The card and epoxy still isn't stiff enough to produce stronger LF to drive The room but is ok ish for near field listening.
Maybe the epoxy will go harder over time ? But I suspect not.
I will have to do some listening with the panel rolling off at about 150hz or so to assess the sound in room , from my now crowded cupboard !
Steve.
Maybe the epoxy will go harder over time ? But I suspect not.
I will have to do some listening with the panel rolling off at about 150hz or so to assess the sound in room , from my now crowded cupboard !
Steve.
First measurements on my AV DMLs with dual center exciters
See attached screen shot.
Followers of my recent posts will remember, I've begun experimenting with two exciters on each panel. One exciter produces the respective left or right channel and the other produces center channel information (a "dual center channel" as per the Magnepan idea I discussed earlier).
The upper curve (darker blue) shows (on each channel) th left or right feed + dual center feed configuration.
The lower curve (teal) shows each of the pair of exciters playing full range (no dedicated center channel information). Each exciter has a dedicated amplification channel.
The shape of the curves looks pretty similar, but that would make sense given that center channel often just mixes the stereo channels and that (if memory serves) REW just plays the same thing out of each speaker when it does a sweep.
I can't account for the difference in dBs between the curves. All the REW settings remained the same during the measurements. The microphone (UMIK-1) stayed in the same place too.
Looks like everything between 90Hz and 10KHz falls with a 15 dB range.
Listening to actual music, both configurations sound compelling and engage me with music. They both have great soundstage and "air".
I might prefer the dual center configuration, but recognize I might have some confirmation bias ;-)
That said, I won't listen to music on this AV system. I've begun experimenting with this dual center channel approach to better hear dialog and other movie center channel information.
I've done nothing yet to address room acoustics.
I do have a miniDSP SHD that I can plug in and use to develop some room tuning with DIRAC Live that I can then manually implement in the AV DSP.
Any thoughts and guidance appreciated.
-- Andreas
P.S. I have four Eminece OB-A15neo: 15” Neodymium Woofers (from an old PureAudioProject build I did some years ago) sitting in boxes. I've thought to repurpose them (maybe just a pair as subwoofers for this AV system. Any suggestions appreciated.
-- A
See attached screen shot.
Followers of my recent posts will remember, I've begun experimenting with two exciters on each panel. One exciter produces the respective left or right channel and the other produces center channel information (a "dual center channel" as per the Magnepan idea I discussed earlier).
The upper curve (darker blue) shows (on each channel) th left or right feed + dual center feed configuration.
The lower curve (teal) shows each of the pair of exciters playing full range (no dedicated center channel information). Each exciter has a dedicated amplification channel.
The shape of the curves looks pretty similar, but that would make sense given that center channel often just mixes the stereo channels and that (if memory serves) REW just plays the same thing out of each speaker when it does a sweep.
I can't account for the difference in dBs between the curves. All the REW settings remained the same during the measurements. The microphone (UMIK-1) stayed in the same place too.
Looks like everything between 90Hz and 10KHz falls with a 15 dB range.
Listening to actual music, both configurations sound compelling and engage me with music. They both have great soundstage and "air".
I might prefer the dual center configuration, but recognize I might have some confirmation bias ;-)
That said, I won't listen to music on this AV system. I've begun experimenting with this dual center channel approach to better hear dialog and other movie center channel information.
I've done nothing yet to address room acoustics.
I do have a miniDSP SHD that I can plug in and use to develop some room tuning with DIRAC Live that I can then manually implement in the AV DSP.
Any thoughts and guidance appreciated.
-- Andreas
P.S. I have four Eminece OB-A15neo: 15” Neodymium Woofers (from an old PureAudioProject build I did some years ago) sitting in boxes. I've thought to repurpose them (maybe just a pair as subwoofers for this AV system. Any suggestions appreciated.
-- A
Attachments
Steve --
UMIK-1 (USB microphone) to PC
PC to miniDSP SHD by USB.
I've currently configured the SHD with:
Each of the 4 amps connects to an exciter.
I take measurements with an REW sweep.
This configuration won't do Dolby ATMOS, but it does attempt to apply some of Hafler's early thinking on producing a synthetic center channel for voice.
