Room is an open room around 7m x 10m (23ft x 32ft). It is a living room/ home office..... it's complicated.
I have my music on the entire day to keep me sane ( so music is important). It's most prominently on at night when everyone is asleep. It's definitely more worthwhile importing, in Singapore there isn't a shop that specialises in these kinds of things. I have heard Tannoys and I really love them, and in an ideal world I would go for a westminster, but I have yet to locate said ideal world.
Acoustics are poor in this room. I listen to mostly jazz. Acid/ funk in the day, classic vocals (ella fitzegerald/ nina simone, Miles Davis) at night.
I have my music on the entire day to keep me sane ( so music is important). It's most prominently on at night when everyone is asleep. It's definitely more worthwhile importing, in Singapore there isn't a shop that specialises in these kinds of things. I have heard Tannoys and I really love them, and in an ideal world I would go for a westminster, but I have yet to locate said ideal world.
Acoustics are poor in this room. I listen to mostly jazz. Acid/ funk in the day, classic vocals (ella fitzegerald/ nina simone, Miles Davis) at night.
Last edited:
Hi, Buffedupboy!
What kfringe and others are alluding to is coaxials can be extremely difficult to design around ... if you care about flat response, freedom from coloration, things like that.
The big problem is the horn in most coaxials, including the 604 Duplex, is too small for the job. The horn also blocks and obscures the cone, which creates unwanted reflections between the outside of the horn and the surface of the cone. This results in two problems: the cone has additional roughness in response that wouldn't be there in a non-coaxial speaker, and the too-small horn is forced to run right through its cutoff region.
So both drivers have severe variations in response that would not be there in an equivalent, non-coaxial pair of drivers. If the speaker is not coaxial, the horn can be increased in size by 1.5X to 2X, so it's not forced into the cutoff region, where response is rough and directivity control is lost.
If the horn is removed from the middle of the cone, the response immediately smooths out, since a very large reflector has been removed. Cones are not improved when you put big, reflective, random-shaped objects in the middle of them.
Most awkward of all, the crossover falls right in the middle of the trouble zone. Peak-to-dip response variations of 10 to 15 dB are not unusual when you look carefully at each of the drivers by itself. That's why published responses of coax drivers from prosound manufacturers typically show the summed response, not the awful truth of the drivers by themselves.
Given these severe constraints on coaxial drivers, the horn that is used has a critical effect on the crossover region. If it has even slightly better performance in the cutoff region, that makes crossover design considerably easier. Changes in the exterior shape of the horn also affect what the woofer cone is doing, in an unpredictable way.
GM, not me, is the expert here. If he says the multicell version of the 604 is the way to go, I would take him seriously. Physically, a short exponential horn with a few vanes in it may look like a multicell, but the performance is completely different than the real thing. Since a coax, by design, must extract every last inch of performance out of the compact horn, even a small improvement is significant.
It might sound faintly ridiculous, but the best solution for most coaxial drivers would be an additional midrange driver, so neither the bass driver nor horn is used in the troublesome 800 Hz to 2.4 kHz range. But a 7" to 10" mid driver sitting above (or below) a coax would look kind of stupid, and you'd never be able to explain to your audiophile friends what the funny-looking driver is actually doing.
What kfringe and others are alluding to is coaxials can be extremely difficult to design around ... if you care about flat response, freedom from coloration, things like that.
The big problem is the horn in most coaxials, including the 604 Duplex, is too small for the job. The horn also blocks and obscures the cone, which creates unwanted reflections between the outside of the horn and the surface of the cone. This results in two problems: the cone has additional roughness in response that wouldn't be there in a non-coaxial speaker, and the too-small horn is forced to run right through its cutoff region.
So both drivers have severe variations in response that would not be there in an equivalent, non-coaxial pair of drivers. If the speaker is not coaxial, the horn can be increased in size by 1.5X to 2X, so it's not forced into the cutoff region, where response is rough and directivity control is lost.
If the horn is removed from the middle of the cone, the response immediately smooths out, since a very large reflector has been removed. Cones are not improved when you put big, reflective, random-shaped objects in the middle of them.
