About the listening angle/height ...
One "mistake" which I have made, and Troels (my soul brother) also is the target listening location vs. actual.
When I designed the SNR-1 my target measurements and listening location was on tweeter axis, but like Troels sometimes discovers, the actual best listening is on the mid-woofer axis. At tweeter level the midrange is too subdued.
My point to this is, it would be super recommended that the builder do a test case if possible before committing to an angle and height.
Best,
E
One "mistake" which I have made, and Troels (my soul brother) also is the target listening location vs. actual.
When I designed the SNR-1 my target measurements and listening location was on tweeter axis, but like Troels sometimes discovers, the actual best listening is on the mid-woofer axis. At tweeter level the midrange is too subdued.
My point to this is, it would be super recommended that the builder do a test case if possible before committing to an angle and height.
Best,
E
I am sure the scan speaks will sound great, I have never used them myself as they are optimised for ported boxes so not my cup of tea, but hey loads of guys love them!
I love the clarity of Pro drivers with light cones / powerful motors with fabric surrounds, the low Qts versions are perfect for sealed box / DSP Eq / active crossover.
Audiophile drivers tend to have heavier cones lower power motors and rubber surrounds.... This tends to result in higher Qts.
I look for the following as a guide:
(1) Low Mms / high Bl. This tends to give a low Qts.
(2) Fabric surround. (points 1 & 2 usually result in a very clean CSD plot.
(3) Large voice coil (low power compression)
(4) Even frequency response over the desired bandwidth (doesn't matter so much about rough peaks / troughs above or below this as I use DSP with 48dB slope crossovers)
(5) Copper wire single layer voice coil with Kapton former.
The attached sim is of a great value Pro driver, 8 inch Beyma which works wonders in a 10 litre sealed box, crossed over at 2.2KHz to a Beyma TPL and a nice Precision Devices bass driver at 100Hz and you have one killer system.
Cheers
A.
I love the clarity of Pro drivers with light cones / powerful motors with fabric surrounds, the low Qts versions are perfect for sealed box / DSP Eq / active crossover.
Audiophile drivers tend to have heavier cones lower power motors and rubber surrounds.... This tends to result in higher Qts.
I look for the following as a guide:
(1) Low Mms / high Bl. This tends to give a low Qts.
(2) Fabric surround. (points 1 & 2 usually result in a very clean CSD plot.
(3) Large voice coil (low power compression)
(4) Even frequency response over the desired bandwidth (doesn't matter so much about rough peaks / troughs above or below this as I use DSP with 48dB slope crossovers)
(5) Copper wire single layer voice coil with Kapton former.
The attached sim is of a great value Pro driver, 8 inch Beyma which works wonders in a 10 litre sealed box, crossed over at 2.2KHz to a Beyma TPL and a nice Precision Devices bass driver at 100Hz and you have one killer system.
Cheers
A.
It's actually not a problem if the tweeter is at ear height and the baffle is tilted. You will just be listening a little off-axis. And the xo will be designed for that too. Plus it helps to physically bring the acoustic centers closer to alignment which can be a plus but isn't absolutely necessary with delay available in the digital domain. Your choice really. Certainly it makes the build more complex but I happen to like the look of the tilt.
No, they are not actually sims. They are physical models done quickly and cheaply with cardboard and measurements to determine the best response in the real world. Or, I was just going through Troels' site because I know I've seen him with pictures doing this and I found this, which uses your same tweeter and baffle style. Can you copy his exact top baffle details or within about 10% margin of error? That should work.
Hehe waaaay ahead of you
Kinda like what you describe right?
I love sketching, thats my second most favorite part of DIY projects.. i sketch everything i make
For simplicity i think i wil just keep the wall thickness the same.. unless you think my bass passage in the sides here are to narrow?
I was going to comment before on the nice and quick sketches you've being doing. Must say I'm a little jealous.
