[scorpion]

Sorry for the confusion. Figures are for the part of H-lenght in front of the baffle as can be set in MJK's worksheet and the total lenght was set to 15".
Yes we will see if this simulation can stand the test equally good as Martin's Goldwood predictions.

Thanks Mark, For IBs I have to put them in the roof in my living room so that will take time, the U-dipole will be a neat packet 18x18 inches and  12 " deep. With these units I think I will be able to produce 100 dB at 20 Hz without bottoming so I will try them also in my HT, homebuilt with Fostex 167, 107 and 103 units.

/Erling

[markC]

I'll be interested in how the U baffles turn out as it's easy for me to remove the IB's and try a different config.

[klykx]

MJK,

What happened to the lowther baffles painted to "blend in" with the decor?

cheers

Klykx

[MJK]

What happened to the lowther baffles painted to "blend in" with the decor?

Klykx,

I salvaged all of the parts like the drivers, binding posts, hinges, wires, and the connectors and set the big pieces of plywood aside for building something else. I will probably build a smaller Lowther OB system that can be easily moved in and out of my room by one person.

Martin

[MJK]

If I was willing to give up on the separate toeing of the OB - what problems
would I run into if I made a single unit approx 40 x 20 with a H on the bottom
part and the OB on the top?  do they need to be isolated?  Should I be worried
that the OB will be vibrating off the H woofer?

Sorry Jeff, it has been a hectic couple of days.

I think a single piece baffle construction would work fine. You could build an OB and than add a back for an U frame and then a front to form an H frame and see which alignment works best in your room. I would not be too worried about vibration from the 18" woofer, it is only powered up to 100 or 200 Hz and these size plywood panels probably would only start to vibrate at higher frequencies.

Martin

[scorpion]

As for the AE_IB15 optimized in 9" U-frame crossover 75 Hz 2nd order LR it will be enough to boost, no EQ will be needed. X-max 18.5 mm, for 95 dB at 30 Hz it will move +- 6 mm:



/Erling

[planet10]

I use the dBx Driverack Pro but the Behringer DCX would also work

I don't know about the dBX, but the Beringer i had, althou very useful for dialing in XO details (& measuring) was not very good sounding.

dave

[doug s.]

pardon the doofus question:  how do you connect the 4 binding posts shown to the 2 on my receiver?

if your receiver has preamp outputs & main inputs, you could use it by adding an outboard active x-over and a stereo amp.  feed the receiver's pre outs to the inputs on the x-over, then feed the low pass & hi-pass outs from the x-over to the receiver and the stereo amp.  depending on the quality of the receiver's amp, you could either use its amp for the woofers and the new amp for the jordans, or get a sub amp for the woofers & use the receiver's amp for the jordans...

doug s.

[holdent]

Erling
"As for the AE_IB15 optimized in 9" U-frame crossover 75 Hz 2nd order LR it will be enough to boost, no EQ will be needed. X-max 18.5 mm, for 95 dB at 30 Hz it will move +- 6 mm"

I was also intrigued by the AESpeakers IB15 and simulated it in both U- and H-frames.  In the end I chose not to go that route because either you have to roll them off early (such as 75Hz as you use) or they need substantial EQ.  Either way you need substantial power.  At the 95dB level for 30Hz you noted you'll be pushing 100 watts. 

I recall thinking that they made a a more OB-specific driver but couldn't find it on the website.

[MJK]

I recall thinking that they made a a more OB-specific driver but couldn't find it on the website.

Lambda Acoustics made an OB version that had a higher Qts. I found it referenced in a thread at DIYaudio and it looked interesting. It was a historic page that somebody dug up. The price was $249 each (back in those days) which seemed a little high and while I am sure the build quality was top notch I could not justify that cost when there were drivers in the $60 range that would do the same job.

Martin

[Saurav]

I recall thinking that they made a a more OB-specific driver but couldn't find it on the website.

There's the Lambda Dipole drivers that are in the higher price range, then there's the AE OB12/15 which are based off the IB12/15 range and are priced lower. I believe I have the first pair of OB12s in existence. I'm enjoying them a lot. I don't have a good enough setup to take T/S or distortion measurements, but I did run some simple FR measurements. This post has T/S numbers from AE, and my FR measurements:

http://www.aespeakers.com/phpbb2/viewtopic.php?f=14&t=1697

And you're right, it's probably not yet on their website.

[scorpion]

I think I will use Martin's argument: you certainly won't, with the exception of 'Also sprach Zarathustra', play 30 Hz continously for longer periods. So the effect take-out will be momentary and within the reach of well designed 100 watt SS amps like my NAD 216. I was impressed to get a so flat response all the way down to 20 Hz from such a small dipole enclosure. The IB efficiency will be 87 dB/watt. 

AE Lambda Dipole 15 figures are given within this thread: http://www.aespeakers.com/phpbb2/viewtopic.php?f=3&t=536 .

/Erling

[scorpion]

The wonderful thing with MJK's models is that you can go on simulating and simulating.

Now this is what I will build when the AE IB15s arrive. U-dipole, inner measures 24 x 24 inches, 11 inches deep (corresponding to timber I have at home), a bit more spacious than I thought from the beginning but this is what it will produce with 100 watts input and crossover set to 75 Hz LR 12 dB/oct and +- 7mm cone movement at 30 Hz:



The 1 watt response is like this:



So you see the trade off is rather between side lenght and efficiency than U-lenght and efficiency, calculation is at 32" height and 1 m from U-dipole

/Erling

[John_E_Janowitz]

I looked at the AE Dipole 15, the specs look impressive. But the price was shown as $249 each. The $129 (or 4 for $100) is for the AE IB 15 which has a lower Qts. So if we used one of the AE drivers, we would need EQ plus boost for the IB 15 or just boost for the Dipole 15. If we did that right we could approach the performance of the $60 Goldwood or Alpha 15A drivers. Why would we spend more money and jump through more hoops to get to the same place? Since the driver only works from fs to 100 Hz, I am not convinced that paying more is going to dramatically increase the bass quality or quantity.

Hi Martin,

The IB15 with an F3 of 16hz and Qts of just under .7 models in an infinite baffle to have an F3 just under 20hz.  When you put this driver on an open baffle, you have to compensate for the baffle rolloff itself, but there i no EQ needed.  At 18.5mm Xmax, the IB15 can displace 3.07L of air peak to peak where the Alpha15 A can displace only .65L.  You would need 4.7 Alpha15A's to equal the output at low frequencies you can get from the single IB15.  If this is a full range system intended to get down to say 30hz, the Alpha15's are not very practical.  Buying 5 of them vs one IB15 isn't practical and either is mounting them on a baffle.  Also the IB15's have the full copper sleeve on the pole which will linearize inductance,decrease distortion, and keeps power compression low.  They will sound much better as excursion gets increased and have no problems with the power required to compensate for the baffle rolloff.

The Dipole15's add even more linearity as they are a 12mm underhung design.  I wouldn't suggest them for a subwoofer specific system only up to 100hz, but in a 3way system I'm not aware of any options that are better.  They can play cleanly up to 2KHz or so.  Inductance is lower than most 5" drivers.  Again you'd need over 3 of the Alpha 15's to match output to the single Dipole15.  When you factor this in, along with all the benefits in distortion, they are not overly expensive.  If you're trying to reproduce the low B of a bass guitar at reasonably high levels and do it cleanly, this is the way to go. 

John

[John_E_Janowitz]

Erling
"As for the AE_IB15 optimized in 9" U-frame crossover 75 Hz 2nd order LR it will be enough to boost, no EQ will be needed. X-max 18.5 mm, for 95 dB at 30 Hz it will move +- 6 mm"

I was also intrigued by the AESpeakers IB15 and simulated it in both U- and H-frames.  In the end I chose not to go that route because either you have to roll them off early (such as 75Hz as you use) or they need substantial EQ.  Either way you need substantial power.  At the 95dB level for 30Hz you noted you'll be pushing 100 watts. 

I recall thinking that they made a a more OB-specific driver but couldn't find it on the website.

I think something must be off in the modeling here if you are seeing 100W to get 95dB at 30hz.  Erling's model shows +/- 6mm at 30hz.  To move an IB15 to this excursion at 30hz requires approximately 45W.  The rolloff of the baffle is responsible for the system being about 9dB down at this point.   You can apply as much as 400W while keeping the driver within Xmax limits.  This should give you approximately 105dB at 30hz with a single driver on this baffle.  In comparison again, the Alpha15A is excursion limited to only 91dB maximum at 30hz. It requires only about 15w to reach this level, but you're looking at a huge difference in maximum output capability.

