http://sound.westhost.com/index2.html
I highly recommend Rod Elliots site if you don't know it already; excellent explanation of baffle step, cross-overs and more importantly power compression
I highly recommend Rod Elliots site if you don't know it already; excellent explanation of baffle step, cross-overs and more importantly power compression
That monacor is reasonably flat between 150 and 500, but the crossover needs an octave flat on both sides so the best X-O point would be half way between at 250/280
These woofers ought to be very well suited fore closed, but would need a rather large box, rated as 6ohm with lowish 4ohm dc-resistance, but priced fairly at Madisound
Although they are danish designed by former ScanSpeak guys, some have questioned the quality, about which I know nothing...my interest is at the moment focused on AE drivers
http://www.sbacoustics.com/product.php
Although they are danish designed by former ScanSpeak guys, some have questioned the quality, about which I know nothing...my interest is at the moment focused on AE drivers
http://www.sbacoustics.com/product.php
Glenn-
Here is link to a nice sealed box calculator:
http://www.carstereo.com/help/Articles.cfm?id=30
The monocor 15" delivers (theoretically, if the parameters are correct) an F3 and Fcb of about 54Hz. Not real deep, but it is a pretty small box.
Lots of questions and info to digest here -- some of which I have suggestions for, but I don't have time to post right now.
JJ
Here is link to a nice sealed box calculator:
http://www.carstereo.com/help/Articles.cfm?id=30
The monocor 15" delivers (theoretically, if the parameters are correct) an F3 and Fcb of about 54Hz. Not real deep, but it is a pretty small box.
Lots of questions and info to digest here -- some of which I have suggestions for, but I don't have time to post right now.
JJ
jupiterjune said:
That's exactly what I just found in WinISD 🙂
For a nice Qtc of 0.707 the box volume calculates pretty close to the manufacturers spec of 55L at 53.58L, with Fsc at 56.5Hz.
For such a monster of a speaker that is woeful
This speaker has a EBP of ~70. You can of course make the box twice as big, but that has a less than commesurate effect of Fsc and the transient response is neutered.I've run a sim on the a 12" Shiva-X with a EBP of ~45, and that is fantastic in a closed box. Definately what I would use if I was building a subwoofer.
All this seems a pretty good confirmation of R. Smalls rule for a driver EBP of ~50 for closed box use.
Cheers,
Glen
Attachments
Moondog55 said:
After further investigation, that Monacor with a sealed box Fsc of ~55Hz is now not an option.
I hear what you are saying about going too high with a big speaker.
At this stage I think I'll be better falling back on either the WWWWMT, with a low bass crossover to avoid lobing effects of the four 8-ohm 8" woofer array, or a WWMT with a pair of high quality 4-ohm 8" drivers.
BTW, do you know of any other distributors of drivers here in Oz?
Cheers,
Glen
Re: Half space sensitivity
Half space is essentially an infinite baffle, so you would subtract your baffle step size from that number to get the mid band system sensitivity. The tweeter would be the 83 dB tweeter, expect to use more efficient mid and tweeters and pads which will give you some room to shape the response as needed and tailor it to differing room absorbtion.
Regarding voltage sensitivity:
Here's how I do it, turns out these 1W/1m figures are really the voltage required to produce one Watt on the rated impedance. So, if the driver is rated as 4 ohm, or has a DCR of slightly less than 4 ohms, they produced the 1W/m figure by driving it with 2 Volts. 8 ohm rated they drove it with 2.83 Volts. So to translate a 1W/m 4 ohm rating to 2.83V sensitivity you apply this correction factor:
20 log (2.83/2) or + 3dB to the 1W/m rating.
This is ususally how it is done, and I therefore do a sanity check against any measured data that they provide, and by also calculating it from the T&S parameters as there are sometimes errors in the figures provided.
Take your Monacor 15" for example which is rated for 89 dB/w, with 2 4 ohm VCs. They should have put .5 W into each coil, or 1.41 V, in parallel you still have 1.41 V and a 2 ohm load of course. Then the 2.83 Volt sens is:
20 log (2.83/1.41) or + 6dB for a total of 95 dB/2.83 V
This is in the passband of course.
Also let me comment that without pulling out a simulator, I'd estimate that driver in the suggested 55L box to have an Fc of about 40 Hz, good for pop music, but not a world class system. You want to be at 30 or below and you need more cone mass to do that which would knock down the sensitivity. Go to a bigger box, or LT this driver if you use it.
