My apologies for my flare up with "foot in mouth syndrome". Now that I've read and understand the methodology behind the lower chamber I still have some reservations. Judging from his graphs it introduces it's own problems.......peaks @ 130,110, and 60...... a massive valley @ 90 (though he does partially blame this on room modes). If it sounds good to you, it's good. IMO (take no offense) that space would be better utilized housing a 8-10" driver built to play flat, opposed to exciting a chamber that's very much dependent on material choice for the diy'er. He does elude to using a pair of subs for true impact on the low end (his own contradiction), so again IMO.....take it with a grain of salt, he hasn't created a stand alone speaker in the Zen OB.
Last edited:
What resonances in the bass?
Here's the Decware published graph (still "controversial" until proven right or wrong).
I don't see any evidence of the big peaks, valleys and suck-outs alluded to in a previous post. Are we looking at the same graphs?
BTW, the system -3dB point is at about 45Hz....... it DOES reach 38Hz but at -6dB
This curve is quite similar to what an acoustic suspension (sealed cabinet) system would give if the Fb is at 45Hz with a Qb of about 0.7
On paper it would seem "bass-shy" but not so much if you look at how it interacts with an average room, where this type of curve can be quite favorable.
If the "average room" usual main bass mode is in the 40 to 45Hz range, this speaker may well play above the reference level at 40Hz and even have useful bass in the low 30's or even high 20's... so no, I don't think bass-shy would be a likely description.
Only drawback that I see to this system is that with an Xmax of 5mm this driver is likely to start modulating heavily at anything over about 2 watts... I doubt it'll still sound as clean when pushed hard. I don't think that resonator will do much to keep the driver from hitting peak excursions easily at its fundamental resonance. i.e. the driver is behaving as if in free air at 70Hz. My bet is under 100dB peaks before excursion limitations become audible.
Not enough volume for the "Loudiophile" contingent (myself included). I'd think a sub would be a good idea to help out in the low bass, for sure.
Deckert's resonator looks like a really interesting re-discovery of the resonators used in the Imperial folded horn he references in his site. As applied here, and if the curves are real, it sure looks like it works! A pretty cool application IMO.
Here's the Decware published graph (still "controversial" until proven right or wrong).
I don't see any evidence of the big peaks, valleys and suck-outs alluded to in a previous post. Are we looking at the same graphs?
BTW, the system -3dB point is at about 45Hz....... it DOES reach 38Hz but at -6dB
This curve is quite similar to what an acoustic suspension (sealed cabinet) system would give if the Fb is at 45Hz with a Qb of about 0.7
On paper it would seem "bass-shy" but not so much if you look at how it interacts with an average room, where this type of curve can be quite favorable.
If the "average room" usual main bass mode is in the 40 to 45Hz range, this speaker may well play above the reference level at 40Hz and even have useful bass in the low 30's or even high 20's... so no, I don't think bass-shy would be a likely description.
Only drawback that I see to this system is that with an Xmax of 5mm this driver is likely to start modulating heavily at anything over about 2 watts... I doubt it'll still sound as clean when pushed hard. I don't think that resonator will do much to keep the driver from hitting peak excursions easily at its fundamental resonance. i.e. the driver is behaving as if in free air at 70Hz. My bet is under 100dB peaks before excursion limitations become audible.
Not enough volume for the "Loudiophile" contingent (myself included). I'd think a sub would be a good idea to help out in the low bass, for sure.
Deckert's resonator looks like a really interesting re-discovery of the resonators used in the Imperial folded horn he references in his site. As applied here, and if the curves are real, it sure looks like it works! A pretty cool application IMO.
Decware's New FRX Full range Drivers!
For those who who aren't afraid of expensive full range drivers, you should check out Steve from Decware recent modification of Dayton's relatively inexpensive new full range driver which lists at $119ea. Steve has essentially added a type of magnetic equalization circuit that magnetically couples current from the amplifier to the driver's voice coil. He then renamed these drivers the FRX and charges $599ea for them!
Those who know me here, know I always refrain from commenting on an audio component's value or sound "if" I haven't heard it for myself and that happens to be the case here. However that said, I'm comfortable stating if they work as described there's a chance they might be very sonically special indeed...
