Open Source "Tower XL"

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This would still have the top and bottom of the bass enclosure at the same positions, still parallel to each other. I don't think this would help with the standing waves, with the first mode at approximately 140 Hz.

I attached a drawing of my earlier suggestion of a slanting separator between the mid and bass chambers. Without the parallel top / bottom, the standing waves in the bass chamber would be less pronounced. Also, they are not focused to discrete frequencies, which is maybe even more important. The first mode would be "smeared" across a frequency range of about 140 to 200 Hz (from eyeballing the dimensions in my drawing). Spreading the resonances will be much less audible than focusing the whole mode at one single frequency.

The only option I see that might work to deal with the internal resonance would be to use an internal Helmholtz absorber (IHA). An IHA is a small box inside the bass chamber, with a bass-reflex port coupling this small volume to the main bass chamber. The IHA volume/port is tuned to the resonance frequency of the bass box, and the IHA volume is strongly stuffed with a lot of damping material. The IHA volume then couples to the resonance in question and converts its sound energy to heat. IHAs are often used in several designs of the German DIY magazine Hobby HiFi, and they seem to work ok.
 

Attachments

  • ostxl_midsep.png
    ostxl_midsep.png
    70.5 KB · Views: 383
I added the mid range section to the new drawing (the red part) it is around 11 liters, and it is possible to add an opening to the back wall to try open or closed option as Paul suggested ( good to hear from you Paul)

as for unparalleled top and bottom, we can cut some boards with different height and attach them in order to make an angled surface (the yellow part), it will take about 10 liters but compromises..

please see the attachments and let me know what you think.

attachment.php
 

Attachments

  • TXL V3 II Drawing.jpg
    TXL V3 II Drawing.jpg
    168.1 KB · Views: 560
But I might be too obsessed with the golden ratio thing.

Yes. Golden Ratio is just an approximate hypothesis from way before any proper simulation software existed, a nifty idea, but not very helpful.
It can actually work against the most optimal compromise for a specific solution, since you'd most likely focus more on the not-so-important Golden Ratio vs actually making a proper box and put drivers in more optimal locations relative to internal modes.

But that's just my opinion.

I really wish you could hear a proper front firing slot port design someday. ;)

/taunt off
 
I added the mid range section to the new drawing (the red part) it is around 11 liters, and it is possible to add an opening to the back wall to try open or closed option as Paul suggested ( good to hear from you Paul)

as for unparalleled top and bottom, we can cut some boards with different height and attach them in order to make an angled surface (the yellow part), it will take about 10 liters but compromises..

please see the attachments and let me know what you think.

Your version of the "unparalleled" top and bottom is the right idea, but not enough if you ask me. The difference between the internal box height at the front and at the height is about 10cm, which is about 8% of the total height. This means the modes / standing waves along the height of the box will be focused within a relatively narrow frequency range of 8% compared to the version with parallel top / bottom: fundamental mode of the parallel version is about 140 Hz, so your "unparalleled" version would spread this to about 140 Hz to 150 Hz, which is still a rather narrow frequency band.

I would suggest to try to spread the modes as much as possible. Take another look at my earlier drawing, where the difference between the internal heights at the front and the rear is much larger. My version also takes care of the midrange chamber, so it would be easier to make. Or am I missing something that is not to like about it?

Yes. Golden Ratio is just an approximate hypothesis from way before any proper simulation software existed, a nifty idea, but not very helpful.
It can actually work against the most optimal compromise for a specific solution, since you'd most likely focus more on the not-so-important Golden Ratio vs actually making a proper box and put drivers in more optimal locations relative to internal modes.

I agree 100% !
 
I prefer the proposal of Matthias with the non-parallel separation between woofer and midrange. It is applied in many commercial speakers. But it occupies a lot of volume for the midrange for this speaker enclosure, around 30 L. With a large mlidrange enclosure, the internal resonances become lower too and are more difficult to damp.
Maybe making a smaller midrange cabinet in some way, around 10 L volume.
I have done a trial in the attached drawing with the green separations. And maybe the extra volume for the woofer at the top can be filled with some dampening material.
Just an idea...
 

Attachments

  • 20190624 Drawing cabinet.JPG
    20190624 Drawing cabinet.JPG
    83.3 KB · Views: 274
Last edited:
Yep it is a compromise land :D Not sure the slanted panel is worth the trouble for the woofer standing waves. Mid box like #840 or #843 and put polyfill for the whole top part from woofer up might have more effect than a slanted top panel. Anyone with extra ply/mdf and time for a prototype box?:)
 
I prefer the proposal of Matthias with the non-parallel separation between woofer and midrange. It is applied in many commercial speakers. But it occupies a lot of volume for the midrange for this speaker enclosure, around 30 L. With a large mlidrange enclosure, the internal resonances become lower too and are more difficult to damp.
Maybe making a smaller midrange cabinet in some way, around 10 L volume.
I have done a trial in the attached drawing with the green separations. And maybe the extra volume for the woofer at the top can be filled with some dampening material.
Just an idea...