One Shiva-X is 84.4 dB 1W/1m but it has two 8 ohm coils, one half W into each is 2 V put in parallel the voltage is the same so at 2.83 V we have obviously + 3dB or 87.4 dB /2.83 V. Add a second driver and you add 6 dB to the voltage sensitivity 3 for drawing twice the power, and 3 because their volume velocities add as in phase SPL vectors. The total is then 93.4 dB/2.83 V
The pair have a lower Fs (very slightly), lower total Vas, and about 5 times the VD of one of the 15" Monacors.
You've got I think 800 W into 2 ohms, do you want drivers that will be loafing along or one that is slightly stressed when the system is pushed? Either way the voice coils will get HOT resulting in compression, etc.
Given how critical you are about amplifier distortion, and that this is a new design, I would think that you'd seek out drivers with some effort to reduce distortion. XBL motors, or other tricks, and copper to reduce inductance modulation effects. Something to think about. The AE drivers are also quite excellent.
The large signal performance is quite important especially when you have around 1 KW to work with.
I expect that you'll end up with a woofer XO of about 200 to 300 Hz, perhaps as low as 150 Hz. I also think that you'll end up with a 4 way. The woofer XO in the NHT 3.3 is about 125 Hz I believe and the crossover schematic is available through the link that I provided. I suggest taking a look to get an idea of what is involved. It is a good higher order 4 way design example.
Pete B.
G.Kleinschmidt said:
Half space is essentially an infinite baffle, so you would subtract your baffle step size from that number to get the mid band system sensitivity. The tweeter would be the 83 dB tweeter, expect to use more efficient mid and tweeters and pads which will give you some room to shape the response as needed and tailor it to differing room absorbtion.
Regarding voltage sensitivity:
Here's how I do it, turns out these 1W/1m figures are really the voltage required to produce one Watt on the rated impedance. So, if the driver is rated as 4 ohm, or has a DCR of slightly less than 4 ohms, they produced the 1W/m figure by driving it with 2 Volts. 8 ohm rated they drove it with 2.83 Volts. So to translate a 1W/m 4 ohm rating to 2.83V sensitivity you apply this correction factor:
20 log (2.83/2) or + 3dB to the 1W/m rating.
This is ususally how it is done, and I therefore do a sanity check against any measured data that they provide, and by also calculating it from the T&S parameters as there are sometimes errors in the figures provided.
Take your Monacor 15" for example which is rated for 89 dB/w, with 2 4 ohm VCs. They should have put .5 W into each coil, or 1.41 V, in parallel you still have 1.41 V and a 2 ohm load of course. Then the 2.83 Volt sens is:
20 log (2.83/1.41) or + 6dB for a total of 95 dB/2.83 V
This is in the passband of course.
Also let me comment that without pulling out a simulator, I'd estimate that driver in the suggested 55L box to have an Fc of about 40 Hz, good for pop music, but not a world class system. You want to be at 30 or below and you need more cone mass to do that which would knock down the sensitivity. Go to a bigger box, or LT this driver if you use it.
One Shiva-X is 84.4 dB 1W/1m but it has two 8 ohm coils, one half W into each is 2 V put in parallel the voltage is the same so at 2.83 V we have obviously + 3dB or 87.4 dB /2.83 V. Add a second driver and you add 6 dB to the voltage sensitivity 3 for drawing twice the power, and 3 because their volume velocities add as in phase SPL vectors. The total is then 93.4 dB/2.83 V
The pair have a lower Fs (very slightly), lower total Vas, and about 5 times the VD of one of the 15" Monacors.
You've got I think 800 W into 2 ohms, do you want drivers that will be loafing along or one that is slightly stressed when the system is pushed? Either way the voice coils will get HOT resulting in compression, etc.
Given how critical you are about amplifier distortion, and that this is a new design, I would think that you'd seek out drivers with some effort to reduce distortion. XBL motors, or other tricks, and copper to reduce inductance modulation effects. Something to think about. The AE drivers are also quite excellent.
The large signal performance is quite important especially when you have around 1 KW to work with.
I expect that you'll end up with a woofer XO of about 200 to 300 Hz, perhaps as low as 150 Hz. I also think that you'll end up with a 4 way. The woofer XO in the NHT 3.3 is about 125 Hz I believe and the crossover schematic is available through the link that I provided. I suggest taking a look to get an idea of what is involved. It is a good higher order 4 way design example.
Pete B.
G.Kleinschmidt said:
The problem with most 8" woofers is that they usually do not have a low enough Fs, obviously Fc will always be higher than Fs. There might be some auto sound units.
