Sounds like I have to make a decision. Thanks all.
P.S. Thingyness, the mass dumax report was an old one, there is a more recent one on their site. The Xsus does not indicate max suspension limits, rather it indicates where resistance to movement is 4X the neutral position. I found this out when phoning them. In their more recent report, even this number has expanded. It goes to a physical limit of 30. Regardless, it may be a mute point.
Good information, thank you.
P.S. Thingyness, the mass dumax report was an old one, there is a more recent one on their site. The Xsus does not indicate max suspension limits, rather it indicates where resistance to movement is 4X the neutral position. I found this out when phoning them. In their more recent report, even this number has expanded. It goes to a physical limit of 30. Regardless, it may be a mute point.
Good information, thank you.
You are correct in that it does not represent the max suspension limits, but it should be used to represent the maximum usable suspension limits.
Keep in mind that Fs is inversely proportional to the square root of Cms - so that if your Cms quadruples (as is the case when you hit the DUMAX suspension limits), then at that finite point in time, your Fs is now double what it normally was. For the audiomobile, this would be around 44hz. That'll throw your alignment off pretty good, I'd think.
Also note that as the suspension stiffens, you'll get even less output with a given input power - as you approach even 200% of the suspension limit, it will start to compress the peaks in the material, and you lose dynamic range. As you hit 400%, this phenomenon gets even worse. This is why I like the Blueprint for your application, as even if you only plan to use 20mm of excursion, the Blueprint's Cms/Kms Vs. X curve is near ruler flat in that region.
Keep in mind that Fs is inversely proportional to the square root of Cms - so that if your Cms quadruples (as is the case when you hit the DUMAX suspension limits), then at that finite point in time, your Fs is now double what it normally was. For the audiomobile, this would be around 44hz. That'll throw your alignment off pretty good, I'd think.
Also note that as the suspension stiffens, you'll get even less output with a given input power - as you approach even 200% of the suspension limit, it will start to compress the peaks in the material, and you lose dynamic range. As you hit 400%, this phenomenon gets even worse. This is why I like the Blueprint for your application, as even if you only plan to use 20mm of excursion, the Blueprint's Cms/Kms Vs. X curve is near ruler flat in that region.
Equalizing the rolloff.
I think everyone agrees that a driver like the 1203 has an early rolloff. Other drivers with higher EBP values tend to not do this, hence the recommendation for large EBP in small sealed apps.
The question is, if I am using multiband parametric Equalization (like the BFD), which is prefered? a shallow early rolloff, like .5 Q, or a .707 optimally configured type?
The EQ is goin to have to attempt to approximate an LT. I may try to develop a spreadsheet that does this. Is there any interest (or maybe it's already been done). That will at least provide a starting point for tuning a room, given the model or even actual sweep measurements. I've never tuned (EQ'd) a room before with para EQ, bandwidth adjustments, etc.
I think everyone agrees that a driver like the 1203 has an early rolloff. Other drivers with higher EBP values tend to not do this, hence the recommendation for large EBP in small sealed apps.
The question is, if I am using multiband parametric Equalization (like the BFD), which is prefered? a shallow early rolloff, like .5 Q, or a .707 optimally configured type?
The EQ is goin to have to attempt to approximate an LT. I may try to develop a spreadsheet that does this. Is there any interest (or maybe it's already been done). That will at least provide a starting point for tuning a room, given the model or even actual sweep measurements. I've never tuned (EQ'd) a room before with para EQ, bandwidth adjustments, etc.
Don't forget the room's transfer curve.
The 1203 does indeed have a relatively early rolloff in a sealed enclosure, due to its low Qts.
Having said that, the low Qts also buys you more efficiency, all other things being relatively equal, so the amount of EQ required may not be as bad as it sounds. If the driver had a Qts of 0.4/0.5 for a shallower rolloff, it would likely be down in the 84/85dB/W/M range. It may require less EQ, but as a whole, would likely give very similar amounts of acoustic output once EQ'd flat.
I personally prefer very shallow, relatively early rolloffs, because the room gain transfer curve of most rooms will serve to make this very close to flat in the average sized room - with no electronic EQ at all!
