DYI horns- ugly but it ain't done yet

[D-MAN]

Well, here is a pic of one my prototypes. I know that they look horrible, but their not finished yet. I was thinking about painting them, so I started filling with putty, which further adds to the already unpleasant look...

When one takes the time to study the Klipschorn, one comes to appreciate what went in to them, and the ingenuity of the design itself. It is virtually perfect for what it is.

The reason that I went my own direction is that I want to experiment with various driver configurations which the Klipschorn is not intended to support.

This is my first whack at designing a horn, and I have a newer design which I have not yet built as this one proved too complicated and took too much wood IMO (and too much experience in woodworking for me, but I'm learning).

I specifically wanted a top-loading design that could be used for back-loading a full range driver for experimentation. Also the back chamber is such that there is internal room for a mix of drivers, like 2-12"'s like the Jubillee and such. Since I set this up for the traditional 3-way, which I love the sound of, I am considering whether to go for a Khorn cabinet and call it good. But for now, I have an experimental platform to play around with.

I took the approach that everything mattered in the horn channel, so the finished side is towards the horn channel, rather than on the outside. If I had it to do over, maybe I would rethink that. In these, the back chamber is 3/4" and the rest is 5/8" 7-ply finished one-side AB no hollow exterior plywood. Takes 3 sheets per cabinet when all is said and done. OUCH!

The overall horn length is 59 inches in these, which is about 5 or 6 inches shorter than a Klipschorn measuring center of channel. Results in slightly less mouth area of I estimate to be less than around 1/2" difference compared to a Khorn overall mouth area in square inches. I imagine that it is not a hearable difference. The horn is designed for 40Hz. Due to the design, it is a couple of inches "deeper" than a Khorn.

They are top-loading; there is an access panel on the top. I am running K33E's in them. The back chamber is approx. 5000 cu. in. in keeping with PWK's design for the same reason. The horn is relatively the same as the Khorn, except that it folds differently.

I did the top "cabinet" as a frame. The mid and tweets sit on a stand. I am using ALK crossovers and non-Klipsch horns and drivers on the top end.

This is an unfinished prototype, so please do not beat me up - yes, even I must admit that they are UGLY. Believe it or not, they look better in person than in the picture...

I will get them completely finished, furniture quality - someday.

But they DO sound pretty nice...

DM

[D-MAN]

Here's another pic showing the upper horns without the frame, if I have the right one (I should have renamed them first, I guess!). Again, sorry about the size; this digital stuff is all new to me. I feel pretty good even getting it to post!

The horn is a P-Audio 400Hz fc with a Selenium D305 driver.

The tweet is a Eminence APT-50.

DM

[sunnysal]

no first pic...

[D-MAN]

Ok, lost the first one because it was too big. I resized this one. Sorry all - it's a learning experience for me.

Dm

[mikebse2a3]

Hi D-MAN

Thanks for showing your work. Your doing great!

please share more pictures and details as you explore this.

mikeb

[sunnysal]

looks great! please describe how you selected the new horns and drivers, are they better than stock k-horn drivers and horns? in what ways? you should not ashamed of the build quality they look quite nice! thanks, tony

[D-MAN]

Thanks, you guys, you are too kind.

Well, to answer Sunnysal's question, I don't think that the different horns and drivers I am using are "better" than the Klipsch versions per se, (well maybe the tweeter to a degree). Also I think that the Klipsch midrange horn is quite a value after they lowered the price.

Here's some of my upper horn criteria:

1) get a horn tweeter capable of at least 20K at 104db sensitivity.

2) get a midrange horn both shorter and less wide so that it may be turned and positioned as desired separate from the top cabinet; it had to go down to at least 400Hz fc - I've tried 500Hz fc 60x40 but ended up with 400Hz fc 90x40 as being my fav. It is possible to use a 500Hz fc horn satisfactorily, but there are complications ; also I feel that the Klipsch mid-horn is too wide and too deep for this purpose.