I think people tend to complicate things with open baffles a lot.  One thing to note is that the baffle doesn't care which driver is mounted to it.  The the baffle rolloff will be the same no matter which driver is used.  So knowing the rolloff of the baffle itself, you can simply find what the rolloff of the baffle itself will be and then apply this to the curve from any driver mounted in an infinite baffle alignment. 

John

[Magnetar]

This is pretty much true save the amount of in room free space loading gain and mutual coupling gain if you are using multiple drivers. My preference is still the sound of drivers between .7 and .85 QTS mounted on a flat board, no wings - (use lots of them - biampe/triamped system!) with a good sub below 40 cycles. In order to get a driver with this QTS range down that low you either have to EQ or use a group of them in a line. I prefer the group because of lower distortion, higher power handling and much higher efficiency. Several inexpensive 10-12-15 inch drivers may not cost any more either.


[/quote]


I think people tend to complicate things with open baffles a lot.  One thing to note is that the baffle doesn't care which driver is mounted to it.  The the baffle rolloff will be the same no matter which driver is used.  So knowing the rolloff of the baffle itself, you can simply find what the rolloff of the baffle itself will be and then apply this to the curve from any driver mounted in an infinite baffle alignment. 

John
[/quote]

[MJK]

John,

I substituted the AE IB 15 for the Eminence Alpha 15A in the Lowther DX2 OB system that we discussed in another thread. This is an idea that had been discussed and both an active and passive crossover option were designed and the responses posted in the thread. This design is typical of the way I design OB speaker systems.

Bottom line is that in my opinion the Alpha 15A is a better driver for this style of OB design then the AE IB 15. Granted the AE IB 15 is probably a better designed, better built, and a more robust driver compared to the budget minded Alpha 15A but in my opinion it falls short on a number of points.

1) The Qts is too low, I really like to see a Qts > 1 for a woofer that is going to be used in a passive mode on an OB.

2) The low fs is wasted, as you correctly stated above the baffle is going to dominate the low frequency roll off and for a 24" wide by 48" tall OB you are just not going to get much below 40 to 50 Hz. Having a super low fs is not an advantage, I would trade a higher fs to achieve a higher Qts.

3) The efficiency is so low that a significant amount of boost is required to get enough SPL, even fomr a pair of IB 15's per side, to match with the Lowther DX2. A pair of Alpha 15A drivers can be matched with the Lowther and still use a passive crossover, no boost is required.

4) I don't belive that a huge Xmax is necessarily a requirement. I have used two Alpha 15A drivers in an OB and played the system at uncomfortably loud levels and still had to touch the Alpha 15A drivers to prove to myself that they are really moving. My experience says it will take a very out of the ordinary recording at played extreme SPL's to exceed the Xmax of the Alpha 15A drivers.

5) The cost difference is almost double, you can buy two Alpha 15A drivers for the price of one AE IB 15 drivers. The cost of the OB 15 seems to be even higher based on the old web page somebody referenced at DIYaudio.

I am sure your drivers are very nice and I have read nothing but glowing reports. But for OB use I would rather have a higher fs, a higher Qts, and a higher SPL/W/m. The IB 15 has the eye opening stats but I just don't see them as an advantage for the types of OB systems I favor, I can not justify paying more then the $50 to $60 it costs for an Eminence of Goldwood woofer. That is only my opinion and others may look at OB design very differently.

Martin

[John_E_Janowitz]

I substituted the AE IB 15 for the Eminence Alpha 15A in the Lowther DX2 OB system that we discussed in another thread.

I am curious if this was simply just a model or if you had built and measured one.  As I'll show below, there is much more to look at than just the small signal models.

1) The Qts is too low, I really like to see a Qts > 1 for a woofer that is going to be used in a passive mode on an OB.

This is one school of thought on this here.  When the Dipole15 was designed by Lambda back about 6 yrs ago, there were discussions based on both schools of thought.  One is that the Qts of 1.0 or higher is beneficial to extend the low end response.  You essentially get a bump in the driver response above Fs that will be countered by the rolloff from the baffle.  This sums flat, extending response a little lower.  While this is a good think, the other school of thought is that a Qts of 1.0 is just extremely loose and "flabby" sounding as some have commented.  The other school of thought, including the likes of Linkwitz want a Qts in the .5 range.  Their argument is that a small amount of EQ can compensate for the additional rolloff, but with the greater than 1.0 Q driver there is nothing that can be done to better damp the system.  The IB15 wasn't originally intended for Dipole/Open Baffle use, but the Q does fall right within both of those 2 ranges.  The OB15 version which has stiffer spider and lower mass will end up right around .75 for a Q.  

