I hardly ever hear any music recorded with a dummy-head (unfortunately). Like I said, if you want headphones to sound like loudspeakers, add lots of room interactions and diffraction effects (including the sound that bends around your head), phase anomalies from being a couple of centimetres further from the left speaker than from the right speaker... And that's not including crossovers and other limitations of loudspeakers.phase_accurate said:
Most stereo recordings are deliberately widened with huge phase delays and unrealistic amplitude differences just so that when played through loudspeakers they don't sound boringly mono. Play such compromised sound through headphones and of course they'll sound worse. My original point with the comparison was that ribbon tweeters aren't necessarily any cleaner than other tweeters, they just reduce many of the imaging problems caused by loudspeakers.
CM
Loudspeakers DO NOT necessarily cause imaging problems !!!
If a recording is produced to be played through loudspeakers (= intensity stereophony) then it is definitely fine to hear the signal from the left speaker with your right ear and vice-versa, since the system is RELYING on that !!!!
The electrostatic Stax headphones I mentioned before don't get even close to my speakers in terms of imaging. And I am usually listening in an acousttically very bad and reverberant living room. Before you say this might be the cause that I am fooled to beleive it is that way (because if the listening environment) I must disappoint you: I know they can do even better (i.e. I've tried it out).
Regards
Charles
If a recording is produced to be played through loudspeakers (= intensity stereophony) then it is definitely fine to hear the signal from the left speaker with your right ear and vice-versa, since the system is RELYING on that !!!!
The electrostatic Stax headphones I mentioned before don't get even close to my speakers in terms of imaging. And I am usually listening in an acousttically very bad and reverberant living room. Before you say this might be the cause that I am fooled to beleive it is that way (because if the listening environment) I must disappoint you: I know they can do even better (i.e. I've tried it out).
Regards
Charles
The best tweets are indeed plasma based.
The second best though is tricky - its not a good true ribbon, its a piezo by Audax. (but decompression can be an issue with these balloons depending on air pressure.)
VERY close third would be the TAD true ribbon. (hmm, sony made a really nice ribbon (silver) for the car about a decade ago - maybe that one would rank here as well.)
Then true ribbons
Then good planars
Then good electrostats
as for best for the money? probably the Fountek JP 2, but the problem here is that most will never realize its true potential because they pad the crap out of 'em (a real No-No).
To all would be ribbon utilizers - Do NOT resistivly pad a ribbon more than 2 dB. Instead use a transformer attenuator (and preferably a good one with a good core). Additionally when pairing with mid's always select a driver with a very low x-max and very low mass with and extended freq. response in the treble on an off-axis. i.e. your mid driver should probably span a range of 600 to 7kHz (crossed over with the ribbon at about 4.5 to 8kHz depending on the slope).
The second best though is tricky - its not a good true ribbon, its a piezo by Audax. (but decompression can be an issue with these balloons depending on air pressure.)
VERY close third would be the TAD true ribbon. (hmm, sony made a really nice ribbon (silver) for the car about a decade ago - maybe that one would rank here as well.)
Then true ribbons
Then good planars
Then good electrostats
as for best for the money? probably the Fountek JP 2, but the problem here is that most will never realize its true potential because they pad the crap out of 'em (a real No-No).
To all would be ribbon utilizers - Do NOT resistivly pad a ribbon more than 2 dB. Instead use a transformer attenuator (and preferably a good one with a good core). Additionally when pairing with mid's always select a driver with a very low x-max and very low mass with and extended freq. response in the treble on an off-axis. i.e. your mid driver should probably span a range of 600 to 7kHz (crossed over with the ribbon at about 4.5 to 8kHz depending on the slope).
There is only one plasma tweet that I know of offered for sale and its been mentioned already, (besides its absurdly expensive) - better off with the audax piezo.
JP2's are 118 US at madisound.com - they are a chinese ribbon very similar to the mid-sized arum cantus ribbon (for about a 3rd the price). 99 dB.