That is very close to what I was thinking and will work fine, still giving you options for what you may want to do with it. Looks like lot's of space on the sides.However, it's probably still larger than the absolute minimum you can get away with and since woofer volume is at a premium I would still try to reduce it. So I would do the same tapering to the top and bottoms that you've done on the sides. Or to save just a little more space, do that and make it circular too - see my very quick and rough partial sketch below (it got too complex for me to fill in the whole thing
).I have gotten hooked on the sealed system now as well.. It look sbadass with two woofers anyway
I'm not sure if you think you need more power for more drivers but in fact when you wire them in parallel, you actually get an extra 6dB out of the same amount of power. Impedance is halved however so you have to be sure that those amps are happy with that load. This also has the added benefit of reducing distortion and increasing headroom. Also forgot that 2 woofers doubles the surface and will mean a little more impact to the bass.
Two words - test box. You could make a quick and dirty mock up with the Hypex amps and with the boost and see how they do. I would prefer to be a little larger than that though, 33-35L?
Yes, at 1m for 1 speaker. 98dB at xmax should be fine for those listening conditions.
Attachments
# erik - Good to know! then it might even be perfect this way
I think i will simply build one cabinet first as i already have two 18W i can build a full speaker only by buying a 12 MU and a FA123. Then i can cry over the second speaker when i am done cheearing over the first one xD
# alex - good to know! one of my friends is also building a pair of speakers, and that may be exactly what he needs!
#jReave - I also like the tilt! it also somehow makes it look less agressive in the room i think
No i am aware that it is actual models, but doo you need a cabinet of your expected size behind it or?
Thanks for the tip, I now coppied his baffel design and tweeter placement so i should be good i guess
Thanks! i have been dooing 3D drawings for 7 years now in the same software
Glad to hear it looks feasable. It gives 2,4L for the mid and 29,5 L for the woofer. I think i will stick with this as I think I may actually be able to realize it. If i start dooing what you sketched i think my crafting skills will be insufficient.. :/
In the manual for the Fusion Amps it says it can deliver 125W at 4 ohm, so two paralel 8 ohms should give 4 ohms right? I can just connect them i paralel for the woofer chanel right? no need for a fourth channel right?
Is there even a reason to use two woofers when i only need SPL's at max 90 dB? It does increase the build cost to use two in each speaker, but is it worth it then it is worth it
I think i will simply build one cabinet first as i already have two 18W i can build a full speaker only by buying a 12 MU and a FA123. Then i can cry over the second speaker when i am done cheearing over the first one xD
# alex - good to know! one of my friends is also building a pair of speakers, and that may be exactly what he needs!
#jReave - I also like the tilt! it also somehow makes it look less agressive in the room i think
No i am aware that it is actual models, but doo you need a cabinet of your expected size behind it or?
Thanks for the tip, I now coppied his baffel design and tweeter placement so i should be good i guess
Thanks! i have been dooing 3D drawings for 7 years now in the same software
Glad to hear it looks feasable. It gives 2,4L for the mid and 29,5 L for the woofer. I think i will stick with this as I think I may actually be able to realize it. If i start dooing what you sketched i think my crafting skills will be insufficient.. :/
In the manual for the Fusion Amps it says it can deliver 125W at 4 ohm, so two paralel 8 ohms should give 4 ohms right? I can just connect them i paralel for the woofer chanel right? no need for a fourth channel right?
Is there even a reason to use two woofers when i only need SPL's at max 90 dB? It does increase the build cost to use two in each speaker, but is it worth it then it is worth it
Would it be a completely crazy idea to make two sets of front baffle plates. One set with two holes for two woofers and one set only with holes for one woofer.
This way i could first build up the two full speakers with one woofer in each and then when I have the money (or need the extra SPL) switch to the other baffel plate and install the second woofer? The single woofer would have the full 29 L of wolume though. But from winISD it seems like it can go just as deep and still deliver 83dB at 3 m distance without exceeding x-max of 6,5 mm.
What exactly is the difference between the linear 6,5 mm x-max and the 11 mm mechancal x-max. My guess would be that above 6,5 mm the sound quality decreases and above 11 mm wil damage the driver right? Just to be sure!
This way i could first build up the two full speakers with one woofer in each and then when I have the money (or need the extra SPL) switch to the other baffel plate and install the second woofer? The single woofer would have the full 29 L of wolume though. But from winISD it seems like it can go just as deep and still deliver 83dB at 3 m distance without exceeding x-max of 6,5 mm.