Thetubeguy1954 (Tom Scata)
For those who who aren't afraid of expensive full range drivers, you should check out Steve from Decware recent modification of Dayton's relatively inexpensive new full range driver which lists at $119ea. Steve has essentially added a type of magnetic equalization circuit that magnetically couples current from the amplifier to the driver's voice coil. He then renamed these drivers the FRX and charges $599ea for them!
Those who know me here, know I always refrain from commenting on an audio component's value or sound "if" I haven't heard it for myself and that happens to be the case here. However that said, I'm comfortable stating if they work as described there's a chance they might be very sonically special indeed...
Thetubeguy1954 (Tom Scata)
I love the look of that frequency response graph
Ruler flat from 45 Hz to 20 KHz at 98 dB? Throw a 16 to 60 Hz tapped horn with that puppy and my journey is over.
However, I also am quite aware that the "last speaker you'll ever buy" was never the last. I hope that the FRX does exactly what it says it does--wisdom indicates otherwise though.
I don't care if he got 45 to 20 KHz flat at 98dB with a toilet paper roll and a rubber band--I would buy it. Since audio is steeped in religion, I will have to get many reports of the Jesus full range actually performing the way the charts indicate before I throw in $1K+ into the FRX offering plate.
Ruler flat from 45 Hz to 20 KHz at 98 dB? Throw a 16 to 60 Hz tapped horn with that puppy and my journey is over.
However, I also am quite aware that the "last speaker you'll ever buy" was never the last. I hope that the FRX does exactly what it says it does--wisdom indicates otherwise though.
I don't care if he got 45 to 20 KHz flat at 98dB with a toilet paper roll and a rubber band--I would buy it. Since audio is steeped in religion, I will have to get many reports of the Jesus full range actually performing the way the charts indicate before I throw in $1K+ into the FRX offering plate.
I love the look of that frequency response graph
Ruler flat from 45 Hz to 20 KHz at 98 dB? Throw a 16 to 60 Hz tapped horn with that puppy and my journey is over.
However, I also am quite aware that the "last speaker you'll ever buy" was never the last. I hope that the FRX does exactly what it says it does--wisdom indicates otherwise though.
I don't care if he got 45 to 20 KHz flat at 98dB with a toilet paper roll and a rubber band--I would buy it. Since audio is steeped in religion, I will have to get many reports of the Jesus full range actually performing the way the charts indicate before I throw in $1K+ into the FRX offering plate.
well said 18Hurts...my thoughts exactly
Not jumping in just yet...
In previous posts I've tried to give an idea of what Mr Deckert MAY be doing in his mods of the Dayton PS220 and then re-naming as the FRX. The key word is MAY, as in I can't be sure. But I think the analysis is pretty much on target.
Something needs to be pointed out:
even though FRX on axis response looks very, very good, there are other factors that count: how wide and even is the off-axis response at 15, 30, 45, 60 and 75 degrees?
The averaged frontal hemispheric response is often a better indicator of perceived fidelity than just the on-axis. And yes, this applies even when you toe the speakers in to aim at your ears. Your ears will hear a blend of the on-axis and reverberant contributions, and BOTH have to be very good.
(Have a look at my website at HolisticAudio dot com to see how we've gone about dealing with the off-axis).
So even if it DOES measure really flat on axis, this does not in itself predict perfect performance. But I will say this: I would get one of these over a Lowther of comparable price.
And there is also a built-in limitation (which applies even more to Lowther, Fostex, et.al.):
5mm of Xmax and a fundamental resonance of 70Hz in an open baffle load is going to allow only a very limited dynamic range... my guess is a max of 95dB when used as a full-range as in the Zen OB... that's OK for many types of music... but not for full scale classical, or loud rock for example
Unless his Zen resonator is providing some magical form of load to the back of the cone, that cone is still going have its fundamental resonance at around 70Hz, which means it's going to rapidly hit full scale excursion in the 50 to 120Hz zone, right in the middle of the lower bass range. Oops.
Yes, the suspension mod may help a little...but not enough:
Deckert has probably modded the surround by adding an extra foam surround to the existing one. The modded surround calculates to be much stiffer, about 5X less compliance than the original, this is why the resonance goes up to 70Hz or so from the stock unit's 37Hz.