If you make your subwoofer flat down to 16hz anechoic, then in the average room, you're going to have a greatly exaggerated sub-bass region, as the room gain curve gives a heavy boost to the very low frequencies - many people fail to take this into account when designing a sub.
If you do indeed decide to go the parametric EQ route, the best route is likely to acquire a decent SPL meter or some other measurement device, set it up in your listening position, and calibrate your EQ for optimum flatness at your listening position.
When seeing the anechoic F3 of a Qtc < 0.5 alignment, many people are immediately scared off when they see 60 or 70hz - they think that there's no way that alignment could possibly produce any bass. Put it in a room though, and you'll likely see the in-room F6 be well down in the sub-bass range, and that's sans EQ.
- Rick
The 1203 does indeed have a relatively early rolloff in a sealed enclosure, due to its low Qts.
Having said that, the low Qts also buys you more efficiency, all other things being relatively equal, so the amount of EQ required may not be as bad as it sounds. If the driver had a Qts of 0.4/0.5 for a shallower rolloff, it would likely be down in the 84/85dB/W/M range. It may require less EQ, but as a whole, would likely give very similar amounts of acoustic output once EQ'd flat.
I personally prefer very shallow, relatively early rolloffs, because the room gain transfer curve of most rooms will serve to make this very close to flat in the average sized room - with no electronic EQ at all!
If you make your subwoofer flat down to 16hz anechoic, then in the average room, you're going to have a greatly exaggerated sub-bass region, as the room gain curve gives a heavy boost to the very low frequencies - many people fail to take this into account when designing a sub.
If you do indeed decide to go the parametric EQ route, the best route is likely to acquire a decent SPL meter or some other measurement device, set it up in your listening position, and calibrate your EQ for optimum flatness at your listening position.
When seeing the anechoic F3 of a Qtc < 0.5 alignment, many people are immediately scared off when they see 60 or 70hz - they think that there's no way that alignment could possibly produce any bass. Put it in a room though, and you'll likely see the in-room F6 be well down in the sub-bass range, and that's sans EQ.
- Rick
With my intention of stereo subs from one amp, my NAD can deliver 250 per side into 8 Ohms. Dynamic into a 4 Ohm load is 800W, and 1000W into 2Ohm.
Using a 1203 in a 40 L box, what kind of SPL can be expected at 20 HZ (per speaker, no corner loading)? My room is 15x30x8 with 2 hallways.
To estimate power requirements, I use linkwits spreadsheet. According to this, my amp will get 12mm excursion (RMS power) and 18-20 mm on the Dynamic end.
Do you agree with this?
Looking at the 1503 specs, it may be more efficient, yes?
Using a 1203 in a 40 L box, what kind of SPL can be expected at 20 HZ (per speaker, no corner loading)? My room is 15x30x8 with 2 hallways.
To estimate power requirements, I use linkwits spreadsheet. According to this, my amp will get 12mm excursion (RMS power) and 18-20 mm on the Dynamic end.
Do you agree with this?
Looking at the 1503 specs, it may be more efficient, yes?
Download Unibox, and you can figure it out yourself (the SPL @ 20hz, that is)
http://www.danbbs.dk/~ko/ubdwnld.htm
The excursion calculations sound approximately correct.
However, you are indeed correct that the 1503 is more efficient, and the 1803 is even more so.
If you have the little extra money, and a little extra room for a slightly larger enclosure, ignore the 1203 and get the 1503 (or ideally) the 1803 - the extra efficiency really helps, as does the extra volume displacement.
Personally, unless I had to cross the woofer over relatively high (above 100hz), or unless I needed a *really* tiny enclosure, I'd never buy anything but the 1803 - it's so much better at minimal extra cost.
- Rick
http://www.danbbs.dk/~ko/ubdwnld.htm
The excursion calculations sound approximately correct.
However, you are indeed correct that the 1503 is more efficient, and the 1803 is even more so.
If you have the little extra money, and a little extra room for a slightly larger enclosure, ignore the 1203 and get the 1503 (or ideally) the 1803 - the extra efficiency really helps, as does the extra volume displacement.
