So the idea is that the tapering quarter wave transmission line from the rear of the driver would supplement the front horn's bass response.
Why is this combination not used, what are the cons of the design? Hornresp has not been helpful.
Why is this combination not used, what are the cons of the design? Hornresp has not been helpful.
the TML won´t reach the SPL of the front horn, better take two driver in a bass horn only one with FL horn, second with BR ~30 Hz
They are used as it happens, albeit more often with a vented box or bass horn on the back rather than a TL. The Altec VOTT enclosures are examples of the former, the Tannoy Westminster is an example of the latter. Typically the front horn is used to supply gain in the lower midrange, from the point the bass enclosure rolls off, either to fill in output between that and the driver's lower mass corner frequency, or to balance off what would otherwise be a rising response above that region.
If you have a high Qts driver, you can use a sealed alignment with Qtc~1.0 and design a short conical front horn to fill-in the output up to the baffle-step or thereabout, keeping system efficiency up. Hornresp is the perfect tool to model this IMO.
I like the looks of that second enclosure from the left on this 1979 Fostex brochure:
I would try to match the square throat of the horn more closely to the driver however. Mouth roundovers wouldn't hurt either.. 🙂
I like the looks of that second enclosure from the left on this 1979 Fostex brochure:
I would try to match the square throat of the horn more closely to the driver however. Mouth roundovers wouldn't hurt either.. 🙂
I played with that idea in a Hornresp few years ago and this is short story.
I went with fully stuffed (Augspurgers) type of TL. If you wish to add horn to that you need drivers that work well in a large(r) closed box and still reach good efficiency. My choice were Seas FA22RCZ and Visaton B200.
After designing a TL, I added a front waveguide. To be clear, my understanding of waveguide is horn shorter than 1/4 of cutoff wavelenght, usually 1/7 or 1/8.
Main idea was to boost the response below baffle step and compensate for HF rise.
And here problems starts. Lower you go waveguide becomes bigger and finally you end up with something like Altec VOTT. Another problem is, with suitable drivers gain in overal efficiency by horn/waveguide is small, few dB only. They just dont have enough "motor sthenght" to fully benefit from horn loading.
This is just my findings, if someone had better results and ideas please say. I guess you could try with with some other form of TL, like ML-TL for eg.
I went with fully stuffed (Augspurgers) type of TL. If you wish to add horn to that you need drivers that work well in a large(r) closed box and still reach good efficiency. My choice were Seas FA22RCZ and Visaton B200.
After designing a TL, I added a front waveguide. To be clear, my understanding of waveguide is horn shorter than 1/4 of cutoff wavelenght, usually 1/7 or 1/8.
Main idea was to boost the response below baffle step and compensate for HF rise.
And here problems starts. Lower you go waveguide becomes bigger and finally you end up with something like Altec VOTT. Another problem is, with suitable drivers gain in overal efficiency by horn/waveguide is small, few dB only. They just dont have enough "motor sthenght" to fully benefit from horn loading.
This is just my findings, if someone had better results and ideas please say. I guess you could try with with some other form of TL, like ML-TL for eg.
Davor D,
I agree with the issues you describe in the design of such an enclosure. Were I to seriously design/build one, a few decibels of gain would be all I need. It is helpful to have the bass alignment provide a slight hump so as not to have too large a step where the horn unloads. And to avoid ending up with an A5 or A7, a higher Fs driver allows that bass hump to be in the low to mid 100Hz range, which can then be 'stitched' to the horn in the ~200Hz range. Of course you won't be getting low-bass from that box.
I don't have it here, but I recently simulated something reasonable for a Faital Pro 6FE100 (Fs~87Hz / Qt~0.9); ~20L sealed (Qtc~1.05 / f3~95hz) with a square ~12" mouth, I forget length, perhaps 6"-8". I have to install Wine on this here Linux laptop so I can run HR and other useful tools.
I agree with the issues you describe in the design of such an enclosure. Were I to seriously design/build one, a few decibels of gain would be all I need. It is helpful to have the bass alignment provide a slight hump so as not to have too large a step where the horn unloads. And to avoid ending up with an A5 or A7, a higher Fs driver allows that bass hump to be in the low to mid 100Hz range, which can then be 'stitched' to the horn in the ~200Hz range. Of course you won't be getting low-bass from that box.
I don't have it here, but I recently simulated something reasonable for a Faital Pro 6FE100 (Fs~87Hz / Qt~0.9); ~20L sealed (Qtc~1.05 / f3~95hz) with a square ~12" mouth, I forget length, perhaps 6"-8". I have to install Wine on this here Linux laptop so I can run HR and other useful tools.
