I have some theories about why it could be a mixed bag. For example, one good way to get very bad measurements for a single unit vs. two is to have both of them together for both of those measurements and powering only one to produce the "single" response. They must be physically separated, and the unpowered one needs to be shorted out if its nearby.
Ivan has said that THs don't drop low corner in multiples, and that's one of the selling points of them- you get all the extension from one box. I think that's a fairly reputable source and if I can find where he said that I'll post it here.
It makes intuitive sense to me because THs tend to have that saddle response unless they're well damped. The low end is one of those peaks, so it's basically already there. On the other hand, an undersized FLH tends towards a peaky response with a continuous rolloff that begins before Fc where the slope steepens. According to somebody, who I cannot remember, the drop in Fc from multiple FLH is present but typically a couple hz at most and fairly insignificant overall. The big deal is that the passband above Fc flattens out which means most of the efficiency gains are in the low end.
Only repeating what I read from sources I thought were reliable. I'll link them here if I can find the posts.
It makes intuitive sense to me because THs tend to have that saddle response unless they're well damped. The low end is one of those peaks, so it's basically already there. On the other hand, an undersized FLH tends towards a peaky response with a continuous rolloff that begins before Fc where the slope steepens. According to somebody, who I cannot remember, the drop in Fc from multiple FLH is present but typically a couple hz at most and fairly insignificant overall. The big deal is that the passband above Fc flattens out which means most of the efficiency gains are in the low end.
Only repeating what I read from sources I thought were reliable. I'll link them here if I can find the posts.
Last edited:
I have measured and (repeatedly :^)posted the same thing Ivan Beaver and Tom Danley have, that the low corner does not drop in TH multiples, and that the low corner drop in FLH is not much either.
As you can see in the measurements of the 60 x 60 x 30 inch DSL BC218, (hardly an "undersized" FLH), the low corner hardly decreases from one to two cabinets, and the upper response actually increases several dB more than the 6 dB low corner increase one would expect from doubling cabinets and power.
The measurements also show the difference between being set as an FLH (flat) and a BC (boundary control), the BC increases level between 40-70 Hz, but rolls off response above 120 Hz.
Art
Attachments
The SPL is so low because I am a bit of a noob and didn't know how to input 2.83V. I just chose a volume that didn't result in clipping of the mic during measurement with REW.
I know that the 1w/1m sensitivity of the Xoc1 TH-18 is about 104dB (using B&C 18SW115).
Use freeware program that can generate 60 hertz on your phone or laptop.
Take cheap multimeter at output of amplifier and measure 2.83 volts.
Simple.
You can also measure the 2.83 volts at the driver terminals or speaker terminals.
That is one watt into 8 ohms.
4 ohms is 2.00 volts. One watt into 4 ohms that is.
The roll off at about 120hz is not a naturel system roll off. It`s probably that some acoustic energy is reflected back to the throat and causing an additional low pass filter.
gr. Marcel
Since I started this thread about
Since I started this thread about one year ago I have been thinking about and simulating my quarterwave resonator enhanced tapped pipe/horn idea.
Finally I was pushed over the edge into the deep waters of sawdust and real life results by my friend Anders Martinsson with whom I collaboratively developed the THAM series of Tapped Horns.
He kind of said "its enough threorizing now. Build something! You have two Beyma 12P80Nd just laying around slowly withering away by neglect - use them!"
If pushed into a corner it has a theoretical efficiency of 110 dB/0,22 watts @ 44 Hz.
That is all theoretical but I can say that is a beast in many ways.
It is very powerful, dynamic and clean. Like most tapped pipes it has a very tactile and physical sound quality.
Since I started this thread about one year ago I have been thinking about and simulating my quarterwave resonator enhanced tapped pipe/horn idea.
Finally I was pushed over the edge into the deep waters of sawdust and real life results by my friend Anders Martinsson with whom I collaboratively developed the THAM series of Tapped Horns.
He kind of said "its enough threorizing now. Build something! You have two Beyma 12P80Nd just laying around slowly withering away by neglect - use them!"
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
If pushed into a corner it has a theoretical efficiency of 110 dB/0,22 watts @ 44 Hz.
That is all theoretical but I can say that is a beast in many ways.
It is very powerful, dynamic and clean. Like most tapped pipes it has a very tactile and physical sound quality.
Tapped horn bouncing steel pipe - Youtube
A video about why you never build a large tapped pipe with a series tuned QW resonator from 22 mm particle board. This is with approximately 10 - 20 watts.
54 mm birch plywood would be much better. 25 mm stainless steel would be even better again.
I don´t have many pictures, since this is only a Proof of Concept. I built it from cheap 22 mm particle board just to verify basic performance - knowing that the particle board would flex violently and add a nasty resonant coloration to the midbas.
Very high efficiency is indicative of very high pressures, and cascaded
quarter-wave resonators can really build huge pressures inside the box.
I have never seen particle board flex like this, except when I intentionally try to destroy it with a sledgehammer.
With my small 1,4 watt single stage NFB class A amp it can play loud enough that my wife believed our neighbors had a garden party as she walk from the car towards our house. I was playing some music inside our house.
Cheers,
Johannes
A video about why you never build a large tapped pipe with a series tuned QW resonator from 22 mm particle board. This is with approximately 10 - 20 watts.
54 mm birch plywood would be much better. 25 mm stainless steel would be even better again.
I don´t have many pictures, since this is only a Proof of Concept. I built it from cheap 22 mm particle board just to verify basic performance - knowing that the particle board would flex violently and add a nasty resonant coloration to the midbas.
Very high efficiency is indicative of very high pressures, and cascaded
quarter-wave resonators can really build huge pressures inside the box.
I have never seen particle board flex like this, except when I intentionally try to destroy it with a sledgehammer.
With my small 1,4 watt single stage NFB class A amp it can play loud enough that my wife believed our neighbors had a garden party as she walk from the car towards our house. I was playing some music inside our house.
Cheers,
Johannes
A cross-brace down the center of the mouth should cure most of that panel flex.
Did you verify the alignment with a measurement of the build's impedance curve?
I will probably double up on wall-thickness on the top and bottom and the pieces around the drivers. I have not glued the top piece yet. I am going to lift the lid of the box and brace and improve it a lot. I have even been contemplating milling down a track around the inner circumference of the mouth and gluing in a frame made from welded cross-braced 8 x 50 mm steel flat bar.
It really needs strength and stiffness. It might seem excessive, but it is not.
But as I said earlier, it is only a cheap proof of concept. I have already evaluated it and there is some things I want to change. It is tuned to low. It kinds of dives down into the abyss of deep bass with a furiosity that can be downright scary. It is very physical and dominating. Everything rattles and buzzes. I am afraid it will tear the plasterboards of the studs if I play loud near a corner.
It feels like holding your hand close to a basreflex-port - but over my whole body when standing 2 feet from the horn mouth.
Cheers,
Johannes
It really needs strength and stiffness. It might seem excessive, but it is not.
But as I said earlier, it is only a cheap proof of concept. I have already evaluated it and there is some things I want to change. It is tuned to low. It kinds of dives down into the abyss of deep bass with a furiosity that can be downright scary. It is very physical and dominating. Everything rattles and buzzes. I am afraid it will tear the plasterboards of the studs if I play loud near a corner.
It feels like holding your hand close to a basreflex-port - but over my whole body when standing 2 feet from the horn mouth.
Cheers,
Johannes
An externally hosted image should be here but it was not working when we last tested it.
Here is the effect of a hot voicecoils - thermal dynamic compression effecting the spl-respons in a basreflexbox. 2 Beyma 18P1000Fe-v2 in a 360 liter BR with a 540 cm2 port 30 cm long.
An externally hosted image should be here but it was not working when we last tested it.
Here is the same amount of thermal compression in a single THAM12 with a Beyma 12P80Nd.
An externally hosted image should be here but it was not working when we last tested it.
Here is my quarterwave resonator enhanced tapped pipe with the same 10 ohm series resistance per driver - simulating thermal power compression and its effect on spl-respons. Black line is with 10 ohm series resistance per driver.
In every example I have increases the drive voltage to level the two spl-respons curves so it is easier to compare.
This might not seem like a very important factor for home audio, but for professional use where power compression is a real fact, it is important. As can be seen both the BR and the tapped horn will need EQ adjustments to keep a set sound-character as the voice coils heat up. My modified tapped pipe does not change its character much despite hard sustained professional use.
Add this to the very high power conversion efficiency with very high impedance over large part of the passband - drawing less current and generating less heat in the first place and you got a very dynamic high spl capable horn. Hornresp does not show this effect if you do not actively look for it, but in real life it often adds considerably to the max spl. We have all heard how basreflex boxes turn into "one-note-wonder boxes" playing an exaggerated port-resonance with lots of turbulent and distortion in live venues.
Cheers,
Johannes
Last edited:
An externally hosted image should be here but it was not working when we last tested it.
Here is with and without the quarter wave resonant chamber.
It does add a lot of efficiency. +7,5 dB @ 74 Hz compared to without.
At 98 Hz there is peak in Power Conversion efficiency. 107,2dB from 0,217 watts (73,6% efficiency) with the quarter wave resonant chamber to 104,9 dB from 1,251 watts (7,56% efficiency) with a straight tapped pipe.
That last image speaks volumes (no pun intended) the efficiency boost alone is impressive, add to that the power compression characteristics and pass band range (seemingly greater then TH), sum it all up and it makes for a very interesting design principle indeed.
As this principle triggered my interest in a big way I'm presently spending a lot of time sketching on a 10" version together with Johannes using the Faital Pro 10HP1020 as the example driver, a really evil little device to be sure, and if limiting the lower range to ~50Hz (simulated) it can be folded into a external volume of ~108dm3, or 600x300x600mm (WxHxD).
The folding initially seems to be scalable as well so let's see where we can take this, the tricky part is to deal with the ~600-700mm long resonance chamber and still get normal box-like proportions, I have not yet fully grasped all of the parameters to be balanced so it is a learning process.
A big thank you to Johannes for figuring out this design principle, or combination of previously isolated principles resulting in something to my knowledge previously unpublished that seems to have a lot going for it.
As this principle triggered my interest in a big way I'm presently spending a lot of time sketching on a 10" version together with Johannes using the Faital Pro 10HP1020 as the example driver, a really evil little device to be sure, and if limiting the lower range to ~50Hz (simulated) it can be folded into a external volume of ~108dm3, or 600x300x600mm (WxHxD).
The folding initially seems to be scalable as well so let's see where we can take this, the tricky part is to deal with the ~600-700mm long resonance chamber and still get normal box-like proportions, I have not yet fully grasped all of the parameters to be balanced so it is a learning process.
A big thank you to Johannes for figuring out this design principle, or combination of previously isolated principles resulting in something to my knowledge previously unpublished that seems to have a lot going for it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.