3) phenolic diaphragms

I chose the drivers based on their specs, and those were closely based on the Klipsch drivers.

I specifically did not want to build a Klone because I had a pair of those way back when so I knew some of the shortfalls, such as you couldn't turn the upper horns to face the listening position without turning the top cabinet, etc. Also I did not want the EV T-35 tweeters as I had them in the 70's and didn't particularily like them even then; also they don't make them anymore even if I wanted to.

I was careful to stay with the same sensitivity ratings as the Khorn drivers. The match between the upper horns and the bass horn output is the maker-or-breaker here. Anyone who has played around with other drivers will back me up on this. I think that that is the reason that the Klipschorn is such a jewel, to tell you the truth. The crossover is going to make or break it, too, and keeping in line with the Khorn meant that I could use various Khorn crossovers as desired.

The midrange is a 2" throat with a sensitity of 110 db and is 8 Ohms vs. the Klipsch 1" throat at 16 Ohms, both are rated at a 400Hz fc (or so). It is not any more efficient, but it is capable of handling more power because it has a larger voice coil and magnet structure. The 2" throat dictated that I find a horn capable of supporting it and to find one of an appropriate size for the application and the dispersion pattern that I wanted. That is a very narrow field and the only one that I have found is the P-audio 4525 90 x 40 degree 2" throat w/400Hz cutoff. It is 17" wide and 9.5" tall and about 10" deep and is glass filled resin. I think that it is too high to fit into a stock Khorn upper cabinet.

So, aside from the size and the increased upper frequency range of the setup, it still is pretty much in line with what you would hear from a Khorn. The settings on the midrange autoformer is the "normal" setting on the Khorn. I would LIKE to say that they sound better than the stock Khorn, they certainly cost more, but I really can't really say that for sure. I've only heard these in my house and all real Khorns elsewhere. You would really have to get them in the same room and A/B them to actually tell, I think. The fact that I can "aim" the narrower dispersion upper horns inside the upper cabinet without moving the upper cabinet will, of course, give me the edge in a Klipschorn shootout, but that's just my opinion, too.

Another interesting thing that I have learned is, yes, you can put just about any old woofer in a Khorn - AND it can definately increase the bass output (get rid of that subwoofer), but then one has to change the upper drivers to match the same increased output and then make or at least modify the crossover to deal with it as they are all inter-related and none of them will line up with each other. This is the time consuming and expensive part. I could not find a woofer that does what the K33 does for the money transparency-wise. Something in the mid and upper bass from a K33 just blends in with the midrange that is the key to the whole Klispchorn thing, IMO. PWK already did most of the work! All we can do is tweak it a bit here and there...

DM

[sunnysal]

thanks for taking the time to explain how you chose the drivers and horns. understanding that helps us understand the goal of the project and then later we can put it into proper perspective (along with you) as you post listening notes (your next task...lol) and tweak the crossover. again thanks and congratulations on your new horns! tony

[D-MAN]

Here's a shot from the top of the upper horns inside the upper cabinet "frame".

DM

[D-MAN]

Forgot to point out something; note the wiring.

I had a fuse blow due to nothing at all and decided to get rid of the fuses completely. But when experimenting, I try not to burn any bridges, so that I can always go back to "the way it was".

I originally just made a 12 ga. 2-inch "fuse" and stuck it in the holder; it made a noticable difference in sound quality, enough to start me experimenting...

I have jumped the fuse holder with a few inches of 12 ga. "Sound King" copper speaker wire. This made a noticable improvement to the upper bass. These wires are soldered in. I did not change the crossover except for the addition of one solder joint. The major drawback with this stuff is that it doesn't take to solder very well.

I then made a run to the woofer from the crossover using the same stuff, and it cleared up some mid-bass "muddiness" quite a bit. It increased the clarity of the bass quite a bit.

When I tried a run of the same stuff to the midrange driver, it sounded "too soft" (it lost what I would call it's "edge"). So I went back to the Home Despot 12 ga. that you see in the pic with the "silver" and copper colored strands for both the midrange driver.