Personally, unless I had to cross the woofer over relatively high (above 100hz), or unless I needed a *really* tiny enclosure, I'd never buy anything but the 1803 - it's so much better at minimal extra cost.
- Rick
I agree with Rick about not worrying about a low QTS. The problem I have seen is that woofers whose value is over .40 tend to muddy the output or boom. I think it is important for a small enclosure to select a driver with high excursion and power ratings along with a low QTS, and use a parametric EQ to compensate. I recommend using a sine wave generator to adjust the equalization by ear. These can be downloaded as freeware or easily built by a DIYer.
For my subwoofer, I modified an automobile EQ/amp to be able to boost/cut frequencies of 30,45,60,75,90,105,125 Hz. I chose values to complement the unequalized response of my subwoofer.
I also recommend using a lot of polyfill insulation in the enclosure, but be certain to give the woofers some breathing room.
I should also mention that I have added a low pass filter to the output of this EQ/amp. Then I run that output into a few hundred watt amplifier to drive the subwoofer.
For my subwoofer, I modified an automobile EQ/amp to be able to boost/cut frequencies of 30,45,60,75,90,105,125 Hz. I chose values to complement the unequalized response of my subwoofer.
I also recommend using a lot of polyfill insulation in the enclosure, but be certain to give the woofers some breathing room.
I should also mention that I have added a low pass filter to the output of this EQ/amp. Then I run that output into a few hundred watt amplifier to drive the subwoofer.
scanned through old-fashioned Bullock tables to proof this statement from a buddy who is speaker designer. He claimed that at a Q_ts of 0.4, the recommended enclosure volume reaches V_as as a minimum.
So if a driver has low V_as and Q_ts of 0.4, V_as is equal to enclosure volume, else (below AND above 0.40 ) the enclosure is bigger than V_as.
Have a small 2way vented speaker to prove this (can be used as closed speaker, too; parameters fit): true, measured 60Hz from a 4 litres enclosure.
The Stryke HE-15 driver sounds excellent and is almost bulletproof, but virtually needs either a 3000W amp, or a cabinet with many expensive passive radiators to really show what it can do.
For the price, I'd still much sooner have the BPD1803 - the Vd of the 1803 will be as much or more than the stryke, and the extra 3dB of efficiency really does help matters.
Also, consider that for the price of an HE-15 and PRs, you could get two 1803's - which would give you 3dB MORE efficiency, and a huge Vd advantage.
- Rick
For the price, I'd still much sooner have the BPD1803 - the Vd of the 1803 will be as much or more than the stryke, and the extra 3dB of efficiency really does help matters.
Also, consider that for the price of an HE-15 and PRs, you could get two 1803's - which would give you 3dB MORE efficiency, and a huge Vd advantage.
- Rick
I'v got to keep my volume under 2 cubes in each cabinet. I downloaded unibox as suggested. The 1503 (and 1803) would require more power (on paper) than the HE. And needs more volume. HE at 1.5-2 ohm load, 1503 at 4 ohm. The HE makes much better use of my NAD, even if it is slightly less efficient.
An 18 is out of the question for me, too large for the downfiring stereo cabinet footprint I have in my head. I think 15's are going to be more than enough anyway, remembering that I am integrating them with recycled paradigm mains, 6.5" plus tweet.
Thak you for hanging in there with me.
An 18 is out of the question for me, too large for the downfiring stereo cabinet footprint I have in my head. I think 15's are going to be more than enough anyway, remembering that I am integrating them with recycled paradigm mains, 6.5" plus tweet.
Thak you for hanging in there with me.
First off, I should ask which NAD receiver you have, and how much power it will give to your subwoofers. I would ignore the "dynamic" power ratings that NAD publishes, as the power supplies in them are far too small for this kind of a load, and the power rails will collapse a ton under heavy load. While the dynamic power may be useful at higher frequencies, the low frequencies you plan to run through these require the amp to sustain its "dynamic" power output for a much longer period of time (as the length of each cycle of the waveform is MUCH longer), and the amp will likely end up putting out its full dynamic power for the first part of the cycle, and as its filter capacitors and power supply collapse, it will likely put out much less throughout the last part of the cycle - not the terribly best thing from a sound quality point of view.
