Recently Macondo was altered and converted into… nope, not into a Super Macondo :-) but in a “Larger Macondo”. The Super Macondo it will be when I resolved how to implement time delays along with a low pass flirter at midbass horn, that would enable me to do my 15/40,151 40-50Hz 1-octave horn, complimenting my current upperbass horn.

So far I decided to alter my LF channel by adding 2 more drivers. The objective were following:

• Increase sensitivity of LF section
• To “load” ceiling with LF source (look for my posts admiring the idea of coupling a LF source with ceiling)
• Take the benefits deriving from LF-line arrays further up to the full height of my room.
• Made the upper knee quality of LF section  more "interesting". (The constant problem with sealed box, when the lower octave do much better then upper octaves)
• Make better power handling of my LF section
• To see what would happen...

The project was very simple and relatively painless. The enclosures were made from 1” MDF, braced. Each enclosure, of course sealed, used a pair of my favorite 10-inchers, light-coned, soft-clipping, rippled version of Scan Speak drivers (25W/8565-00). The enclosure tuned for 39Hz, and damped with very ordenary high and mid density 2” furniture foam. No, sexy syntactic dampers, no crazy rubber insertions, no plastic gaskets. Everything is very simple but from my point of few quite effective. The only thing that I allowed myself to do was to treat the corners of the enclosures with expending insulating foam from “Great Staff”, slightly rounding the the roughness of the corners inside the box.

The drivers are serial connected, making theoretical 16R. The main woofer towers are theoretical 8R and the new enclosure meant to be connected in parallel to the existing woofer tower. Why I said “theoretical” because the LF channel of Super Melquiades is crossed at 55-60Hz, first order and the woofer operates mostly near the resonance frequencies when the impedance rises, of course no impedance equalization ever was used. The LF channels of the Supper Milq use a full 6C33C, dissipating 35W, the output transformer 1:9, 700R. So, I presume, it can “swallow” the drop of load impedance and do not loose the “too much loading” status that I would like to maintain.

Everything is set up now, up and running and I am in process to teaching it to sound correct.  I intentioanly extended the ceiling section closer to myself then the rest of the tower trying to synchronize the arriving time as the total high is too high. To my surprise the effect it is quite audible and I need to play more with it as I think that I need to use the “extension closer” much more aggressively.

It is too early to make any conclusions. Prematurely I gained approximately 2.7dB and the “ceiling affect" did work VERY nice, making the listening "space" much larger. When I settle down with my observations I will share more about the results.

Interesting, Very interesting.

I might make some observations now, claming this recent modification as a success. The minor gain of sensitively certainly did not heart but the major advantage was defiantly the “activation of ceiling”. It is hard to describe, it have to be heard. In a way it sounds like Sound coming from two out of phase with each other loudspeakers only it delivered by a system that is completely time-aligned. What it did in my case was a creation some sort of “out of geometry reference” experience, effectively killing, even with more vigorousness ten before, the soundstage as an audio consciousness… and presenting the necessary imaging WITHOUT the soundstage. I mean the soundstage is here but it has no definitive position and location in the room but rather it behaves like an abstraction of presentation instead of the presentation of itself. It was exactly why I was craving for the idea of the “ceiling loading” since I discovered this effect.

The power handling of the LF section got better as well. Scan-Speak suggests that the drivers I use are 100W but I feel that the more like 20-25W. However, the biggest discovery I had with the fact the since I begin to use the second LF section I was not able to increase the plate current on the LF channel of the Super Milq.

This question I asked to the technical people and it was posted at the Triodemafia.com:

*****************************************************

Hi, guys.

I have question to ask, or more prissily I would like to learn about your visions what might be the causes for the situation that I can’t explain.

CIRCUMSTANCES:

I have a dedicated low frequency amp (Supper Milq) that low-passed at 60Hz.. The output stage is a full 6C33C, 230V-150mA driving a sealed 8R array of woofers with no impedance EQ. It runs all the way up to 33R which is fine. The transformer is 450mA, 15H, 15W, DCR secondary 23R and it’s made to load the 6C33C with 700R. The tube kind of overloaded in this setting and I have less power, fewer distortions and all together the sound that I found appropriate or let call it for the sake of given illustration the “correct sound”.