Because of baffle step considerations and x-max I would specifically utilize a true midrange crossed about 600 Hz in a sealed, or open baffle config. (depending on how steep the high pass slope is for it). My preference is the Audax Pro PR170Z0 aerogel at a reasonable cost (in fact this driver, like the JP2's is worth FAR more than the asking price). Also consider the other Pro Audax midranges, Focal's pro kevelar mid, and PHL-Audio's high eff. 6.5 offerings.
For either of the fostex units your talking about padding down the tweet a LOT. I'd prob. find another tweeter - if your running those drivers full-range then I'd just as soon use a cheap piezo and a cap (quite nice if crossed very high.. >15kHz).
I personally don't like biamping in the midrange. Also, if your going full range with a super tweet then there really is no reason for bi-amping. Now biamping for the bass is another matter - there you can get a better dampening factor which is preferable for the higher mass bass units - allowing you to use a low dampening factor amp on the mid and tweet.
JP2's are 118 US at madisound.com - they are a chinese ribbon very similar to the mid-sized arum cantus ribbon (for about a 3rd the price). 99 dB.
Because of baffle step considerations and x-max I would specifically utilize a true midrange crossed about 600 Hz in a sealed, or open baffle config. (depending on how steep the high pass slope is for it). My preference is the Audax Pro PR170Z0 aerogel at a reasonable cost (in fact this driver, like the JP2's is worth FAR more than the asking price). Also consider the other Pro Audax midranges, Focal's pro kevelar mid, and PHL-Audio's high eff. 6.5 offerings.
For either of the fostex units your talking about padding down the tweet a LOT. I'd prob. find another tweeter - if your running those drivers full-range then I'd just as soon use a cheap piezo and a cap (quite nice if crossed very high.. >15kHz).
I personally don't like biamping in the midrange. Also, if your going full range with a super tweet then there really is no reason for bi-amping. Now biamping for the bass is another matter - there you can get a better dampening factor which is preferable for the higher mass bass units - allowing you to use a low dampening factor amp on the mid and tweet.
ScottG said:
There are a number of reasons, one being that the tweeter doesn't have to be padded. the other that the amp can be optomized for HF.
IMO, a mid should go down to at least 300 Hz, and up beyond 5 k to get the XO out of the critical midrange. A wider range is even better. My bipoles will have no baffle-step. I;m hoping to get away with a 6-7k XO at the top (strict "MTM rules" say it should be closer to 4 k).
dave
My Aurum Cantus 2Si is rated exactly the same efficiency as my Fostex 208 E Sigma- at least if you don't count the resonance peak (I'll deal with that) we'll see in my Basszilla when it's running again.......
Also the Fostex is a very light and efficient and low x-max and will cover the wide range ScottG specifies. I think it might be the ticket. The only possible downside is maybe the dispersion is a bit narrow at high freqs since it is 8". However on the positive side, I can run it down to about 200 hz.
The Audax/ Fountek sounds like a very good combo. The Audax's have always been considred a bargain, and it does seem like they will match the ribbons well. Have you actually auditioned the combo?
Also the Fostex is a very light and efficient and low x-max and will cover the wide range ScottG specifies. I think it might be the ticket. The only possible downside is maybe the dispersion is a bit narrow at high freqs since it is 8". However on the positive side, I can run it down to about 200 hz.
The Audax/ Fountek sounds like a very good combo. The Audax's have always been considred a bargain, and it does seem like they will match the ribbons well. Have you actually auditioned the combo?
decent midrange
Thought I'd mention another alternative for the inclusion in the affordable and good sounding high efficiency mid category:
http://www.northernsound.net/Sales/speakers/raw/bc/woofers.html
Good things reported. Power handling/durability - pro use - no concern.
While I have not heard the various B&C drivers, I have seen the construction and it is very good.
Their larger woofers are also very well made.