Not to mention that the 1.5 ohm load presented by paralleling two coils of the HE-15 may be too much even for the NAD, and you'll may find the protection circuits activating more often than not. If this ends up being the case, you'll have to run them in series, giving you 6 ohms, which will end up being much worse from a power utilization point of view than the BPD drivers were.
Seriously though, I think you're underestimating the amount of power required to do the linkwitz transform justice. The suggested cabinet for an HE-15 is a 6 cubic foot passive radiator enclosure, which is way, way WAY more efficient than a tiny sealed box in the low end. Still, even with a ton of extra efficiency gained by this, many people were unsatisfied with 300, 500, 800, and even 1000W RMS! (Note that most of these figures were produced by amps that could actually put out that amount of power continuously - it wasn't "peak" or dynamic power.) Many people ended up buying something along the lines of a crown K2, which put out 2000W bridged into the HE-15. Even with THAT, the HE15 still wouldn't reach its limits.
(for reference, here's the K2 datasheet)
http://www.crownaudio.com/pdf/amps/131429.pdf
I'm not trying to dissuade you here, just trying to make you aware of the enormous *continuous* power requirements that are required for the linkwitz transform to be of any use at high volume levels.
I have a feeling that if you power these off a NAD receiver or integrated amp, it'll sound fine with any material over 40hz or so, but unless you have the volume nice and low, you'll push the amp so far into clipping as soon as you make use of the linkwitz transform (down around 20-30hz or so), that you'll be hearing more distortion products than actual undistorted output.
As for your question re: the paradigms - you might want to consider sealing the vents - with a 50hz vented rolloff, the sealed rolloff will likely start around 80hz, which is right around where most subwoofers are crossed. There's no point in having the vent active below your passband, as it just adds pipe resonances and other garbage that aren't needed. This will also provide a useful reduction in intermodulation distortion as the woofer cone won't have to move as much with any given input.
Also, there is NO such thing as too much swept volume - ever. The more you have, the less you need to use, and the more you have in reserve. The less you need to use, the lower the distortion's going to be. =)
- Rick
Not to mention that the 1.5 ohm load presented by paralleling two coils of the HE-15 may be too much even for the NAD, and you'll may find the protection circuits activating more often than not. If this ends up being the case, you'll have to run them in series, giving you 6 ohms, which will end up being much worse from a power utilization point of view than the BPD drivers were.
Seriously though, I think you're underestimating the amount of power required to do the linkwitz transform justice. The suggested cabinet for an HE-15 is a 6 cubic foot passive radiator enclosure, which is way, way WAY more efficient than a tiny sealed box in the low end. Still, even with a ton of extra efficiency gained by this, many people were unsatisfied with 300, 500, 800, and even 1000W RMS! (Note that most of these figures were produced by amps that could actually put out that amount of power continuously - it wasn't "peak" or dynamic power.) Many people ended up buying something along the lines of a crown K2, which put out 2000W bridged into the HE-15. Even with THAT, the HE15 still wouldn't reach its limits.
(for reference, here's the K2 datasheet)
http://www.crownaudio.com/pdf/amps/131429.pdf
I'm not trying to dissuade you here, just trying to make you aware of the enormous *continuous* power requirements that are required for the linkwitz transform to be of any use at high volume levels.
I have a feeling that if you power these off a NAD receiver or integrated amp, it'll sound fine with any material over 40hz or so, but unless you have the volume nice and low, you'll push the amp so far into clipping as soon as you make use of the linkwitz transform (down around 20-30hz or so), that you'll be hearing more distortion products than actual undistorted output.
As for your question re: the paradigms - you might want to consider sealing the vents - with a 50hz vented rolloff, the sealed rolloff will likely start around 80hz, which is right around where most subwoofers are crossed. There's no point in having the vent active below your passband, as it just adds pipe resonances and other garbage that aren't needed. This will also provide a useful reduction in intermodulation distortion as the woofer cone won't have to move as much with any given input.
Also, there is NO such thing as too much swept volume - ever. The more you have, the less you need to use, and the more you have in reserve. The less you need to use, the lower the distortion's going to be. =)
- Rick
Thanks Thingyness.
