The quest for a new DIY-speaker kit

You learn a lot just by following the evolution.

I knew all of that so I did not learned anything new from that. But you answered none of my questions so if you still claim it to be a horn then please share the horn k, expansion and simulation. Or, at least show on which alleged horn it's based. Because the extreme narrowing is typical for a TQWT, a Voigt pipe or a TML, depending on how generous you want to outline the principle. So anything but a horn!

What is not a horn?

"The horn serves to improve the coupling efficiency between the speaker driver and the air. The horn can be thought of as an "acoustic transformer" that provides impedance matching between the relatively dense diaphragm material and the less-dense air. The result is greater acoustic output power from a given driver." - Wikipedia

Since the frugals don't have a higher spl than a equally sized TML I conclude they are, in fact, not a horn.

Every modern design borrows from the pioneers.

Nothing wrong with that - but that doesn't make anything a horn!
 
Tang Band horn

A quick estimate suggests any response below 80-115 Hz is “transmission line” action (sitting on the floor). The ripple is a sihn it is turing into an undamped transmision line.

Screenshot 2024-09-26 at 17.02.10.png


dave
 
Last edited:
"The horn serves to improve the coupling efficiency between the speaker driver and the air. The horn can be thought of as an "acoustic transformer" that provides impedance matching between the relatively dense diaphragm material and the less-dense air. The result is greater acoustic output power from a given driver." - Wikipedia

Since the frugals don't have a higher spl than a equally sized TML I conclude they are, in fact, not a horn.

Neither do any of the 4 you linked (so by your definition are not horns?). In a BLH the sensitivity of the system is whatever the midband of the driver outputs — the part that is not horn loaded..

The gain comes in the bass. In many to bring the bottom of a low Q driver up to the midband levels or help with BSC. Some people do like oo much bottom.

Frugel-Horn are not well suited to really low Q drivers, the best drivers for these are also happy in reflex boxes. Excess bass is intentionally damped down by the polyfluff. Adjust depending on driver, placement, amplifier, and taste.

dave
 
A quick estimate suggests any response below 80-115 Hz is “transmission line” action (sitting on the floor). The ripple is a sihn it is turing into an undamped transmision line.

.. And..? Does that explain why your speakers are a horn? Hint :it doesn't. At least this speaker works as a horn above the frequency. Which you can also see at the spl. And the coupling to the ground is something you ignored at your estimate on the TB horn.
 
Neither do any of the 4 you linked (so by your definition are not horns?). In a BLH the sensitivity of the system is whatever the midband of the driver outputs — the part that is not horn loaded..

They are. You seem to miss that they actually are at the level of the midrange - after the linearisation. I think someone should tell you how to recognize a horn speaker. The frugly can't be a horn by principle and nothings wrong with that but you can't claim that they are. It's wrong and you know it - which makes it a deception, a lie.
 
I'm sorry for you because you are every time deeply hurt if someone noticed your speakers aren't what you claim they are. You provoked me numerous times to show I'd be wrong and every time it shows your claims are wrong. That's denial. And just wrong. And that's not how a moderator should behave. Did I go through all your speakers and said what all of them is wrong? No, I didn't. But with your avoiding any admission of faults and instead just trying to get back with any straw you are able to grab, no matter how flimsy, is not a way a Moderator should act. At all. Everyone makes mistakes but most admit it and try to learn from it. You don't. You defend your mistakes, wrong claims and deny as much as you can. Even after it's clearly wrong. That's the true definition of being stuck in a corner.

With such a behavior you should resign as moderator.
 
Just to quietly drop in here, since Dave asked me to (been away for a week). I haven't read everything as I don't have time, but a few general facts & some related to the FHs, FWIW:
  • A horn is a pipe that expands toward the terminus. Period. It may or may not be impedance matched down to its QW cutoff frequency (determined by the actual terminus size when any relevant intended boundary loading is factored in). Ideally it is, many are not. In the latter case, the BW betwen the QW cutoff frequency & the point at which the effective terminus size results in impedance matching is supplied by resonant action. Some never are, over their functional operating BW -some of hm's interesting designs are good examples of that. The point of commonality is that they all to some extent share some basic 1/2 wave characteristics. Rocket science this ain't but it's a point / a few points often overlooked.
  • In simplistic terms, a bass horn should only be functional up to the mass-corner frequency of the drive unit, providing a/ an effective acoustic LP is designed in (at least 2nd order, preferably 3rd order), b/ that mass corner frequency is no higher than 300Hz, and c/ [related to b/ previously] the expansion path length is not so great as to introduce audible GD in the intended operating BW. If it is, then you are going to be either accepting that as a compromise over xyz BW defined by the expansion pathlength and the acoustic LP, need to shorten your bass horn & accept a higher tuning as a result, or provide an acoustic LP at a lower frequency and introduce some form of short, BW limited midrange horn to 'fill in' the resulting amplitude gap, which is what you see in something like a Tannoy Westminster & other compound designs.
  • Some bass horn designs may have abrupt (or not so abrupt) discontinuities deliberately inserted into their expansion path to provide a low-pass filter effect. This can be achieved in various ways, ranging from bend profiles to dedicated features. This goes back to the dawn of hi-fi and horn design, so is a very long way from being either news or even unusual. An efficent designer (in my view, so this sentence is purely opinion & you can form your own views about how little / much that may be worth) will always attempt to achieve 'the most with the least' assuming there are no other considerations at work, so when possible, they may exploit a folding scheme or the features of an OA configuration to help provide that if / when desired.
  • The amount of gain needed (note all horns that expand in only one dimension [e.g. have parallel sidewalls] are technically parabolic if you do the maths) should be determined in the design process & for practical purposes within the target range of driver specification / specifications is essentially a function of the QW tuning, the available enclosure bulk and the intended gain boundary operating conditions (the 'actual' bulk). Ideally you'll provide as much as you can & then damp out what you don't require, on the principle that too much is always a good thing to have as you can do something about it. Unfortunately, that tends to run into the brick wall of practical realities, so when designing a generic enclosure where size is limited, I personally aim to provide a reasonably balanced output within the design requirements as there isn't sufficient bulk to provide an excess unless tuning is raised to some unreasonably high frequency.
 
Last edited:
You've made my point. He can't explain what a horn is.

Not so. Dave knows what horns are far better than most. He's been working with them since the '70s. And it looks to me like he's actually made most of the above points already -I've simply expressed them differently & added a couple. This isn't Technical Writing 101, we're not awarding marks out of 10 for grammar or means of expression, here (which is a matter of opinion anyway). If you want that, there are dedicated Forums / groups in various places.
 
Is this discussion very useful on this thread?
Probably not.

To me, horns, TL and whatever are simplifications in some way. Ever since we have access to reasonably accurate sim software like HornResp, AJhorn and Akabak/BEM, the academic discussion becomes moot to me.
If practical results are what you require, that's the way, always keeping GIGO in mind, which is where theory can (can -not always, depends what you use & how) help out. Either way, my part is now complete.