These simple tweaks helped out quite a bit. I can't recommend them enough to fellow tweakers beacuase they are cheap and easy to A/B. For those who think that wire is just wire, here, do it for yourself and tell me what you think.

DM

[D-MAN]

I resized the previously posted pic of the upper freq. horns here. There is a optical-illusion going on on the right hand side of the cabinet caused by a shadow.

It shows the relative sizes of the horns as viewed from the direction of the listening position without the grill/frame assembly. The bass horn cabinet is 30-1/2" wide for a reference.

Note that they are pointed slightly outward instead of directly at the listening position. This is the optimal position that I have found for the room limitations imposed as far as achieving imaging and getting an unmuddied response. It is also intended to allow for better sound in other environments of differing size (that is, different distances of separation between the two corners).

I chose specifically to avoid having a "baffle" mounting the horns. The mid sits on its own stand which is positionable within the top grill/frame assembly. I suppose one could say that some diffraction is avoided that way. I read a paper on horn edge diffraction and wanted to experiment. My conclusion is that it works just fine and certainly takes less wood and work...

The triangular 1/2 ply piece that they are sitting on is the access panel to the back chamber. The tweeter is mounted to the side of the midrange horn; I did this initally because I was lazy, but it seems to work just fine. I originally had a different and smaller midhorn and aligned the tweeter vc with the mid driver vc for time-delay correction. I find no sonic difference in the two "versions" and intend to leave it the way it is as you now see it.

The opposite side (the other speaker) is reversed, of course. I put the tweeter on the outside closed to the side wall on each respective speaker for better directivity and a wider dispersion due to the separation. Also the fact that the tweeter is closer to the wall, its reflectivity off of the walls can be better controlled while the reflection is reasonably small, IMO.

DM

[D-MAN]

Here's some technical details and some more of my yammering incessantly about the bass horn.

I cut the throat opening cavity at 6" x 13" vs. the Klipsch 3" x 13". This is for the use of drivers other than the K33E. I have 4 carriage bolts used to mount a driver-mounting board, which allows for the slot to be changed with the motor board/driver assembly.

The Klipschorn and my "thing" both have a 78 sq. in. cross-sectional area at the throat, i.e., the 6 x 13" opening is 1:1. The slot in the driver-mounting (or motor board) is used to alter the capacitance at the throat for upper frequency control. That is why PWK uses the 3x13" slot for the K33E.

I've talked with some who are in favor of the direct-to-baffle-board mounting of the woofer to the horn, but I intentionally avoided that for a couple of reasons, the first of which is I want to experiment with different drivers and such, so I will be mounting and unmounting alot. I used the old Speakerlab-K approach and used a separate motor board that mounts to the horn with wingnuts, which is a time-proven approach that I knew worked.

I've heard it said (wrongly) that the Khorn is specifically designed for the K33E. Which the exception of the stock Klipsch 3x13" throat opening, it JUST AIN'T TRUE!

Actually, the K33E is designed for the Khorn, and that is a compromise between lower bass output and upper frequency ability in a 4 Ohm vc.

The Khorn back chamber is somewhere around 4800+ cu. in. in volume (mine works out to 5011 cu.in). This volume is to counter-balance the amount of air in the horn itself, and ensure that the backward "cushioning" effect of the air is relatively the same as presented by resistance to movement by the air in the horn itself. Lowers intermodulation distortion by matching the reactance of the horn throat. PWK published the formula for calculating this in his 1945 patent for the Khorn.

Therefore, one can use any 15" woofer that one wants to in a Khorn. The parameters that determine what woofer works better are somewhat vague. The "normal" T/S parameters are somewhat not applicable as they tend to change somewhat when placed in the horn. The main things that I have found are the sensitivity. That is dependent on a couple of things. The K33E is 4 Ohms and is driven at 8, which gives somewhat of a boost to it. Where it really shines is the mid and upper bass frequencies. But others can work, too.