ACTIONS:

I recently added to the current array of 4 woofers 2 extra woofers. The very same driver, the identical volume of the enclosures and the almost the same boxes resonation frequencies (41Hz and 39Hz) The new enclosure has total of 16R and connected parallel to the old one bringing the combined impedance down to 6R. As the result, I as expected yielded some power, slightly more “loaded” loading of the output stage at the amplifier level and all other (none-related to this subject) benefit in a listening room. Everything sounds kosher and the bass after the new amplitude adjustment still exactly the same “correct sound”.

PROBLEM:

Using a common sense, since I dropped the load impedance I decided to increase plate current. To my huge surprise the increase of plate current for even 20-30mA killed bass instantaneously. The bass still was there but at 170mA it begin to sound like a noise coming out of fourth order port. Interestingly that the word “instantaneously” was the key mystery word in my last sentence as when I used a single old enclosure with impedance of 8R (out of resonance range) then the effect of worsening of sound with increase plate current does NOT manifest itself AT ALL. I mean I could drive a single tower with even 300mA that dissipate almost 70W on this tube but I never had those “ported” distortions. It would be another problem but it will be not so bad.

WHAT THE HELL IS GOING ON?

Really, why is it so? Should I drop voltage? I still way beyond the max tube dissipation (60W) with 230V and 175mA! I did drive this tube at 60W and it dose fine with one single 8R enclosure. Does the impedance of this tube changes too dramatically when it loaded with lower impedance and driven harder and this change in some mystery way affects the speaker’s damping? Why the amp, or the woofer towers, or the interface between them suddenly do not like the increase of current? Are any further thoughts from the member of the Mafia about my quandary?

I kind of do not need “help” with it as I do like what is going in at 150mA and I can perfectly live with it “as is”. However, I would like to UNDERSTAND WHY my higher current kills bass if I’m so far from the saturation point.

*****************************************************

The further experiments leaded to the discovery that those particular woofers of this particular no-feedback amp have some kind bizarre puss in them that makes them superbly sensitive to dumping factor.  Dima discovered that adding 2.7R resistor in series with 5.5R of the new load does return the amp operation into the former state and does allow me to raise the plate current. The interesting part that when I played before with different amps and different speakers I never ever ever ever ever ever seen such a dramatic sensitively to the dumping factor. So, it looks like the need a new 700 :5.5 low frequency transformer.  I still am contemplating if I need to take out of the LF channel more power then I’m taking out now, as I might perfectly fine to live with the current 150mA. If I decided that I need more power then the transformer will be in order, or perhaps I will loose a few turn on the secondary of my current transformer…

Still, I have no idea why my system became suddenly so super-sensitive in regards to the dumping while I dropped impedance juts for 2.5R.

Ronnie,

Arranging the driver in a line array is a fierily straight forward thing. Also be advised that the arrays will less likely work across a short wall, and you need to use a long wall. Generally it is imposable to advise anything as any rooms are different. My experience instate the if you have a narrow bandwidth suck off then leave it as it and do not fix it – you will have better result then LCR it. Find an average level with your 3 octaves and set this level with your amplifiers. My room has ¾dB more bass on right… witch it kind of work as the basses are mostly on right (unless the orchestras use the stupid Wagnerians sitting positioning). Also, the 100Hz, looking at your horns, does not sound as useable crossover point to me. You most likely have a high pass filer at your bass horns (you should). The horn decay with 2-3 order and the first order high pass kicks in somewhere there as well. So, the bass horn should due quite fast. The bass tower should be to the bass horn with 1-2 order, depends of your room. In my room I use 107Hz second order Bessel (at line level you might write it very precise). Approximately in 2002 I stopped to use the section order and went for first order 65Hz. Disregard the number and looks how the channels talk to each other. Thos driver are not good above 150Hz and the lower you would driver then the better it would be. I would strongly encourage staying with first order and doing the time alignment… generally the piling up driver in the array should not crate the frequencies anomalies, but should do opposite: it should laniaries the response.