You say that the Shiva-X is a subwoofer, are you planning to have an external subwoofer or for this to be a world class system that does not need a subwoofer and can handle organ pedals and cannon shots without stress? I would think it should handle everything without strain.
More XBL drivers:
http://www.creativesound.ca/products.php?category=Drivers&make=CSS
You might want to look into custom drivers from AE if you can't find a good solution.
Pete B.
I think you'll find this LT interesting unless you don't want any line level compensation:
http://sound.westhost.com/linkwitz-transform.htm
You could use 6 or 8 of these 7" woofers with XBL motors to get a nominal < 1 ohm load, an LT is then obviously required to provide a lower effective Fc:
http://www.creativesound.ca/details.php?model=SDX7
The XBL motor reduces IM distortion, and you should be able to run these easily up to 500 Hz. Placing the drivers then becomes an issue, perhaps half on the back or sides?
Don't know if I like this solution, food for thought ...
I should add that the LT will let you experiment with different Fc and Qtc values without changing anything regarding the enclosure. You could also do BSC at line level where it would be easily adjusted.
A must read: http://www.linkwitzlab.com/
Really a must read, you will like this: http://www.linkwitzlab.com/x-sb80-3wy.htm
Pete B.
http://sound.westhost.com/linkwitz-transform.htm
You could use 6 or 8 of these 7" woofers with XBL motors to get a nominal < 1 ohm load, an LT is then obviously required to provide a lower effective Fc:
http://www.creativesound.ca/details.php?model=SDX7
The XBL motor reduces IM distortion, and you should be able to run these easily up to 500 Hz. Placing the drivers then becomes an issue, perhaps half on the back or sides?
Don't know if I like this solution, food for thought ...
I should add that the LT will let you experiment with different Fc and Qtc values without changing anything regarding the enclosure. You could also do BSC at line level where it would be easily adjusted.
A must read: http://www.linkwitzlab.com/
Really a must read, you will like this: http://www.linkwitzlab.com/x-sb80-3wy.htm
Pete B.
G.Kleinschmidt said:
Why don't you build a WWMT Sealed Di-Pole
You would have to build two identical (Perhaps Mirror Imaged except for tweeter) Cabinets in one.
The Rear Firing drivers are flipped 180 degrees out of phase.
Just build two identical Xovers.
The whole system will be half the impedance.
And properly spaced A Dipole speaker will excite less room modes, than standard mono pole.
I bet it would sound Amazing!!!
Antone-
Hi Sumsound, why do you say that the rear firing drivers in a di-pole should be 180 reverse phase??
Have read of a few and thought of this aproach myself but I've never seen this particular phasing issue mentioned.
Hi Tinitus; asked that question myself a while ago, I got the answer that at low frequencies ( Subwoofers ) it doesn't matter which way the drivers are facing ie: monopole/di-pole/ tri-pole/ quadro-pole all sum the same ; counter intuitive, but di-pole mid-bass and midrange a possible yes; I would say the main reason for not doing it is the high cost of the drivers needed.
For a pair of boxes in D'Appolitto lay out though would you really need to have the same directionallity in the rear firing mids?? If not thats a saving of a few hundred dollars, I would say the main reason it's not done ( apart from the cost ) is that very few amplifiers would be able to comfortably handle the low impedance load, something Glen's amp should do fine.
regards Ted
Have read of a few and thought of this aproach myself but I've never seen this particular phasing issue mentioned.
Hi Tinitus; asked that question myself a while ago, I got the answer that at low frequencies ( Subwoofers ) it doesn't matter which way the drivers are facing ie: monopole/di-pole/ tri-pole/ quadro-pole all sum the same ; counter intuitive, but di-pole mid-bass and midrange a possible yes; I would say the main reason for not doing it is the high cost of the drivers needed.
For a pair of boxes in D'Appolitto lay out though would you really need to have the same directionallity in the rear firing mids?? If not thats a saving of a few hundred dollars, I would say the main reason it's not done ( apart from the cost ) is that very few amplifiers would be able to comfortably handle the low impedance load, something Glen's amp should do fine.
regards Ted
I think it depends on the speaker. In a lot of "bookshelf" speakers that are designed to sit right against the rear wall, the answer is no, because they are only radiating into halfspace.
For a speaker with a lot of bass woofers such that it is almost a line array (not sure if 4 8" woofers would qualify) -- you may need less than a full 6 dB of baffle step compensation. (something I read about line arrays, but I've never tested it.)