All the other parameter changes seem consistent with just stiffening the surround. This damps out cone resonances at the edges, and may reduce distortion.
Yet, even if it has radically less distortion in the linear range, Xmax is still Xmax. I predict about 95 to 98dB usable peaks in the bass, enough for many, but certainly not most situations.
So I would say this driver (modded OR stock) is a great candidate for using with a pair of Eminence Alpha 15's in a tall open baffle, crossed in at 200Hz or so... now we're talking... a full range DIY OB system for less than the cost of a single FRX !!
Now about those highish prices: while I believe Decware's price may still be "fair" given his labor, experimentation, experience, etc. plus the added cost of the components and fancy housing he's adding on the back.... uh, well.... $595 is still a LOT of bucks for an 8" driver. (Don't get me started on Lowther's pricing... that's just crazy!)
So, you won't find me lining up to buy FRX anytime soon. It may well be that very similar results can be obtained for far less money.
It looks like it would primarily require adding an 8" foam suspension ring to remove/control the edge resonances on the cone. Dayton sells an 8" suspension repair kit for about $25 that looks like it would do the job. Then, add an appropriate shelf network and voila, something that should be pretty close to the FRX, for a whole lot less.
If I do it and it works well, perhaps I would offer the modded Dayton drivers with optimized shelf networks at a price that would be a little more approachable, maybe $270. I have the same measurement suite Decware is using, Fuzzmeasure 3, and a calibrated mic to measure with, so results will be verified by measurement.
Only I won't add the fancy tube on the back and call it a fancy name like SFR (the Stealth Full-Range). Hmmm. But I could charge more if I did....nah.
Disclosure:
About the projected selling price: I'm presuming there may be some failed attempts in the process, at $100 per driver that adds up quick! Thus, I would want to amortize that cost, along with my labor and time invested in doing this. I don't work for free, but I think this might be a valuable service to offer. So $270 if it works well.
I'd like to know if any of you would be interested in this experiment, just drop me a note here or at Holisticaudio@gmail.com.
All the Best
In previous posts I've tried to give an idea of what Mr Deckert MAY be doing in his mods of the Dayton PS220 and then re-naming as the FRX. The key word is MAY, as in I can't be sure. But I think the analysis is pretty much on target.
Something needs to be pointed out:
even though FRX on axis response looks very, very good, there are other factors that count: how wide and even is the off-axis response at 15, 30, 45, 60 and 75 degrees?
The averaged frontal hemispheric response is often a better indicator of perceived fidelity than just the on-axis. And yes, this applies even when you toe the speakers in to aim at your ears. Your ears will hear a blend of the on-axis and reverberant contributions, and BOTH have to be very good.
(Have a look at my website at HolisticAudio dot com to see how we've gone about dealing with the off-axis).
So even if it DOES measure really flat on axis, this does not in itself predict perfect performance. But I will say this: I would get one of these over a Lowther of comparable price.
And there is also a built-in limitation (which applies even more to Lowther, Fostex, et.al.):
5mm of Xmax and a fundamental resonance of 70Hz in an open baffle load is going to allow only a very limited dynamic range... my guess is a max of 95dB when used as a full-range as in the Zen OB... that's OK for many types of music... but not for full scale classical, or loud rock for example
Unless his Zen resonator is providing some magical form of load to the back of the cone, that cone is still going have its fundamental resonance at around 70Hz, which means it's going to rapidly hit full scale excursion in the 50 to 120Hz zone, right in the middle of the lower bass range. Oops.
Yes, the suspension mod may help a little...but not enough:
Deckert has probably modded the surround by adding an extra foam surround to the existing one. The modded surround calculates to be much stiffer, about 5X less compliance than the original, this is why the resonance goes up to 70Hz or so from the stock unit's 37Hz.
All the other parameter changes seem consistent with just stiffening the surround. This damps out cone resonances at the edges, and may reduce distortion.
Yet, even if it has radically less distortion in the linear range, Xmax is still Xmax. I predict about 95 to 98dB usable peaks in the bass, enough for many, but certainly not most situations.