Your post really summarizes what my concerns have been all along. I'd like to do the small sealed box, but using an alignment that minimzes the EQ needed for low 20 HZ extension. This has been my argument all along for trying to stay away from high EBP drivers, like the 1203.
The less EQ, the more realistic my amp is. I really don't want to get another amp. It is NAD208, very stable even at 1 ohm (from reading reviews). It may be underrated.
But paying attention to this now, may improve todays performance, even though I may get a bigger Amp down the road, especially after putting in mega hours into this project.
We're almost there.
Your post really summarizes what my concerns have been all along. I'd like to do the small sealed box, but using an alignment that minimzes the EQ needed for low 20 HZ extension. This has been my argument all along for trying to stay away from high EBP drivers, like the 1203.
The less EQ, the more realistic my amp is. I really don't want to get another amp. It is NAD208, very stable even at 1 ohm (from reading reviews). It may be underrated.
But paying attention to this now, may improve todays performance, even though I may get a bigger Amp down the road, especially after putting in mega hours into this project.
We're almost there.
The clipping at 8 ohms is voltage clipping.
In theory, the nad could instantaneously put out 1200W into 2 ohms before voltage clipping, but internal current limiting, along with the collapse of the power supply rails will limit the output to far less than that in practice.
Luckily, your amp isn't as low-powered as I had feared - it may indeed provide decent performance with this setup.
Just know that in your quest for a very small enclosure, and ultra-extended bass, you have thrown efficiency to the wind and will require a ton of power - there's no way around it.
Also, ignore the amount of EQ needed - it's irrelevant. What's more important is the absolute power level required to put out a certain SPL at a certain frequency.
For example, let's assume the blueprint puts out 80dB at 20hz with 1 watt in your enclosure, and the stryke puts out 78dB at 20hz with 1 watt in your enclosure.
Now, let's assume the upper limit of its efficiency comes at 80hz, where the blueprint puts out 92dB with one watt, where the stryke puts out 89dB.
In theory, the stryke would then require 11dB of EQ versus 12dB for the blueprint. Does this matter? Not at all. No matter how much EQ you have to use, the blueprint will still put out more power at 20hz for each watt you throw at it than the stryke.
Imagine that we had a hypothetical driver that was horrifically efficient (say 120dB/W/M - which can't exist but is fine for our thought experiment), but started rolling off very early (say 200hz), and below 80hz followed the exact same rolloff as the numbers i've cooked up for the blueprint, settling to 80dB at one watt at 20hz.
With this theoretical driver, we now require 40db (!) of EQ, but this isn't a bad thing in the slightest. We still have the same efficiency on the low end, but we just need more EQ because our efficiency on the high end is so much better.
Drivers with high EBP's that you'd choose for this application tend to have a low Qts, since it's in the denominator for the EBP equation. A low Qts corresponds to a low Qes, becuase in almost all drivers (tweeters notwithstanding), Qes dominates the Qts since Qms is relatively high. Since Qes is low, and efficiency increases as Qes decreases, the efficiency of a driver with a high EBP tends to be high as well.
This will mean it tends to roll off sooner, but since it had a higher efficiency to start with, it may well indeed be more efficient down low as well, despite the earlier rolloff. This is what matters, *not* a simple dB amount of EQ that you have to apply.
If you like, you can think of EQ not as "boosting" the low end, but "cutting" the high end down to meet the low efficiency in the infrasonic region.
Pick the lowest frequency you expect your sub to produce, and pick that as your reference point, and pick a driver and enclosure such that you get the most output at this frequency with a given power level. Once you do this, you can think of your EQ as "cutting" the high end of the driver, and giving it *less* power as frequency increases. Looking at it this way, if you have two drivers and the low end efficiency is the same, then a greater amount of EQ is in fact *better*, since you now can "cut" the higher frequencies more, and have to output less power to the driver to produce them.
With our hypothetical driver that requires 40dB of EQ, this means that we need 40dB less power if we want the driver to produce 200hz! This means that if you're running the amp at a nominal 200W at 20hz, then the driver is getting in the milliwatt range at 200hz - certainly makes things much easier on the amp and driver, no?
