Scottmoose said:
Hi,
where the worst ripple occurs is dependent on the size of the circular baffle. You may want to look into Murphy's article on diffraction loss - he's showing the Olson graphs for spheric, cylindric (round baffle) and cubic (square baffle) enclosures. Round and square baffles with centered drivers are clearly the worst case from an accoustical point of view.
I.E for a round baffle with 12" diameter (ca. 30 cm) the first deep ripple will occure at ca, 2000 Hz, a very sensitive region. But then, 6.5" fullrange speaker will already be beaming in the 2000 Hz area, so does it even "see" the baffle?
So, it would be really interessting to know more facts about the graph Nelson Pass has posted - driver and baffle size at least.
best, LC
Tolvans Edge allows to model diffration for different baffle sizes and shapes.
The Edge is a program I've used in the past, but not enough as I've never bothered with false baffles, and the VSC of regular enclosures is easy enough to predict. Something I've been meaning to look at for a while though, but time is the problem. It'll go on the agenda, along with K-Slots etc.
lovechild said:
Murphy does not thou show a number of very significant plots from Olson page 23. The full page again:
Of particular note is the hemisphere which shows the importance of the trailing edge, and the truncated pyramid with trailing edges which comes a close second to the sphere and a way easier shape for someone with flat sheets of material and a saw to build.
dave
Attached are the dimensions of the baffles Olson used above... by adding a scaling factor to the results above you can get an idea of what your baffle will do. ie in the case of the latest build of the Frugel-Horn we should have something similar to 1.10 J but with all the values on the x-axis doubled. The same speaker with no supra-baffle will be more like 1.10 I but with the x-axis trabled. As pointed out, the saving grace is that our full-ranges will be beaming at least somewhat at these frequencies so that will ameriolate things somewhat.
dave
dave
Attachments
I find this discussion of the proper shape of a false baffle very interesting. I was first introduced to the concept with the various circular false baffles applied to TQWT's. These circular plates are supposed to improve wave launch and other claims. I've not run any experiments myself or seen the results of experiments of others, so who knows? I've always thought that they are too thin, usually 3/4", to help with edge defraction, and besides, by the time one uses a driver of any size, say 6" and larger, the driver is beaming sufficiently at the frequencies where edge defraction is important, it may well be a moot point. I use a false baffle in my designs, but it is primarily cosmetic. Since it is smaller than the cabinet width, the only real effect is to increase the edge radius.
From the diagrams Dave presented, the best practical false baffle shape is a truncated pyramid with the chamfer on the order of 6". But wait a minute! If we extend the bottom edge to the floor, there will be no defraction from that edge. Then, doesn't this say that the best edge shape is simply to take a normal rectangular box and put a large chamfer (or round-over) on the edges? And then, aren't we back to the age-old discussion of how much round-over we need on the edge of a cabinet? IIRC, the point of diminishing returns begins around 3" or so.
Bob
From the diagrams Dave presented, the best practical false baffle shape is a truncated pyramid with the chamfer on the order of 6". But wait a minute! If we extend the bottom edge to the floor, there will be no defraction from that edge. Then, doesn't this say that the best edge shape is simply to take a normal rectangular box and put a large chamfer (or round-over) on the edges? And then, aren't we back to the age-old discussion of how much round-over we need on the edge of a cabinet? IIRC, the point of diminishing returns begins around 3" or so.
Bob
Bob Brines said:
Too some extent... in the case of the Frugel-Horn & the RonHorns with supraBaffles part of the idea thou is to push the bafflestep down to where additional gain from the horn counters the BS eliminating the need for any electrical compensation.
This also does bring up the point that a BS Circuit will lift the flattened lower portion of the curve to the same level as the falltened HF portion, but it does nothing to the ripple which is just shifted upward in level by the filer.
dave
Wow,
thanks Dave for posting all these graphs! I knew there's more but never seen most of them.
I admit, I have a problem (prejudice?) with false-or supra baffles. I know enought to see they make sense for pushing the BS to where the horn gain fills up. But isn't that kind of like fighting the devil with the help belzebub?
I don't know actually ... maybe a series of systematic messurements could clearify. As a friend of mine sayed: "Music is art, audio is engineering".
best, LC
thanks Dave for posting all these graphs! I knew there's more but never seen most of them.
I admit, I have a problem (prejudice?) with false-or supra baffles. I know enought to see they make sense for pushing the BS to where the horn gain fills up. But isn't that kind of like fighting the devil with the help belzebub?
I don't know actually ... maybe a series of systematic messurements could clearify. As a friend of mine sayed: "Music is art, audio is engineering".
best, LC
lovechild said:
But making the box acceptably pretty is industrial art... a designer friend of mine, said after seeing the current Fugel-Horn that it was the most striking speaker he had ever seen and that we should tart it up a bit, pop it into a New York gallery and charge $6500 as a piece of functional art. That got my brain turning....
dave
This thread split off from:
Terry Cain's BIB -why does it work and does anyone have those Fostex Craft Handbooks?
http://www.diyaudio.com/forums/showthread.php?s=&threadid=66173
The next 2 posts have been editing to reflect their start of this thread... the complete original posts are here:
http://www.diyaudio.com/forums/showthread.php?postid=1004866#post1004866
dave
Terry Cain's BIB -why does it work and does anyone have those Fostex Craft Handbooks?
http://www.diyaudio.com/forums/showthread.php?s=&threadid=66173
The next 2 posts have been editing to reflect their start of this thread... the complete original posts are here:
http://www.diyaudio.com/forums/showthread.php?postid=1004866#post1004866
dave
lovechild said:
If not done with care it well could... for instance a wider baffle will push any of the baffle-step ripple lower in frequency, where beaming may not minimize it -- ie a thin circular disk as supraBaffle. It also has to be looked at as an opportunity -- it gives room for a large fall-way (round-over or bevel) that may not be possible with a cabinet where the edges of the driver are already close to the cab edges. It also opens up huge opportunity for artistic/cosmetic treatment.
An example of widly artistic is here http://www.diyaudio.com/forums/showthread.php?postid=981600#post981600
One would look at these and say they might be of a nature to introduce more problems than they fix, but Josh says the speaker sounds better with them ... no wallflowers his build (pun intended). One could guess that they may be creating such a crowd of ripples that they average out -- much like NXTs do to get flat FR out of a bunch of resonant modes.
An externally hosted image should be here but it was not working when we last tested it.
What is important is to get the pluses outweighing the negatives.
dave
McFaBs said:
Greets!
The baffle's effective area dictates its F3 frequency, so its size is a function of the driver's output WRT the BLH's in-room response, ergo scaling one from another design is an exercise in futility since the driver's mass corner can be considerably different.
Close, I was referring to the driver baffle's inner surface. IOW if you space the driver > ~3.4" away from the inside of the cab's baffle, then you run the risk of creating waveguide (WG) mouth reflections between it and its 'throat' (St) at WLs long enough to cause audible modulation of the driver if not well damped, something we're trying to avoid as much as practical to maximize the horn's LF/midbass gain.
When it comes to maximizing any high performance system, be it a race motor, horn baffle, jet wing profile, whatever, 'GOD' is in the details, though considering how frustrating it can be it's probably more appropriate to give the 'DEVIL' his due.
GM
Scottmoose said:
That's my point - there's something going on that makes this
a better-than-acceptable sound in spite of the curve. If you
take the same driver and mount it in a large, irregular, non
frisbee baffle, it doesn't sound the same (obviously) but it
also doesn't sound as good.
Offhand, I have no clear idea why that might be.
With regard to ZV9 or F3's, they are pretty happy driving these
speakers.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.