So I would say this driver (modded OR stock) is a great candidate for using with a pair of Eminence Alpha 15's in a tall open baffle, crossed in at 200Hz or so... now we're talking... a full range DIY OB system for less than the cost of a single FRX !!
Now about those highish prices: while I believe Decware's price may still be "fair" given his labor, experimentation, experience, etc. plus the added cost of the components and fancy housing he's adding on the back.... uh, well.... $595 is still a LOT of bucks for an 8" driver. (Don't get me started on Lowther's pricing... that's just crazy!)
So, you won't find me lining up to buy FRX anytime soon. It may well be that very similar results can be obtained for far less money.
It looks like it would primarily require adding an 8" foam suspension ring to remove/control the edge resonances on the cone. Dayton sells an 8" suspension repair kit for about $25 that looks like it would do the job. Then, add an appropriate shelf network and voila, something that should be pretty close to the FRX, for a whole lot less.
If I do it and it works well, perhaps I would offer the modded Dayton drivers with optimized shelf networks at a price that would be a little more approachable, maybe $270. I have the same measurement suite Decware is using, Fuzzmeasure 3, and a calibrated mic to measure with, so results will be verified by measurement.
Only I won't add the fancy tube on the back and call it a fancy name like SFR (the Stealth Full-Range). Hmmm. But I could charge more if I did....nah.
Disclosure:
About the projected selling price: I'm presuming there may be some failed attempts in the process, at $100 per driver that adds up quick! Thus, I would want to amortize that cost, along with my labor and time invested in doing this. I don't work for free, but I think this might be a valuable service to offer. So $270 if it works well.
I'd like to know if any of you would be interested in this experiment, just drop me a note here or at Holisticaudio@gmail.com.
All the Best
Toole's research indicates that the on-axis without this additional info does not tell us much.
dave
I doubt that the cone gets a surface treatment without the whizzer also. If like his FE206 derivative, he doesn't use much (acrylic i think).
Of late, Joachim has reported significant smoothing of the HF of the "budget" 8" SEAS FR by treating just the whizzer (with a commercial speaker coatong). Bob Brines killed the HF response of a set of FE166En (mod-podge (PVA) with a high dilutionm that likely goes deep into the paper), and i regularily coat whizzer cones (mostly Fostex/a dilution of mod-podge that sits mostly on the paper surface, and only part of more extensive treatment) and smooth out HF response.
So we know that whizzer response can be altered.
dave
Gentlemen, behave. Any more posts similar to the ones I have just deleted will lead to further action. Jack, the "Report Post" button is there for a reason, use it.
I appologize if I called Steve names, it was provoked by someone here who clearly stated that there was subwoofer playing. That person was wrong, and I was wrong, or too quick.
I did my own experiments with output trafo, and found out it did affect the sound quite a bit, making a tilt in the response, but the same exact peaks remained.
I neved had a problem with Steve's pricing. Whoever wants to pay, its his decision. Not even single my post touched the pricing.
What I have a problem with is that ruler flat response...
And it may not be the whizzer...
More about the topic: how did Decware get it so flat?
As Dave pointed out, treatment of the whizzer cone can certainly affect its performance.
AND.... it may not be the whizzer! It could be the cone itself, exhibiting multiple bell-mode resonances. If you look at the Dayton sheet, the resonant peaks and valleys appear to be fairly closely linked:
1) big peak at 3.5KHz + small peak @ 7KHz
2) Dip @ 9KHz + Peak @ 18KHz Note: often a "dip" shows up in waterfall plots as energy storage with delayed release
3) Peak @11KHz + Peak @23KHz
I've seen VERY similar patterns from other wide-bandwidth drivers that do not have whizzer cones.
In my HolisticAudio H1 speakers I'm using matched pairs of 6.5" drivers, full range without whizzers, inverted in Quasi-Walsh mode. (See HolisticAudio.com for details) When measured untreated, on axis, these full-range drivers exhibit similar response peaks to what PartsExpress/Dayton show on their data sheet for the PS180 and PS220, with the PS220 being what the Decware FRX is based on.
When treated to make the cones stiffer, the peaks often get WORSE, not better.