Not suggesting as a substitute to the F103a, but an interesting 3 way could be managed with the ribbon for hf, one of these, and your 15" of choice.
Tim
Thought I'd mention another alternative for the inclusion in the affordable and good sounding high efficiency mid category:
http://www.northernsound.net/Sales/speakers/raw/bc/woofers.html
Good things reported. Power handling/durability - pro use - no concern.
While I have not heard the various B&C drivers, I have seen the construction and it is very good.
Their larger woofers are also very well made.
Not suggesting as a substitute to the F103a, but an interesting 3 way could be managed with the ribbon for hf, one of these, and your 15" of choice.
Tim
Further
The units of interest were the 6PE13 and the 8PE21, for example.
The feedback was on the 6", but the 8" looks good, too.
I still like the JBL 2123 10" mid, but dispersion is a not a good match for anything above 2K Hz or so.
BTW, I have a pair of the AC-G2SI, too, which await further crossover efforts or possible mid changes. I have a Altec Bi-Flex 12" 412C for these now, which measure pretty well out past 7K, with the exception of a peak in the response around 2K. Interesting driver with two cones joined at the hip, the small center cone of about 41/2" inserted into the the 12" with a suspension between.
Destined for open baffles.
Tim
The units of interest were the 6PE13 and the 8PE21, for example.
The feedback was on the 6", but the 8" looks good, too.
I still like the JBL 2123 10" mid, but dispersion is a not a good match for anything above 2K Hz or so.
BTW, I have a pair of the AC-G2SI, too, which await further crossover efforts or possible mid changes. I have a Altec Bi-Flex 12" 412C for these now, which measure pretty well out past 7K, with the exception of a peak in the response around 2K. Interesting driver with two cones joined at the hip, the small center cone of about 41/2" inserted into the the 12" with a suspension between.
Destined for open baffles.
Tim
Altec 16" Bi-Flex
Yes, the 12" is the same construction. Sounds very nice, with just a hint of the peak I mentioned. About 96dB for the 12", will pair well with the ribbons above 3K or so.
16" frame is substantial indeed. 604s used that big frame.
The driver on top of the big box (onken-style) is an Altec 416Z (16 ohm). Needs very large enclosure, but do not sound bad at all in the onkens at 7ft^3.
Tim
Yes, the 12" is the same construction. Sounds very nice, with just a hint of the peak I mentioned. About 96dB for the 12", will pair well with the ribbons above 3K or so.
16" frame is substantial indeed. 604s used that big frame.
The driver on top of the big box (onken-style) is an Altec 416Z (16 ohm). Needs very large enclosure, but do not sound bad at all in the onkens at 7ft^3.
Tim
ScottG said:
my goal is to make a system that has very very high WAF. that means it has to occupy very little or no floor space and be very unobtrusive. to do this i have made one compromise. the system need not operate above 100Hz. That will be managd by a sub
amplification would be via either a HT/AV amp or if I get it going a EL84 PPP AB1 amp (30W) so the system must have a sensitivity above 90db/1m/1w. I figure one i get 5 channels going i would get levels in the order of 95db/3W in most of the room (3m x 6m).
This is an extension of a query my wife put to me after we visited a friend who uses B&W VM1 (center and rear) and FPM5 (Left and Right) supported by a B&W 10" sub (with a 52" plasma and a Onkyo HT/AV amp) and some more ciritcal listening done via a Manger 2 way and NAD silverline electronics.
