Sharp edges are done by people who do not understand physics yet, or put cost and aesthetics over sound. It is very hard to put veneer over round edges and grill cloths without a refractive or reflective edge takes some creativity. My experience is flat bevels don't work anywhere near as well as a 3/4 inch radius on all 12 edges of a box. Yes, all 12 seemed to help. Surprised me.
I have not found it to matter with subs. 1/2 inch over sharp is dramatic. 3/4 works better than 1/2 inch. It seems to be enough as I can get response within half a dB of modeled at 1 meter with smooth off-axis fall-off. Voicing is just fine, and imaging wider than the speakers with some height and good depth. Would a larger radius do better? I will find out when I save up for a inch radius bit. IT will start to push the width of the baffle a bit, so that means other factors.
Don't take my word for it. Just do a simple test. Build some simple prototypes with and without a big radius. If your ears are healthy the results will be clear. Expect to do a lot of work to bring the crossovers to where the voicing is at all close. Maybe an active crossover would make the test go quicker.
I have not found it to matter with subs. 1/2 inch over sharp is dramatic. 3/4 works better than 1/2 inch. It seems to be enough as I can get response within half a dB of modeled at 1 meter with smooth off-axis fall-off. Voicing is just fine, and imaging wider than the speakers with some height and good depth. Would a larger radius do better? I will find out when I save up for a inch radius bit. IT will start to push the width of the baffle a bit, so that means other factors.
Don't take my word for it. Just do a simple test. Build some simple prototypes with and without a big radius. If your ears are healthy the results will be clear. Expect to do a lot of work to bring the crossovers to where the voicing is at all close. Maybe an active crossover would make the test go quicker.
Check out this monster!
C1192 1-1/4"r Roundover Bit, 1/2" Shank
I have a 1" that I used on an omni.......... I'm just about finished with it. I will post a thread about it, with pictures.
im with Charlie on this. I dont doubt edge diffraction occurs, but in sims with boxsim (fine i.e. 256 or 512points) it doesnt have as much impact as i expected. A 2cm mitre helped a little but not as much as offsetting the tweeter. I found it puzzling to say the least. I considered that the resolution may not be great enough, but with 25 or 50 points per octave im doubtful of that too. I do however believe that an 'optimum' chamfer size can be found from simulation, where the ripples are benign, and that it is not necessarily 'bigger is better.
Edge diffraction may not be all that audible, but it may be audible enough to produce a cue that sound is coming from a box.
Just thinking here.....
Thanks Melo. It will be on order nest payday. Doug mentioned it too. The other ones I found were well over $100! This month's budget was a new set of eye glasses which cost more than my last speaker build. I still want to order a set with yellow tint for night driving. Us old farts need to knock down that blue glare from HID lights so we don't run into you.
I'm always here, lurking, ready to pounce, but I do have more projects than speakers. I have most of what I need to build an amp I designed (my first try). With a 50 year old house, there is always something. In my spare time, I just picked up a semi-antique Chinese style dining set that needs some restoration. Should be a busy winter, not that I will cut anyone here any slack.
I'm always here, lurking, ready to pounce, but I do have more projects than speakers. I have most of what I need to build an amp I designed (my first try). With a 50 year old house, there is always something. In my spare time, I just picked up a semi-antique Chinese style dining set that needs some restoration. Should be a busy winter, not that I will cut anyone here any slack.
i think that a narrow angle resonance is more likely, but this is just loosely thinking. I visualise the difference in the front lobe radiation with a narrow and wide baffle. Whilst not the only factor, i imagine that wider baffles maintain a narrower lobe to a lower freq, ignoring the other factors, before diffraction splays out the radiation. This would jccount for (in my minds eye) the impression that the sound sticks to the baffle. In narrow baffles, the same effect wasnt nearly as apparent to me in the midrange, rather the emphasis seems shifted to the tweeter instead. This subjectively gave me the impression of more diffuse recessed mids and more forward treble, in contrast to a wide baffle where it felt like everything was more recessed. Of course this is just my opinion and attempt to explain my experiences.
if I remember, someone said that when they made ball enclosures (like the top of a b&w 802) that the speakers seemed to vanish. But a ball would have a huge peak due to the standing waves !!!!!!!!!!! An egg would work well.