That's roughly what the Altec 816 & related enclosures aimed to achieve, albeit through venting the LF alignment rather than via a driver with underdamped electro-mechanical characteristics. Works OK within the target context. The Fostex box above is evidently (one of the many) variations on that theme. As far as TLs go, it depends what you define as such, but if you select an aperiodic or near-aperiodic alignment then you're inherently in a low-gain scenario, so while it isn't exactly a contradiction in terms, there are very limited scenarios in which that configuration would be of much use. If you broaden the TL definition to include resonant QW types, then it include boxes like the Westminster, where the bass load is an expanding pipe (horn) and gives broader band gain.
Last edited:
The Voigt Domestic Corner Horn had a substantial FLH down to about 100Hz, and a "labyrinth" at the back. I don't know what the design of the labyrinth was, but the bass output matched the output from the horn. The driver was a Voigt, which essentially is the present day Lowther PM driver. I would imagine the corner loading was essential for the bass output.
IIRC it was a fairly short single (acoustically speaking) fold that vented out of the base; good for about 60Hz or so. The indirect front horn loading is one reason why the drivers had that rising response up to ~8KHz, or the losses would have attenuated a lot of the upper midband, lower treble region. Outright HF extension was a bit limited, but that wasn't a big deal (for the same reason as slightly limited LF) since most of the contemporary sources couldn't make it much past 12KHz. Hence the so-called 'Lowther shout' for many (not all) of their units when used as direct radiators.
I think any system resonance is an abomination because it is contrary to the purpose of building systems to reproduce whatever is thrown at them and that have no characteristics of their own. Builders need to ask how to benefit (or at least, suppress) the rear wave of a Rice-Kellogg driver, not how can I build using the same standard set of box concepts, as OP suggests.
That's why there's so much enthusiasm for dipoles that spray the rear wave around the back, except maybe in the lowest bass.
My approach (just for woofing) was a 17-foot, two-folds, labyrinth with lots of stuffing. Non-critical to build and no tuning. The the driver and whatever is coming out of the exit port are 7 feet apart and each works into a corner for highest radiation resistance. But could be used full-range with a different geometry.
https://www.diyaudio.com/forums/subwoofers/322730-pipe-sequester-rear-wave.html
That's why there's so much enthusiasm for dipoles that spray the rear wave around the back, except maybe in the lowest bass.
My approach (just for woofing) was a 17-foot, two-folds, labyrinth with lots of stuffing. Non-critical to build and no tuning. The the driver and whatever is coming out of the exit port are 7 feet apart and each works into a corner for highest radiation resistance. But could be used full-range with a different geometry.
https://www.diyaudio.com/forums/subwoofers/322730-pipe-sequester-rear-wave.html
Last edited:
What you're describing is a compound horn and is I believe the default setting on Hornresp. I spent years designing exclusively this idea, with RLH, though not TL. In fact if you check my posts from, like 10 years ago you'll find lots of them, even built a few. FR drivers with rising HF work great because as Scottmoose mentioned the front horn mainly effects the upper midrange, I'd say up to about 2.5k or so. No offense but I think bentoronto is making the same mistake I made for many years and insisting that the design criteria that he is most enthusiastic about and the goals that he has are the same as everyone else's. Every design has its advantages and disadvantages, while as audiophiles we like to think perfect reproduction is our ultimate goal and should be everyone's if we really think about it enjoying ourselves and loving music is the real goal. So the kid with a 15" sub and a 1000 watt amp in his Civic is just as valid as anyone, perhaps not for this forum but nonetheless a hobbiest. While horns do produce resonances that are hard to quell they reproduce dynamics that hard if not impossible to recreate any other way. In my experience dynamics are far more important to realism than flat frequency response, low distortion, or any other metric, but then, that's just my experience.
Last edited:
Sure, anybody can choose any path in audio-life they want (within the usual limits that John Stuart Mill clarified). OK with me.
But you are quite right when you add incidentally at the end of your take-down screed "perhaps not for this forum". So why are you dissing me for wanting sort of what almost everybody "for this forum" wants (that you call vaguely "perfect reproduction")?
If OP had wanted some kind of special sound, as we all agree he is entitled to, "for this forum" he has to put in his opening post "...BTW, I want the treble to really screech" or "I want lots of distortion to sound like my guitar". Folks might ask why, but nobody would dispute his wishes.
True horns are great and can play loud and clean. But other than max clean loudness, whatever are you talking about when you like "dynamics"? Do you mean max loudness? Or do you mean dynamic range which includes S/N of the whole system (and traffic noise, A/C noise, etc too) and controlled by the volume control on an amp? Or maybe the ever-elusive "fast bass"?