Lot's of video recordings still synthesize center channels information in the same way.
Surround speakers (those literally on the side walls, back walls, and even ceilings these days) often just delay the front stereo information rather than have separate recorded channels.
All this processing makes me long for the days of direct to disk recordings ;-(
As I proceed, I expect to tweak the center channel outputs, e.g., using highpass and lowpass filters to limit the center signal to typical human voice frequencies.
I can set up to 4 preset configurations on the SHD and switch between them with the remote.
I can certainly envision different center channel settings for:
You get the idea.
I've also begun exploring putting together a Home Theater PC (HTPC) that could support Dolby ATMOS, or whatever the current surround thing used.
I haven't found an an obvious or simple way forward for a DIYer. The trick to do this:
Alternatively, I could spring for a AV processor, but all of them have far more functionality than I really need or want and cost silly amounts of money in my opinion.
(I know I could spend a few hundred dollars on a cheap AV receiver, but that doesn't seem like much fun).
-- Andreas
UMIK-1 (USB microphone) to PC
PC to miniDSP SHD by USB.
I've currently configured the SHD with:
- Left & Right stereo channels &
- A mono mix as the two center channels
Each of the 4 amps connects to an exciter.
I take measurements with an REW sweep.
This configuration won't do Dolby ATMOS, but it does attempt to apply some of Hafler's early thinking on producing a synthetic center channel for voice.
Lot's of video recordings still synthesize center channels information in the same way.
Surround speakers (those literally on the side walls, back walls, and even ceilings these days) often just delay the front stereo information rather than have separate recorded channels.
All this processing makes me long for the days of direct to disk recordings ;-(
As I proceed, I expect to tweak the center channel outputs, e.g., using highpass and lowpass filters to limit the center signal to typical human voice frequencies.
I can set up to 4 preset configurations on the SHD and switch between them with the remote.
I can certainly envision different center channel settings for:
- dialog focused films or shows and
- sports and/or blockbuster movies (I don't watch much of either my my adult kids do when they visit)
You get the idea.
I've also begun exploring putting together a Home Theater PC (HTPC) that could support Dolby ATMOS, or whatever the current surround thing used.
I haven't found an an obvious or simple way forward for a DIYer. The trick to do this:
- Build a "quiet" PC (e.g., heat sinks rather than fans - they look like amps) and
- Find a PC sound card that supports Dolby ATMOS.
Alternatively, I could spring for a AV processor, but all of them have far more functionality than I really need or want and cost silly amounts of money in my opinion.
(I know I could spend a few hundred dollars on a cheap AV receiver, but that doesn't seem like much fun).
-- Andreas
Fanless PC's are slick.
NT-TX2000 Fanless Media Case with Heatpipe CPU Cooler
This one is very cool looking and compact:
Streacom FC8 ALPHA Series Fanless Home Theatre Aluminium Chassis
NT-TX2000 Fanless Media Case with Heatpipe CPU Cooler
This one is very cool looking and compact:
Streacom FC8 ALPHA Series Fanless Home Theatre Aluminium Chassis
Last edited:
aagas.
It depends on how the signals are processed.
Is the processor in the computer or part of the minidsp ?
Does it process the all 4 channels of does it just add a third hafler channel ?
It would seem something is going on as you are applying a mono signal to all channels I presume ?
Steve.
It depends on how the signals are processed.
Is the processor in the computer or part of the minidsp ?
Does it process the all 4 channels of does it just add a third hafler channel ?
It would seem something is going on as you are applying a mono signal to all channels I presume ?
Steve.
Steve -
I currently have optical from my smart TV to the miniDSP SHD, which then:
See above.
To generate the sweep, REW either sends either the left or right signal to both speakers or a synthesized left + right signal.
- Andreas
I currently have optical from my smart TV to the miniDSP SHD, which then:
- Extracts the left and right stereo signals from the digital stream and
- Synthesizes the "Hafler" center channel signals.
See above.
To generate the sweep, REW either sends either the left or right signal to both speakers or a synthesized left + right signal.
- Andreas
Steve -
I currently have optical from my smart TV to the miniDSP SHD, which then:
See above.
To generate the sweep, REW either sends either the left or right signal to both speakers or a synthesized left + right signal.