I think felt work better than a 1/2" round over.
True Audio TechTopics: Diffraction Loss
or something like this
Fujitsu Ten Eclipse TD712z loudspeaker, $7k/pair 4.7" drivers (ouch)
Fujitsu Ten Eclipse TD712z loudspeaker | Stereophile.com
Then again, that would be a minimal baffle, lol
Norman
I think felt work better than a 1/2" round over.
True Audio TechTopics: Diffraction Loss
or something like this
Fujitsu Ten Eclipse TD712z loudspeaker, $7k/pair 4.7" drivers (ouch)
Fujitsu Ten Eclipse TD712z loudspeaker | Stereophile.com
Then again, that would be a minimal baffle, lol
Norman
Yeah I love grizzly bits! I used to buy freud bits, but I've realized that they all wear out just as fast.
I wear blu blockers on my motorcycle, they make the world look like the aftermath of nuclear fallout, oh and you can't see yellow lights!
me too, had it for years, never used it
it does work, even mounted on back and sides have effect
and it works ok without
Those are the darker brown lenses. I have had them too ( prescription) and found they are great for glare and I had no problem with yellow lights. I have yellow clip-ons for night and they are fantastic. If you can't see yellow lights well enough, maybe you should have your color perception tested.
What needs to be understood is the blue end causes the glare and as you get older, your eyes can't adjust fast enough. It is a big problem to those of us over 50. At night, you see more mono-chromatic anyway, so just being yellow tinted gives no glare and better contrast.
Back to speakers.
Felt is handy but it can cause too much falloff in off-axis. So, like everything else, it is a tool that can be used well, or used not so well. Careful positioning, on axis, very effective! I have wondered what an OB with thick felt would sound like. Long time reflections with few short reflections. McMaster Carr still has half inch. Hmmmm.
Harry Olson results
The discussion on this thread suggests to me that many are unaware of the experimental results obtained by Harry Olson, though there are selective references to some of them. He built a selection of enclosure shapes and measured the frequency response ripples caused by the various edge configurations. The results are pretty compelling (to me, anyway).
The discussion on this thread suggests to me that many are unaware of the experimental results obtained by Harry Olson, though there are selective references to some of them. He built a selection of enclosure shapes and measured the frequency response ripples caused by the various edge configurations. The results are pretty compelling (to me, anyway).
Many of use are well aware, which is why it is so much a mystery why so many continue to build sharp edge boxes. (The OP's point) After all, was Olson's original work not published some time in the 30's? Theories: #1 Deaf, so they don't care, #2 Aesthetics is more important than acoustics, #3 They are too cheap, #4 They are stupid, #5 They think we are stupid.
These papers, along with some commentary are also on the True Audio site as has been mentioned at least by me a dozen times in this forum.
I know there were newer studies that may have addressed the radius and other baffle shapes but I have no direct reference to them. One by him, and some by others. Did Geedes publish one?
Before everyone runs out and builds a sphere for a box, I will say, I have NEVER seen one that deals with the driver to baffle surface transition without a sharp edge totally defeating the purpose. Someday I may build one.
These papers, along with some commentary are also on the True Audio site as has been mentioned at least by me a dozen times in this forum.
I know there were newer studies that may have addressed the radius and other baffle shapes but I have no direct reference to them. One by him, and some by others. Did Geedes publish one?
Before everyone runs out and builds a sphere for a box, I will say, I have NEVER seen one that deals with the driver to baffle surface transition without a sharp edge totally defeating the purpose. Someday I may build one.
Even with those who know we need big radius or big chamfers on the edges; simple lack of money is ( I would say ) the main reason. Very few of us build boxes with walls and front baffles thick enough to allow a 100mm 45/60 degree edge.
Too be honest tho, except for critical listening in perfect environments; edge effects are not the most noticeable artifacts, the room after all dominates
Too be honest tho, except for critical listening in perfect environments; edge effects are not the most noticeable artifacts, the room after all dominates
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.