What exactly is the difference between the linear 6,5 mm x-max and the 11 mm mechancal x-max. My guess would be that above 6,5 mm the sound quality decreases and above 11 mm wil damage the driver right? Just to be sure!
According to this simulator:
Speakerbuilder Pro 2.0
Two 18Ws wired in parallel in a closed box will give a Qtc of 0.81 and Fb= 62.54 Hz (0.1 Ohm in series)
I would say, without doubt, that is not a good way of using those 18W.
The engineer behind the driver recommends around 28-30 Liter vented. Cannot remember if that is before or after damping ie. the damping will "simulate" a bigger cabinet.
Speakerbuilder Pro 2.0
Two 18Ws wired in parallel in a closed box will give a Qtc of 0.81 and Fb= 62.54 Hz (0.1 Ohm in series)
I would say, without doubt, that is not a good way of using those 18W.
The engineer behind the driver recommends around 28-30 Liter vented. Cannot remember if that is before or after damping ie. the damping will "simulate" a bigger cabinet.
Last edited:
#Rocky - if i use two, then it would be in a sealed cabinet not ported
You Are right about The tests before building. However i wont have acces to The workshop before Christmas, so i am just trying to learn and prepare as much as possible beforehand
#alex - would me a dramatik difference between two 14l Chambers and one 28l chamber? My current design will be a bit hard to separate into two equal rooms..
You Are right about The tests before building. However i wont have acces to The workshop before Christmas, so i am just trying to learn and prepare as much as possible beforehand
#alex - would me a dramatik difference between two 14l Chambers and one 28l chamber? My current design will be a bit hard to separate into two equal rooms..
Danner,
Some random responses:
- for cardboard modeling, yes, putting the mid in a box would be beneficial. The tweeter is closed back so that one doesn't matter.
- 2.4L for the mid seems reasonable. I'm actually not sure I'd go much smaller than that. Might depend on the stuffing strategy.
- all correct on parallel woofers. No extra channel needed in the amp, just connect the drivers together in parallel for 4 ohms and 125W gives you plenty of headroom in the amp department.
- As to 1 or 2 woofers, running real world tests may be the only accurate way to know for certain but here's a rough look at running the numbers.
One 18W sealed will allow you 92dB at 1m without exceeding xmax at 40Hz in about 15L sealed, xmax being the distance that the voice coil is capable of travelling while still remaining inside the magnetic gap. Distortion and compression etc. increase as you push it beyond this point. Actually a little bit too before this point as well.
So, if your listening average is 85dB at 3-4m, you still want to allow room for musical peaks in the driver (that's already covered in the amp). So add another 3-6dB for a 3-4m max SPL of 88-91dB. Often forgotten though is the 6dB of baffle step loss, but you actually get that right back when you include the 2nd speaker. So those 2 cancel out.
Now SPL decreases as you move further away from the source (6dB/m? I can't quite remember the exact amount), but it doesn't decrease quite so much in a closed space as it does in open space. Pulling out my trusty SPL meter, I just measured about an 8dB loss in my room at 3m vs 1m. So let's ballpark it at a 7-10dB loss at 3-4m. Adding that to the previous total of 88-91dB will give you what you more or less need from a single driver at 1m, about 95-101dB. Just a little bit shy of the max 92dB at 40Hz that the sims suggest. Now if I haven't lost you so far, there's actually 1 more factor to consider: 40Hz is your desired F3 frequency, meaning it's your -3dB point on the bottom end. So that subtracts another 3dB from your requirements, so now I'd call it at about 92-98dB at 1m. Which means 1 x 18W per speaker gives you about 0-3dB of headroom listening at an 85dB average, whereas 2 x 18W will give you up to 6-9dB of headroom (varies with 3m or 4m listening distance).
- 2 baffles? That could work. I'm wondering if you couldn't make something work with your current speakers though? Stuff the port and drop some 2x4's or anything else solid to reduce the box volume down to about 14-15L, bypass the existing xo and then connect it to the Hypex amps with the bass boost activated. Perhaps that's easier said than done though. Just a thought.