- Andreas
I currently have optical from my smart TV to the miniDSP SHD, which then:
- Extracts the left and right stereo signals from the digital stream and
- Synthesizes the "Hafler" center channel signals.
See above.
To generate the sweep, REW either sends either the left or right signal to both speakers or a synthesized left + right signal.
- Andreas
Steve,
In reverse order...
For clarity, I don't have any "surround channels" -- no speakers to the sides, back, or above the listeners. No subwoofers yet either.
I only have the pair of plexiglass panels, with 2 exciters on each producing (well intending to produce):
Does this "...improve the intelligibility on the center channel..."?
It varies greatly vis-a-vis source material.
Dialog
On the whole the dual center approach just better engages me in the material.
Movie soundtracks
That said, I still think I can optimize the center output, possibly:
As I've mentioned in a prior post, I could replace my miniDSP SHD with a dedicated AV processor. The sophisticated ones should seamlessly adapt to different source material. Some high end AV processors have built in DSP. Some even have Dirac Live, but such devices seem to start at $6,000 and can go up much more than that.
They additionally, have far more functionality than I really need. I don't anticipate ever needing multi-room sound or 16 channels.
My ideal solution now... miniDSP would offer a firmware|software upgrade to the SHD, which enabled one to apply Dolby ATMOS and the like, but this all goes to them "licensing" the Dolby ATMOS decoder.
I believe the availability of these decoders in cheap mass-market AV receivers goes to volume licensing pricing.
Lower volume "high-end" AV processors probably have to pay far more for the decoder licensing.
Short of that, I hack what I currently have or try to build an AV HTPC (but it seems the decoder licensing issue affects this too).
Any suggestion on ways forward appreciated.
- Andreas
In reverse order...
For clarity, I don't have any "surround channels" -- no speakers to the sides, back, or above the listeners. No subwoofers yet either.
I only have the pair of plexiglass panels, with 2 exciters on each producing (well intending to produce):
Left stereo - Center | Center - Right stereo
Does this "...improve the intelligibility on the center channel..."?
It varies greatly vis-a-vis source material.
Dialog
To subjectively evaluate dialog I play old episodes of Aaron Sorkin's, The West Wing or The News Room. I don't know if any shows regularly packed more words into any 1 hour episode. They also provide a good test, because the dialog often occurs between two, three, or four characters, while they walk through noisy environments.
For me, raised in the US, British and Australian TV dramas also provide a good test. Dialog driven and I typically have to work a bit to "catch" the accents.
In all of these cases, the center channel most certainly makes this information more intelligible.
Prior to setting this up, I would turn on the subtitles when I watched these kinds of shows.
I don't need to do this anymore (or least not nearly as much).
The added center doesn't dramatically change the overall experience, just (in this kind of material) clearer dialog.
Also of note: when the grand The West Wing theme song kicks-in or the show has Yo-Yo Ma playing Bach cello sonatas or the soundtrack has Jeff Buckley singing Leonard Cohen's Hallelujah I don't feel I miss anything in the full rich stereo reproduction.
For me, raised in the US, British and Australian TV dramas also provide a good test. Dialog driven and I typically have to work a bit to "catch" the accents.
In all of these cases, the center channel most certainly makes this information more intelligible.
Prior to setting this up, I would turn on the subtitles when I watched these kinds of shows.
I don't need to do this anymore (or least not nearly as much).
The added center doesn't dramatically change the overall experience, just (in this kind of material) clearer dialog.
Also of note: when the grand The West Wing theme song kicks-in or the show has Yo-Yo Ma playing Bach cello sonatas or the soundtrack has Jeff Buckley singing Leonard Cohen's Hallelujah I don't feel I miss anything in the full rich stereo reproduction.
On the whole the dual center approach just better engages me in the material.
Movie soundtracks
I'd include in this contemporary high-end TV series such as AppleTV's Foundation.
I think the quality of my current system depends on the sound production quality of these soundtracks.
The quality seems to vary enormously. As I understand it, many of these films and shows get recorded in stereo, then all the center and surround gets synthesized in sound production.
Some (maybe THX) may actually have multi channel recording way beyond stereo.