Unless a pattern like Dave's Tri-Foil is used, or something like ScanSpeak's slit paper "cut n' glue" is applied, as a general trend, stiffening alone is not a cure for those resonances.
I think most of what we are seeing with the Dayton must be caused by bell-mode resonances where the cone interfaces with the surround. The abrupt change in mechanical impedance and damping creates reflections in the cone and bell-mode edge resonances.
So while it's probable Deckert is treating the whizzer as well, I think the MAIN reason for seeing those peaks diminish is the 2nd ROLL added to the suspension. Deckert has written about this in his papers about modding Fostex drivers, it's a technique that seems to have been pioneered by Lowther.
So... the Dayton driver:
The stock suspension on the Dayton is an inverted roll made of very soft and compliant rubber. When coupled with a stiff cone, it will ring like a bell at the edges, with multiple resonant modes all the way into the HF region. Often, the suspension itself is one of the major "ringing" culprits!
If you make the cone out of soft polymer, that interface is better mechanically, which is why poly cones often have very smooth response. (IMO they sound dull and dead, but they sure do measure nicely!)
Now if we insist on using a stiff paper cone.. (which to me usually sounds much better than any poly cone) if the suspension is soft... you've got a bell ringing! And Dayton's measurements look like...surprise!....a bell ringing.
BUT... when you add the 2nd outer roll suspension, things change:
A) The trapped air-pocket forces the suspension to stay more linear, and it will inherently damp its tendency to ring. (This is explained somewhere in the Lowther threads).
B) Adding the 2nd roll radically lowers the compliance of the suspension. (Makes it much stiffer). The mechanical impedance is better matched to that of the cone, and the energy doesn't get reflected nearly as much, instead it gets trapped and dissipated in the suspension. If my calculation is correct, to attain the changes in Thiele-Small parameters published for the FRX vs the Dayton, the new suspension is now 5X stiffer! That's a HUGE change in mechanical impedance and damping!
Would this change those peaks? YES! Radically!
C) Remember the B&W white paper about their soft kevlar weave for the midrange driver? They eliminated the suspension and made the cone very lossy to absorb these reflections.
Well, proper mechanical impedance matching eliminates reflections. Making the suspension stiffer to better match the mechanical impedance of the cone is in all likelihood the reason for the disappearance of those peaks.
So, if I were a bettin' man, I would put my money on the 2nd roll of that suspension as being the main reason those HF peaks have gone away. If those measurements are real, and I'm thinking they are (for that ONE sample!)... well, that's just good execution of a clever concept lifted from Lowther.
And now a BIG caveat: just how REPEATABLE are those measurements going to be?
How many of Decware's FRX production drivers are going to measure like the published curves? This is a non-trivial issue!
I take great pains to match the drivers in my speakers. It's a lot of work to find drivers that operate like twins when going full-range. Out of 30 pairs of drivers, I get only about 10 pairs that match up well. Then after modding, maybe 6 to 8 pairs.... whew.....
If Deckert has to contend with similar issues, it's going to be interesting to see if the production drivers can match up. If he manages to do this consistently, well, that'll be interesting indeed.
More about the topic: how did Decware get it so flat?
As Dave pointed out, treatment of the whizzer cone can certainly affect its performance.
AND.... it may not be the whizzer! It could be the cone itself, exhibiting multiple bell-mode resonances. If you look at the Dayton sheet, the resonant peaks and valleys appear to be fairly closely linked:
1) big peak at 3.5KHz + small peak @ 7KHz
2) Dip @ 9KHz + Peak @ 18KHz Note: often a "dip" shows up in waterfall plots as energy storage with delayed release
3) Peak @11KHz + Peak @23KHz
I've seen VERY similar patterns from other wide-bandwidth drivers that do not have whizzer cones.
In my HolisticAudio H1 speakers I'm using matched pairs of 6.5" drivers, full range without whizzers, inverted in Quasi-Walsh mode. (See HolisticAudio.com for details) When measured untreated, on axis, these full-range drivers exhibit similar response peaks to what PartsExpress/Dayton show on their data sheet for the PS180 and PS220, with the PS220 being what the Decware FRX is based on.