I've got the Audax (but not the JP2) - and it is excellent (purchased one for testing a couple of years ago but haven't gotten around to using it in a design yet (thinking omni for the Audax with a B&G planar right now but I don't have the time to do anything about it - maybe in a year?). And I've heard the Raven R2 (and the JP2 is probably quite similar), Though I haven't heard the Raven and the Audax together, I found them subjectivly similar.
as for how much bandwidth a driver should take depends a LOT on the crossover (and the avg. power input). While I agree on removing crossover artifacts from a design, it doesn't mean you can't have a crossover right in the middle of the midrange.. it just needs to be done right. In fact, to get that seemeless sound with outstanding drivers virtually requires a crossover in that region. (Manger is an exception.) Note however that what I consider a crossover may not be what you consider a crossover - I don't neccesarily mean the -3db point. (afterall is that acoustic or electrical.. or both?, - though it rarely is it both).
I've never heard an active crossover implementation that offered a seemless sound. There are a multitude of factors that will deny you this quality - simply saying I'll customize an amp for each driver is typically little more than wishfull thinking - what you'll end up with is hearing both drivers and both amps.. not music. (Of course this isn't to say you won't enjoy the end result.) Now just bi-amping (with a passive after the amp) *can* be good - look to Sakuma-San's work for more info.
So rather than continuing on with what I "wouldn't" do, I'll give you an example of what I would do and why.
Utilize the Audax driver I've mentioned in sealed configuration, with an aperiodic vent, 3db down at about 400 Hz using a low "source impeadance" amp (i.e. a tube amp - SET, or OTL, or SE-OTL with no global negative feedback). No electrical crossover (i.e. run "full-range").
Okay first of all why the driver?
VERY low mass relative to SD (or its surface area).
VERY low excursion.
VERY low distortion.
VERY clean decay. (and no break-up problems.)
VERY high force factor for such low mass.
Relativly high upper freq. limit.
Decent off-axis performance within 30 degrees.
All of these attributes are HIGHLY desirable when combining with a low mass tweeter because the tweeter typically shares similar attributes. Furthermore these attributes work exceptionally well with amplifers that have low dampening factors - allowing the amplifier to do its "thing" (modulate the driver with ease), and the driver its "thing" (modulate the air with ease). Subjectivly this allows more holographic imaging while still retaining exellent transparency. (interestingly the only drivers I've heard that buck the low mass trend with regard to these amps are accuton's.)
For crossover purposes utilize the driver at a 30 degree angle - *remember when loaded to the amplifier type mentioned, your driver's impeadance factors into to your gain*. Also remember that a line-source like a 5 inch ribbon will run a little "hotter" than measurement for real-word use. Effectivly then our low-pass "crossover" has a 3db point around 4.5 kHz (allowing for an acceptable first-order filter on the tweeter slightly above this freq.. provided your not looking for an ultimate spl speaker - which practically speaking is incongruant with the amplifer type chosen). For the high-pass section of this driver (i.e. the lower freq. crossover region) again remember to factor in gain via the amplifer and the driver's impeadance. Utilizing a small sealed enclosure with a theoretical 12 db rate will shift the driver's impeadance "bump" up - in this case we want it at about 500 Hz for the peak - so create the appropriate volume to achieve this. The rising impeadance will be further dampened (lowering the "Q") with the aperiodic vent. This should give us a real-world -3db point at about 400 Hz (which is about perfect with regard to baffle step - not just baffle boundry loss but also power delivery). The shape of the enclosure should be similar in size to your head for HRTF's (which naturally will experience baffle step loss).
NOTE: All of the above depends HEAVILY (for freq. linearity) on measuring the driver with the amplifier to be used. (My figures are obviously just guesstimates. If you want to get a theoretical figure for gain with "source impeadance" amps then look at Bruce Rosenblit's article on amplifier output impeadance.
A good match for a bass unit would be the Eminence Kappa Pro 10.
