Audiolics Anonymous Chapter 26
Nothin' Much To Say
Article by Bill Gaw

  Well, believe it or not, for the first time in two plus years I have very little to say. Since I don't get paid by the word, I won't pad the column out like some writers for other high end journals. That'll leave more time for you folks to tweak your systems, or maybe even listen to some music. After last month's marathon column in which I reviewed several tweaks, I have nothing to report about new equipment. There are several pieces of equipment scheduled, but none have shown up. That'll teach me not to bunch product into one column and then have to go looking for topics the next month, but I was just so excited about my findings that I couldn't wait to get the information out.

For the first time in several years I have actually been happy with my system for two full months. I know, that's heresy for a tweaker! I should be drummed out of Audiolics Anonymous, but remember who founded this organization... and its purpose was to maximize the effects of and give meaning to the changes we make in our systems. The improvements that were made in my system by the Walker Audio High Definition Link and the Sound Application CF-XE Line Conditioner have satisfied (for now) my search for the Holy Grail. Never thought I'd get to this point, but I have. My electrical and RFI problems, my two biggest bugaboos, have been solved. Having said that, I must confess that there are a couple of further tweaks this month made possible by the CF-XE unit. There, I've confessed.

First, immediately after completing last month's column, I paid for and kept the CF-XE unit and then, per Mr. Weil's suggestion, begged him for a second CF-XE unit to isolate the digital and video from my analog side. Did this make another $5,000 improvement? Not quite, but close. It did clean up the analog side to the point where I don't have to turn off all of the video and digital equipment to get pristine analog sound. Now I can go back and forth between analog, digital and video, without having to fuss over whether one will affect the other. While the single units' individual sockets are isolated from each other, I guess the isolation is finite, as separating the digital from analog does take away a very slight cloud or haze from analog when the digital is turned on. Believe me, you have to hear the difference. Again, its funny how you think everything just sounds or looks perfect until an actual improvement occurs. Then you just sit there and mumble to yourself how can it be? How was I able to stand the sound before this. Its like wearing a set of glasses for years, going to the optometrist and getting a new prescription... and finding that things weren't as clear as possible before. Problem is, once you see (or hear) clearer you can't go back.

And I must say the greatest improvement with two CF-XE units, at least in my system, was not in the audio but the video side. My Electrohome projector suddenly puts out a much more saturated picture that is cleaner with fewer artifacts. Its almost as if the projector's power supply has been strengthened. Focus has tightened, and the color saturation has been boosted, without the inevitable color shifts that occur when you fiddle with the controls.

For instance, I watched "Band of Brothers" in HDTV last night from the HBO C-Band HDTV feed in 1080I, and the video was as close to the best film reproduction as I've seen, and the audio beat anything heard in the best theaters. Just don't understand why using two CF-XE units has affected the video to such a great extent over using one unit, yet thankful that it does. Anyway, I've been tempted, seen the difference, and have dug into the retirement fund for the moolah to purchase the second unit. Maybe the temptation of Eve wasn't an apple, but a new piece of audio equipment.

The second tweak this month was to rearrange my main speakers and subs. To refresh your memories, the main speakers consist of an Edgar 18" round tractrix mid-high horn with TAD 4001 driver from 350 to 20,000 Hz., mated to a 6 ft. long hyperbolic-elliptic rectangular straight woofer horn with two EV-12L drivers from 55 to 350 Hz., with a 2x3x4 cu. ft. subwoofer with four NHT 12" drivers from 16 to 55 HZ. Each uses a 24 dB fourth order active crossover. Obviously the combo takes up a tremendous amount of space in my 16x28x14 ft. room, and the trick is to place them in such a way as to maximize the sound without taking up the entire room. In addition, I was using the Toshiba and MSB power supplies, which were taking up more space and also giving off fan noise. By placing them behind the horns and padding the area with Sonex, I was able to decrease the fan noise by about 10 dB, which was very helpful, but took up the space behind the woofer horn.

While horns have a great reputation for their dynamics, ability to play both quietly without noise and loudly without distortion, and their ability to use low power amps, they have a bad reputation for soundstaging, imaging and sometimes a honkiness. My studies have shown that these problems relate to the time alignment of the drivers, and their crossovers, the curve type of the horn, and room placement.

First, most horns are far from time aligned and most cross over in the midrange where time alignment is most important. This is the nature of the beast, as mid horns usually have a length of one to two feet, depending on how low they go, versus a bass horn which can be five to fifteen feet long. Again depending on the lowest wave length one wishes to obtain. Also, the horn's low end cutoff should match the driver's for lowest distortion. Obviously a ten foot long straight horn with a ten foot square mouth would take up a considerable amount of room space, and thus most bass horns are labyrinths. Labyrinth types take up less space but give problems with time alignment with the mid horn. Also, the twists and turns eat up and distort the higher frequencies passed by the horn, and the walls tend to flex more, which also causes distortion.

This is especially so in single driver horns, such as Lowther's, where the rear firing bass component has to travel through a several foot long labyrinth as compared to the forward firing mid-high component, which has no horn. I think one of the reasons that the Beauhorn (reviewed here and also here) sounds so good for a Lowther horn is that the driver has both a back and front horn (which loads the driver equally on both sides and decreases the time discrepancy between the mids and bass as both travel through horns).

Second, for some reason with multi-driver horns, the designers rely on the natural 12 dB rolloff of the horns and second or third order passive crossovers. This screws up the timing component of the signal, and makes the drivers work out of their effective range, as only the horn is rolling off the amplitude, while the driver is still trying to reproduce the lower frequencies.