- Regarding the sealed volume for a single 18W, Madison actually recommends 12.5-18L but I'm not sure if that's based on experience or just from running ScanSpeak's numbers through a box sim program. Zaph's measurements suggest a slightly larger box is reqiured. As someone else mentioned, it's what happens on the bottom end when trying to add the bass boost that matters most since the air inside a closed box is acting like an internal spring on the cone and a smaller box equals a stronger spring. I have no direct experience in this area but my gut feeling is that Q=.8 isn't so extreme that it's going to create problems. On the other hand, maybe adding a higher resistance aperiodic port to the sealed woofers would solve any problems with too stiff a spring in a smaller sealed box. Anyone with any experience or knowledge on that?
- I don't see any particular need for 2 separate boxes for 2 sealed woofers. I'm wondering what the advantage would be. Certainly in a situation like this where internal volume is at a premium, it doesn't make much sense to me to add another piece of wood to divide the chambers.
Some random responses:
- for cardboard modeling, yes, putting the mid in a box would be beneficial. The tweeter is closed back so that one doesn't matter.
- 2.4L for the mid seems reasonable. I'm actually not sure I'd go much smaller than that. Might depend on the stuffing strategy.
- all correct on parallel woofers. No extra channel needed in the amp, just connect the drivers together in parallel for 4 ohms and 125W gives you plenty of headroom in the amp department.
- As to 1 or 2 woofers, running real world tests may be the only accurate way to know for certain but here's a rough look at running the numbers.
One 18W sealed will allow you 92dB at 1m without exceeding xmax at 40Hz in about 15L sealed, xmax being the distance that the voice coil is capable of travelling while still remaining inside the magnetic gap. Distortion and compression etc. increase as you push it beyond this point. Actually a little bit too before this point as well.
So, if your listening average is 85dB at 3-4m, you still want to allow room for musical peaks in the driver (that's already covered in the amp). So add another 3-6dB for a 3-4m max SPL of 88-91dB. Often forgotten though is the 6dB of baffle step loss, but you actually get that right back when you include the 2nd speaker. So those 2 cancel out.
Now SPL decreases as you move further away from the source (6dB/m? I can't quite remember the exact amount), but it doesn't decrease quite so much in a closed space as it does in open space. Pulling out my trusty SPL meter, I just measured about an 8dB loss in my room at 3m vs 1m. So let's ballpark it at a 7-10dB loss at 3-4m. Adding that to the previous total of 88-91dB will give you what you more or less need from a single driver at 1m, about 95-101dB. Just a little bit shy of the max 92dB at 40Hz that the sims suggest. Now if I haven't lost you so far, there's actually 1 more factor to consider: 40Hz is your desired F3 frequency, meaning it's your -3dB point on the bottom end. So that subtracts another 3dB from your requirements, so now I'd call it at about 92-98dB at 1m. Which means 1 x 18W per speaker gives you about 0-3dB of headroom listening at an 85dB average, whereas 2 x 18W will give you up to 6-9dB of headroom (varies with 3m or 4m listening distance).
- 2 baffles? That could work. I'm wondering if you couldn't make something work with your current speakers though? Stuff the port and drop some 2x4's or anything else solid to reduce the box volume down to about 14-15L, bypass the existing xo and then connect it to the Hypex amps with the bass boost activated. Perhaps that's easier said than done though. Just a thought.
- Regarding the sealed volume for a single 18W, Madison actually recommends 12.5-18L but I'm not sure if that's based on experience or just from running ScanSpeak's numbers through a box sim program. Zaph's measurements suggest a slightly larger box is reqiured. As someone else mentioned, it's what happens on the bottom end when trying to add the bass boost that matters most since the air inside a closed box is acting like an internal spring on the cone and a smaller box equals a stronger spring. I have no direct experience in this area but my gut feeling is that Q=.8 isn't so extreme that it's going to create problems. On the other hand, maybe adding a higher resistance aperiodic port to the sealed woofers would solve any problems with too stiff a spring in a smaller sealed box. Anyone with any experience or knowledge on that?
- I don't see any particular need for 2 separate boxes for 2 sealed woofers. I'm wondering what the advantage would be. Certainly in a situation like this where internal volume is at a premium, it doesn't make much sense to me to add another piece of wood to divide the chambers.
Just want to check that you know with the active amps Danner is going to be using, the sealed LF response is going to be boosted to taste? Should improve a normal sealed response considerably.