Most awkward of all, the crossover falls right in the middle of the trouble zone. Peak-to-dip response variations of 10 to 15 dB are not unusual when you look carefully at each of the drivers by itself. That's why published responses of coax drivers from prosound manufacturers typically show the summed response, not the awful truth of the drivers by themselves.
Given these severe constraints on coaxial drivers, the horn that is used has a critical effect on the crossover region. If it has even slightly better performance in the cutoff region, that makes crossover design considerably easier. Changes in the exterior shape of the horn also affect what the woofer cone is doing, in an unpredictable way.
GM, not me, is the expert here. If he says the multicell version of the 604 is the way to go, I would take him seriously. Physically, a short exponential horn with a few vanes in it may look like a multicell, but the performance is completely different than the real thing. Since a coax, by design, must extract every last inch of performance out of the compact horn, even a small improvement is significant.
It might sound faintly ridiculous, but the best solution for most coaxial drivers would be an additional midrange driver, so neither the bass driver nor horn is used in the troublesome 800 Hz to 2.4 kHz range. But a 7" to 10" mid driver sitting above (or below) a coax would look kind of stupid, and you'd never be able to explain to your audiophile friends what the funny-looking driver is actually doing.
Last edited:
for my twisted mind both are certainly same fun ( even if I saw just one pair of 604 in entire life) , but for novice , I can bet that Tannoys are more friendly to implement and more HiFi-sh , at least slightly
Ah... Tannoys. I don't think they deserve the blanket recommendation.
If you're looking to DIY, there's no source as convenient as GPA on the Tannoy side. If you can get a pair of HPD 385a over to Lockwood for servicing and the hard edge treatment, sure, it's a great idea. If the crossovers are still good, it's a great idea. Blind ordering a couple of random Tannoys of dubious provenance off the internet? Designing your own crossovers and enclosures? As a first project? That's going to be a bad trip.
I say that with love, of course.
BINGO! and trust me.... I walked down that road and turned back immediately.
Leaning towards the 8H ever more now....
Leaning towards the 8H ever more now....
Hah. You should see my room!
But seriously, you can bet a 604 is going to choke on that funk. GM may have a trick or seven about cabinets, room loading, and tricks to get the system to sound respectable, but, short of that, the 604 is not going to give you the fat bottom that you'll want.
Tulip WG series is (almost ) safe bet , regarding Tanns
ceramic magnet , xover much simpler then for Pepperpot WG , difference in sound vs. old ones important only to Tann worshipers
, in reality being strictly on personal taste level
however , I'm not advocating them - merely mentioned them as viable choice to consider , regarding possible difference in buying from Eu or from USA
though , it's probably irrelevant difference , looking from Singapore
ceramic magnet , xover much simpler then for Pepperpot WG , difference in sound vs. old ones important only to Tann worshipers
, in reality being strictly on personal taste levelhowever , I'm not advocating them - merely mentioned them as viable choice to consider , regarding possible difference in buying from Eu or from USA
though , it's probably irrelevant difference , looking from Singapore
Last edited:
Agreed (on my sixth pair now). I don't even mind the plastic cones from the little DMTs all that much.
I may have a coax problem. It's a personality flaw.
If you're listening to funk, go for the alnico. That extra bass response is going to make a difference there.
Ok, I'm dead set on the 8H-III and onwards to designing the cabinet. Remember, I have no idea what I am doing, I have no experience in electrical/ electronics.
1) If I purchase the crossover from GPA, do they come fully soldered/ assembled? Or will I need to mount the parts on some type of circuit board?
2) Where will said crossover sit? Am I to mount it on the back or the floor of the cab?
3) I'm going to mount the driver from inside the cab (don't wanna see screws). Any disadvantage to this? (I'm going to chamfer the edge of the opening).
4) The current thought is to have 3/4" ply on all sides and 1" on the front panel. Is there any benefit to have it 2" on the front?
5) I'm planning to have a bassport cut on the front in a rectangle, much like the 612 cab. Will I be crazy if I had an internal ring around the bassport to increase the depth of the opening much like a bass pipe to increase the bass output?