Third, in order not to take up an entire room, and still get a decent low end, bass horns have to be placed close to floors and walls, or in corners. Each increase in boundary it's placed next to decreases by half the size of the horn necessary to produce the same low frequency range. Thus, if it's against a wall and floor, the horn can be 1/4th the size necessary to obtain the same low frequency extension of a horn suspended in the air. Also, being against the floor gets rid of lower mid range-upper bass combing of the floor bounce effect seen in normal cone speakers leading to frequency anomalies. Thus giving a better upper bass reproduction. The problem with placing the bass horn against the wall is that it then screws up the imaging of the mid horn, which is stuck in a corner or against the side wall. This can be ameliorated by using some sound absorbing material against the side wall at the first reflection point of the mid horn.

Finally, while hyperbolic or elliptical or conical curves are fine for low frequency reproduction as they make for ease of cabinet construction, with mid horns, these curves cause honkiness. I have found only a Tractrix curve will produce clean, sweet mids without distortion.

So, where does that leave me? First, a straight bass horn is a must. Each turn in a labyrinth degrades the sound, especially if the horn has to extend into the mid region. Of course, this means you need a very large room to accommodate them, and a very accommodating wife. Straight also make it easier to time align with the other drivers.

Second, the horns should be crossed over with even, high order crossovers at or above their natural rolloff points so that the rolloff is actually being done by the crossover and not the horn. This keeps the driver and the horn functioning in its natural low distortion range. Obviously this does present the problem of passive crossovers having multiple parts which can degrade the sound.

Therefore, third, active crossovers, preferably fourth order, with multiple low powered amps should be used for best sound reproduction, possibly building the active crossovers directly into the amps and/or preamps for lowest parts counts.

Fourth, the mid horn should be properly time aligned with the bass horn by using an MLSSA program or by ear. The problem with this is that the mid horn then is usually not in the optimum room position for imaging.

Fifth, all walls of the horn need to be very well damped, preferably with sand or various combinations of wood, bitumen, etc., as they need to transport tremendous sound energy without flexing or vibrating. Again this presents a weight problem. See why designers went away from horns when high powered amplification became a possibility. Too bad, because when horns are done properly, they have the best imaging, soundstaging, dynamics, etc., with the lowest distortion of any speaker-amp combination, including the exotics. Trust me.

So how does the above relate to my latest adventure? The woofer is a quarter horn, and thus the mouth has to be placed against the side wall and floor to get the correct boost in the bass, which also alleviates the usual speaker problem of cancellation of the mid bass due to side wall and floor reflection interference. The mid horn needs to be out in the room to decrease reflections from the walls, and the first reflection point between the horn, side wall and listener has to be damped. Also, best meshing between the two with my system occurs when the drivers for the two horns are time aligned to less than 1/2 inch and the drivers are at least 10 feet away from the listener. In addition, I've found that the subs mesh best when they are within one to two feet of the woofers. As they couldn't be placed behind the woofers because of the power supplies, I had to place them next to the woofers, which brought them out into the room where they were a tremendous eye sore. As they blocked the sound of the mid horn there, I had to move the mid horn on top of the subwoofers, which screwed up the time alignment.

Thus it is difficult for all parameters to be met. For instance, the mid horn has to rest on the woofer directly over its throat, which brings the mid horn within 1 foot of the side wall, and resting on top of the woofer horn. Thus, early reflections occur between the mid horn and the side wall and woofer cabinet. This has been alleviated by building a 6 inch thick chamber against the wall and filling it with Sonex, then covering it with a thick wool tapestry with music scenes, which damped the first reflection and added a little class to my music room. But I still had a problem with decreased depth of field and a loosening of the image with the mid horn on top of the subwoofer.

Well, the CF-XE units have now solved this problem also. Now I don't need the AC power supplies. Thus the room ambient noise is much less, and the room corners behind the woofers were freed up. That gave me the idea of moving the woofer cabinets a little further out into the room, and placing the subs behind them in the corners. Of course I got one great backache moving the 250 LB subwoofers and the 600 plus lb. woofers, and it took a while to time align everything again using my IMP unit, but wonder of wonders, I've solved at least the clutter problem.

With the sub directly behind the woofer, and the mid horn resting on top of the woofer throat, the room looks twice its previous size, which is great for the WAF. Also, the imaging and soundstage have improved, with a slight tightening of the individual instrument's sizes, and a deepening of the soundstage, probably secondary to the mid horn coming further out into the room. Thus, not only have the CF-XE units gotten rid of my electrical and RFI problems, but they've also cut down the background room noise, and allowed a better speaker placement. How can you beat that? Next, I'm going to try removing my expensive AC cords one by one to see if the CF-XE unit will allow cheap power cords. Who knows, if the unit can replace twelve expensive cords it may pay for itself plus a few bucks. I'll let you know. Gee, can you tell how excited I am about this unit? It's the best thing since the invention of sliced bread.

Well, that's it for this month. See I told you it would be a quick column. And I did tell my editor last month that I wouldn't have anything of value to say. So I was right. Hopefully next month I'll have some news on some new SET amps being made for me by Jack Eliano of Electraprint Audio, and being demo'ed at the San Francisco Tube show. They will have his latest circuitry invention, which eliminates all caps in the signal path, and improves isolation of the power supply from the signal path.