When treated to make the cones stiffer, the peaks often get WORSE, not better.
Unless a pattern like Dave's Tri-Foil is used, or something like ScanSpeak's slit paper "cut n' glue" is applied, as a general trend, stiffening alone is not a cure for those resonances.
I think most of what we are seeing with the Dayton must be caused by bell-mode resonances where the cone interfaces with the surround. The abrupt change in mechanical impedance and damping creates reflections in the cone and bell-mode edge resonances.
So while it's probable Deckert is treating the whizzer as well, I think the MAIN reason for seeing those peaks diminish is the 2nd ROLL added to the suspension. Deckert has written about this in his papers about modding Fostex drivers, it's a technique that seems to have been pioneered by Lowther.
So... the Dayton driver:
The stock suspension on the Dayton is an inverted roll made of very soft and compliant rubber. When coupled with a stiff cone, it will ring like a bell at the edges, with multiple resonant modes all the way into the HF region. Often, the suspension itself is one of the major "ringing" culprits!
If you make the cone out of soft polymer, that interface is better mechanically, which is why poly cones often have very smooth response. (IMO they sound dull and dead, but they sure do measure nicely!)
Now if we insist on using a stiff paper cone.. (which to me usually sounds much better than any poly cone) if the suspension is soft... you've got a bell ringing! And Dayton's measurements look like...surprise!....a bell ringing.
BUT... when you add the 2nd outer roll suspension, things change:
A) The trapped air-pocket forces the suspension to stay more linear, and it will inherently damp its tendency to ring. (This is explained somewhere in the Lowther threads).
B) Adding the 2nd roll radically lowers the compliance of the suspension. (Makes it much stiffer). The mechanical impedance is better matched to that of the cone, and the energy doesn't get reflected nearly as much, instead it gets trapped and dissipated in the suspension. If my calculation is correct, to attain the changes in Thiele-Small parameters published for the FRX vs the Dayton, the new suspension is now 5X stiffer! That's a HUGE change in mechanical impedance and damping!
Would this change those peaks? YES! Radically!
C) Remember the B&W white paper about their soft kevlar weave for the midrange driver? They eliminated the suspension and made the cone very lossy to absorb these reflections.
Well, proper mechanical impedance matching eliminates reflections. Making the suspension stiffer to better match the mechanical impedance of the cone is in all likelihood the reason for the disappearance of those peaks.
So, if I were a bettin' man, I would put my money on the 2nd roll of that suspension as being the main reason those HF peaks have gone away. If those measurements are real, and I'm thinking they are (for that ONE sample!)... well, that's just good execution of a clever concept lifted from Lowther.
And now a BIG caveat: just how REPEATABLE are those measurements going to be?
How many of Decware's FRX production drivers are going to measure like the published curves? This is a non-trivial issue!
I take great pains to match the drivers in my speakers. It's a lot of work to find drivers that operate like twins when going full-range. Out of 30 pairs of drivers, I get only about 10 pairs that match up well. Then after modding, maybe 6 to 8 pairs.... whew.....
If Deckert has to contend with similar issues, it's going to be interesting to see if the production drivers can match up. If he manages to do this consistently, well, that'll be interesting indeed.
Further experiments verify what you said here Jack.
Using an old 12" Magnavox console speaker that rang like the skin of a drum, I first doubled the original paper surround with a foam gasket. It was easy to hear the difference in resonance simply by tapping the cone with a finger. Frequency response measurements changed, which sadly I did not save. The speaker became slower in transients, but more refined.
Finally the speaker was modified by cutting the paper and foam surround out and replacing them with a surround of 2mm sheepskin chamois, as seen here:
This completely relieved the tension of the paper cone and the drum skin sound it had. The compliant leather surround restored transient response, and surprisingly, reduced resonances some more.
Regarding the transformer used for FR shaping. Steve mentioned that the transformer he used would completely prevent any signal from passing to the driver if no load was present on the opposite windings. That makes me think that he may have used a transformer with much more capacitive coupling than an output transformer would have. A 120V:240V stepdown or 120V:120 isolation transformer for instance. The isolation / stepdown transformers I have tested so far are superior sounding than the output transformers I have tested, in that they do not saturate and distort on transients. The trade off is that the entire musical signal has to be shaped by the passive filters coupled to those windings, the inductance of the transformer itself having less influence on the FR.
