Welcome! Need support, you got it. Or share your ideas and experiences.
Return to Planar Speaker Asylum
98.26.54.17
In Reply to: RE: Interesting observations posted by josh358 on September 10, 2012 at 15:07:14
Interesting link you refer to at musicanddesign.com. I'm still processing information from 'WBF Discussions with Roger Sanders' you suggested at an earlier time. More horse sense in that thread than most. And has had a good deal to do with why I want to get away from itty, bitty woofer box format at some point. Perhaps the Open Baffle H or W frame might be the practical answer. Or perhaps I'm overlooking one of the obvious solutions in the DW1/DWM speakers but they're not really built to get you to the 20's. And I still would like to implement a stereo sub arrangement.
The upside is that I have lots of time to look at the idea before I make a move. The single 15" Velodyne does a credible job of it most of the time and I'm happy I didn't lose it. It still needs to be retired in favor of something that is an order of maginitude better.
Oops, I apparently have guests so I'll have to continue this later...
Tom Waits, Cooler than you for thirty years.
Follow Ups:
That's quite a thread, isn't it? I had the tab open for something like a month before I'd read all the way through it.
I don't know of anyone who's compared an open baffle dynamic with the DWM's. I'm sure the guys at Magnepan have, though.
One possible issue I can think of is that even in an open baffle arrangement, dynamics are going to sound different, they just have a different aural signature. And since with dynamic woofers you pretty much have to use a sharp crossover that could result in audible discontinuity even with a dipole approach. Not only is the DWM in the same driver family, but it's a low mass driver so can be crossed over with a single pole crossover, or at a higher frequency, as in the approx. 300 Hz crossover that's used with the Minis.
Wendell also points out, rightly I think, that the midbass is more of a limit on the performance of the typical system than the deep bass, and this is something the DWM's can help with. He's had trouble getting that message out despite the fact that many respected acousticians and designers are on record as saying this.
I'm thinking though that if you went with the DWM's you'd still want to use an LF sub. What about keeping the Velodyne for bottom octave duty? Also, the bottom octave may not have modal problems in all rooms. Dipoles I think become less attractive when you're dealing with the bottom octave (actually it seems you have to go down to 14 Hz to get the full spatial benefit), because judging by Linkwitz's results anyway you're going to run out of amp power in a critically damped alignment or Xmax in either a critically damped or underdamped one.
Ya it is. I appreciated the link to that, Josh. And I still need to go back and absorb the discussion on digital audio. Roger is good at presenting information in a way that is useful and not difficult to absorb. I would like to hear one of his systems for myself. I do not doubt that it's very good.
As far as using an OB H-frame like the GR it would seem intuitive that a dipole bass transducer has potential to match with Magnepan's bass panels. The DWM's should be a nice fit specificly because of the uniform driver behavior. Being a bit of a gear hound it's hard to let go of that 20 Hz. placeholder that's supposed represent full bass reproduction. But I am startinng to suspect that it isn't quite the gold standard that I may have thought. It gets trickier when the wavelengths near the dimensions of listening room. And far more difficult to visualize for me. Corret me if I have forgotten but isn't it the 1/4 wavelengths to the wall(s) that starts to play havoc with bass nodes? Wish I can express the concept a little better than that.
I can see the appeal of using a sub/bass driver at the listening position. Perhaps it's a potent solution to the problem but it would seem to imply a DSP type of delay to get the impulse response right. This is part of the hingepoint to the way that I think about the problem. If we had single transducers that could do it all it would make the task all so easy. Dividing up the spectrum between multiple drivers becomes necessary but opens up all the problems that go with that. To my mind if you can't get the impulse response near perfect then there's always going to be tell tales that go with it. I have to think that DSP is the tool to solve much of these problems. I'm trying to delay launching into that for various reasons. A lot of it is trying to make the effort far more focussed with time, money and knowing more about what I'm doing before I go off half cocked with it. And I am thinking/writing out loud again...
I like Rogers reasoning with his speaker system concerning a transmission line bass implementation. It avoids the problem with the sealed box air spring effect. And although it is a type of ported box design it's a little less compromised than the simpler ported tube builds. I think I can place one behind each speaker and use the vertical tube as a absorber/defractor. Perhaps what little if any localization occurrs at those frequencies would play well with the 2.6's if place directly behind them. In an 11x13 room floor space is hard to come by so there is that appeal to the idea. I can try a couple of tubes behind the speakers before I commit to the build to judge if it's a bad idea to start with.
Tom Waits, Cooler than you for thirty years.
Actually, I'm not even sure that 20 Hz is low enough. That figure has been used for years, but it apparently reflects the limitations of the early experimental apparatus rather than the actual low frequency limit of our ears. I've read that response down to 14 Hz is necessary for optimum recreation of ambiance.