2) The low fs is wasted, as you correctly stated above the baffle is going to dominate the low frequency roll off and for a 24" wide by 48" tall OB you are just not going to get much below 40 to 50 Hz. Having a super low fs is not an advantage, I would trade a higher fs to achieve a higher Qts.

The above depends on your goals.  With 2 drivers per channel and a decent amount of power you have plenty of ability to EQ and get lower in frequency.  Most people I talk to want to reach down to around 30hz to be able to cover a low B on the bass guitar.  Again there is a huge difference in the output levels.  The IB15 again wasn't originally intended for this use, people just started using them and were very happy.  We then did the OB15 variant that has a 27hz Fs and is 89.5dB 1W.  A pair of these is 97.5dB at 2.83V, about 3dB more than the Alpha15.  So with a pair of the OB15's you'd have about 3dB more sensitivity than the single Alpha15 and you would have 19.7dB more maximum output capability at 30hz.  At 36hz and below the OB15 pair becomes more efficient.  The only drawback is that you would require an additional 2.5dB of EQ centered at 53Hz, to equal the bump in response given by the Alpha15 at that point.  MUCH more output capability, better damping, lower distortion, at only the cost of 2.5dB of EQ that can be applied by one parametric band.

3) The efficiency is so low that a significant amount of boost is required to get enough SPL, even fomr a pair of IB 15's per side, to match with the Lowther DX2. A pair of Alpha 15A drivers can be matched with the Lowther and still use a passive crossover, no boost is required.

I think here we need to define what you mean by boost so some can understand.  In an open baffle there are 2 two things that require EQ.  First is the rolloff of the baffle.  This is defined by the dimensions of the baffle and doesnt' change no matter what driver you put in.  Second is to EQ for the response of the driver.   If the driver begins to roll off higher in frequency, this combines with the rolloff of the baffle and you're needing to EQ to account for the sum of both of these 2 rolloffs.  

In the case of the IB15, this driver is going to be only 1dB down at 25Hz.  The OB15 is 3.6dB down at 25Hz.  The Alpha15A is going to be 7dB down at 25Hz.   If you don't need to go that low, look at 30hz.  The Alpha15 is 4dB down, IB15 .6dB down and OB15 2.26dB down.  Now you can look at whatever the rolloff from the baffle is and add it to any of those numbers and you get the total amount of EQ you need.  

If 30hz is your low end limit you actually need more EQ for the Alpha15 to be flat to that point.  From 37hz on up to about 140hz the Alpha15 does have a bump due to the high Qts.  This is often what is preferred in the open baffle alignments, but in reality as stated above, the maximum amount of boost from this high Q bump is about 2.5dB at 53Hz.  In reality this is a quite minor benefit at the expense again of the less damped sound.  

One thing to note also is that the "claimed efficiency" of the alpha15 is not in reality anywhere near what the calculated efficiency is.  IT calculates out to 92.7dB 1W vs the 97dB published, based on the parameters.  At 2.83V it comes in at 94dB.  A pair of the IB15's in parallel at at 2.83V comes in at 93.3dB, not even a full dB less.  The pair of OB15, 97.2dB.  A pair of Alpha15's is 100dB at 2.83V.  In any active system, these numbers are all very workable.

4) I don't belive that a huge Xmax is necessarily a requirement. I have used two Alpha 15A drivers in an OB and played the system at uncomfortably loud levels and still had to touch the Alpha 15A drivers to prove to myself that they are really moving. My experience says it will take a very out of the ordinary recording at played extreme SPL's to exceed the Xmax of the Alpha 15A drivers.

 

Again, you stated above your goals are 40-50hz.  If you're targeting a full range music system to not need a subwoofer, you're going to want to get close to 30hz.  To play at a certain volume level requires a given amount of displacement.  There is no way around that.  I can find you various recordings with bass guitar, keyboard, etc from all genre's of music that have material at those frequencies.   With the Alpha15A's, they have 3.8mm Xmax and claimed 8.4mm overall suspension travel.  You also have a very peaky BL curve.  The suspension begins to get very stiff about 4mm.  By 3.8mm your suspension has nearly doubled in stiffness and you have only 70% of the Bl that you did at the rest position.  By 5mm it is exponentially worse.  You're not going to move the driver that much because you simply don't have enough force to do so.  You can turn it up louder and louder and make things uncomfortable, but at those lowest frequencies, they just don't get much louder as the driver can't move any farther.  The driver is compressing those lowest notes.  It's how they keep the drivers from bottoming out and being damaged.   Harmonic distortion is also skyrocketing as this happens.  In these cases the fundamental note you're playing is often significantly lower in level than the harmonics.  

Another factor to look at is what happens to the Q as you get out to Xmax.  At 3.8mm with the Alpha15, Bl is dropped to 70% of the rest value.  This means that the Qts is now approximately 2.2 vs the 1.26 it starts with.  Your efficiency has dropped by 3dB and the peak at Fs is now sharper.  Add in the effects of the stiffer suspension at the higher excursion and the Fs has gone up to 58hz and Qts is now at 3.06.  Now this is simplified a little but it illustrates the point of what happens to the force, compliance, and loss of output.  When you use a short Xmax driver with very high Q already, anywhere near it's Xmax, things get pretty wild pretty quickly.  

Now compare that to a driver with higher Xmax, a more broad BL curve, similar Q over a higher excursion range, etc.  Clean output and much more headroom.  You can expect that the parameters are going to stay 95% the same out to about 10mm or so each way.  You may not even need the full Xmax, but when keeping a driver operating within a linear range, I think the benefits are quite clear.

5) The cost difference is almost double, you can buy two Alpha 15A drivers for the price of one AE IB 15 drivers. The cost of the OB 15 seems to be even higher based on the old web page somebody referenced at DIYaudio.

Yes, you can buy 2 of the Alpha15A's for the price of one IB15.  The OB15 is I believe $20 more as we do some other tricks to the surround to smooth out the respone.  You can essentially take the OB15 up to 1KHz or higher if necessary.  In any case, if 30hz is your goal for low end output, you'll still be excursion limited to about 6dB less with a pair of the Alpha15A's.  Again all of the above applies to the issues as you are getting the Alpha15's near their Xmax, much sooner than a single IB15 or OB15 would.

I am sure your drivers are very nice and I have read nothing but glowing reports. But for OB use I would rather have a higher fs, a higher Qts, and a higher SPL/W/m. The IB 15 has the eye opening stats but I just don't see them as an advantage for the types of OB systems I favor, I can not justify paying more then the $50 to $60 it costs for an Eminence of Goldwood woofer. That is only my opinion and others may look at OB design very differently.

That's the joys of DIY, you can do exactly what you desire.  Some may share the same desires as you, and some may look at things from a different perspective.  People can build what they feel is necessary though, not rely on others to tell them what to do.  I just wanted to make sure I could point out things from a different standpoint.  

John

[Graham Maynard]

Hi John,

What a good case you made for the AE drivers.

However there is one major point I wish to pick up on - efficiency/sensitivity.
Yes parallel AE drivers will perform well on OB with the more sensitive widerange drivers, but they will need greater,and thus separate amplification of say 250W, whereas 2x Alpha will be fine with just a single 50W amplifier driving everything.

I agree with Martin that the low Fs is wasted - and a similar AE driver would actually perform better on an OB with Fs a little higher, say with a lighter cone.  If LF power is available I would suggest choosing a driver with an Fs circa 25Hz;  if not 30 to 40Hz is a good start.
I also agree with Martin that once you get to 15" you can get satisfactory 'in-room' LF listening levels without needing more than say 4mm of X.max, especially with paralleled drivers.  Of course we need to be realistic about desired as opposed to satisfactory listening levels.

Hi Martin,

As you know I am one who likes lower Qes LF drivers for OB.
The Alpha-15A does not satisfy my requirement in this regard, but I wonder if when two are parallel connected in a U or H frame - does the additional LF loading and mutual coupling not reduce their Q ? 
Have their T-S characteristics, or even just their new 'system' resonant frequency, ever been measured in such an arrangement ?
I suggest that the natural driver Q might be reduced when thus used, more so than would be a driver which already has a low Q ?

Further to this discussion, I would suggest that the AE driver sound will not integrate as well as an Alpha with a widerange on an OB, and that using 2 in parallel will not improve this aspect.

Cheers ........ Graham.

[MJK]

The Alpha-15A does not satisfy my requirement in this regard, but I wonder if when two are parallel connected in a U or H frame - does the additional LF loading and mutual coupling not reduce their Q ? 

When you put the Alpha 15A in a U or H frame you essentially add moving mass, the air in the pipe, to the system. The result is that the efficiency goes down, the fs goes down, and the Qts goes up.

Martin