Just a couple more options to add the the heap of ideas talked about here.
Last edited:
How about correcting in DSP?
Gee, I would have thought most any decent audio transformer should work for this, the speaker would only see the DCR of the winding + the reflected impedance of the correction network, virtually the same as just having a regular high freq shelving network in series with the speaker (e.g a 1.3mH aircoil inductor in parallel with a 6 ohm resistor, in series w. the Dayton PS 220).
Still, I'm interested in checking out the coupling transformer idea, if nothing more than to check out Deckert's claim that it actually sounds better, and also to compare with the digital EQ I've been working with, a modded Behringer EQ2496. Do you have some suggestions for where to get similar trafos? The 2:1 for example?
How about Digital EQ instead? on my H1 speakers I've had really great results using Core Audio functions on Mac during playback (that, or using a modified Behringer DEQ 2496) to correct for frequency aberrations.
The GREAT sonic benefit of this is that we get direct drive on the voice coil and IMO it sounds much better than using any in-line components I've heard. Layers and layers of veiling fall away.
And even tho' Deckert's approach is "magnetically coupled" .... the transformer DCR + reflected impedance of the frequency shaping elements will still show up IN SERIES with the driver. It's bound to create some losses and some veiling. With the DSP and direct amp drive, you are likely to get big gains in overall transparency vs Deckert's approach. But it's worth comparing, that's for sure.
BTW it is TRUE the Behringer analog section sounds really bad. It is also equally true that when properly modded, the direct DAC output is stellar. Sendler does extensive mods on these, I might offer some entry level mods also.
Or, you could use it like I do in my top systems, stock unit unmodified, digital in/digital out and use a separate DAC.... extremely transparent sound.
No discernible difference with it inserted and set flat but with EQ engaged vs not in the loop at all... and this was the case with ANY of the GoldenEars I've had visit, none noticed a difference either here or at their place. That's a pretty good level of transparency.
Oh wait, I exaggerate, one DID hear a difference. He liked the sound better with the Behringer IN the loop!
Also worthy of notice, the sound of the "direct out" MODDED DAC on the Behringer is so good that it is better than many reference DACs...for about $300 for the 2496 on ebay and less than $200 for the direct out mod you get a reference quality DAC and also an astonishingly good equalizer... sure beats paying big $$$ to get a really good outboard DAC... and also saves on the interconnects!... and no more monster size (and price!) inductors and capacitors for frequency shaping on your full-range speakers. Many, many advantages to this approach. To me the chief one is the transparency. For Fostex or Lowther owners, this is a real difference maker.
And this way you can tinker with it and get it ultra-flat or change the curve to get the sound the way YOU like it. Seriously folks, with today's quality DSP there is no reason to mess with passive components which by comparison really veil the sound. And there is no reason to put up with thin, over-etched sound from your otherwise great Lowthers and Fostex! Now that I've lived with this approach for over 2 years, and never felt it to be veiled or lacking in any way, well, I'm probably not going back to the old ways, magnetically coupled as they may be!
The isolation / stepdown transformers I have tested so far are superior sounding than the output transformers I have tested, in that they do not saturate and distort on transients. The trade off is that the entire musical signal has to be shaped by the passive filters coupled to those windings
Just a couple more options to add the the heap of ideas talked about here.
Gee, I would have thought most any decent audio transformer should work for this, the speaker would only see the DCR of the winding + the reflected impedance of the correction network, virtually the same as just having a regular high freq shelving network in series with the speaker (e.g a 1.3mH aircoil inductor in parallel with a 6 ohm resistor, in series w. the Dayton PS 220).
Still, I'm interested in checking out the coupling transformer idea, if nothing more than to check out Deckert's claim that it actually sounds better, and also to compare with the digital EQ I've been working with, a modded Behringer EQ2496. Do you have some suggestions for where to get similar trafos? The 2:1 for example?
How about Digital EQ instead? on my H1 speakers I've had really great results using Core Audio functions on Mac during playback (that, or using a modified Behringer DEQ 2496) to correct for frequency aberrations.
