Fine with thread hijacking. I'm quite amused that no one has found anything definitive on dispersion constraints or otherwise in relation to throat diameter. Regarding higher order modes they are interesting but a secondary consideration compared to all the easy to measure properties.
I got the price of the Autotech SEOS-30 as 600 euro/pair and SEOS-24 as 500 euro/pair. My current plan is to use the 18" waveguides I have and something like the Beyma SMC-60. This should get low enough to match my slot loaded mid-bass. The all out solution with a coaxial compression driver and the big horns I will consider doing based on the outcome of this trial as the cost is many times this lesser solution. Will have to wait a month or so though as I'm still building subs.
I got the price of the Autotech SEOS-30 as 600 euro/pair and SEOS-24 as 500 euro/pair. My current plan is to use the 18" waveguides I have and something like the Beyma SMC-60. This should get low enough to match my slot loaded mid-bass. The all out solution with a coaxial compression driver and the big horns I will consider doing based on the outcome of this trial as the cost is many times this lesser solution. Will have to wait a month or so though as I'm still building subs.
You could make ONE long term INVESTMENT now in a BMS coaxial compression driver.
The 2" BMS 4592Nd sells for about 500-Euros in Germany.
The 1.4" BMS 4594Nd-HE is the newest design, and uses a longer phase cone and larger tweeter magnet. ---uses the same area annular midrange diaphragm as 4592Nd and produces simular low freq SPL. Study this one.
The SEOS-30 supports 1.4" , 1.5", and 2" compression drivers.
The SEOS-24 supports 1.4" compression drivers.
=====Dear Santa=====
The BIG EASY ... SEOS-24 waveguide with the new BMS 4594HE_ND coaxial compression driver to cover 500-20,000Hz with excellent pattern control. One 15" or 18" woofer offers high efficiency with a large... but easy to build big cabinet
Attachments
Well that's the plan but I'm building the smaller scale version first to test the ideas as its 1/4 of the cost. The SEOS-30 with 2" throat would be my horn of choice as I would ultimately like the Celestion Axi2050 if it became available for non OEM:
Celestion Axi2050 - AxiPeriodic Compression Driver
Its distortion performance at low frequencies is much better than the BMS coaxes:
YouTube
Or even TD18+ MTM
:SEOS-24, BMS 4594ND, Dual AE TD18H+ - Speaker Design - Data-Bass Forums
Last edited:
For a given power level output a smaller throat horn propagation wave has more abrupt slowing and widening where the throat transverses into the widening flare of the horn than a larger throat horn. This more abrupt slowing cause a more pronounced reduction in the Directivity Index than a larger throat horn. Also see post https://www.diyaudio.com/forums/mul...1.html?highlight=throat+impedance#post5611771 .
Another way to do it is look at the off axis frequency response of the horn that you want to use. If the off axis frequency response does not fall off the cliff build the speaker.
Thanks DT
Sorry, I thought we covered that. There are still some assumptions made/guesswork with horns for DIYers, especially until we build (or buy) and measure. These are not shots in the dark though. Do you use 'hornresp.exe' ?
You are deciding on the size of compression driver? 1" is a sweet spot. Will play loud crossed below 1kHz in a domestic environment. Crossing below 1k has its benefits as I see it. It can be tempting to use a larger driver to get another octave, don't compromise the 4-5 above it.
Some see HOM as a buzzword but I see it as knowing what is happening in the horn. There are some fundamental questions that you want to know first, such as what do you begin with and where do you want the sound to end up. It's a means to an end. Some horns will give a constant stream of diffraction and release their directivity smoothly and sometimes this can be a good thing, usually at lower frequencies. This also increases loading, again sometimes desirable at lower frequencies.
Well said. Angels on the head of a pin.
Moreover:
People don't seem to realize that when Earl postulated this that:
-effectively any HOM's are only in relation to very high SPL's. Quite likely higher than 100db at 2 meters (in-room/small-room).
IF you listen really loudly (or are quite far from your loudspeakers AND are in a very absorptive room above 600 Hz), AND have material with a high dynamic range, then you might get peaks of HOM.. but I'm not sure that short-term transients of this nature would be audible in this context.
HOM's (assuming they exist as predicted) are generally a sound reinforcement concern (potentially), not really a domestic HiFi concern.
Last edited:
This is for PA use as well as for home. On the subject of higher order modes how would the horns in this design of speaker do:
Soundgear - Speakers - Amplifiers - DSP
I heard four recently arrayed for 120 degree coverage and didn't notice any distortions despite the high SPL.
I do read and understand the posts I just prefer quantitative statements and sometimes in text statements can have the wrong emphasis. What I was trying to get at was that getting an accurate prediction of a horn driver combinations polar response would need FE simulation? (there is no simple method of prediction?)
I have designed bass speakers using hornresp but only looking at power response, I understand it has had some directivity prediction added but haven't had a play with this yet.
Soundgear - Speakers - Amplifiers - DSP
I heard four recently arrayed for 120 degree coverage and didn't notice any distortions despite the high SPL.
I do read and understand the posts I just prefer quantitative statements and sometimes in text statements can have the wrong emphasis. What I was trying to get at was that getting an accurate prediction of a horn driver combinations polar response would need FE simulation? (there is no simple method of prediction?)
I have designed bass speakers using hornresp but only looking at power response, I understand it has had some directivity prediction added but haven't had a play with this yet.
Turbulence is is not going to happen in a horn or waveguide because the "flow" is too small. The only place that turbulence is a factor is in woofer ports. And the friction at the walls should be negligible. The sound waves are never perfectly spherical in any device, but they are more so in some devices than others. Sharp edges create diffraction and the less diffraction there is the closer the waves are to spherical.
Diffraction is the cause of HOM's. Perhaps you are confusing the term "turbulence" with diffraction.
Well spotted
I see how the directivity can be controlled by the wavefront at the throat.
I find it hard to understand how the directivity can be "opened up" by a waveguide, at least usefully in practice.
Some of wave optics is counterintuitive and Earl is an expert but I have some concerns and questions.
What are the limits? (presumably there are limits).
Where is it demonstrated that the measured DI is not due to non-planar throat wavefront?
I haven't found theory or specific experimental data from Earl on this, or references, have you?
Best wishes
David
Yes, it is non-intuitive, but happens none-the-less.
Just look at the DI posted for my speakers. The polar response is much wider that a 1" piston would have, hence, the waveguide does indeed pull out the wavefront. It is true that as the frequency goes up the wavefront at the throat is becoming less and less coherent, so this then becomes the "limit" that you ask about. As to the the non-planar throat wavefront, this cannot be the case because the frequencies are not yet high enough to cause a problem. If you test the polar from a 1" driver without a waveguide you will clearly see that it beams. Put on a waveguide and it does not.
The "theory" is the core of the concept of a waveguide. It is well presented in my papers.
Last edited:
The air nonlinearity is indeed present, but it is very low order (inaudible) and low compared to the nonlinearities in the drivers.
You have HOMs completely wrong. They are linear, no THD or IMD at all. The HOMs are delayed in time and hence create nonminimum phase aberrations, which are linear, however we have a nonlinear perception of them. The sound worse at high SPL than they do at low SPL, but the aberrations are completely linear.
You do have this pretty well completely incorrect.
12 years later there has been very little movement on new nonlinear distortion metrics, simply because it turns out that nonlinear distortion in audio is not a big issue. No one cares. Fund us and we'll gladly do the studies. No funding and you just have to take what you can get and glean as much as possible from what is there. Of course, it helps if you understand what is there correctly.
I don't think that this is true, at least not in my experience. Horns that will have high HOM's always sound bad to me. Ones where one would expect low HOMs sound much better. Proof? No, not really, it works so no one has much interest in "proof". The "proof" is in the pudding as they say.
Last edited:
Yea, still not correct. Since the directivity and wave propagation in a waveguide is linear, the statement "For a given power level output"would imply that they are not linear. The directivity cannot depend on the "power level".
Thanks for the reply, this is clearly the crux.
Do you have a polar for a 1" driver compared to that same driver on a WG?
That would be the clearest possible demonstration and very educational.
Best wishes
David
Hello,
I will address a couple of you posts, first this one.
I make no claims in the post that you quote.
I only reference and quote your 2006 Conference Paper, if there is anything that is “completely incorrect” it is your Conference Paper.
Next there is your 4-13-2008 ALMA-Las vegas Power Point presentation.
https://www.homecinemaguru.com/wp-c...mprovements-in-Compression-Driver-Systems.pdf
Any reference to delayed THD and IMD is now missing.
Slide 19 documents that the “New Metric” now labeled the “GedLee Metric” was coined in 2001. Also on slide 19 is the claim that the new metric “was shown to be significantly more accurate at predicting the subjective perception of nonlinear effects”.
Now it has been 17 years without objective documentation of the “New Metric” now labeled the “GedLee Metric”. We know what THD is, it is well documented. We know what IMD is, it is well documented. “GedLee Metric”, not so much.
How many Angeles dancing on a pin?
Thanks DT
I did find this:
http://www.gedlee.com/Papers/THD_.pdf
Last edited:
How to Perform a Nonlinear Distortion Analysis of a Loudspeaker Driver | COMSOL Blog
and Klippel's recent work seems to be looking at non linearity....