As for the 8 inch fostex - yup, by increasing the sd your allowing yourself lower freq.s for the same level of excusion. Remember though that transition between the drivers (at higher freq.s) will probably result in comb filtering, and its freq.s anomolies will likely be a problem (with regard to freq. linearity).. but I'm not sure you'll find it objectionable (i.e. try it and see if you like it).
bummer about the Audax piezo..
as for how much bandwidth a driver should take depends a LOT on the crossover (and the avg. power input). While I agree on removing crossover artifacts from a design, it doesn't mean you can't have a crossover right in the middle of the midrange.. it just needs to be done right. In fact, to get that seemeless sound with outstanding drivers virtually requires a crossover in that region. (Manger is an exception.) Note however that what I consider a crossover may not be what you consider a crossover - I don't neccesarily mean the -3db point. (afterall is that acoustic or electrical.. or both?, - though it rarely is it both).
I've never heard an active crossover implementation that offered a seemless sound. There are a multitude of factors that will deny you this quality - simply saying I'll customize an amp for each driver is typically little more than wishfull thinking - what you'll end up with is hearing both drivers and both amps.. not music. (Of course this isn't to say you won't enjoy the end result.) Now just bi-amping (with a passive after the amp) *can* be good - look to Sakuma-San's work for more info.
So rather than continuing on with what I "wouldn't" do, I'll give you an example of what I would do and why.
Utilize the Audax driver I've mentioned in sealed configuration, with an aperiodic vent, 3db down at about 400 Hz using a low "source impeadance" amp (i.e. a tube amp - SET, or OTL, or SE-OTL with no global negative feedback). No electrical crossover (i.e. run "full-range").
Okay first of all why the driver?
VERY low mass relative to SD (or its surface area).
VERY low excursion.
VERY low distortion.
VERY clean decay. (and no break-up problems.)
VERY high force factor for such low mass.
Relativly high upper freq. limit.
Decent off-axis performance within 30 degrees.
All of these attributes are HIGHLY desirable when combining with a low mass tweeter because the tweeter typically shares similar attributes. Furthermore these attributes work exceptionally well with amplifers that have low dampening factors - allowing the amplifier to do its "thing" (modulate the driver with ease), and the driver its "thing" (modulate the air with ease). Subjectivly this allows more holographic imaging while still retaining exellent transparency. (interestingly the only drivers I've heard that buck the low mass trend with regard to these amps are accuton's.)
For crossover purposes utilize the driver at a 30 degree angle - *remember when loaded to the amplifier type mentioned, your driver's impeadance factors into to your gain*. Also remember that a line-source like a 5 inch ribbon will run a little "hotter" than measurement for real-word use. Effectivly then our low-pass "crossover" has a 3db point around 4.5 kHz (allowing for an acceptable first-order filter on the tweeter slightly above this freq.. provided your not looking for an ultimate spl speaker - which practically speaking is incongruant with the amplifer type chosen). For the high-pass section of this driver (i.e. the lower freq. crossover region) again remember to factor in gain via the amplifer and the driver's impeadance. Utilizing a small sealed enclosure with a theoretical 12 db rate will shift the driver's impeadance "bump" up - in this case we want it at about 500 Hz for the peak - so create the appropriate volume to achieve this. The rising impeadance will be further dampened (lowering the "Q") with the aperiodic vent. This should give us a real-world -3db point at about 400 Hz (which is about perfect with regard to baffle step - not just baffle boundry loss but also power delivery). The shape of the enclosure should be similar in size to your head for HRTF's (which naturally will experience baffle step loss).
NOTE: All of the above depends HEAVILY (for freq. linearity) on measuring the driver with the amplifier to be used. (My figures are obviously just guesstimates. If you want to get a theoretical figure for gain with "source impeadance" amps then look at Bruce Rosenblit's article on amplifier output impeadance.
A good match for a bass unit would be the Eminence Kappa Pro 10.
As for the 8 inch fostex - yup, by increasing the sd your allowing yourself lower freq.s for the same level of excusion. Remember though that transition between the drivers (at higher freq.s) will probably result in comb filtering, and its freq.s anomolies will likely be a problem (with regard to freq. linearity).. but I'm not sure you'll find it objectionable (i.e. try it and see if you like it).
bummer about the Audax piezo..
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.