The GREAT sonic benefit of this is that we get direct drive on the voice coil and IMO it sounds much better than using any in-line components I've heard. Layers and layers of veiling fall away.
And even tho' Deckert's approach is "magnetically coupled" .... the transformer DCR + reflected impedance of the frequency shaping elements will still show up IN SERIES with the driver. It's bound to create some losses and some veiling. With the DSP and direct amp drive, you are likely to get big gains in overall transparency vs Deckert's approach. But it's worth comparing, that's for sure.
BTW it is TRUE the Behringer analog section sounds really bad. It is also equally true that when properly modded, the direct DAC output is stellar. Sendler does extensive mods on these, I might offer some entry level mods also.
Or, you could use it like I do in my top systems, stock unit unmodified, digital in/digital out and use a separate DAC.... extremely transparent sound.
No discernible difference with it inserted and set flat but with EQ engaged vs not in the loop at all... and this was the case with ANY of the GoldenEars I've had visit, none noticed a difference either here or at their place. That's a pretty good level of transparency.
Oh wait, I exaggerate, one DID hear a difference. He liked the sound better with the Behringer IN the loop!
Also worthy of notice, the sound of the "direct out" MODDED DAC on the Behringer is so good that it is better than many reference DACs...for about $300 for the 2496 on ebay and less than $200 for the direct out mod you get a reference quality DAC and also an astonishingly good equalizer... sure beats paying big $$$ to get a really good outboard DAC... and also saves on the interconnects!... and no more monster size (and price!) inductors and capacitors for frequency shaping on your full-range speakers. Many, many advantages to this approach. To me the chief one is the transparency. For Fostex or Lowther owners, this is a real difference maker.
And this way you can tinker with it and get it ultra-flat or change the curve to get the sound the way YOU like it. Seriously folks, with today's quality DSP there is no reason to mess with passive components which by comparison really veil the sound. And there is no reason to put up with thin, over-etched sound from your otherwise great Lowthers and Fostex! Now that I've lived with this approach for over 2 years, and never felt it to be veiled or lacking in any way, well, I'm probably not going back to the old ways, magnetically coupled as they may be!
there is a member that used sucj EQ on the stock Dayton 7 reported very good results.
What are you using on your mac.
dave
DSP on Mac
The least expensive way I've found to access the CoreAudio functions on the Mac is by using Audio Hijack Pro (AHP)...it's about $35 USD for the license
The CoreAudio allows me to build in the correct filter and delay functions to create specific correction filters which go way beyond regular or parametric EQ for my speakers. I've actually been able to correct for time related issues. Not ready to share this part yet...
AHP allows me to divert the audio stream system-wide or input by input, apply processing and then feed it back into the application input, such as iTunes. It also allows for cool stuff like ripping directly from movie soundtracks. Works really well.
On my system I use iTunes, PureMusic or Amarra and AHP feeds into any one of them flawlessly. Doing all the processing in AHP has allowed me to compare between these different playback engines. Current favorite: Amarra 2.1, have not yet heard 2.2
Amazingly enough, the Behringer in digital I/O mode seems to be just as transparent as doing everything inside the Mac. Digital used to suck. Now it seems we're finally gettin' somewhere good!
The least expensive way I've found to access the CoreAudio functions on the Mac is by using Audio Hijack Pro (AHP)...it's about $35 USD for the license
The CoreAudio allows me to build in the correct filter and delay functions to create specific correction filters which go way beyond regular or parametric EQ for my speakers. I've actually been able to correct for time related issues. Not ready to share this part yet...
AHP allows me to divert the audio stream system-wide or input by input, apply processing and then feed it back into the application input, such as iTunes. It also allows for cool stuff like ripping directly from movie soundtracks. Works really well.
On my system I use iTunes, PureMusic or Amarra and AHP feeds into any one of them flawlessly. Doing all the processing in AHP has allowed me to compare between these different playback engines. Current favorite: Amarra 2.1, have not yet heard 2.2
Amazingly enough, the Behringer in digital I/O mode seems to be just as transparent as doing everything inside the Mac. Digital used to suck. Now it seems we're finally gettin' somewhere good!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.