So -- nothing against the bottom octave (and a half), it's just that the midbass is more important. Which it is. And it's a very problematic area because of proximity effects. The quarter wavelength you mention would give you a half wavelength delay, which would result in a complete null with the direct sound, so you're spot on, however, this is for omnis -- dipoles cancel at a full wavelength delay since the backwave has inverted polarity, which means the speaker is half a wavelength from the surface.
I think it's a fair bet that dipole woofers will match dipole speakers more closely, though I'm not sure how significant that is unless you're listening way off axis. Line source vs. point source is likely very significant for full height line sources, since in the former case the bass drops off as 1/R and in the latter as 1/R^2. In practice, this means that you can balance them for only one listening distance, which may or may not matter in any given installation. To make matters a little more complex, Maggies aren't quite full height and point source speakers (dipole or not) behave as line sources below the spatial Nyquist frequency, which for an 8' tall room and the woofer on the floor would correspond to a 32' wavelength but with the woofer 4' off the floor a 16' wavelength. So the woofer's transition from point to line occurs within the audible range. (It's actually a bit more complicated than that owing to reconstruction errors.)
I think you're right that DSP would be needed to get a minimum phase solution to that woofer response deviations, though it's possible you could do it with minimum phase analog filters. Kind of a complicated curve, though, and DSP can automate the process. It does add to the complexity but I think you really need DSP for the best bass anyway, even with trapping and careful placement.
I've heard some good things about transmission line bass. I have heard that you can hear the transition in his hybrids, but that's probably unavoidable with any dynamic/planar combination, if nothing else cone breakup sounds different than stretched membrane resonances.
I have no strong opinion about the limits of LF audibility. I'll need to break out the test disc again but I know I can't hear a 20 Hz. sinewave on the Velodyne. I wish I would have thought to try the other LF tracks in the 20's to see what how low I can reasonably hear. Perhaps later today now that its come up. Of course that doesn't mean that the subaudible stuff has nothing to add to the picture.
Good point about backwave cancelation with dipoles. I had forgotten about that. You mentioned previously about a passive bass array. Since I have an untasked set of 1.5's to play with perhaps I could try a temporary mounting at the 1/3 room length points. The idea would be to use the side walls to be an extension of the baffle on one side of each speaker. Keep the current LF crossover point and equalize the 1.5's only and see what that's like. I can't be sure what to expect from that or even if it's a good idea or not. Just something to ponder while I'm thinking about possibilties. I'm pretty sure that I want to use a stereo sub arrangement when the time comes. Hemholtz's GR's should be significantly better than my current set up but I'll never really know till I try it. Looking at the cabinet dimensions it comes in at 13" x 23.5" which is pretty easy to work with. Hmmm... that's attractive. Could always start with one and go from there. If I can only get into it without making myself half nuts in the process. There does seem to be a point where I question my sanity for obsessing over this stuff.
Tom Waits, Cooler than you for thirty years.
I learned about being able to hear below 20 Hz from the Eminent site. Of course, he has a reason to say that since he's selling rotary woofers, but I read about the 14 Hz limit elsewhere so I don't think it's just him, though.
Just another thing I've never had time to research more deeply. And most of us don't have access to subs with that kind of performance, mine certainly won't do it. Maybe I could ask the guy across the street with the boom car. :-)
Your side wall idea will definitely work, it's the same baffle-width-doubling principle they use with the DWM and the Tympani IVa split woofer arrangement. I've tried it with the MMG's. However, you'll end up with very boomy bass because the 1.5's woofers are tuned for use away from the side walls. So you'd have to use EQ to get flat response.
Still, it would be an interesting experiment.
Open baffle GR's would definitely be high on my short list of subwoofer approaches.
The general consensus seems to be that stereo is important in subs for audiophile applications, although I suspect that some of that is room and placement sensitive (based on a paper I read on subwoofer localization and also my personal experience). However, most recorded bass is essentially mono, this was done because you can't cut stereo bass on an LP. So I think the effect may have more to do with higher frequencies that are still being reproduced by the woofer owing to the finite crossover slope. You can hear stuff until it's more than 20 dB down so if you have a crossover of 80 Hz, say, even with a four pole crossover you're going to be hearing some signals up to 160 Hz from the woofer.
I have a test disc with recorded sine waves from 10 hz on up to 25 hz,
Each individual sine wave, starting at 10 hz can be heard easily with
my rotary subwoofer. It is simply not true that one can not hear below
20 hz; ones hearing does not just stop at 20 hz as it has been reported
in many audio articles. The problem is conventional subwoofers are
incapable of reproducing these infra frequencies with authority.
One could use many subwoofers in a infinite baffle arrangement (I have
heard 8 18 inch woofers that are also capable with clean and powerful bass
that is easily heard down to 10 hz.
Below 10 hz I can still feel and hear these infra sounds, but this is where the
rotary is king and low frequency transients are stunning.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: