2-way and 3-way high efficiency driver suggestions for Troels TL speakers

I see a huge problem with harmonic resonances, which will affect the frequency response of the speaker. This can be ignored in subwoofers, but not in 3 and 2 way systems.
Well, this is inherent to every TL, isn't it. You exaturate the problem. And unless you go for taper ratio in the order of 14:1 there is not much you can do. Resonances are present in every practical design. Troeals does not write the xo frequency, but looking at my sims above it has to be below 200Hz and that plus some small degree of dumping is enough to control the harmonic resonances..
You say the tunnel geometry is not optimally designed. That is a claim without any justification from your side. Sure, one could say make the taper ratio 14:1 nad you get better results, but we are talking about practical designs. Practical is the key word here. And how do you know what the "optimal geometry" is in this case? Hornresp does not even allow to simulate bends. Are you gonna be building different tunnels and be taking the measurements? Good luck with that, there is an infinite number of possiblities.

I do not understand why you critisize his every design. Sure, one would make changes here and there, but Troels is certainly experienced. And his approach to every design is very detailed. He takes open baffle measurements, makes a prototype box and again takes measurements. Frankly, I wouldn't be bothered to do that with my every design. Look at the photos from construction and open baffle setup and other mid stages of building the box. He puts a lot of efort at every stage.
 
Well, this is inherent to every TL, isn't it. You exaturate the problem. And unless you go for taper ratio in the order of 14:1 there is not much you can do. Resonances are present in every practical design. Troeals does not write the xo frequency, but looking at my sims above it has to be below 200Hz and that plus some small degree of dumping is enough to control the harmonic resonances..
You say the tunnel geometry is not optimally designed. That is a claim without any justification from your side. Sure, one could say make the taper ratio 14:1 nad you get better results, but we are talking about practical designs. Practical is the key word here. And how do you know what the "optimal geometry" is in this case? Hornresp does not even allow to simulate bends. Are you gonna be building different tunnels and be taking the measurements? Good luck with that, there is an infinite number of possiblities.

I do not understand why you critisize his every design. Sure, one would make changes here and there, but Troels is certainly experienced. And his approach to every design is very detailed. He takes open baffle measurements, makes a prototype box and again takes measurements. Frankly, I wouldn't be bothered to do that with my every design. Look at the photos from construction and open baffle setup and other mid stages of building the box. He puts a lot of efort at every stage.
Every project is a set of compromises. You have to decide for yourself what compromises you accept. If you think TL Troels' designs are good enough for you, that's fine.
For me, they are not well designed, because they contain a lot of errors, which are based on the lack of optimization of the tunnel geometry as well as the way of damping and the type of damping materials.
 
Every project is a set of compromises. You have to decide for yourself what compromises you accept. If you think TL Troels' designs are good enough for you, that's fine.
For me, they are not well designed, because they contain a lot of errors, which are based on the lack of optimization of the tunnel geometry as well as the way of damping and the type of damping materials.
As I said, I am looking for inspiration. Please suggest a design of a similar enclosure volume that does not have errors.
 
I'm afraid there's not much you can do with an enclosure whose tunnel geometry is not optimally designed. Of course, you can add damping, in order to minimize resonances, but this will simultaneously weaken the radiation of the quarter-wave resonance. Besides, the design of the tunnel itself in general already does not allow for much change in its damping.

Troels does not show a measurement of the tunnel response in his designs. If you are curious to see what it looks like, along with its harmonic resonances, you can make a measurement of the tunnel in the near field. Place a measurement microphone at the very beginning of the tunnel output and take a frequency measurement.

Thanks. I will try and see if i can get a measurement set up next weekend.

...
Troeals does not write the xo frequency, but looking at my sims above it has to be below 200Hz and that plus some small degree of dumping is enough to control the harmonic resonances..
...

Looking at what i have on hand here, i would not be surprised at all if the lower crossover is somewhere around 200Hz
 
  • Like
Reactions: mV8
I suggest analyzing the MJK example, which allows optimal use of the quarter-wave enclosure.

http://www.quarter-wave.com/TLs/TL_Alignments.pdf
I believe we all know the theory. Neither Hornresp or MJK alignment tables confirm your claims of wrongly designed TL. They both confirm the opposite.
Any practical examples of correctly designed TL enclosures in a box of a similar size?

@kaan Thanks. I will try and see if i can get a measurement set up next weekend.
@kaan I am curious of your findings. Please keep me updated. Although, Troels does show the response curves on his website so I do not expect any surprises. If I can make a request, could you also measure near field response of the woofer and TL's output separately?
 
Last edited:
  • Like
Reactions: kaan
As I said, I am looking for inspiration. Please suggest a design of a similar enclosure volume that does not have errors.
Your Hornresp simulation has the path set to 0, if you set it to about 60cm it will be a better match with the Ekta design, and the dip 200-300Hz will be more flat. The Ekta-TL has about 40l volume, right? Here is a TL simulation with the sb-acoustics sb16pfc25-8 with 32l and f3=35Hz. It's not as loud but looks much better to me(depending on the goal):
tl-sb16.png
 
  • Like
Reactions: kgrlee
I second rertrobaer's suggestion.

I've played with many TLs but always went back to simple bass reflex. Simple bass reflex ALWAYS had better bass with a smaller box. The TLs, quarter-wave boxes, bla bla always needed a (sometimes MUCH) bigger box for the same performance and you had to do loadsa stuff to mitigate the resonances.

Once you have chosen your bass unit, build a simple classical reflex. Most sim programmes will allow you to do this easily. Use a fairly large port (eg 150mm) unless it would make the port too long. If your -3dB point is between 30 & 45 Hz, this will give you practically all the good stuff of TLs, bla bla.

You can then stuff the whole internal volume with Fibreglass insulation. This will give you the rest of the TL good stuff. Tuning will be out but sound will be good.