Welll, actually, looking at your drawing, I just ha another idea... the part where your green lines do not cover my red line could be the opening of the internal Helmholtz absorber (IHA). This would allow combining the non-parallel walls (to smear out the resonances / modes) with the Helmholtz absorber; best of both worlds! In addition, the volume of the Helmholtz absorber becomes part of the main woofer box at frequencies that are lower than the Helmholtz tuning frequency, because the port is transparent for those frequencies. This means the IHA does not "waste" the box volume for the woofer chamber.

Let me know if I need to make a new drawing to explain this idea a bit better.
 
Oups! I meant to write "The only OTHER option I see ... would be an internal Helmholtz absorber (IHA)."

Well, it seems to me a good idea to add an Helmholtz absorber on that place in the enclosure.

Now we're firmly back in Yamaha NS5000 territory.
I am a bit uncertain how that would work in a BR, I know it works in sealed boxes because the modes are more pressurized, no alternate path to choose. But in a BR I think the math for taking into account port behaviour at varying SPL might be a bit more complicated.
 
Now we're firmly back in Yamaha NS5000 territory.
I am a bit uncertain how that would work in a BR, I know it works in sealed boxes because the modes are more pressurized, no alternate path to choose. But in a BR I think the math for taking into account port behaviour at varying SPL might be a bit more complicated.

The Yamaha NS5000 uses "resonating tubes", like organ pipe resonances. This is different from a Helmholtz resonator.

As I wrote earlier, the German DIY magazine Hobby Hifi frequently uses the Helmholtz resonators to "kill" the lowest modes of the bass chambers with bass reflex loudspeakers. Their idea is to avoid those modes fro leaking out via the bass reflex port of the bass chamber. This would not make a lot of sense with sealed boxes, because they don't have a "leak" (=bass reflex port).
 
Thanks for the replies,
Matthias, I forgot to reply to your post about the midrange section, one of the main challenges We have here is that the woofer requires a very large box and these dimensions is like the largest I can get away with (WAF is the most influential factor here :D), and that mid section is taking a huge chunck out of the available volume for the woofer. I also forgot to mention the IHA idea, I never heard of it and no experience with, I also tried to search for it with no luck, I would greatly appreciate it if you can help us with that, it looks like the best solution if it does not compromises any volume from the box, and maybe we can XO the mid lower to avoid the the second octave of modes.

@Paul, The green mid chamber is looking good, it maybe more difficult to build but I already started to buy some tools and should be able to pull it off, my only concern is that since the HF horn has a long throat it would be adifficult fit there, i m not at home right now but I will measure once I get home.

@KaffiMann, good to hear from you.
 
Yep it is a compromise land :D Not sure the slanted panel is worth the trouble for the woofer standing waves. Mid box like #840 or #843 and put polyfill for the whole top part from woofer up might have more effect than a slanted top panel. Anyone with extra ply/mdf and time for a prototype box?:)

good to have you onboard, haha extra time would be difficult to find, but I m up for the task, but to avoid wasting wood and more importantly time consuming measurments and testings, i m trying to put as much thought and simulation into this as we can before start building anything hoping we can get away with only one build, otherwise I'm itching to start building something, so itchy infact that I was planning to rebuild cabinets for my current speakers :D.
 
Thanks for the replies,
Matthias, I forgot to reply to your post about the midrange section, one of the main challenges We have here is that the woofer requires a very large box and these dimensions is like the largest I can get away with (WAF is the most influential factor here :D), and that mid section is taking a huge chunck out of the available volume for the woofer. I also forgot to mention the IHA idea, I never heard of it and no experience with, I also tried to search for it with no luck, I would greatly appreciate it if you can help us with that, it looks like the best solution if it does not compromises any volume from the box

I have to say that I have no personal experience with IHA either, but the idea makes a lot of sense to me. The measurements shown in Hobby Hifi also seem to support this.

I attached a new drawing. The midrange volume is now quite a bit smaller (I'd guess it's about 14 L or so). The volume of the internal Helmholtz absorber (IHA) is now larger (about 20 L), but remember that at low frequencies the IHA volume is part ob the bass chamber. The volume for the bass tuning is therefore the full box volume minus the midrange volume (14 L). In other words, the IHA does not work against WAF.

The tuning of the IHA works like this:
  • The IHA volume is about 20 L (estimated from my drawing)
  • The IHA port is just a hole in the separator, and the port length is equal to the thickness of the wood (I am assuming 22 mm)
  • The port diameter needs to be set such that the IHA is tuned to about 150 Hz (the lowest mode)

Using these numbers and this calculator, I get a port diameter of about 18 cm.*

Note that these are just very rough numbers to illustrate the idea. You'll have to make a proper drawing to get the volumes right. And I am pretty sure you'll have to tweak the IHA by trial and error by taking measurements from the prototype speaker (woofer impedance, nearfield acoustics of the woofer and bass reflex port).


