Yep. The EQ curve looks horrific. It does sound better though. But this is hardly the ideal way to get good sound, IMO. The multiple subs approach would probably work, but space is at a premium and there is no way I will be "allowed" to add more speakers. The bass boxes are pretty big anyway (6 ft3) and when the massive 30" dia horns go on top, the speakers will dominate the room. I think that will make the speakers as big as the refrigerator, putting me firmly in the 'looney audiophile' category 
Last edited:
I also have a space problem. It occurs to me that hanging a woofer from the ceiling above my video screen might work pretty good. I'm thinking about a 3 ft. long generally tubular enclosure, with at least an 8 inch woof in each end, in phase, so their vibrations will cancel out mechanically. For a "flanking" woofer to work, it has to be in a substantially different place in the room, and if you've already got two woofs on the floor, it makes good sense to me to have a 3rd woof up high somewhere.Yep. The EQ curve looks horrific. It does sound better though. But this is hardly the ideal way to get good sound, IMO. The multiple subs approach would probably work, but space is at a premium and there is no way I will be "allowed" to add more speakers. The bass boxes are pretty big anyway (6 ft3) and when the massive 30" dia horns go on top, the speakers will dominate the room. I think that will make the speakers as big as the refrigerator, putting me firmly in the 'looney audiophile' category
Hi,
If you want to do some listening tests with shaped tone bursts, you might want to try a set of low frequency test signal sequences I made about 10 years ago (man, I'm getting old, time passes by faster and faster) : http://casakustik.de/down/testsignale_v0.9.21.zip
(Sorry, no english documentation, but the CD-index pic (see attachment) gives the basic information).
Main intent was, originally, to have a very ease to use test CD (if you had a CD-text feature in the player) to dial in EQ for small live venues at the time I was a (semipro) live mix / club sound engineer, but it is well suited for home use, too.
The sequences are organized as 4 sets of bursts per test frequency (1/24th oct steps), with raised cosine edges, burst length continuously decreasing in each set.
1st set has no polarity flip beetween bursts (left pic),
2nd set has polarity flip after each burst (right pic)
3rd and 4th are like 1st and 2nd, but with the right channel inverted (out-of phase).
Keep volume low (signals are at very high digital level)!
The key is to also listen to the silenced parts in between, how they are smeared, and the overall temporal character (besides the obvious level differences).
It is a pretty annoying test, you might think your speakers and room are total crap, but to our rescue the problems with real music are less pronounces, but still noticable. Try listen to the signals with headphones first.
Have fun!
While this is a MP3-download of a complete CD content, all that is really needed is the two little generator proggies (DOS, 16bit, source code provided) which build a set of 98 .WAV files. A pink or brown noise signal is also assumed to have at hand for the first track.
- Klaus
Attachments
Speaking of Laurie Fincham, he made an appearance on the "Home Theater Geeks" podcast recently which I just got around to watching, and he talked quite a bit about things relevant to this thread including steered line arrays, so for those interested in watching it you'll find it here:
Home Theater Geeks 80 | TWiT.TV
Here's a quick list of time indexes to sections I thought were relevant to this thread, although there is plenty of other interesting stuff in there as well:
21:30 - answering the question "can the room only hurt or can it improve the sound", (me paraphrasing here) he talks about Toole's findings, our ability to listen "through" the room, of room EQ only being valid for "low" frequencies, with the speakers natural response being most important at high frequencies, and doubt about the efficacy of auto room EQ systems.
28:30 - mention of the Quad Electrostatics
29:50 - efficiency claims of horns and horn characteristics
32:00 - application of beam forming arrays to cinema
35:00 - talks about beam forming/steered arrays to control directivity and sound source location
39:00 mention of KEF Uni-Q / Tannoy / KEF blade concentric driver approaches
42:00 mention of Don Keele's CBT array
45:00 mention of Fraunhofer Iosono arrays
48:00 use of left-right crosstalk cancellation with narrowly spaced speakers
52:00 Sound bars
54:30 Ear brain perception and adaption to the sound environment, exercising restraint with application of EQ (avoiding trying to correct every small peak and dip, only correct "what matters")
58:00 Headphone surround and localization
As an aside there are also a lot of other well known names from the audio and video industries interviewed on that podcast, especially in the very early episodes, some of which are audio only. In fact diyaudio's own Nelson Pass was one of the early guests of the show, and listening to that episode was how I discovered diyaudio.com
There is some really interesting stuff there if you dig back through some of the older episodes.
Last edited:
Hi Klaus,
Thinking about this I really like the simplicity of the idea. Can you tell us more about the 15" coax system (model, xo, directivity, max SPL, etc.)?
