Inside the kill zone (nearfield) the comb filtering is tolerable. It doesn't just flatten out because of all the sources - your ears/brain does the rest. You wouldn't be able to hear a 5dB dip if you tried. Even several closely spaced ones. The louder it is the less yoiu notice, and the bigger the stack the further back it reaches before it makes that transition to farfield.
Away from the stack, from a moving vehicle or with a lot of wind it's a totally different story. It will have that classic "phasey" sound. You're not supposed to be that far away when the party's going on.
Those pics are incredible to me... I also agree with the FLH. Never heard of the super bass horn but it looks great. Doing some googling for it lead to some interesting info and the name "King Earthquake". There also is the 12pi: Pi Speakers - Push/Pull Basshorn Subs
The SBH looks like a heck of lot easier to build than than the Labhorn. I've always envisioned the guys here would someday put their heads together to best the LABhorn, but they're all old farts (like me!) that don't like EDM nor want to move the big SOB FLHs.
In any event, the key attribute to the FLH is that the low corner gets lower and louder in multiples and with what you're doing that would be ideal.
The SBH looks like a heck of lot easier to build than than the Labhorn. I've always envisioned the guys here would someday put their heads together to best the LABhorn, but they're all old farts (like me!) that don't like EDM nor want to move the big SOB FLHs.
In any event, the key attribute to the FLH is that the low corner gets lower and louder in multiples and with what you're doing that would be ideal.
I agree, where I live I am pretty sure it is illegal to paint your house any color other than dark tan, light tan or tan or drive anything other than a grey 4-door. I can't imagine the kind of fun that party would be
I've always envisioned the guys here would someday put their heads together to best the LABhorn, but they're all old farts (like me!) that don't like EDM nor want to move the big SOB FLHs.
There's the time and resource commitment required (no one can build just one - you need 8) for something that gets used rather infrequently. And then where to put all of them until the labhorns are sold - they would stay until the replacements are complete
That's why I do have 4 TH18's - they are very practical everyday cab.
I'm pretty sure I'll be an even older fart when I get around to building 'the big one'. If I ever get old and decrepit enough that I can't even wheel the TH18's around anymore, just shoot me.
Those pics are incredible to me... I also agree with the FLH. Never heard of the super bass horn but it looks great. Doing some googling for it lead to some interesting info and the name "King Earthquake". There also is the 12pi: Pi Speakers - Push/Pull Basshorn Subs
The SBH looks like a heck of lot easier to build than than the Labhorn. I've always envisioned the guys here would someday put their heads together to best the LABhorn, but they're all old farts (like me!) that don't like EDM nor want to move the big SOB FLHs.
In any event, the key attribute to the FLH is that the low corner gets lower and louder in multiples and with what you're doing that would be ideal.
The size of the SBH is a problem for me... I have Yorkville 1208's. I wanted something smaller than those. I'm 49, I'm not into stacking 175 lb boxes like I use to.
I saw a hybrid horn... but wasn't sure how they sounded.
I'm currently reviewing this thread for inclusion of data about this design in to a summary sheet, and I ran across this. FWIW, I've measured changes in the impedance curve of my TH as the loading on it changed, and if there are changes in the impedance curve, then there will be corresponding changes in the FR.
I performed impedance measurements on my POC3 with it raised in the air, on the floor, against a wall and in a corner.
If you agree that changing the loading on one TH as I have done is the equivalent of doubling or quadrupling the number of THs (which is an assumption made many times in simulating these builds in HornResp), and that changes in the impedance curve imply changes in the frequency response, then if my measurements show changes in the impedance curve when I change the loading on a TH, that means that the frequency response will also change when the loading is changed, or when the number of stacked THs is changed.
Again, going through my fact-checking exercise here, but the premature rolloff in this FR graph suggests that there is some leakage around the driver. See the following link for more details: http://www.diyaudio.com/forums/subwoofers/275969-when-things-dont-go-quite-planned.html
I ran into the same issue with one of my builds, and this was partly caused by making the mistake of setting the cutout in the baffle to be the same as the driver's published cutout requirements, even though the driver not going to be mounted through the baffle. The cutout should actually be at least a few mm smaller than the published requirement, to minimize the possibility of any leakage around the driver. E.g. if the driver's published cutout requirement is 43 cm, make the actual cutout 42.5 cm, or even 42 cm if you're sure that the driver's surround won't hit the baffle at high excursion.
Exactly, and one can make deductions from those measurements, if you understand what they mean. In this case, if the measured impedance curve of the DUT changes, then the frequency response of the DUT MUST ALSO CHANGE. You can't have one without the other.