I'd welcome "test" recommendations to try.
I've tried watching and listening to AppleTV's Foundation series, with the center channel off vs on.
I can't discern any difference (once I've adjusted volume) in the grand sounds from the music soundtrack or action scenes.
I currently prefer having the center channel on as it does provide better some better clarity with respect to dialog.
I think the quality of my current system depends on the sound production quality of these soundtracks.
The quality seems to vary enormously. As I understand it, many of these films and shows get recorded in stereo, then all the center and surround gets synthesized in sound production.
Some (maybe THX) may actually have multi channel recording way beyond stereo.
I'd welcome "test" recommendations to try.
I've tried watching and listening to AppleTV's Foundation series, with the center channel off vs on.
I can't discern any difference (once I've adjusted volume) in the grand sounds from the music soundtrack or action scenes.
I currently prefer having the center channel on as it does provide better some better clarity with respect to dialog.
That said, I still think I can optimize the center output, possibly:
- Bracketing the frequency range to typical human voice range?
- Placing a bit of acoustic foam within the bookcase behind the hanging DMLs?
- Placing something to diffuse rather than dampen any of the back wave output?
- Running Dirac Live?
As I've mentioned in a prior post, I could replace my miniDSP SHD with a dedicated AV processor. The sophisticated ones should seamlessly adapt to different source material. Some high end AV processors have built in DSP. Some even have Dirac Live, but such devices seem to start at $6,000 and can go up much more than that.
They additionally, have far more functionality than I really need. I don't anticipate ever needing multi-room sound or 16 channels.
My ideal solution now... miniDSP would offer a firmware|software upgrade to the SHD, which enabled one to apply Dolby ATMOS and the like, but this all goes to them "licensing" the Dolby ATMOS decoder.
I believe the availability of these decoders in cheap mass-market AV receivers goes to volume licensing pricing.
Lower volume "high-end" AV processors probably have to pay far more for the decoder licensing.
Short of that, I hack what I currently have or try to build an AV HTPC (but it seems the decoder licensing issue affects this too).
Any suggestion on ways forward appreciated.
- Andreas
aagas.
When I was testing my card panels on my surround sound in my front room ,I turned the centre speaker off so that I could hear the panels in the front only.
It takes a bit of getting used to because sounds sound so realistic.
On one film I was watching on tv,can't remember the film,there was a romantic encounter on a busy street,it was a very tense moment.
All of a sudden someone started to play a cello ,now normally this would have been floating around in the background.
But it suddenly sounded as if someone was sitting next to them playing this very moving cello , as if the camera could pan to the right and you would see this person playing,no more than a foot of so away,you could hear the bow rasping on the strings
It sounded fantastic, but it was a little distracting, probably not what the director was going for ?
I can't remember the film but that moment is now burnt in my mind.
My normal speakers wouldn't have done this,it would have sounded like a normal surround sound music effect , not to be noticed ,but to be felt.
Is this a good thing for surround sound , I don't know, if you don't listen to a dml When mixing the sound track , you will not know what the dml will bring out in the mix.
They are very revealing of recording mistakes , in music and film.
I've lost track of all the things I've heard in music recordings , people knocking into things, whispering, doors opening(for ages would go out and check to see if someone was in the house,I was sure there was someone walking around when I knew I was on my own ) only to eventually find it was in the background on that CD .
Steve.
When I was testing my card panels on my surround sound in my front room ,I turned the centre speaker off so that I could hear the panels in the front only.
It takes a bit of getting used to because sounds sound so realistic.
On one film I was watching on tv,can't remember the film,there was a romantic encounter on a busy street,it was a very tense moment.
All of a sudden someone started to play a cello ,now normally this would have been floating around in the background.
But it suddenly sounded as if someone was sitting next to them playing this very moving cello , as if the camera could pan to the right and you would see this person playing,no more than a foot of so away,you could hear the bow rasping on the strings
It sounded fantastic, but it was a little distracting, probably not what the director was going for ?
I can't remember the film but that moment is now burnt in my mind.
My normal speakers wouldn't have done this,it would have sounded like a normal surround sound music effect , not to be noticed ,but to be felt.