My experience is the more 25W8565 drivers you use the better it would be: less excursion for each drivers, more linear, more sensitively, better room loading, lass HF off the axes and so on. I personably was searching for quite whale better driver but was not able to fine it. I like the 25W8565 very much. The SD1 motor is very smart, the suspension is very interesting (pay attention that the suspension has the same “tone” as the cone), he cone is light and very “sound”, the driver is 10” and sub 20Hz resonance. In addition the driver “clips” when stressed very-very soft, almost like 300B SET. If you can find a 10” driver with Fs=20Hz, good tone, light cone that still can “bend” at 20Hz then let me know and would very much embrace it. So far I did not see other contestant.

However be advised that it you use no feedback amp and with really high output impedance then the 25W8565 is very-very freaky and require very-very precise loading. In fact it is more sensitive then anything else I have ever seen. If you use tubes then pick the very correct plate current and the plate loading as 1Ohm off will toss the sealed 25W8565 in not good sound. If you use SS then running the bias and voltage set to the mode of the output stage that the 25W8565 appreciate. One again the 25W8565 is very-very sensitive. I affect the quality of bass but of cause not the frequency response…

Rgs,
The Cat

Ronnie wrote:

Listening to the horns, it seems that they indeed die somewhere below 120Hz. I was surprised, and a little disappointed that it seemed so steep. Good to hear that it's to be expected!!Yes, I'm using a 30-something-Hz 1st order high pass on them too.

Yes, what happens with horns is very similar to what depends with boats. When you do boats you always experience the “2 feet deficiency” as you fell that if you have a boat 2 feet longer then you would be able to feet everything…. Of course the longer boat does not help, as it will have also the “2 feet deficiency”. The very same happens with horns, as we always feel that it is were a few inches larger and a few Hz lower then it would be more fun…. It is the miserable human nature. We Cats are so much better! We juts hate everything regardless what it is….

Ronnie wrote:

The LF crossover point has been tuned by ear to 60Hz (1st order), and I was beginning to think that perhaps that should result in a dip around 100Hz. I guess I'll have to sit down and learn the crossover maths ;-)

Hm…. Forget about the math. You need a RTA and a good calibrated microphone in the listening position.

Ronnie wrote:

A propos that, My big-mouth-fostex-horns should be high-passed at the same frequency as the big horns are low-passed, right? I'm using 1st order 950Hz HP and LP.

Yes , most likely so.

Ronnie wrote:

There are just so many variables: LF-box positions, LF-volume (on the amp), XO point (pots in the line-level XO). The slightest change can screw everything up. especially turning the volume knob on the big-horn-controller-amp a fraction of a millimeter can be devastating or perfect. I've got those knobs glued in approximately the right place. :-)

Actually if you milliamp then you should get rig off all those knobs and use fixed voltage dividers at line level (and perhaps the  filters in there as well)

Ronnie wrote:

Anyway. The LF is wavy and that's why I've resorted to listening to beautiful sinesweeps tonight, trying to find the most even LF, and go on from there.

You will find a good setting, Ronnie, it might take for a while. Do not forget that your room might be no LF symmetrical and it is fine to use different dB and different cut offs for different channel. In-phase the LF and upperbass and sink any deeps with moving the LF cat off on right.  BTW, be advised that linearity of separate channels not always does the linearity of stereo mix…. BYW, if I were you and if the real-estate would allow, then I would place the speakers twice wider. Move them as wide unit the mid image become “thin”. Also try to move the bass arrays further from the horns but at the same distance from the geometrical “sweet spot”.

The Cat

Ronnie wrote:

I'm curious about how you set up your horns in the room. From photos it looks as if they are fairly close together, compared to the wide setting you suggested I try, while our room dimensions are fairly similar? When I started pushing the horns apart as you suggested, I didn't notice a point where midrange would thin out, but the whole picture would certainly open up and become clearer in every way, like a painting on the bellow of an accordion as it's opened. The stage/sound-landscape would sort of overtake the room, or become one with it... As if the room no longer had any chance of compromising it (although there were of course still room mode problems with my single woofer per side!). Unscientifically it seemed that every room-induced reflection-problem would become percentually smaller, because the stage was life-sized. Like spilling a drop of paint on a 1x1 inch picture is worse than on a 100 x 100 inch one. I noticed you wished for a bigger room after finishing the 250Hz horn, but it looks to me as if there is room to push the speakers 1-2 meters more apart. Perhaps not related to the width dimension? Would be interesting to know a little more about your chosen speaker positions, and perhaps see what it all looks like from the listening position.

I always felt that it is very strange that some people try to propose universal rules for speakers positioning, disregarding the specifics, even as big as speakers topology. Even with the same topology there are too many differences: drivers, crossovers, shapes and many others that make the rules too much confusing or sometime not even applicable. It is possible to write up a correct methodology for a giver speaker but not for a variety of speaker. Therefore I do not think that it make any sense to share my personal experience with Macondo as other people use different speakers and therefore their mileage will very much wary…

Still, I would like to bring up some very general and not specific to any horn design points that I feel any horn person should take under consideration. I do not pretend that it would be compute guide and juts dump whatever I have atop of my mind:

1) Imaging. It is important to recognize difference between imaging and soundstage.

2) Soundstage should not be considered at all at lower levels of sonic assessments.  If to accept the following evaluation schema:

http://www.goodsoundclub.com/TreeItem.aspx?postID=50#50

then soundstage being to be important from 5th level and up. If you concern about Soundstage while your mind operate at firs assessment level then you are a fool or an audio-reviewer.

3) Imaging is supremely important element but not for listening music but for audio evolutions. Imaging is a bogus artificial element and it hardly useful (with very minor exception of few selected compositions that use space allocations). However, (very big however!!!) imaging is a phenomenal indicator of “something else”. The differences  or imperfection of system imaging always suggest that something is wrong and let to mange that “something wrong” using imaging as monitoring tool.

4) The greatest thing even was written about speakers positioning is following:

http://www.goodsoundclub.com/TreeItem.aspx?postID=994#994

This is the real high-end of the speakers positioning and it should be persuaded AFTER the rudimentary correct position was accomplished. The DPoLS is superbly difficult to accomplish but if one does it then it overwrites whatever one knows about sound redaction. I am not kidding.

5) With horns everything is much more completed as horns change a lot while you tow them in or out. Any tandem of driver/horn has a fixed angle (that I call “Horn Critical Angle”) under which the channel performs at it’s best. If you go further then the Horn Critical Angle then you lose HF and get amplitude deviations. If you go sharper then the Horn Critical Angle then you might get too brutal sound and high deviation in the channel’s transient evenness across the horn’s band pass.  This is the most critical with compression drivers that very seldomly should shot directly to ears.

6) Horns should be setup with tweeters turned off. Do not worry about sound – pay attention only in imaging. If introduction of tweeters changed imaging then you use tweeters inappropriately.

7) Buy, steal, borrow or whatever Lamm L1 or L1 preamp. You would need it only for one weekend then you might let it go. I did not see anything in Audio that might be use as remotely comparable educational tool about power of imaging or how it “might be”

8) If you use multi-channels horn then discard evaluations my mono material: the different channels should produce near identical size of the center image and it should be place at near identical distance into the death of presentation.

9) Severally and anal-retentively record each step you do during positioning, literally keep dairy. You will make mistake and will return back to the refused settings. Make your returns reproducible.

10) If you are in horns then imaging VERY strongly connected with tonal balance (ironically it is not affect the “Absolute Tone” ™). While you chase the rudimental positing then take tonal deviations under consideration. However, after the loudspeakers are positioned, the system set and you go have guts to go for DPoLS then discard tonal balance. If you found DPoLS then it will overwrite everything.

11) While you in search for speaker position then use the principles of setting focus on cameras: move forward to the selected direction until is it better, better, even more better, good, very good, excellent, perfect and then until it become less excellent. From that position you know where it was the best.

12) Never, under any circumstance angle horns or allow any none-parallelling of horns.

13) Electricity does screw up everything including imaging. If you decided to spent time with horns moving and electricity go south then shut down your system and do something else.

14)  If you as the result of your efforts accidentally (it does happen) moved your speakers and hit DPoLS (trust me: you will know it) then do NOT touch that speaker anymore!  Document and map this location with unimaginable level of precision and details. Then as a tribute to my foretelling about DPoLS send a $50 donation to your local shelter for homeless Cats and declare that audio is over. If you there then you know what I mean.

Ok, I need to run grab a bite and it was all that I had atop of my hungry head…