- Rick
In theory, the nad could instantaneously put out 1200W into 2 ohms before voltage clipping, but internal current limiting, along with the collapse of the power supply rails will limit the output to far less than that in practice.
Luckily, your amp isn't as low-powered as I had feared - it may indeed provide decent performance with this setup.
Just know that in your quest for a very small enclosure, and ultra-extended bass, you have thrown efficiency to the wind and will require a ton of power - there's no way around it.
Also, ignore the amount of EQ needed - it's irrelevant. What's more important is the absolute power level required to put out a certain SPL at a certain frequency.
For example, let's assume the blueprint puts out 80dB at 20hz with 1 watt in your enclosure, and the stryke puts out 78dB at 20hz with 1 watt in your enclosure.
Now, let's assume the upper limit of its efficiency comes at 80hz, where the blueprint puts out 92dB with one watt, where the stryke puts out 89dB.
In theory, the stryke would then require 11dB of EQ versus 12dB for the blueprint. Does this matter? Not at all. No matter how much EQ you have to use, the blueprint will still put out more power at 20hz for each watt you throw at it than the stryke.
Imagine that we had a hypothetical driver that was horrifically efficient (say 120dB/W/M - which can't exist but is fine for our thought experiment), but started rolling off very early (say 200hz), and below 80hz followed the exact same rolloff as the numbers i've cooked up for the blueprint, settling to 80dB at one watt at 20hz.
With this theoretical driver, we now require 40db (!) of EQ, but this isn't a bad thing in the slightest. We still have the same efficiency on the low end, but we just need more EQ because our efficiency on the high end is so much better.
Drivers with high EBP's that you'd choose for this application tend to have a low Qts, since it's in the denominator for the EBP equation. A low Qts corresponds to a low Qes, becuase in almost all drivers (tweeters notwithstanding), Qes dominates the Qts since Qms is relatively high. Since Qes is low, and efficiency increases as Qes decreases, the efficiency of a driver with a high EBP tends to be high as well.
This will mean it tends to roll off sooner, but since it had a higher efficiency to start with, it may well indeed be more efficient down low as well, despite the earlier rolloff. This is what matters, *not* a simple dB amount of EQ that you have to apply.
If you like, you can think of EQ not as "boosting" the low end, but "cutting" the high end down to meet the low efficiency in the infrasonic region.
Pick the lowest frequency you expect your sub to produce, and pick that as your reference point, and pick a driver and enclosure such that you get the most output at this frequency with a given power level. Once you do this, you can think of your EQ as "cutting" the high end of the driver, and giving it *less* power as frequency increases. Looking at it this way, if you have two drivers and the low end efficiency is the same, then a greater amount of EQ is in fact *better*, since you now can "cut" the higher frequencies more, and have to output less power to the driver to produce them.
With our hypothetical driver that requires 40dB of EQ, this means that we need 40dB less power if we want the driver to produce 200hz! This means that if you're running the amp at a nominal 200W at 20hz, then the driver is getting in the milliwatt range at 200hz - certainly makes things much easier on the amp and driver, no?
- Rick
Rick,
Right before I read that post, I was doing what you said. I have decided to just look at the 20 HZ SPL with 300W in my 2 cube box. Both are about 100 dB. The HE-15 just 1/2 dB better. The 1203 also comes in about the same, the biggest diff is that it is closer to the excursion limit, which makes sense.
In all the drivers alignments run from .5 to .6 for the specified volume.
Now in theory, doubling to 600W will present 3 dB gain. Yup, plug it in and sure enough.
Your argument is proven, explanation, excellent! Thanks for taking the time.
Now maybe another NAD will come my way.
Michael
Right before I read that post, I was doing what you said. I have decided to just look at the 20 HZ SPL with 300W in my 2 cube box. Both are about 100 dB. The HE-15 just 1/2 dB better. The 1203 also comes in about the same, the biggest diff is that it is closer to the excursion limit, which makes sense.
In all the drivers alignments run from .5 to .6 for the specified volume.
Now in theory, doubling to 600W will present 3 dB gain. Yup, plug it in and sure enough.
Your argument is proven, explanation, excellent! Thanks for taking the time.
Now maybe another NAD will come my way.
Michael
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.