Just to screw with your head a little more , actually just to provide you with as many reasonable options as possible, how about this:
Use the 18W as your mid and go with 2 of these for your woofers - Satori MW19P-8.
Took me a while to find something with at least equivalent quality and just about the exact same dimensions as the 18W and that will also give you pretty much the exact same LF response as the 18W sealed but will do it comfortably in half the volume, so 20L for 2 x 19P vs 39L for 2 x 18W, Q=.7 for both. Two of the Satoris still provide you with 98dB at 1m at 40Hz at xmax although they do require just a little bit more power to do so, 30W instead of 20W for the ScanSpeaks. Still not a problem when you have 125W available. And 4 x 19P will actually be cheaper than 2 x 18W and 2 x 12MU. Plus and minuses both ways. Just something to think about anyways.
, actually just to provide you with as many reasonable options as possible, how about this:Use the 18W as your mid and go with 2 of these for your woofers - Satori MW19P-8.
Took me a while to find something with at least equivalent quality and just about the exact same dimensions as the 18W and that will also give you pretty much the exact same LF response as the 18W sealed but will do it comfortably in half the volume, so 20L for 2 x 19P vs 39L for 2 x 18W, Q=.7 for both. Two of the Satoris still provide you with 98dB at 1m at 40Hz at xmax although they do require just a little bit more power to do so, 30W instead of 20W for the ScanSpeaks. Still not a problem when you have 125W available. And 4 x 19P will actually be cheaper than 2 x 18W and 2 x 12MU. Plus and minuses both ways. Just something to think about anyways.
Last edited:
"Conventional" loading Vs DSP
Three words.... DSP, DSP and DSP!
With all due respect to the old school "Loudspeaker design cookbook" brigade and the countless clones based on passive crossover / MDF box ported box with bass/mid and a tweeter.... There is a better way.
All the "classic" loudspeaker design books / papers and software simulations based on them are well out of date.
Modern DSP and computing power combined with efficient high power amplifiers and driver design (material science) with dramatically increased power handling open up new design possibilities.
Three words.... DSP, DSP and DSP!
With all due respect to the old school "Loudspeaker design cookbook" brigade and the countless clones based on passive crossover / MDF box ported box with bass/mid and a tweeter.... There is a better way.
All the "classic" loudspeaker design books / papers and software simulations based on them are well out of date.
Modern DSP and computing power combined with efficient high power amplifiers and driver design (material science) with dramatically increased power handling open up new design possibilities.
Can do one single enclosure of 28 litres....Two separate 14 litre enclosures would be slightly better but not massive.
#rocky - Does it still need the bigger cabinet with DSP tuning to boost the lower frequencies?
I am also aiming at the aperiodic tuning for my mid
#alex - Good to hear. It makes sense that when the driversare seperated then they cannot affect one another,but if this project is ever gonna happen then i think i have to stop adding complexities xD
#jReave - There is no doubt that if i had to start from scratch, then i would have gone for the Satori woofers.. I am still slightly considdering it (then it could reduce the speaker size again). I am not so tempted to use the 18W as a mid driver.. One of the reasons i want to go 3-way is to have a smaller mid, as it seems like it gives a better soundstage/less directivity. At least this is how i interpret TG's articles.
Part of me knows the right thing to do would be to use tha satori since they are more suited for this size speaker. Another part of me have always been in love with the revelators and really wnats to keep them...
But thanks for the idea!
Forcardboard models - could i just use my current speakers with a cardboard front? or does it have a big affect that the drivers then won't beflush with the cardboard?
Good to hear, then i think the design of the cabinet is somewhat settled!
It is not a issue that the cap is shaped like a long tube at the rear?
Good to know i got it right with two woofers!
Thankyou very much for the epxplenation on SPL. I am pretty sure i understood it all. It also does make sense!
It seems like the only right thing to do here is to get a hold on two Fusion Amps and use my existing speakers for tests. They happen to have a woofer volume of 21L. So if i block the port I should be able to do some good tests of bass response and SPL. The initial volume will correspond to two woofers in 42L so i could gradually fill the cabinet with stuff to reduce the volume down to my target and see the difference - Great idea!
Again thanks for all the inputs!