If the sensitivity of the bass array is 96dB @ 2.83V/m, I would try to find a mid or mid array that will give about 92dB to 93db @ 2.83V/m. Same goes for the tweeter. This gives you some room to move in case you only need 3 or 4 dB of baffle step compensation. If you do need the full 6dB, you can pad down the mid and tweeter 2-3 dBs.
JJ
Re: Re: Half space sensitivity
Fantastic post!
Lots of constructive advice here.
You are right about the virtues of the XBL motor drivers and the limited bass output of smaller woofers.
Most woofers I've looked at (even as large as 15”) simply so not have a sufficiently low EBP (fs/Qes) to operate in closed box satisfactorily with an Fsc substantially below 50Hz.
I was also wrong in my previous comments about bandwidth of the Shiva-X subwoofers - the XBL motor they use has exceptionally low inductance.
Here is what I am considering at the moment for the bass:
http://www.audiomarketplace.com.au/...acturer_id,0/option,com_virtuemart/Itemid,49/
This is a 12” Shiva-X with dual 8-ohm voice coils. It has a high Qes of 0.55 and a free air resonance of 20Hz, giving it a fantastically low EBP of 36. Performance in a sealed enclosure is fantastic.
I can use two of these per side with the coils in parallel. That will give me my 2-ohms nominal impedance.
At 84.4dB it is relatively inefficient. Two units in parallel will give me 90.4dB SPL. This makes mid-range selection easier.
At this stage I am still seeking to avoid the complication of a 4-way system.
I am thinking of spending the dough on a high quality ribbon tweeter. There are units out there with fantastic sensitivity and the ability to work with a crossover as low as 2kHz.
I would like to cross the twin 12” Shiva-X’s at approximately 250Hz. So, with a high quality ribbon tweetie coming in a 2kHz, all I need is a relatively efficient, high quality bass-mid-range driver to work in the middle (maybe a 6" or 6.5" unit). That seems a perfectly reasonable proposition.
I guess such a system could be considered an augmented 2-way 🙂
The layout could be either:
T
M
W
W
Or (my current preference):
W
MT
W
Cheers,
Glen
PB2 said:
Fantastic post!
Lots of constructive advice here.
You are right about the virtues of the XBL motor drivers and the limited bass output of smaller woofers.
Most woofers I've looked at (even as large as 15”) simply so not have a sufficiently low EBP (fs/Qes) to operate in closed box satisfactorily with an Fsc substantially below 50Hz.
I was also wrong in my previous comments about bandwidth of the Shiva-X subwoofers - the XBL motor they use has exceptionally low inductance.
Here is what I am considering at the moment for the bass:
http://www.audiomarketplace.com.au/...acturer_id,0/option,com_virtuemart/Itemid,49/
This is a 12” Shiva-X with dual 8-ohm voice coils. It has a high Qes of 0.55 and a free air resonance of 20Hz, giving it a fantastically low EBP of 36. Performance in a sealed enclosure is fantastic.
I can use two of these per side with the coils in parallel. That will give me my 2-ohms nominal impedance.
At 84.4dB it is relatively inefficient. Two units in parallel will give me 90.4dB SPL. This makes mid-range selection easier.
At this stage I am still seeking to avoid the complication of a 4-way system.
I am thinking of spending the dough on a high quality ribbon tweeter. There are units out there with fantastic sensitivity and the ability to work with a crossover as low as 2kHz.
I would like to cross the twin 12” Shiva-X’s at approximately 250Hz. So, with a high quality ribbon tweetie coming in a 2kHz, all I need is a relatively efficient, high quality bass-mid-range driver to work in the middle (maybe a 6" or 6.5" unit). That seems a perfectly reasonable proposition.
I guess such a system could be considered an augmented 2-way 🙂
The layout could be either:
T
M
W
W
Or (my current preference):
W
MT
W
Cheers,
Glen
Glen perhaps you should read thru this other thread I started a while ago, as a learning exercise it was very instructive
Re: Re: Half space sensitivity
Pete, the specification sheet for the Shiva-X 12” gives the sensitivity of 84.4dB (1W/1m) with both coils in parallel for 4 ohms nominal.
Here is what is performs like in a 70L closed box. Qtc is quite acceptable at 0.804 and Fsc is 33.87Hz
Cheers,
Glen
PB2 said:
Pete, the specification sheet for the Shiva-X 12” gives the sensitivity of 84.4dB (1W/1m) with both coils in parallel for 4 ohms nominal.
Here is what is performs like in a 70L closed box. Qtc is quite acceptable at 0.804 and Fsc is 33.87Hz
Cheers,
Glen
Attachments
- Status
- Not open for further replies.