Hi gurus...
I've got CSS FR125S mounted in a pair of 2002 Mission 780 7 liter bookshelves. Although I like the timbre and resolution, the imaging in my small concrete-walled room of 9' x 11' x 9' (depth, width, height) seems less than desirable.
So I've been wondering if a project like Tom Zurowski's PAWO would give me the magic imaging. http://www.audio-resolution.com/zhorn/fr125s.html
I've read in many articles saying that narrow baffle would give pin-point imaging. But a wide baffle is easier to place since it is reported that these speakers don't mind being close to walls. Since I'm listen in a small room, I'd like to use them against the walls. Hence the Bipolar MLTL, with placement requirement to be well off the walls, would not be something I'd like to experiment with at the moment. So what would be my best option to obtain good soundstaging? Narrow or wide baffle?
Thanks in advance for any opinions!
Cheers!
I've got CSS FR125S mounted in a pair of 2002 Mission 780 7 liter bookshelves. Although I like the timbre and resolution, the imaging in my small concrete-walled room of 9' x 11' x 9' (depth, width, height) seems less than desirable.
So I've been wondering if a project like Tom Zurowski's PAWO would give me the magic imaging. http://www.audio-resolution.com/zhorn/fr125s.html
I've read in many articles saying that narrow baffle would give pin-point imaging. But a wide baffle is easier to place since it is reported that these speakers don't mind being close to walls. Since I'm listen in a small room, I'd like to use them against the walls. Hence the Bipolar MLTL, with placement requirement to be well off the walls, would not be something I'd like to experiment with at the moment. So what would be my best option to obtain good soundstaging? Narrow or wide baffle?
Thanks in advance for any opinions!
Cheers!
Hi Bro,
Your room is similar to mine. I ended up with the front mains in the corners, at 45 deg. I biult triagular cabinets and they sit right in the corner. The center of the driver is about 1.4 metres from the floor.
This configuration gives a wide sweetspot, and the axis cross in front of the usual listening area. Imaging is good, with no need for centre fill.
If you use a subwoofer, place it in one of the front corners, not in the centre.
My baffles are 450mm wide. Is that wide or narrow? with a 10" driver, it's average in my book. Because it's in a corner, it doesn't matter.
Geoff.
Your room is similar to mine. I ended up with the front mains in the corners, at 45 deg. I biult triagular cabinets and they sit right in the corner. The center of the driver is about 1.4 metres from the floor.
This configuration gives a wide sweetspot, and the axis cross in front of the usual listening area. Imaging is good, with no need for centre fill.
If you use a subwoofer, place it in one of the front corners, not in the centre.
My baffles are 450mm wide. Is that wide or narrow? with a 10" driver, it's average in my book. Because it's in a corner, it doesn't matter.
Geoff.
I'd go with wide. Imaging isn't quite as precise, but how many times do you get pin-point imaging at a live concert? IMO, wide baffles provide the more realistic sound, plus, if you design them right, you can push baffle-step down to the point where room-gain takes over, so if you design in a slightly more powerful bass than you'd normally have, you can compensate for it that way, no BSC circuits needed.
There is no noticable baffle step. The sides are against the walls. The left cabinet is on the desk, the right is on a bookshelf. So the only physical place for a step to occour is from the driver to the top of the cabinet. 260 mm from the centre of the driver. Anything that spills over the top will reflect back into the listening area. Which would occur at about 230 Hz, and only 3db at the most. At that frequency there would be slight rise in radiation restistance of the cone (due to diameter/wavelength) before it falls away.
The bass is generated under the desk. It's fairly good, but struggles to keep up to the effiency of the Wharfedales. I have plans in that area, but I need to secure a panel in the back of the desk first - it rattles.
And then rejuvinate the big amp in the workshop.
In regard to imaging and baffle width, the driver plays an important role too. Deep apex and ribbed cones seem to give beter control than curvilinear cones. Less flex in the cone.
Cheers,
Geoff
The bass is generated under the desk. It's fairly good, but struggles to keep up to the effiency of the Wharfedales. I have plans in that area, but I need to secure a panel in the back of the desk first - it rattles.
And then rejuvinate the big amp in the workshop.
In regard to imaging and baffle width, the driver plays an important role too. Deep apex and ribbed cones seem to give beter control than curvilinear cones. Less flex in the cone.
Cheers,
Geoff
Good to see you're having success. I also notice you're using a sub. That's interesting as it must give you a degree of control.
Have you noticed any concerns with the corner behaving as a waveguide at the lows? Apart from room mode issues, do you feel it is feasible to create a system that fits right into the corner and also extends right down to 20Hz?Scottmoose said:
I never put my WB designs in a corner so I can't comment on that from personal experience I'm afraid.
From a general perspective, yes, of course it's possible to design for corner loading down to low frequencies. In fact, it's a positive benefit. There've been plenty of speaker designs for just this purpose, both DIY and commerical -think of all those huge Tannoy corner cabinets for example. For myself, I've built corner horns for a friend that were ~flat to 18Hz, and they work well to this day. These use the corners as a waveguide, and also to increase mouth are and acoustic impedence, while reducing pressure. Active Eq is a good idea as every room is different, but if you don't like this approach, you can go one of two ways. Engineer a smooth roll-off at a fairly high frequency into the design to allow room-gain to lift it ~flat. Or, design too much gain into it, and then damp it down. I always used to go with the former, but I generally now follow GM's advice to go with too much, as it;s always easier to damp down and over-eager bass than to boost something you don't have in the first place.
Regards
Scott
From a general perspective, yes, of course it's possible to design for corner loading down to low frequencies. In fact, it's a positive benefit. There've been plenty of speaker designs for just this purpose, both DIY and commerical -think of all those huge Tannoy corner cabinets for example. For myself, I've built corner horns for a friend that were ~flat to 18Hz, and they work well to this day. These use the corners as a waveguide, and also to increase mouth are and acoustic impedence, while reducing pressure. Active Eq is a good idea as every room is different, but if you don't like this approach, you can go one of two ways. Engineer a smooth roll-off at a fairly high frequency into the design to allow room-gain to lift it ~flat. Or, design too much gain into it, and then damp it down. I always used to go with the former, but I generally now follow GM's advice to go with too much, as it;s always easier to damp down and over-eager bass than to boost something you don't have in the first place.
Regards
Scott
I've been contemplating an infinite baffle and these replies have been encouraging, thanks.
I feel as if using the corner as an extension to the baffle would give an effect similar to the common baffle step, only in reverse. If a common baffle step causes a loss of 6dB of SPL due to the transition from half-space to full-space radiation (before the room's boundaries come in at the low end), then, corner loading would be like going from half space (on the baffle) to quarter space with an increase of 6dB (before the ceiling and floor come in).
If I'm not mistaken, room gain is an independent phenomena. I figure that if I make the baffle large enough so that the step happens low, it may meet up with the room gain and produce an interesting inverse rolloff that could be matched. (I wish there was more info about on room gain. )
I've been contemplating a baffle of a few feet across, but haven't done the math yet. I have thought of hanging a transmission line behind it. From the work done in recent years on TLs, it would seem they can be quite happy set up for a lean response, even working far above the driver's fs if desired.
I will do a search for GM's contribution to this, and look forward to reading it.
I feel as if using the corner as an extension to the baffle would give an effect similar to the common baffle step, only in reverse. If a common baffle step causes a loss of 6dB of SPL due to the transition from half-space to full-space radiation (before the room's boundaries come in at the low end), then, corner loading would be like going from half space (on the baffle) to quarter space with an increase of 6dB (before the ceiling and floor come in).
If I'm not mistaken, room gain is an independent phenomena. I figure that if I make the baffle large enough so that the step happens low, it may meet up with the room gain and produce an interesting inverse rolloff that could be matched. (I wish there was more info about on room gain.
)I've been contemplating a baffle of a few feet across, but haven't done the math yet. I have thought of hanging a transmission line behind it. From the work done in recent years on TLs, it would seem they can be quite happy set up for a lean response, even working far above the driver's fs if desired.
I will do a search for GM's contribution to this, and look forward to reading it.
Corner cabinets have plenty of advantages. Rigidity being one. Everything is triagulated. The rear panels are at 45 deg to the baffle: min internal reflection. I haven't loaded any dampening material yet. Don't think I need to. And you are radiating into 90 degrees, not 180.
I had the bass in the centre of the wall, couldn't cancel the phase shift.
Standing waves usually occur in the corners of a room. Placing the bass in a corner reduces that also.
Wharfedale had a few models based on corner cabs. One was for the 12W and W3 combination, and one for the Super 10RSDD. A ducted port in the corner bled into the absolute corner of the room. The cabinets were on legs about 6 inches high. My old man had a recording of a pipe organ somewhere in the US, in a concert hall. The bass pedals rattled windows. And from a Pioneer amp with P-P 6BM8s providing 3 watts per channel. I still have those.
The pair I currently use are modified - shorter voice coils to improve efficiency and transient response further, and stiffer spiders, and a 3" dome radiator coupled to the voice coil. I used to x-over at 500 hz @6db/oct, with a 10dB attenuater paired to Altec Lansing 4118As. (93dB/w) They are quite happy at 180 Hz@12dB/oct.
Geoff
I had the bass in the centre of the wall, couldn't cancel the phase shift.
Standing waves usually occur in the corners of a room. Placing the bass in a corner reduces that also.
Wharfedale had a few models based on corner cabs. One was for the 12W and W3 combination, and one for the Super 10RSDD. A ducted port in the corner bled into the absolute corner of the room. The cabinets were on legs about 6 inches high. My old man had a recording of a pipe organ somewhere in the US, in a concert hall. The bass pedals rattled windows. And from a Pioneer amp with P-P 6BM8s providing 3 watts per channel. I still have those.
The pair I currently use are modified - shorter voice coils to improve efficiency and transient response further, and stiffer spiders, and a 3" dome radiator coupled to the voice coil. I used to x-over at 500 hz @6db/oct, with a 10dB attenuater paired to Altec Lansing 4118As. (93dB/w) They are quite happy at 180 Hz@12dB/oct.
Geoff
Hello Indm,
It will all balance out in the wash. As the wavelength gets longer and spill occurs - where? into the room.
That eliminates the loss you would otherwise have. The cone gets a bit more resistance to work into, so the impeadance doesn't rise as quick towards resonance.
Already starting to flatten the Z curve, and haven't decided on the cabinet.
Did you say the resonance of your drivers is 38 Hz? If so I'd go for TLs.
In the corners, you only need 1 panel, the baffle, cut it at 90" and wide enough so the TLs x-sectional area is that of the piston area of the cone. Seal it at floor level, and the walls. Then experiment with the dampening.
Well, thats one option.
Cheers.
PS, did you see my post on "plastic tls"
It will all balance out in the wash. As the wavelength gets longer and spill occurs - where? into the room.
That eliminates the loss you would otherwise have. The cone gets a bit more resistance to work into, so the impeadance doesn't rise as quick towards resonance.
Already starting to flatten the Z curve, and haven't decided on the cabinet.
Did you say the resonance of your drivers is 38 Hz? If so I'd go for TLs.
In the corners, you only need 1 panel, the baffle, cut it at 90" and wide enough so the TLs x-sectional area is that of the piston area of the cone. Seal it at floor level, and the walls. Then experiment with the dampening.
Well, thats one option.
Cheers.
PS, did you see my post on "plastic tls"
I'd advise reading Martin King's alignment tables (& all his other articles for that matter) before embarking on a TL. see www.quarter-wave.com
Hehe, great
I was once so into these, oh, maybe ten years ago. I had a stock of 100mm pipe, corner pieces and splitters used for changing to dual pipes to effectively taper the line. Used to have the stuff everywhere. My wife used to think they were a little, well, eccentric but she didn't let on . I used 4" Magnavox paper cones that were almost full range. They were only $25 from Dick Smith but they were also used at that time in an MTM by a professional mob I can't recall just now, and sold for many 100's.
The baffle step range I often encounter seems largely to be from the low 100's of Hz to the low 1000's. I think in my design I would like to bring this down a little by having the flat part of the baffle a little larger than I usually use. I'm assuming the frequencies would be the same only the step would be reversed (a reasonable assumption I presume).
My woofer's fs is 28Hz. The standard TL, IIRC rolls off at 60 odd Hz. I have seen other design methods that provide higher cut-offs if it turns out that I need that. It feels good that I don't need to look at extended alignments as if there is no other way to get bass.
Edit: Scottmoose, thanks, I will
I was once so into these, oh, maybe ten years ago. I had a stock of 100mm pipe, corner pieces and splitters used for changing to dual pipes to effectively taper the line. Used to have the stuff everywhere. My wife used to think they were a little, well, eccentric but she didn't let on
. I used 4" Magnavox paper cones that were almost full range. They were only $25 from Dick Smith but they were also used at that time in an MTM by a professional mob I can't recall just now, and sold for many 100's.The baffle step range I often encounter seems largely to be from the low 100's of Hz to the low 1000's. I think in my design I would like to bring this down a little by having the flat part of the baffle a little larger than I usually use. I'm assuming the frequencies would be the same only the step would be reversed (a reasonable assumption I presume).
My woofer's fs is 28Hz. The standard TL, IIRC rolls off at 60 odd Hz. I have seen other design methods that provide higher cut-offs if it turns out that I need that. It feels good that I don't need to look at extended alignments as if there is no other way to get bass.
Edit: Scottmoose, thanks, I will
You won't regret it. Forget the old rules of thumb -they're just that; old rules, now superceeded, and there's no reason to use them anymore. Martin's alignment tables provide the first step (Augsperger's did something similar, but he left it there; Martin has continued working in the field), his MathCad worksheets allow the accurate (and I mean accurate) modelling of QW and TL enclosures. Guesswork does not come into it anymore. Thank goodness.
Hi Scottmoose,
Some old rules of thumb don't change. The velocity of sound through
the atmosphere being one.
Martin"s (and other's) work is what keeps this technology advancing.
But we don't have to follow the leader. Take on board what they have discoverd, and see if we can't improve on it.
That's why I am playing with my miniture TL's. I don't need another set of speakers. The only measurements I have used on this project are from a tape measure and vernier calipers. My guess at the Fr was wrong, seems like I was 1/2 an octave high. Found that with a tunable TL, music and ears.
Still don't know what the Fr is, but I know the 1/4 wavelength in the pipe I am using. The velocity of sound in a pipe is faster than that in the open. How much faster depends on the diameter.
When I do start measurements, it will be with square waves, not sine.
Cheers
Some old rules of thumb don't change. The velocity of sound through
the atmosphere being one.
Martin"s (and other's) work is what keeps this technology advancing.
But we don't have to follow the leader. Take on board what they have discoverd, and see if we can't improve on it.
That's why I am playing with my miniture TL's. I don't need another set of speakers. The only measurements I have used on this project are from a tape measure and vernier calipers. My guess at the Fr was wrong, seems like I was 1/2 an octave high. Found that with a tunable TL, music and ears.
Still don't know what the Fr is, but I know the 1/4 wavelength in the pipe I am using. The velocity of sound in a pipe is faster than that in the open. How much faster depends on the diameter.
When I do start measurements, it will be with square waves, not sine.
Cheers
Scottmoose,
I looked at a comparison that MJK showed for measured vs calculated. The similarity is truly uncanny. Room gain, well, I may be able to best guess that.
The state of the art in tls really has moved in the last few years
Do you have a preference for the type of quarter wave enclosure to use?
Geoff H,
Your view is a bold one. Getting your hands dirty is certainly a good way to learn the old school practical knowledge. I can respect that
I looked at a comparison that MJK showed for measured vs calculated. The similarity is truly uncanny. Room gain, well, I may be able to best guess that.
The state of the art in tls really has moved in the last few years
Do you have a preference for the type of quarter wave enclosure to use?
Geoff H,
Your view is a bold one. Getting your hands dirty is certainly a good way to learn the old school practical knowledge. I can respect that
Good question Martin.
One of the physical aspects is pressure. Speed of sound at sea level (14psi) is said to be 1129 feet/second.
In the open atmosphere, that pressure wave expands into a 3 dimensional atmosphere.