Markus,
Beyma 15XA38Nd (my current avatar, too), quite a hell of driver in all regards (price too). Not true HiFi of course. It is possible to mount you own horn/WG to the HF driver, discovered that after removing the bug screen (of the cone driver).
Some details (german) here : http://www.aktives-hoeren.de/viewtopic.php?f=6&t=2174
There are some commercial box speakers based on this driver and its smaller 12" cousin, the guy who's using the 12" encouraged me to give the 15" a try.
It was only recently that I found Emerald Physics (via that link) who use a (rather cheap) pro coax in OB for a commercial product getting rave reviews, all in all a quite out-of-the-box product, literally.
I don't know if it will be a keeper, as I'm about to try 15" reflex + 1"CD on elliptic OSWG soon. This will certainly have more "room punch" and close to perfect treble, but the low/mid clarity and freedom of room overload with the nude 15" might be hard to beat.
- Klaus
Beyma 15XA38Nd (my current avatar, too), quite a hell of driver in all regards (price too). Not true HiFi of course. It is possible to mount you own horn/WG to the HF driver, discovered that after removing the bug screen (of the cone driver).
Some details (german) here : http://www.aktives-hoeren.de/viewtopic.php?f=6&t=2174
There are some commercial box speakers based on this driver and its smaller 12" cousin, the guy who's using the 12" encouraged me to give the 15" a try.
It was only recently that I found Emerald Physics (via that link) who use a (rather cheap) pro coax in OB for a commercial product getting rave reviews, all in all a quite out-of-the-box product, literally.
I don't know if it will be a keeper, as I'm about to try 15" reflex + 1"CD on elliptic OSWG soon. This will certainly have more "room punch" and close to perfect treble, but the low/mid clarity and freedom of room overload with the nude 15" might be hard to beat.
- Klaus
Last edited:
Hi Klaus,
tak a look: interesting bare 12'' coaxial (whizzer cone FR) driver experiments with measurements by Radugazon in another diyaudio thread from page 67 - post #669 on:
simple experiment
regards,
graaf
tak a look: interesting bare 12'' coaxial (whizzer cone FR) driver experiments with measurements by Radugazon in another diyaudio thread from page 67 - post #669 on:
simple experiment
regards,
graaf
Klaus, thanks. Did you measure off-axis response? What driver is in the Emerald Physics?
Those are P.Audio drivers and I know them quite well. Cheap(ish) in price, but not in build quality or engineering. I've heard or worked with the 8, 10, 12 and 18" versions. Very dynamic drivers but you need good crossover skills to get them to sound more Hi-Fi than PA.
Thanks for the info, Pano. In no way I meant the P.Audio drivers are bad, when I wrote they are cheap.
I know that they make some excellent drivers over there in Thailand
Same goes for some China / Hong Kong driver makers.
However, as long as I can afford the prices I tend to support european workmanship but I might try one of the cheap far-east coax drivers (there some 12"/15" for <$100 retail).
- Klaus
Last edited:
Yes... they look pretty horrible, as ragged as it gets. On-axis and nearby not being any smoother, that is. I tried a bit to get off-axis nicer but I felt it does not do much better in anything essential (it' like polishing the motor of a fully chopped Harley
). What is important, though, is to let the drivers work coincident (that is, using the proper delay for the LF section, and linphase filters, in my case) to get optimzed summing on-axis and lowest off-axis levels. With a passive/analog XO this is hard work and my hat is off to those who go into the trouble of taming this driver with a passive XO succesfully.- Klaus
Unfortunately that's what I had expected. I've never seen a really good coax with SPL capabilities.
Pano, are the P.Audio's any better off-axis?
Well I said when I finished my new crossovers I'd do the pink noise test again. I seem to have got a much more uniform result than with the old crossover. Pics attached. 1st is the latest in/out of phase measurement (again It's late so I didn't try and optimise). The second pic is a spectrogram measurement taken at the same point. Looks pretty nasty below 500Hz (though I'm not really sure how to interpret it).....
finally the third pic is the previous comparison.
The big peaks in response at 40Hz and 560 Hz are interesting, I'll have to see if they are room nodes. Certainly I don't think that they are the speakers! The 40Hz one is puzzling as the speakers should be about -10db by then.
Tony.
finally the third pic is the previous comparison.
The big peaks in response at 40Hz and 560 Hz are interesting, I'll have to see if they are room nodes. Certainly I don't think that they are the speakers! The 40Hz one is puzzling as the speakers should be about -10db by then.
Tony.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.