Then we will have to see when it will be possible to get a hold of two fusion amps xD
I am also aiming at the aperiodic tuning for my mid
#alex - Good to hear. It makes sense that when the driversare seperated then they cannot affect one another,but if this project is ever gonna happen then i think i have to stop adding complexities xD
#jReave - There is no doubt that if i had to start from scratch, then i would have gone for the Satori woofers.. I am still slightly considdering it (then it could reduce the speaker size again). I am not so tempted to use the 18W as a mid driver.. One of the reasons i want to go 3-way is to have a smaller mid, as it seems like it gives a better soundstage/less directivity. At least this is how i interpret TG's articles.
Part of me knows the right thing to do would be to use tha satori since they are more suited for this size speaker. Another part of me have always been in love with the revelators and really wnats to keep them...
But thanks for the idea!
Forcardboard models - could i just use my current speakers with a cardboard front? or does it have a big affect that the drivers then won't beflush with the cardboard?
Good to hear, then i think the design of the cabinet is somewhat settled!
It is not a issue that the cap is shaped like a long tube at the rear?
Good to know i got it right with two woofers!
Thankyou very much for the epxplenation on SPL. I am pretty sure i understood it all. It also does make sense!
It seems like the only right thing to do here is to get a hold on two Fusion Amps and use my existing speakers for tests. They happen to have a woofer volume of 21L. So if i block the port I should be able to do some good tests of bass response and SPL. The initial volume will correspond to two woofers in 42L so i could gradually fill the cabinet with stuff to reduce the volume down to my target and see the difference - Great idea!
Again thanks for all the inputs!
Then we will have to see when it will be possible to get a hold of two fusion amps xD
I have used DSP for quite some years now. I cannot make a passive crossover - but I have pretty good idea of how to make a digital crossover.
I believe in combining the best of both worlds. Saying that old knowledge is Well out of date is IMO a way of blindfolding yourself for new knowledge. Better is to stay open.
Have you actually built two cabinets - one that is proper sized on paper - and one that is 'undersized'? - and compared how they Sound ?
No doubt that you can get a lot of bass with DSP boost - but is the quality the same? What happens to the stepresponse ?
There must be a reason that we dont see a lot of speakercompanies using an undersized box.
Another question:
When i use the bassboost to extend my bass response, then i guess that will efectively be the same as lowering the rest right?
In this case i guess that i will in fact have to limit the output of my amplifier to avoid driving the woofer over x-max. In fact i will have to limit it to avoid putting to much power through the drivers no matter what.
As far as i can see the Fusion amp allows the user to set a gain for each driver and a master gain for the whole speaker. This gain is adjusted in db from -96 to +18 dB.
Now how do i know how much to limit my wooferchannel so that if the speaker is put on maximum volume then i still won't damage the woofer?
My best bet would be to lower the gain for the woofer alot. Put the speaker at maximum volume and then slowly raise the gain again until I reach my expected dB at 1 m (minus 1 or 2 for safety). Is there a more scientific way of doing it or is this a good apporach?
With my current setup (passive crossover and a 125W amplifier) i never ever considdered if i was risking over excursion. I luckily just didn't turn it up loud enough i guess. But with this new setup it seems that the risk of overexcursion is higher due to the sealed box and bass boost, or am i wrong about that?
When i use the bassboost to extend my bass response, then i guess that will efectively be the same as lowering the rest right?
In this case i guess that i will in fact have to limit the output of my amplifier to avoid driving the woofer over x-max. In fact i will have to limit it to avoid putting to much power through the drivers no matter what.
As far as i can see the Fusion amp allows the user to set a gain for each driver and a master gain for the whole speaker. This gain is adjusted in db from -96 to +18 dB.
Now how do i know how much to limit my wooferchannel so that if the speaker is put on maximum volume then i still won't damage the woofer?
My best bet would be to lower the gain for the woofer alot. Put the speaker at maximum volume and then slowly raise the gain again until I reach my expected dB at 1 m (minus 1 or 2 for safety). Is there a more scientific way of doing it or is this a good apporach?
With my current setup (passive crossover and a 125W amplifier) i never ever considdered if i was risking over excursion. I luckily just didn't turn it up loud enough i guess. But with this new setup it seems that the risk of overexcursion is higher due to the sealed box and bass boost, or am i wrong about that?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.