6) Will bevelling/ chamfering the bassport have any difference to the sound?
7) I've seen different people add fill to different parts of the cab. I'm planning to add fill to one side, the back and the top of the inside of the cab. I assume there's no hard and fast rule to this?
Thanks to everyone for getting me this far....
1) If I purchase the crossover from GPA, do they come fully soldered/ assembled? Or will I need to mount the parts on some type of circuit board?
2) Where will said crossover sit? Am I to mount it on the back or the floor of the cab?
3) I'm going to mount the driver from inside the cab (don't wanna see screws). Any disadvantage to this? (I'm going to chamfer the edge of the opening).
4) The current thought is to have 3/4" ply on all sides and 1" on the front panel. Is there any benefit to have it 2" on the front?
5) I'm planning to have a bassport cut on the front in a rectangle, much like the 612 cab. Will I be crazy if I had an internal ring around the bassport to increase the depth of the opening much like a bass pipe to increase the bass output?
6) Will bevelling/ chamfering the bassport have any difference to the sound?
7) I've seen different people add fill to different parts of the cab. I'm planning to add fill to one side, the back and the top of the inside of the cab. I assume there's no hard and fast rule to this?
Thanks to everyone for getting me this far....
8H-III good choice.
Yes, the GPA crossover is ready to connect to the driver and go.
Mount the XO to the floor of the cab, use a resilient mount so the cabinet vibes are somewhat isolated, but don't fret over it.
No sonic disadvantage to inside mounting other than the difficulty in mechanically mounting it to the box. Make the edge rounded or something to get rid of the sharp outside edge. You'll need a large enough removable access panel to install the driver, which has the potential to vibrate (make noise) when playing, so think about the best approach to that.
The benefit of thicker baffle is to stiffen the front baffle and add mass, both of which help reduce the audible vibrations of the front panel.
3/4" plywood box will make audible contribution to the sound in such a large box, because the woofer plays in midrange freqs, through the resonant frequencies of the plywood. Consider laminating a layer of MDF to the inside surface of your plywood box using "green glue" damping adhesive between the layers of wood. This will quiet it down a lot. Some internal "window" bracing will help too.
A bass reflex box is a system that mechanically amplifies the LF by using a controlled resonance of the air in the box through a port - the air rings at a designed frequency. So the box volume, the port area and the port length all three are critical in getting the ringing to work as desired. Those three box parameters are specific to the woofer used. So try to use an existing box design for your woofer and stick closely to it. GPA must have box designs up their sleeve for the customers. You can make it any shape as long as it matches the volume, and there's room inside for the port and driver.
Rounding over the port edges does have an effect on SQ. Read about the KEF LS50 about the lengths they went to curve the innards of the port, ditto with Revel Ultima2 speakers. A sharp edge makes turbulence, which affects the resonance of the air spring, this turbulence can be audible if it's bad enough. If your speaker design has port length of 3/4" the thickness of the box material then I wouldn't worry about roundover. There's not enough material to reduce turbulence.
Stuffing a speaker is an art, and there are many approaches, which all sound different. I prefer minimal amount of stuff to absorb some of the midrange reflections. The more stuffing, the stuffier and quieter the sound. The tweeter SPL is not affected by the stuffing, so the woofer gets quieter but not the tweeter, so the tweeter level is usually adjusted after the stuffing is done. In your case with crossover pre-built just don't use too much stuffing. Keep it away from the reflex port. Try 2" on all surfaces, that's not a lot for a big box. FG is a very good for that if you can find it thin enough. 4" thick might be too much. You have to iterate and adjust to taste.
Yes, the GPA crossover is ready to connect to the driver and go.
Mount the XO to the floor of the cab, use a resilient mount so the cabinet vibes are somewhat isolated, but don't fret over it.
No sonic disadvantage to inside mounting other than the difficulty in mechanically mounting it to the box. Make the edge rounded or something to get rid of the sharp outside edge. You'll need a large enough removable access panel to install the driver, which has the potential to vibrate (make noise) when playing, so think about the best approach to that.
The benefit of thicker baffle is to stiffen the front baffle and add mass, both of which help reduce the audible vibrations of the front panel.