I experimented with putting gallon jugs of water (231 cu. in. volume each) for displacement into the backchamber to see what happened and to verify that I had the optimal size for the application. I went up to 4 gallons per cabinet in my tests.

The results are that the lower frequency response moves upward as the backchamber volume is decreased, which makes sense. This would have the effect of causing unequal excursion and increase modulation distortion, but I couln't hear it, per se. I also think that the upper frequency limit also moves up correspondingly, but that is just by ear, so its an impression rather than a fact. It was just an experiment that convinced me that the back chamber was appropriately sized.

Bear in mind that the free-standing parameters no longer apply when placed in the horn, for instance, the nominally 96.5 db sensitivity of the K33E now becomes 104db in the bass horn. Is this a combination of 4 Ohms "overdriven" to a degree along with the physical amplification provided by the horn itself, or the matching of the increased reactance provided by the 3x13" opening and the inductance of the voice coil combined with the 4 Ohm "easier-to-push" damping seen by the amplifier? I haven't got the faintest idea as to why it works so well. There is some element of "synergy" going on, but what it is exactly, I don't know.

When one uses an 8 Ohm driver in the horn, the crossover will be effected, of course. Also I have found that the wider throat cavity opening is desirable in all cases of 8 Ohm drivers that I have tried. In trying to figure this out, I decided that it has to do with the voice coil inductance (higher, of course) and its effect on the reactance at the throat imposed by the more-restricted area of the smaller 3x13" slot (which gives it the ability to act as a high-pass acoustic filter). My experiments bear this out. Perhaps someone else could explain it to me?

The impact of changing the woofers causes a ripple effect on the crossover. The all-important blending between the woofer and the midrange is the area of discontent, IMO, generally because one cannot effectively estimate the performance of a given driver ahead of time.

I find that the sensitivity issue to be the most important. There is nothing that one can do that does not effect the overall sound when the woofer output is higher than the midrange output. There is nothing worse than an over-driven midrange (I refer to it as "glare"; the universally irritating "shrill horn sound") trying to keep up with the bass.

Also, why is it that any combination of the midrange autoformer connections is either too much or too little?

Oh, for a continuously adjustable autoformer!

DM

[D-MAN]

Side view of bass bin.

DM

[D-MAN]

I know that Klipschorns do not have insulative material in the back chamber. However, I use some fiberglass in mine, just around the back of the K33E driver and the bottom of the back chamber.

The effect is that ALOT of upper mids actually reflect right through the cone and into the throat without it causing a muddying effect. This can be cleaned up a bit by some insulating material, as long as you don't use too much.

Since the respective back chambers of this "thing" and a Klipschorn are different shapes, your mileage may vary in this, but for me, the insulation stays in.

I would recommend trying it, but use as little as needed or you will over-damp the back chamber and then bass gets muffled and boomy (unbalanced reactivity at the throat).

DM

[D-MAN]

I'm getting a little better at this digital stuff, so here is a pic of the side view with some notations added using "Paint".

One the other cabinet, which is not shown, I added a 3/4" strip down the backside of the back reflector panel to further strengthen it. But I think I'll go with the side-braces as in the pic, next time...

The drawback to all of this is that these things weigh a ton each.

DM

[D-MAN]

KIDS DON'T TRY THIS AT HOME!!

Bass horn diagram (sort of)...

At least it attempts to show the general orientation, etc.

DM

[D-MAN]

Well, I swapped out drivers (again) and took some pics of the back chamber while it was empty. This is looking from the top down with the driver/motor board removed.

The glue is TITE-BOND II, which discolors as sort of a yellowish orange...unsightly but strong.

DM

[D-MAN]

Here is a shot of the throat cavity opening from the top.

DM

[D-MAN]

I should say that I'm not as sloppy as the pictures show with the glue. After it dried, I applied silicone sealant to all internal seams. This is what makes it look particularily nasty. I did the extra step to ensure an air tight cabinet in lieu of having good woodworking skills.

DM

[Colin]

nice thread, looks like a lot of work, thanks!