The dynamic range is not controlled by the volume on your amp. The truth of the matter is that any speaker is a compression device. Max volume is not remotely the same as dynamic range. When you just crank up the volume you also crank up the noise and the net dynamic range remains exactly the same, but the sensitivity rating of a speaker tells you how little it compresses the dynamics, so the higher the sensitivity the higher the dynamic range, period. Speakers with low sensitivity compress the signal and reduces the dynamic range to a greater degree than those with a higher rating. Horn speaker have an inherently higher dynamic range than any other design that I am aware of. It sounds to me that you have a seriously mistaken view of what dynamic range is. If your speakers do not have "Max efficiency" you don't have it, and it is to me the single most important factor in realistic sound reproduction.
The volume control does help define the S/N for all noise sources between the VC pot and your ears, as I made clear previously.
Can you explain the non-linear "compression" factor? Of course, everybody knows about voice coil heating when heavily driven for a while and how over the course of time, "compresses" the output. You don't just mean that?
I am confused about "dynamic range" and "effiiency". Granted, arguments about how to grasp efficiency in this forum, but not about dynamic range that I know of.... unless you are gratuitously adding in amp power when you talk about dynamic range limits. An efficient speaker (and especially true and even pseudo-horns) will play max louder for a given input power and it may well be able to play real loud by virtue of its design.
Can you explain the non-linear "compression" factor? Of course, everybody knows about voice coil heating when heavily driven for a while and how over the course of time, "compresses" the output. You don't just mean that?
I am confused about "dynamic range" and "effiiency". Granted, arguments about how to grasp efficiency in this forum, but not about dynamic range that I know of.... unless you are gratuitously adding in amp power when you talk about dynamic range limits. An efficient speaker (and especially true and even pseudo-horns) will play max louder for a given input power and it may well be able to play real loud by virtue of its design.
Last edited:
Dynamic range is peak/average transient response, which is set by the recording and based on my OTA TV it's basically zero dB, so if I want any 'dynamics' in a movie/musical concert, etc., have to rely on the THX +20 dB for the main channels and +30 dB for the LFE channel (20-120 Hz), which in turn is basically nothing compared to an orchestra's ~108 dB potential.
Heat in the voice coil doesn't only cause distortion and compression when it reaches some critical level called overheating, the more power going through the voice coil the more heat it produces, the more distortion it creates the more nonlinear compression it causes. Plus a horn greatly decreases the amount that the cone has to move, thus reducing mechanical compression due to the non-linearity of the suspension system and so on and so on.
Sorry to high jack the thread, but my only point is that different people have different objectives and find different forms of distortion more or less offensive than others do. Personally bass from anything other than a horn is a serious compromise to me because any other bass loading method I have heard seriously decreases the the dynamics of the the response. I think it becomes less critical as the frequencies rise which is why a horn tweeter with a BR Woofer is just silly to me, but we all have our preferences and as long as the goal is realistic low distortion sound and we are making progress toward that goal they are all valid. I suppose I started all this by going on a silly rant about my preferences, so I'm being a bit hypocritical, sorry. I do however think that there are very sound advantages that make a horn loaded bass enclosure a legitimate high fidelity option.
Sorry to high jack the thread, but my only point is that different people have different objectives and find different forms of distortion more or less offensive than others do. Personally bass from anything other than a horn is a serious compromise to me because any other bass loading method I have heard seriously decreases the the dynamics of the the response. I think it becomes less critical as the frequencies rise which is why a horn tweeter with a BR Woofer is just silly to me, but we all have our preferences and as long as the goal is realistic low distortion sound and we are making progress toward that goal they are all valid. I suppose I started all this by going on a silly rant about my preferences, so I'm being a bit hypocritical, sorry. I do however think that there are very sound advantages that make a horn loaded bass enclosure a legitimate high fidelity option.
Hey Brsanko🙂
What do you mean(exactly) when you say horn? The textbook gigantic, huge version of an expanding path that produces 40 Hz or so, With a mouth and flare rate to accompany the extremely long length requirement etc.
or something similar , a compromise of that idea but taking Other things into consideration and getting the most out of it it’s realistic in size and performance?
Or nothing in particular to do with that at all and just the generic version of pipes horns and ducting that takes into consideration 1/4 wave or similar ?
I always get confused because there’s horn loading, Mass loading, acoustical impedance matching…. and then there’s pipe harmonics and pressure nodes scattered in odd intervals to use as well for given pipe length.
But over time it seems like everything just became a transmission line with either an expanding path or reduction or a chamber,vent discontinuity/ size exchange which serve similar purposes all the way down to the base reflex phase inversion type instead?
I think it's just that we've realized that all these loading topologies are part of a single continuum with no hard and fast cutoff in-between. The extreme cases might look alien to each other, but can be linked within a few steps of transformation.