3/4" plywood box will make audible contribution to the sound in such a large box, because the woofer plays in midrange freqs, through the resonant frequencies of the plywood. Consider laminating a layer of MDF to the inside surface of your plywood box using "green glue" damping adhesive between the layers of wood. This will quiet it down a lot. Some internal "window" bracing will help too.
A bass reflex box is a system that mechanically amplifies the LF by using a controlled resonance of the air in the box through a port - the air rings at a designed frequency. So the box volume, the port area and the port length all three are critical in getting the ringing to work as desired. Those three box parameters are specific to the woofer used. So try to use an existing box design for your woofer and stick closely to it. GPA must have box designs up their sleeve for the customers. You can make it any shape as long as it matches the volume, and there's room inside for the port and driver.
Rounding over the port edges does have an effect on SQ. Read about the KEF LS50 about the lengths they went to curve the innards of the port, ditto with Revel Ultima2 speakers. A sharp edge makes turbulence, which affects the resonance of the air spring, this turbulence can be audible if it's bad enough. If your speaker design has port length of 3/4" the thickness of the box material then I wouldn't worry about roundover. There's not enough material to reduce turbulence.
Stuffing a speaker is an art, and there are many approaches, which all sound different. I prefer minimal amount of stuff to absorb some of the midrange reflections. The more stuffing, the stuffier and quieter the sound. The tweeter SPL is not affected by the stuffing, so the woofer gets quieter but not the tweeter, so the tweeter level is usually adjusted after the stuffing is done. In your case with crossover pre-built just don't use too much stuffing. Keep it away from the reflex port. Try 2" on all surfaces, that's not a lot for a big box. FG is a very good for that if you can find it thin enough. 4" thick might be too much. You have to iterate and adjust to taste.
I'd mount it externally. The GPA crossovers are workmanlike, but it will be easier to play with it later if you don't have to open up the carcass of a refrigerator-sized speaker.
There's probably a benefit, but I'd pay a lot more attention to the internal bracing. A 1" panel is usually sufficient. If you'll be rear-mounting the driver and chamfering the cutout, 2" may make sense.
That's not a crazy way to do things at all. Billfort's 604 cabinets provide an excellent example of this in action.
Yes. Just be careful to take the rounded area into account properly when you size your port.
Unfortunately, no, there's not a good rule. What you've described is usually the recommended starting place for a 612-style cabinet. Since you'll have a removable panel on the back of the box, though, this is something that's very cheap and very easy to change.
Here's the cabinet drawing. Some changes:
1) 1" ply all around except for front baffle, 2 layers making it 2".
2) Not sure of the height of the drivers. Estimating them based on where I am sitting.
3) Distance of the port to driver also taken from Altec's suggestion.
4) Port size is unchanged, except that it flares due to the rounding of the corner. No collar inside, as 2" thickness should be deep enough?
5) I want to make the rear panel accessible for future upgrades. Was thinking of creating a silicone gasket with dried up silicone before installing the panel. Any views on this? How are cabinets normally finished?
6) Gonna mount XO inside on bottom of cab, for aesthetic reasons.
7) Cabinet works out to 9.5 cubic feet/ 278 liters.
May move back to 3/4" ply as it cuts down on the overall size. Let me know what you guys think!
1) 1" ply all around except for front baffle, 2 layers making it 2".
2) Not sure of the height of the drivers. Estimating them based on where I am sitting.
3) Distance of the port to driver also taken from Altec's suggestion.
4) Port size is unchanged, except that it flares due to the rounding of the corner. No collar inside, as 2" thickness should be deep enough?
5) I want to make the rear panel accessible for future upgrades. Was thinking of creating a silicone gasket with dried up silicone before installing the panel. Any views on this? How are cabinets normally finished?
6) Gonna mount XO inside on bottom of cab, for aesthetic reasons.
7) Cabinet works out to 9.5 cubic feet/ 278 liters.
May move back to 3/4" ply as it cuts down on the overall size. Let me know what you guys think!
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
I'm curious, considering all that's been posted, what 'tipped the scales' to choosing the S3? Cost, or.......?
GM
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.