It is a pity the TH18 path* didn't get all the way to a generally accepted as being optimal finalized design. I would go out and buy a pair of 18sw115's tomorrow if the plans were around. I know this is kind of lazy but i just can't justify learning to do the last bit of design my self as I am sure it would take me many many hours to learn and then design.
Even smaller TH bins should outperform BR boxes of around the same size, using the same driver. Sim programs tend not to take into consideration the impact of vent compression effects (significant with BR boxes and tend to start happening even before the onset of audible chuffing), and thermal compression (the layout of a TH results in better cooling for the driver than your typical BR box). There's also a possibility that the S1-S2 section of a TH can also help to reduce overall distortion, but I haven't explored that fully yet by experimentation (I have noticed it in offset TLs, currently my favourite alignment, that have a similar arrangement).
THs do have the disadvantage of having a smaller usable passband, but this usually isn't an issue for boxes designed for subwoofer duty.
The TH18 design can perhaps do with a bit of tweaking however to bring it up to date. Since its original design, we've learned of semi-inductance, and its impact on output at higher bass frequencies. Hornresp now allows the use of the semi-inductance parameters in the sims, so their impact can be taken into the consideration in the design.
THs do have the disadvantage of having a smaller usable passband, but this usually isn't an issue for boxes designed for subwoofer duty.
The TH18 design can perhaps do with a bit of tweaking however to bring it up to date. Since its original design, we've learned of semi-inductance, and its impact on output at higher bass frequencies. Hornresp now allows the use of the semi-inductance parameters in the sims, so their impact can be taken into the consideration in the design.
I always challenged anybody to prove that. Didn´t happen, and that´s why I don´t have tapped horn instead of bassreflex box Yet. Using the same large high-end driver (18DS115, 21DS115), I cannot get to touch the bassreflex box power density with anything.
For PA usage, I´m even considering partially supressing bassreflex port circuit by cutting at (low) tuning freqency, as for the most part, bassreflex only helps with loading from the amplifier, but not so much with efficiency it seems.
From what I´ve seen TH18 is far from being 250l box.
Also with bassreflex box, it´s not a problem to work in 30-100Hz pasband too.
I fail to see the universal advantage of THs. Only where we are not that much constrained by the box size, and we need to squeeze most output per driver, then it works for me too...
For PA usage, I´m even considering partially supressing bassreflex port circuit by cutting at (low) tuning freqency, as for the most part, bassreflex only helps with loading from the amplifier, but not so much with efficiency it seems.
From what I´ve seen TH18 is far from being 250l box.
Also with bassreflex box, it´s not a problem to work in 30-100Hz pasband too.
I fail to see the universal advantage of THs. Only where we are not that much constrained by the box size, and we need to squeeze most output per driver, then it works for me too...
Have you published the details of your findings anywhere? I'd be interested in the observations that you made.
I think he's on about HPF and shoving more drivers in less space, it may have an advantage when the boxes are relatively small. Increased thermal capacity, in exchange for limited extension.
NEO Dan: What´s up with HPF? Yes, I´m also about shoving more drivers in less spece, even if it easily can mean only one driver in comparison...
I have built few tapped horns, and I like the general performance, construction, sound. Yet its power density in "reasonable sizes" was not convincing.
Brian Steele: No, I usually scrap things including boxes, as all becomes less relevant in time. Different drivers, different amps, different circumstances between comparison and measurements. Findings simply do not add up to make a cumulative study full of worthy comparable and presentable data. I also didn´t find any online. It would take some time and money, which I´m already pouring on other findings in audio. Knowledge is expensive...
The observation is, that high displacement volume driver with high motor force gives me better size/performance ratio than what Tapped Horns gives me in small volumes (I.E. The box I can transport alone).
After that finding, I went with 2x very compact B&C 21DS115 bassreflex box.
I have built few tapped horns, and I like the general performance, construction, sound. Yet its power density in "reasonable sizes" was not convincing.
Brian Steele: No, I usually scrap things including boxes, as all becomes less relevant in time. Different drivers, different amps, different circumstances between comparison and measurements. Findings simply do not add up to make a cumulative study full of worthy comparable and presentable data. I also didn´t find any online. It would take some time and money, which I´m already pouring on other findings in audio. Knowledge is expensive...
The observation is, that high displacement volume driver with high motor force gives me better size/performance ratio than what Tapped Horns gives me in small volumes (I.E. The box I can transport alone).
After that finding, I went with 2x very compact B&C 21DS115 bassreflex box.
Last edited:
Yes. If one tunes the box lower, the impedance peak goes lower too. So you are pulling the highest efficiency point lower in the frequency too. That helps temendously. Then you can afford to mitigate port compression compared to conventional bassreflex box approach by cutting around its tuning frequency. As long as the driver has enough available displacement (which 21DS115 arguably has) and distortion stays in the limit, it all helps with making really power dense small box.
I´m still working out the bassreflex port size for no HPF usage at home, and for port compression with HPF at 32-34Hz. for PA. Seems 600cm2 is plenty.
For the time being, it is 125l net box each, port tuned at 30Hz.
Outer box dimensions: 580*600*730mm, wheels mostly hidden in the box footprint.
Raw outer volume of the box is 247,5l. Should have good ergonomics and low weight, so one guy can move/load it. The single box volume and performance is the important constraint in this case...
For the time being, it is 125l net box each, port tuned at 30Hz.
Outer box dimensions: 580*600*730mm, wheels mostly hidden in the box footprint.
Raw outer volume of the box is 247,5l. Should have good ergonomics and low weight, so one guy can move/load it. The single box volume and performance is the important constraint in this case...
Last edited:
Here is current state of the development. Seems I will keep it like that. Small touchup here and there, and then getting these manufactured by pros...
PicFront - dev_set.jpg
PicFront - dev_set.jpg
I still see how that could happen, but not this time! :-D One 21" in one 125-130l box at a time... And I wonder how one could cram 18" or 21" speaker in TH of this size, going down to 28Hz for home usage and 38Hz for PA...
Amps. That´s hard one. Recently replaced QSC RMX1850HD for both subs by two pieces of T.AMP TA 2400 Mk-X. I modified these slightly here and there, adding two pairs of output stage transistors into each, and so on. Peak power output 2500-2600W into 7,95Ohm, 2250W burst, 1850W steady RMS.
For these speakers, it wasn´t smartest choice I must say. These truly deserve some good 3000W amp in D-Class + SMPS. The toroidal AB amp will probably burn 50% of the heat just on the difference between apparent power and real power due to the great phase differences.
Amps. That´s hard one. Recently replaced QSC RMX1850HD for both subs by two pieces of T.AMP TA 2400 Mk-X. I modified these slightly here and there, adding two pairs of output stage transistors into each, and so on. Peak power output 2500-2600W into 7,95Ohm, 2250W burst, 1850W steady RMS.
For these speakers, it wasn´t smartest choice I must say. These truly deserve some good 3000W amp in D-Class + SMPS. The toroidal AB amp will probably burn 50% of the heat just on the difference between apparent power and real power due to the great phase differences.
Last edited:
