Romy the Cat wrote:

Paul S wrote: But doesn't putting the drivers on an inward-firing arc more or less shoot all the sound into a point, like the opposite of the Big Bang? Paul, locating the horn channels on an inward-firing arc and targeting them to the listening point faces absolute prohibition from the perspectives of my Macondo's Axioms. The drivers in nearfield position should be positioned on curve but they must not be angled and shells keep the channels axis strictly parallel

Since the “no tilting rule” is one of the Macondo’s Axioms: (#2 - “All channels shell have strictly parallel axis”) then I would like to defend it.

be wrote:

The time alignment will depend on the geometrics of horn set up and listening position, in the same manner, no matter if the horns are tilted or not. It is simple: delay=(distance of ear to source)/(speed of sound)

It is just logically appears this way but unfortunately it is not so in reality. I do not know well theories behind all of it – my understanding and knowledge has more applied, practitioner outlook to those things – so I declared the harm of tilting as a practical fact but I do not explain the physical nature of this phenomena – I just do not know it. Interestingly I even never asked myself about the underlying reasons. I admit it would be interesting if somebody explain why tilting screw up arrival time but I am too pragmatic to dig in it myself.

be wrote:

I dont see why non tilting should have any advantage over tilting.

Ok, let me to explain. Take MF and HF drivers, position them strictly parallel and perfectly align them from 5 feet and listening position directly on the axis between the drivers.  In this case the driver’s cones will be strictly perpendicular to the main axis and equidistant from the listening position.  Now tilt one of the drivers for 1 degree and recheck the alignment from the same from 5 feet - you will be off. Now re-align the non-parallel drivers again. After you hit the alignment you will see that the equal proximity between the listening position to the driver do not exist anymore and the tilted driver will be in time-aligned setting closer or further then the non-tilted. Let call this tiny change of alignment distance with tilted drivers as the Tilting Excursion

Here is where the whole notion of drivers tilting hits the fan: the changing of listening position distance will demand a different Tilting Excursion. With the parallel drivers the Tilting Excursion juts do not exist. 3 feet, 5 feet, 10 feet – all the same: (distance)/(speed of sound). With the tilted channel you need to introduce a correction – the Tilting Excursion -depending of the listening distance.

The only know to me way to correct the thing is to wear some kind of detector on the listener that will read the distance to the driver and dynamically tilt one of the drivers contingent upon what the listening distance is. One would say: “Well, how about if the listening distance is very firmly set?” Then it should be fine, though I would have my concerns with the alignment at one octave above and before the crossover point in this case and I fell it might be the problem.

Take any loudspeaker with tilted baffle for done tweeters: Grand Slamm, Utopia and so on and take a look how dramatically HF sound changes with them whiles you move closer and further from them. BTW, GOTO officially “do not believe in time alignment” but they recommend tilting the drivers “until you get better sound”. Now you understand what kind BS they propose. BTW, if you have 2 channels with one cone plus 5 degree off the perpendicularity to the main axis and with another cone minus 5 degree off the perpendicularity to the main axis and the channel are time-aligned. Then you will have a perfect listening position no deeper then 1-2mm and you can very much listen in this case in a dentist char. As soon your hear move 1mm off then it will sound like somebody just put a ton of new Black Gates capacitors in your amplifiers….  if you know what I mean. :-)

be wrote:

Actually if one is so close to the horns that one or more of the the throats get obscured from vision by the inside of the horn walls, then the necesarry bending of the sound path will increase the changes of delays, when one is moving around.

I think the diffraction has nothing to do with it as the effect is clearly observable even with the direct radiator dome tweeters.

I.M. Distortion wrote :
      "...#1 from were you listen to system from you should be able to look down the throat of every horn in your system..."

jd wrote : 
      "...Hmmm... I will soon be able to verify this for myself..."


OK, I said soon... Last night I got home and tried angling the horns down toward my listening point.

For those who don’t know, I have made a horn system patterned after Romy’s Macondos. More on that and the differences between the two systems at end of post.

First I should say that the frames which will locate the horns are almost ready. In the mean time, I use a pair of modular metal racks, which I have adjusted for time alignment (not to within .001 of an inch, but as close as is reasonable knowing its temporary). The horns are normally all firing along parallel axes, straight out perpendicular to the rear wall.

So last night I angled the upper horns of right channel down toward listening point, and left the left channel with horns firing along parallel axes. The preamp I use has left and right volume controls, making it easy to listen to one channel at a time. I played a variety of music, trying to stick with mono recordings as much as possible while listining to only one channel.

Here are my preliminary observations listening to the right channel with upper horns angled down, as compared to the left channel with horns firing parallel :

Upper horns angled down (upper horns means 180 Hz fundamentals horn and the tweeter horn) :

1) « Tighter » sound when instrument plays a single note, I would say it was more in focus, but possibly less vibrant.
 
2) Possibly subjectively quieter (want to use one more click on the volume knob… a good sign I guess)

3) Flatter sound (as in less 3D space)

I thought it sounded good enough to take the time to angle the left channel down as well ; its sort of dangerous work due to the weight involved, but I wanted to hear it.

Both channels with upper horns angled down :

1) Game over in 3 seconds… Forget it… The space this system used to depict so fantastically had now gone completely FLATTTTTTTTT.

2) Notes sounded slightly dead and muffled by comparison (reminded me of the very nice box speakers I used to own)

3) Over all less captivating and less involving

This actually scared me ; what if I somehow couldn’t get the magic back (who knows man, maybe I had unknowingly hit the dPols jackpot ?). I put the horns back to parallel alignment, and happily the magic returned.

The new frames will make it far easier and safer to conduct this experiment again, and I will do it just to confirm these results.

I should add that had I never heard the system in its normal parallel firing configuration, I would have probably been impressed by the sound with horns angled down… It did still sound quite good, and I could have been fooled, I am after all an absolute beginner.

About getting this system up and running :

Since moving everything into a room that’s large enough, and putting the upper bass horns into the game, I just sit there in awe. The first night I was literally on my ass in amazement. I told myself no way is this possible… Such warm delicate detial is now filling the space, and what space! Really, I was quite certain that I was just too tired and was imagining this sound. The next night I told myself no way; must have some freaky good electricity here tonight… I gotta get some sleep ! Two weeks later and I am still on my ass. My life has come to a screeching halt. I don’t get anything done… Don’t ever get to sleep before 3 a.m. Don’t leave the house other than for work ; have been pulling on dirty clothes because I just can’t stand the thought of hearing the washing machine in the background… I am quite sure I will forever be single and I just don't care.

Romy you may take this all as a compliment.

Main differences between Romy’s Macondos and my horns, or where I would take the Macondos :

I use a larger top or "fundamentals" horn (180 Hz cut off vs Macondo’s 250 Hz),  which I do not attenuate at all. Consequently, the upper mid range horn (400 Hz ; same as Romy’s) is high-passed electronically at a higher point (4800 Hz 1st order), and the upper bass horn is not asked to play so high.

Relieving the small (400 Hz) horn of some lower frequenqies allows use of the original metal-suspended diaphragms in all four S2s.

I don’t use an injection channel ; My mind could be changed, but I do feel that I get the very nice tone via the above-mentioned diaphragms along with the drivers I use in the upper bass horns (115 Hz cut off ; same as Romy’s) : Old paper cone drivers with big AlNiCo magnets.

For tweeters I use (and love) the Electrovoice T350, which I high-pass at 12 KHz 1st order (well ok, I love one of the two pairs I have… For whatever reason, the other pair is not in the same league). I have zero motivation to try other tweeters.

This system still lacks true dedicated mid-bass horns (pair of 45 Hz straight bass horns "coming soon"), but the upper bass driver goes really low before rolling off, and I do let it go there for now. Also I use a pair of extreme low frequency enclosures, which I currently allow to run up to 90Hz 1st order, so a bit higher than will be the case once the mid-bass horns are done. Romy does have true mid bass (via a pair of line arrays).

The lower bass enclosures are responsible for a lot of the magic (pair of 16 cu ft sealed boxes), and they are very much a part of the system. The drivers have a really easy life… The cones don’t seem to move at all ; but switch them off to hear what they’ve been doing. It doesn’t matter what the music, at the very least they will tell you a lot about the recording space.

I don’t have a pair of Melquaid DSETs. I drive the horns with a pair of Lamm ML2s, running full range. I use a second pair of M1.1s for lower bass. The ML2s are hard to fault. In fact at this point, I can only find positive things to say about them. When I finally identify a good reason (sonic) to do it, I will try building a pair of DSETs… If the 45 Hz horns don’t kill me first.

Jd*

jessie.dazzle wrote:

So last night I angled the upper horns of right channel down toward listening point, and left the left channel with horns firing along parallel axes.

Romy wrote :

"...do not forget also while you do your alignment experiments to listen the system from different distance. Get a reference whatever it is and then make a step back and a step forward... You might eventually found a distance from where the "angled sound" will suddenly "click"..."

To help establish the best listening point in the new place, I decided to use an office type chair with wheels (its also nice to slide up to the equipment rack without having to get up!). When doing the tilted horn experiment I did try moving toward and away from the horns.

"...Try to discard the upperbass changes and pay attention only to the frequencies that you are trying to align..."

Yes, I started off listening to music with a lot of violin and cello for exactly that reason, but then realized that because my fundamentals channel goes quite low (700 Hz 1st order), is not attenuated, and was the most steeply tilted, I was curious to know what was going on around the crossover point between this horn and the upper bass... So I started paying attention to that crossover point as well; though I realize that it is less critical.

"...With the parallel horn’s axises the "click zone" will be way larger. I personally tend to recognize that with parallel horns the imaging still better then with the tilted horns and a listener located in the "click zone"..."

Loss of imaging was by far the most noticeable effect with horns tilted. This loss may be more pronounced in my case because I started off with such strong imaging. This capacity for imaging (with horns parallel) may be due to the fact that the top of the upper horn is positioned about 2 ft below the ceiling, which is higher than normal and opens toward the listener (not parallel to the floor). So the stack of horns, with the upper bass horn on the floor and MF + HF in between, is sort of like a line array.

be wrote :

"...I think that Romys preference for non tilting lies in, that by not tilting the horns, the directivity, beaming or on axsis frequency iregularities are avoided, simply by pointing them where they dont do so much harm, below or above the ears..."