A similar presure wave in a pipe can only expand in one direction, away from the source. That results in an increase in speed.
A similar thing can be observered with wave action on a shore. The swell travels at a given speed, usually about 14 Knots. As it approaches less depth, the wave stands higher due to the pressure against the sea bed, and accellerates to the shore, then breaks. A wave can easilly double it's speed before it breaks.
If you are tuning an exhaust system on a car (which is only sound or pressure waves) you allow for 1500 feet per second - much higher pressure involved - about 80 PSI at the exhaust valve.
Well done on your work by the way. What you have done for TLs is what Thiele and Small did for bass reflex. Removed a lot of uncertainty and replaced it with predictability.
Geoff
One of the physical aspects is pressure. Speed of sound at sea level (14psi) is said to be 1129 feet/second.
In the open atmosphere, that pressure wave expands into a 3 dimensional atmosphere.
A similar presure wave in a pipe can only expand in one direction, away from the source. That results in an increase in speed.
A similar thing can be observered with wave action on a shore. The swell travels at a given speed, usually about 14 Knots. As it approaches less depth, the wave stands higher due to the pressure against the sea bed, and accellerates to the shore, then breaks. A wave can easilly double it's speed before it breaks.
If you are tuning an exhaust system on a car (which is only sound or pressure waves) you allow for 1500 feet per second - much higher pressure involved - about 80 PSI at the exhaust valve.
Well done on your work by the way. What you have done for TLs is what Thiele and Small did for bass reflex. Removed a lot of uncertainty and replaced it with predictability.
Geoff
Hi Martin, more info on the velocity, and I forgot heat, but that doesn't concern us.
http://www.engineeringtoolbox.com/speed-sound-d_82.html
Don't loose sleep over this, I have been known to pedantic at times.
Some would say all the time. I regard it as a compliment.
Geoff
http://www.engineeringtoolbox.com/speed-sound-d_82.html
Don't loose sleep over this, I have been known to pedantic at times.
Some would say all the time. I regard it as a compliment.
Geoff
I doubt Martin is planning to loose any sleep over it Geoff. He has a rather formidable engineering background himself as I imagine his subsequent posts will reveal...
I was refering to the ROT regarding TL design, most of which are now proven to be utterly incorrect, not basic physics. The moving fibre theory for one (groan). Bailey & Bradbury have much to answer for. I fail to see any reason at all why anyone would ever bother to use them, when they can create a far superior system using the application of scientific principles rather than guesswork.
Indm -depends on the driver you're using, and your design goals. Time for a brief resume.
There is a dramatic difference between what is called a Transmission Line (an electrical, not an acoustic term) and a Quarter Wave Resonator. The latter types actually came first; no surprise given that all enclosures that are open at one end and sealed at the other resonate at 1/4 of a frequency which is determined by both length and taper. Straight lines are the only ones where you can simply take a frequency and /4 to get a length. A positive taper, expanding line is a horn, or a variation on that theme, and these need to be considerably longer than a straight line for the same fundametal resonant frequency. Conversely, a negative taper, narrowing line can be considerably shorter. These resonant enclosures are intended to prop up the bass of a driver, and the damping placed in them is there only to absorb the unwanted higher harmonic resonances.
Enter Bailey, who created a massively stuffed pipe, and called it a transmission line. There the design object was completely different; the goal being to damp the entire rear wave of the driver, and to flatten the impedence curve. A worthy goal, but hardly the same thing. Most of the time they were 1/4 wavelength of the tuning frequency in length, so for some reason the monker TL was applied to all quarter-wave cabinets, causing much confusion, particularly as the original TLs were based upon decidedly dodgy physics now proven to be erroneous.
A lot of classic TLs from the heyday of the type, the 1970s, were actually a hybrid between the two variations of resonant and non-resonant, though they tended to lean more toward the physics of the latter while trying to preserve some of the bass reinforcement of the former. Not an easy proceedure using simple ROT.
Stuffing a TL does not increase the effective length of the line, contrary to popular myth, unless you cram it so full of damping that the seams burst; and I'm not entirely sure why anyone would want to do that. FWIW, I have a suspicion that the belief that stuffing increases the effective length of a line stems from a time when it was not realised a negative taper, narrowing line (which were the most popular in the 1970s) will resonate at a lower frequency than a straight line of the same length. So the stuffing got the credit.
