. If I wanted a 7 ft line array, no problem. I can have as many as I want scattered about the house... but they have to look good, they have to look like they belong. A bare plywood prototype smeared with glue and bondo won't make it out of the workshop, it would be "dead on arrival"
Thank you to everyone for your thoughts.... j.
Haha I know, I know. Stupid little sound waves
I cant find heat lock down here, so all my veneering is done with contact cement. I'll leave it at that. To be fair, my comment was a bit incorrect, should have said "its a shame its so much easier to veneer a flat surface"
PSA works too with a roller, I don't know about NZ but Amazon does some pretty good deals on basic PSA veneer with free shipping.Haha I know, I know. Stupid little sound waves
I cant find heat lock down here, so all my veneering is done with contact cement. I'll leave it at that. To be fair, my comment was a bit incorrect, should have said "its a shame its so much easier to veneer a flat surface"
https://www.amazon.com.au/Edge-Supp...efix=walnut+venee,home-improvement,303&sr=1-8
The alternative is to make your own on plain paper backed stuff by using a paint roller or brush to paint on PVA glue, let it dry then iron it on. Home made heatlock. Works pretty well for me but I've never done a huge area of it.
With veneer, we just have to keep in mind the limitations. It bends easily across its grain. It bends poorly with its grain. Since it is inelastic, it cannot be applied over a compound curve, so spherical sections are not doable.
For my next project I will probably use a solid baffle so I can freely profile it and maintain good aesthetics. I am thinking about a new mid-tweeter head unit, using a Bliesma T-34B and a Purifi PTT6.5M04-NFA-01. I want to go further in the direction of diffraction control, and take steps to minimize diffraction on the sides and rear of the cabinet. I have not decided if I want to use a waveguide yet. These are just musings at this point.
For my next project I will probably use a solid baffle so I can freely profile it and maintain good aesthetics. I am thinking about a new mid-tweeter head unit, using a Bliesma T-34B and a Purifi PTT6.5M04-NFA-01. I want to go further in the direction of diffraction control, and take steps to minimize diffraction on the sides and rear of the cabinet. I have not decided if I want to use a waveguide yet. These are just musings at this point.
Just to chime in.
It was a pleasure to hear two complete systems. Thanks for having me over and for listening to my music and pontifications.
Your summary is spot on (I was going to write something... sorry I am a slow poke). Both sounded great! Your TX design is pretty darn invisible to my ears.
The towers are darn nice looking. They look slim in person. I think champagne is the only drink that should be consumed in their presence. I kept on envisioning them at weddings and parties with ice sculptures. The Towers would elevate any event.
The center image of the towers was very wide vs the tight spot on the TX. It made me wonder if the close spacing is a good thing or if something else was going on. I preferred the tighter spot of the TX.
It was a pleasure to hear two complete systems. Thanks for having me over and for listening to my music and pontifications.
Your summary is spot on (I was going to write something... sorry I am a slow poke). Both sounded great! Your TX design is pretty darn invisible to my ears.
The towers are darn nice looking. They look slim in person. I think champagne is the only drink that should be consumed in their presence. I kept on envisioning them at weddings and parties with ice sculptures. The Towers would elevate any event.
The center image of the towers was very wide vs the tight spot on the TX. It made me wonder if the close spacing is a good thing or if something else was going on. I preferred the tighter spot of the TX.
The wider image that HeadShake reports once more suggests that directivity properties where critical in the perception of the systems. Jim, would you like to export the horizontal sonograms of both systems once more, with the span compressed to 30 dB, and in vCad options the ratio for the graphs set to 34 dB instead of 25? Obviously, DI index tells the story too, but anyway, just for fun.
You could always 3D print an enclosure. The technology is pretty nice these days and print quality is much higher. Even just a front baffle with minimal edges has been my current fantasy.