*This would be a very short and wide port. Maybe it would be better to make the IHA volume smaller, which, for the same tuning frequency, would give a smaller cross section for the IHA port.
 

Attachments

  • TXL V3 II Drawing_MB20190624.png
    TXL V3 II Drawing_MB20190624.png
    93.8 KB · Views: 439
thank you Matthias for quick calculation, at first glance I have to check this with 3d model making sure that HF+Horn will fit there, and also what the volume would be with with HF+Horn inside the chamber , the side walls are 19mm for now, and also if the width is enough to have an 18cm hole there, will check this tonight (hopefully).
thanks again.
 
@KaffiMann, good to hear from you.

I never really left this thread, just waiting. :)

The Yamaha NS5000 uses "resonating tubes", like organ pipe resonances. This is different from a Helmholtz resonator.

As I wrote earlier, the German DIY magazine Hobby Hifi frequently uses the Helmholtz resonators to "kill" the lowest modes of the bass chambers with bass reflex loudspeakers. Their idea is to avoid those modes fro leaking out via the bass reflex port of the bass chamber. This would not make a lot of sense with sealed boxes, because they don't have a "leak" (=bass reflex port).

I understand that the principles are derived from two seemingly different base views.
But the behaviour and end results are not that different.

Helmholtz resonator "box inside the box":
Dimensioned so that the volume in combination with the "port" induces a specific "mode frequency" to work as an opposite force to negate the internal modes, much like a regular vented enclosure, but instead utilized to absorb internal modes in a cabinet. Might require some creative thinking in certain designs to place in the actual location of the stronger modes.

The Yamaha resonant pipes seems to me a more elegant solution:
Dimensioned to trigger/work optimally at a specific "mode frequency", then dampened to reduce the mode somewhat through absorption, and also: the other end of the "pipe" ends in a place to counter the pressure variations induced by the problematic "mode frequency".

Both of the solutions are based on some mode induction principle:
"Helmholtz resonance or wind throb is the phenomenon of air resonance in a cavity, such as when one blows across the top of an empty bottle."

And the Yamaha solution has looked to the Organ Pipes, much as you say, but apply a more modern understanding of dampening materials and end termination in different parts of the enclosure.

Both of these would benefit more from a pressurized enclosure IMO, because the "mode" frequency" will sort of look for an escape route IE of conventional physics, pressure reduction. But both can probably work "well enough" in a vented enclosure.

I would think that a Double Bass Reflex enclosure would be far superior to using an internal resonator solution, since the original modes will not be able to escape through another separate chamber with completely different resonance characteristics. However, this will not solve the potential problems of reflections back to the driver and possibly through the cone.

And better than this again IMHO, would be making an enclosure that pushes the resonant behaviour to a frequency high enough, that a few cm of the correct insulation, in exactly the right place, can take care of the problem. Quite possibly high enough that it will be outside the frequency bandwidth of the driver in this particular design, would still need to dampen it though, to make it easier to hand over the job to the next driver in line.
 
hmmm, interesting, I have very limited knowledge about the DBR but I have seen it in some Fostex designs I believe.
it would be nice if you can help with more info on your suggestions so I can make the 3d model, visualizing the design will help having a better understanding of what is going on and maybe decide on which path to follow ;)
 
Last edited:
some updates on the design.
I added the IHA as suggested, the mid chamber is about 13 liters and the IHA is about 23 with the 18cm cutout on the back wall of the IHA and HF Driver+Horn will fit nicely there as well.
this should not complicate the build much, only some 45 degree cuts, I think we passed the point that someone would be able to build this on a kitchen table while back :D

attachment.php
 

Attachments

  • TXL 6-25-19.PNG
    TXL 6-25-19.PNG
    97.8 KB · Views: 544
  • TXL 2 6-25-19.PNG
    TXL 2 6-25-19.PNG
    43.3 KB · Views: 368
hmmm, interesting, I have very limited knowledge about the DBR but I have seen it in some Fostex designs I believe.
it would be nice if you can help with more info on your suggestions so I can make the 3d model, visualizing the design will help having a better understanding of what is going on and maybe decide on which path to follow ;)

Yes, I think it's very interesting. But I was merely trying to sum differences/similarities and suggest alternate solutions. I have no experience at all with DBR.

But I do have a firm preference/bias towards front firing vented enclosures where you take into account internal resonances/modes and rather than avoid thinking about them or deal with them directly, you instead make the box in such a way that these problems require less effort to deal with.
This path is not at all compatible with the fabled "Golden Ratio". And if you have some desired measurements that you want to stick to for your boxes, it's better that you move the ports around and break up some reflections like what you're attempting now, than to make a failed attempt at mixing in other ideals as an afterthought.
I have preferences and opinions, but try my best at being a pragmatic realist, I do not care which solutions are in the mix as long as the end result is good.
The most simple solution is usually better though.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.