Is this a good thing for surround sound , I don't know, if you don't listen to a dml When mixing the sound track , you will not know what the dml will bring out in the mix.
They are very revealing of recording mistakes , in music and film.
I've lost track of all the things I've heard in music recordings , people knocking into things, whispering, doors opening(for ages would go out and check to see if someone was in the house,I was sure there was someone walking around when I knew I was on my own
) only to eventually find it was in the background on that CD .Steve.
Steve,
I've had very much the same experiences and concur that...
Add to this that so much of movie sound does not so much get recorded on filming, but fabricated and synthesized in the studio post "film" production, which creates an entirely artificial (sometimes terrible but sometimes profoundly interesting) experience.
I suspect the sound reproduction in theaters just has far less variation between theaters than it can possibly have across home theaters.
I started playings with setting up (bracketing) a center channel "voice" frequency range and will start sticking some bits of diffusing and absorbing material around the DMLs and the bookcase. Then on for a try at Dirac Live.
Just a thought...
Jean Luc Goddard made a series of short films that made masterful use of naturally recorded sound. They might prove very interesting as test material to evaluate a DML based home theater system.
- Andreas
I've had very much the same experiences and concur that...
Add to this that so much of movie sound does not so much get recorded on filming, but fabricated and synthesized in the studio post "film" production, which creates an entirely artificial (sometimes terrible but sometimes profoundly interesting) experience.
I suspect the sound reproduction in theaters just has far less variation between theaters than it can possibly have across home theaters.
I started playings with setting up (bracketing) a center channel "voice" frequency range and will start sticking some bits of diffusing and absorbing material around the DMLs and the bookcase. Then on for a try at Dirac Live.
Just a thought...
Jean Luc Goddard made a series of short films that made masterful use of naturally recorded sound. They might prove very interesting as test material to evaluate a DML based home theater system.
- Andreas
I ment to photo the phase plug before I cut the panel out of the hoop.
I want this exciter to go on my 5mm x600x380mm xps test panel, so I'm cutting it down slowly to see how it affects the performance.
The phase plug is just some rolled up blu-tack, it's just another way to prevent the cavity resonance in the vocal range .
I have seen designs with out the phase plug as in the second picture.
Steve.
I want this exciter to go on my 5mm x600x380mm xps test panel, so I'm cutting it down slowly to see how it affects the performance.
The phase plug is just some rolled up blu-tack, it's just another way to prevent the cavity resonance in the vocal range .
I have seen designs with out the phase plug as in the second picture.
Steve.
Attachments
KoAP.
It's a pity that the patent has lost its prior art, I gave up reading about half way through.
I get the idea of what is being said ,but am not sure if I like the idea.
You could just look at the prior art and save yourself all those words.
I've read too many patents ,I suppose.
I prefer his idea of an air pump for low frequency driving(very similar to what is now known as an ABR )if only I had the time to try it out.
The way I look at it is, I let the panel do what it does best and leave driving the low frequencies in the room to something with a bit more welly.
Although I must say that the fabric type panels do seem to give a little low end support to my TLS , I can hear the difference when they are playing full range.
Steve.
It's a pity that the patent has lost its prior art, I gave up reading about half way through.
I get the idea of what is being said ,but am not sure if I like the idea.
You could just look at the prior art and save yourself all those words.
I've read too many patents ,I suppose.
I prefer his idea of an air pump for low frequency driving(very similar to what is now known as an ABR )if only I had the time to try it out.
The way I look at it is, I let the panel do what it does best and leave driving the low frequencies in the room to something with a bit more welly.
Although I must say that the fabric type panels do seem to give a little low end support to my TLS , I can hear the difference when they are playing full range.
Steve.
I am now listening to my smaller 5mm epoxy xps panel plus the 2mm wood panel ,very nice.
I'm going to put up the larger 5mm epoxy xps and compare the performance full range.
There is a considerable cavity Noise hump that will need to be taken care off , but otherwise I'm very p!eased with the sound so far.
The epoxy I am using is used for mouldings,so is not as hard as the bonding epoxy I believe ?
Would the harder epoxy be that much harder, I don't know, would it sound better ?
I think I paid about £15 for 16oz which goes a long way when you only use about 15 to 20g per side, depending on panel size.