Do I have a favourite? Yes. Horns (yup, they're QWRs), or ML TQWTs or MLTLs (actually MLQWRs) but that's more to do with the type of drivers I work with than anything else. A resonant enclosure is useful to support the LF regions of a lowish Q driver, which usually need all the help they can get down there. A non resonant line conversely is excellent if you want a well damped system, electrically and acoustically, and is often better for higher Q designs. Generally, you use the cabinet best suited to what you want to achieve. If you're using a traditional, damped TL, or a hybrid between the two resonant and non-resonant types, I'd advise a negative taper line, with either a 3:1 or a 6:1 ratio to keep line length reasonable. If you're going with a hybrid between resonant and non-resonant, leave the last 10in of the line empty, with no stuffing applied. Naturally, everything I do is modelled in Martin's MathCad software, which has no rivals, and is of proven accuracy.
Hope this is of use
Scott
I was refering to the ROT regarding TL design, most of which are now proven to be utterly incorrect, not basic physics. The moving fibre theory for one (groan). Bailey & Bradbury have much to answer for. I fail to see any reason at all why anyone would ever bother to use them, when they can create a far superior system using the application of scientific principles rather than guesswork.
Indm -depends on the driver you're using, and your design goals. Time for a brief resume.
There is a dramatic difference between what is called a Transmission Line (an electrical, not an acoustic term) and a Quarter Wave Resonator. The latter types actually came first; no surprise given that all enclosures that are open at one end and sealed at the other resonate at 1/4 of a frequency which is determined by both length and taper. Straight lines are the only ones where you can simply take a frequency and /4 to get a length. A positive taper, expanding line is a horn, or a variation on that theme, and these need to be considerably longer than a straight line for the same fundametal resonant frequency. Conversely, a negative taper, narrowing line can be considerably shorter. These resonant enclosures are intended to prop up the bass of a driver, and the damping placed in them is there only to absorb the unwanted higher harmonic resonances.
Enter Bailey, who created a massively stuffed pipe, and called it a transmission line. There the design object was completely different; the goal being to damp the entire rear wave of the driver, and to flatten the impedence curve. A worthy goal, but hardly the same thing. Most of the time they were 1/4 wavelength of the tuning frequency in length, so for some reason the monker TL was applied to all quarter-wave cabinets, causing much confusion, particularly as the original TLs were based upon decidedly dodgy physics now proven to be erroneous.
A lot of classic TLs from the heyday of the type, the 1970s, were actually a hybrid between the two variations of resonant and non-resonant, though they tended to lean more toward the physics of the latter while trying to preserve some of the bass reinforcement of the former. Not an easy proceedure using simple ROT.
Stuffing a TL does not increase the effective length of the line, contrary to popular myth, unless you cram it so full of damping that the seams burst; and I'm not entirely sure why anyone would want to do that. FWIW, I have a suspicion that the belief that stuffing increases the effective length of a line stems from a time when it was not realised a negative taper, narrowing line (which were the most popular in the 1970s) will resonate at a lower frequency than a straight line of the same length. So the stuffing got the credit.
Do I have a favourite? Yes. Horns (yup, they're QWRs), or ML TQWTs or MLTLs (actually MLQWRs) but that's more to do with the type of drivers I work with than anything else. A resonant enclosure is useful to support the LF regions of a lowish Q driver, which usually need all the help they can get down there. A non resonant line conversely is excellent if you want a well damped system, electrically and acoustically, and is often better for higher Q designs. Generally, you use the cabinet best suited to what you want to achieve. If you're using a traditional, damped TL, or a hybrid between the two resonant and non-resonant types, I'd advise a negative taper line, with either a 3:1 or a 6:1 ratio to keep line length reasonable. If you're going with a hybrid between resonant and non-resonant, leave the last 10in of the line empty, with no stuffing applied. Naturally, everything I do is modelled in Martin's MathCad software, which has no rivals, and is of proven accuracy.
Hope this is of use
Scott
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.