So the deduction that can be made from my test is either this means that the frequency response WILL change if multiple THs are stacked, or the assumption that you can change the simulation to reflect the output of multiple THs by changing the loading from 2PI to 1PI or 0.5PI is also incorrect. As this is a common assumption made when using HornResp to model horns in general, what would be the basis to conclude that it is somehow incorrect for THs?
The unanswered question here is if the changes to the impedance curve reflect the magnitude of the changes suggested by the corresponding HornResp sim when the loading is changed. That's something I haven't checked yet.
Hi Brian,
Changing the radiation angle (2Pi, 1Pi…) as used in the multiple speakers function in order to calculate the acoustic power, is a 'point source model' that disregards directivity. If you stack multiple speakers they always become directional (which is the point of stacking). In several papers from the late eighties and early nineties about modelling they explain why a 'directional point source model' should be used (today they use hybrid models but fwiw, I don't think that is necessary for designing cabs or analysing virtual subwoofer stacks).
Also an acoustic power response relates to an omni-directional source. Even if the enclosure is directional (like a horn), in a single arrangement you can get away with it. For multiple enclosures a 'sound pressure response' is needed, which is the standard in the PA industry.
Cheers,
Djim
Changing the radiation angle (2Pi, 1Pi…) as used in the multiple speakers function in order to calculate the acoustic power, is a 'point source model' that disregards directivity. If you stack multiple speakers they always become directional (which is the point of stacking). In several papers from the late eighties and early nineties about modelling they explain why a 'directional point source model' should be used (today they use hybrid models but fwiw, I don't think that is necessary for designing cabs or analysing virtual subwoofer stacks).
Also an acoustic power response relates to an omni-directional source. Even if the enclosure is directional (like a horn), in a single arrangement you can get away with it. For multiple enclosures a 'sound pressure response' is needed, which is the standard in the PA industry.
Cheers,
Djim
Last edited:
Hi Brian,
I really want to see the data.
My guess is that the TH, traditional BLH "scoop bin", and even typical ported bins probably go lower when coupled in stacks beyond what directivity modeling would dictate.
I think the box to air interface/load changes from less than 2pi for a single box with typically a fraction of the area required to achieve proper directivity at the LF cutoff to better than 2pi as you have potentially more radiating mouth than needed to make directivity down to the cutoff in a large stack.
I really want to see the data.
My guess is that the TH, traditional BLH "scoop bin", and even typical ported bins probably go lower when coupled in stacks beyond what directivity modeling would dictate.
I think the box to air interface/load changes from less than 2pi for a single box with typically a fraction of the area required to achieve proper directivity at the LF cutoff to better than 2pi as you have potentially more radiating mouth than needed to make directivity down to the cutoff in a large stack.
Hi Dan,
I think you are right, directivity modelling probably doesn’t show a drop of the low corner since it is not related to directivity.
I think it is related to what is called a 'bubble' of the mouth (it works similar for ports) that extends the path outside its physical dimension with a factor 0.6 of the diameter. In a stack that is 0.6 of the diameter of the combined mouths (or ports).
However, since ports are very small in diameter they hardly, if at all, cause a drop of the low corner.
When the port is close to the ground it can cause a drop of the low corner because the ground can also extend the function of the port (or mouth).
Cheers,
Djim
I think you are right, directivity modelling probably doesn’t show a drop of the low corner since it is not related to directivity.
I think it is related to what is called a 'bubble' of the mouth (it works similar for ports) that extends the path outside its physical dimension with a factor 0.6 of the diameter. In a stack that is 0.6 of the diameter of the combined mouths (or ports).
However, since ports are very small in diameter they hardly, if at all, cause a drop of the low corner.
When the port is close to the ground it can cause a drop of the low corner because the ground can also extend the function of the port (or mouth).
Cheers,
Djim
Last edited:
This is probably correct.
I'm going to repeat the measurements tomorrow to confirm (SWMBO will be out, so I won't be disturbing anyone with the DAT's sine sweeps), but I think the greatest change took place when I moved the TH from the table (where there was no surface nearby to the TH's mouth) to the floor, and then only minor changes to Fb (and a bit more to the peak in the passband) moving it from the floor to against a side wall.
Here's something to look at:
1. Blue - POC3 TH on a table, ~3 feet off the ground
2. Red - POC3 TH on the floor, center of the room
Fb dropped by 3% (based on the raw data, it's not that visible in the graph), and there's a noticeable drop in the impedance peak in the middle of the passband.