I've also glued quite a few things around the house as well
Steve.
I'm going to put up the larger 5mm epoxy xps and compare the performance full range.
There is a considerable cavity Noise hump that will need to be taken care off , but otherwise I'm very p!eased with the sound so far.
The epoxy I am using is used for mouldings,so is not as hard as the bonding epoxy I believe ?
Would the harder epoxy be that much harder, I don't know, would it sound better ?
I think I paid about £15 for 16oz which goes a long way when you only use about 15 to 20g per side, depending on panel size.
I've also glued quite a few things around the house as well
Steve.
this thing moves Fast! wanted to share some research and ask some questions as i would like to make a cheap prototype, already have 4 daex25FHE-4 wich will be used in series.
First of all do fundamental frequency and concidencce frecuency depend on the bending stiffness-mass ratio? i got that from chalmers Kuonan Li thesis. If so, to achive enough bass you get a lower point where diffraction becomes "lobulized"? (above that freq bending waves become transversal and there is a phase something).
-About the boundarie condition, it seems it alters the bass and midbass section mainly, being quite more peaky dipy for fixed and free edges, while elastic boundries smoothens the response. Choices here may be rubbers, foams, pitch materials (tar) , polimers, silicones...(look at Basilio Pueo boundarie conditions).
-Then the Kuonan Li thesis showed in a graph (abour the end) highest modal density is achieved when the ratio between length and wide aproaches 1. Also Zenker showed that as the aspec ratio gets bigger than cero isotropic materials show a signifficantly bigger modal density.
Aboout the exciter positioning i found two papers one of zenker, look if you are interested (ask the full names if you want, i dont want this to go forever). But yep, avoid the center.
As interesting things i found, restrained chambers of air behind the plate smoothens the peaks. A couple of M######rs used an array of exciters arround the whole perimeter of a plate to cancel the reflections, and kind of achieved an infinite plate repsonse.
This thing is getting lots of attention, its used un MAPS too, so there are potential or prototypes for wavefielsynthesis. A thing that i dont understand well is abour evanscent waves, when they start to occur, i guess higger than coincidence frequency, and relatively close to the plate?
I've got ideas about everything but material desition is my main concern, XPS, if i did the math righ wich i may not, at 20mm may offer an f0 between 100 and 20000, but it seems suspucious to me. Another one i wished to try is corrugated plastic as its cheap! the idea woud be to use them full range of course, with a sub or something, any recomendations on materials for FR panels?
Regards!
First of all do fundamental frequency and concidencce frecuency depend on the bending stiffness-mass ratio? i got that from chalmers Kuonan Li thesis. If so, to achive enough bass you get a lower point where diffraction becomes "lobulized"? (above that freq bending waves become transversal and there is a phase something).
-About the boundarie condition, it seems it alters the bass and midbass section mainly, being quite more peaky dipy for fixed and free edges, while elastic boundries smoothens the response. Choices here may be rubbers, foams, pitch materials (tar) , polimers, silicones...(look at Basilio Pueo boundarie conditions).
-Then the Kuonan Li thesis showed in a graph (abour the end) highest modal density is achieved when the ratio between length and wide aproaches 1. Also Zenker showed that as the aspec ratio gets bigger than cero isotropic materials show a signifficantly bigger modal density.
Aboout the exciter positioning i found two papers one of zenker, look if you are interested (ask the full names if you want, i dont want this to go forever). But yep, avoid the center.
As interesting things i found, restrained chambers of air behind the plate smoothens the peaks. A couple of M######rs used an array of exciters arround the whole perimeter of a plate to cancel the reflections, and kind of achieved an infinite plate repsonse.
This thing is getting lots of attention, its used un MAPS too, so there are potential or prototypes for wavefielsynthesis. A thing that i dont understand well is abour evanscent waves, when they start to occur, i guess higger than coincidence frequency, and relatively close to the plate?
I've got ideas about everything but material desition is my main concern, XPS, if i did the math righ wich i may not, at 20mm may offer an f0 between 100 and 20000, but it seems suspucious to me. Another one i wished to try is corrugated plastic as its cheap! the idea woud be to use them full range of course, with a sub or something, any recomendations on materials for FR panels?
Regards!