A lot of horn users seem to have these issues. I have not heard any other systems using Vitavox S2 drivers, but I know it is possible to get the S2 to generate on-axis unpleasentries. In the case of the system I am building, one can happily sit straight in the firing axis of any of the horns without experiencing the above-mentioned unpleasantries... No cringing. The reason for this I believe is directly related to the fact that I'm not asking each horn to cover such a wide frequency range. In other words, there is a lot of "horn margin" on either side of the frequency range being produced... Which probably looks like a high percentage of wasted horn to some.

jd*

jessie.dazzle wrote:

I have not heard any other systems using Vitavox S2 drivers, but I know it is possible to get the S2 to generate on-axis unpleasentries. In the case of the system I am building, one can happily sit straight in the firing axis of any of the horns without experiencing the above-mentioned unpleasantries... No cringing. The reason for this I believe is directly related to the fact that I'm not asking each horn to cover such a wide frequency range. In other words, there is a lot of "horn margin" on either side of the frequency range being produced... Which probably looks like a high percentage of wasted horn to some.

Hm… it is would be debatable.  S2 on-axis might be brutal, particularly with originals metal cones and one of the ways to deal with the brutal is using various low-pass filters or transient filters. Turning the horn slightly off-axis, using deeper horns and some other methods are low-pass filters. Loading the out tube harder probably is the most effective transient filter. The S2 is very much sensitive to quality of electricity. Since you are at 12 feet, sine you use the 16R tap on ML2.0 I think you might go away with on-axis shooting. I personally prefer to load the ML2’s output tube slightly idler (8R top) and to turn point the horns to my shoulders instead of my ears.

Be,

I do not think that effect you describe has anything to do with demand of horns have strictly parallel axis. I also came across to some situations where the extremely indirect sound was superbly interesting but it is I feel is very different subject and has very different underlying reasons.

My presumption is that with non-parallel case the different frequencies get time-aligned at different distances, something to what we in photography (my first trade) was called chromatic aberration. If you look at the nature of chromatic aberration in optical system then you will see a perfect model why the unparallel horns do not sound right. If you have a lens and hold it’s axis on the line of infinitely located light source then you have one level of chromatic aberrations. If you then angle the lend 5 degree  for instance  then you will have the very same focusing but the amount of chromatic aberrations will avalanche.

I do not insist that it is right explanation but it does it to me. After all - they are my Axioms… :-)

be wrote:

There is no analogy to chromatic aberration in acoustics of air.

Disagree. Regardless the reason for Chromatic Aberration in context of the given illustration we care ONLY about the arrival time. In case of lens we have the discrepancy of arrival time as a deviation of focusing distance. In case of speakers we pay the penalties of misalignment. Do not forget that when we are talking about crossover aliment we are talking about the arrival of a single wave at the middle of the cut off. Let say that we have 10K cut off and use 10K sinus.  What would happen what we use 7K signal or 13K signal? In case of parallel horns we have no discrepancies and 7K, 10K and 13K will require the same alignment. However, with none- parallel horns you can get a perfect alignment only for a single frequency. In the case of Chromatic Aberration it will be the alignment for a single distance. The speed, distance and time are the very same subjects and in that I see a lot of similarities between optics and sound. With all that said I would like to remind that the “Chromatic Aberration Syndrome” as an explanation for sonic problem of non-parallel horns is juts my working hypnosis. I do not insist that it is so. As I said before I have no definitive enlightenment about the true reasons. The subjective evaluations that I did 7 year ago did indicate that the non-parallel horns are a big taboo. I never care seriously look deeper into the reasons. I feel that Chromatic Aberration is adequate enough explanation to satisfy my interest on the subject.

be wrote:

I suspect that your negative views about on axis listening due to sensitivity to listening position must be to some narrow angle on axis frequency anomaly of the driver/horns you tried.

I do not think so. Pay attention I did not bring any subjective arguments to defend my pro-parallel axis portion. If you experiment with aliment yourself and play with horn tilting during aliment then you might discover some aliment inconsistencies. To keep the horns parallel is about do not have those inconsistencies. How it affects sound? I could propose any theories or to refer to my listening experience but in order you and others agree or acknowledge it we need to sit in the same room and listening the same playback. Since it does not happen I did proposed the easy way to objectively confirm the problems with horn tilting. The only contra-argument that I would consider if somebody would align the channels and then by tilting one of the channels will NOT observe the introduction of mis-alignment.

be wrote:

If you insist to use chromatic aberration as a working hypnosis!!

Oh, no, I do not INSIST to use chromatic aberration as a working hypnosis. It is the explanation to myelsef with which has settled. If you of anybody else would offer something more lucid then I would be glad to contemplate it.

be wrote:

I sugest that those inconsistencies ar due to on axis frequency anomalities of the used horns. Elimminate these, and the inconsistecies will vannish, and the concerned axiom wil be void.

be wrote:

"Since most horns have frequency response problems, at and nearby their axsis, it is advisable