KHorn mod- experiences and a question

[Horatio]

I have not posted on this forum for some time, but, I have reached a point in my messing about that I am writing to ask for input.

The story so far:

I have a pair of Speakerlab K's, factory built. However, these are running K33E woofers, and are using Type-A crossovers. Those two steps alone vastly improved the K's, to the extent that they are, for all practical purposes, indistinguishable, from a sound perspective, from the Klipschorn. This much is well established and shown by others, too.

Most of us are aware of the changes PWK made to the midrange over the years, questing to get it right. Most of us are also aware that our KHorns have a (pick your adjective, here) woody, muddy, closed sound in the lower midrange. This seems to me to be the most common criticism of this speaker, from a sound perspective. One of the objectives leading to the 'Jube was the need to extend the upper end of the bass horn's response, of course. And, many of us amuse ourselves with different changes to the midrange horn: tractrix horn, Altec 311/511/811?, etc. I fall into this category as well.

I have spent some time messing around with the original CD horn family that EV (Keele's design) invented in the mid-70's, and I have a pair of the HR-9040A's (the great white whales) and a pair of the HR90's, the little brothers to the great whites. Of course, these need to be equalized, and I run a biamp setup with the horns equalized per EV's specs. I am using a well respected English Mosfet amp on the bass side, and feeding this directly to the Type A crossover, so it is rolling off per stock Khorn design. The midranges are being fed from a Tripath amp (rediculously good sounding little amps, these things), and they are crossed in using the usual L-R 24 db/octave network. The midrange drivers are EV DH1012H's. The tweeters: I am using the Eminence APT's on the top; I like them better than the EV T-35's I have, although they are a bit hot, and need to be padded down some. It may just be that the T-35's are a bit tired and could use new diaphrams. Right now (you won't believe this), these are coming off that Type-A, along with the bass horn.

Now here's the thing: putting the HR90's in play makes a substantial improvement in the midrange overall quality. The sound stage is wide open and very very stable. You'd expect that, after all, this design pretty well solved (and started the run toward CD designs in all their permutations) the beaming problems with straight shot exponential horns, and offers a very stable set of polar radiation patterns. The HR90's are rated to be used to 800 Hz, and EV states that they can be used to 500 Hz. I use them at 500 Hz. The drivers can go lower, but the horns won't load them much below 500 Hz. The sound is very very nice to my ear. Still have that wooden lower midrange, though, and you'd expect so, since this is the relative notch in the response between the bass section and the midrange (whether the standard exponential horn, or the HR90). This notch is probably worse with this setup, although it is hard to tell, when you A-B the HR90 against the stock 400 Hz exponential. The smooth, wide open sound the HR90 makes it very hard to go back to the stock 400 Hz horn in these systems. The HR90 can be built into a reasonable form factor with respect to the top unit of these speakers: it's not all that much taller than a stock Klipsch setup, considering that the HR90's are pretty good sized horns.

Putting the HR9040A's in play makes yet another improvement: these are rated for use to 400 Hz, and can be used as low as 350 Hz. The DH1012A, while rated to 400 Hz, is conservative: it'll take the full rated power at 400 Hz, but the driver responds nicely well below 400 Hz. Since I have NO plans to feed THAT much power to it in my living room (!), it is reasonable to drop its crossover frequency down to 350 Hz. Used in this way, the woody, muddy character of the sound simply goes away. It is full, powerful and clear. I cannot hear defects in the response through this range, at least whatever defects there may be, they pale in comparison to the standard setup. This arrangement is simply stunning to my ear. FWIW, I use the 1st Nora Jone's release to judge, because her voice is pretty husky in many songs, and the piano and bass give a good test through this range, too. It's very easy to hear the difference changing the midrange horns/drivers makes.

The HR9040A's are nearly 40" wide and about 17-18" high. To get them to fit reasonably to the bass section would mean that I would need to trim the mounting flanges back a bit (well, more than just a bit), just to get them to fit, width-wise. The HR9040A also can be made to fit within a top cabinet that mimics the standard system (e.g., flat front; as big as this horn is, it's not all that deep, really.), although you're now talking about a speaker that's going to end up about as tall as a Jensen Imperial or so. All this can be done, to be sure; but, you know, when I A-B right and left sides, one with the HR9040A and the other with the HR90, it's hard to find fault with the HR90, as long as you are not talking about the lower midrange. It's clear that the HR9040 wins, but, the HR90 just isn't all THAT much of a compromise. And, it's head and shoulders better than the stock midrange, again, to my ears, anyway.

So, my question to you all: is it possible to profitably equalize the bass horn's response to boost that lower midrange region? This'd be another good way to address the issue: I can easily fit the HR90 to this setup and it's not going to be a big deal. Pretty much the same dimensions, etc., etc. So, if a decent parametric EQ can bring this notch up to snuff, I'd contemplate using the HR90. It may also be possible to parametrically EQ the 60-100 Hz range, too, so that this response is a bit better (like the 'Jubes) here...I know, some of you will say, 'why don't you just get a set of 'jubes?', and you have a point. Thing is, the standard issue PWKs are pretty damned good...they just need a little rouge here and there to tart up the response, you know?

Again, can the bass horn be profitably equalized in the manner I've described, or is this not the way to go? Clearly, I have a solution with the HR9040A's, it's just going to run the ragged edge of domestic peace.

Thoughts?

[Edgar]

For goodness sake, please DON'T cut the HR9040As. They're becoming as scarce as Faberge eggs and I'd hate to see a pair destroyed.

I cannot comment on your other questions, except to say that the 9040A might look very strange mounted atop a KHorn or Jubilee bass cabinet, because the "lips" will overhang the front of the bass cabinet by quite a bit. That seems to be the hardest part of living with these horns -- aesthetically it is difficult to match them up with any kind of "box" cabinet.

Greg

[Edgar]

Another possibility. Keele's equations show a mass rolloff frequency of 177 Hz for the K33E woofer. If you can find an old Electro-Voice EVM15L (361 Hz), or a current DL15BFH (314 Hz), you might get a few extra dB above 175 Hz. Note that the current DL15ST (187 Hz) and DL15Y (268 Hz) are not as good as the DL15BFH in this respect. You might also have to adjust the back chamber volume.

There are undoubtedly woofers from other manufacturers that will provide similar performance. I'm just more familiar with the EV products.

Greg

[tigerwoodKhorns]

You mentioned the tractrix horn, why not just build one that goes down to 350 hz? They are relatively compact.

[colterphoto1]

Mr. Delgado has designed a KPT 904 horn that is tractrix, 2" throat, and is relatively shallow compared to the stadium horns. Go to Product Search and look up something like KPT 535

JBL 2365.pdf

[Horatio]

I am not eager to cut the 9040's down, either. Not surprised that they are getting rare: they are enormous! But, I also have to say, they sound really really nice. They do look gangly on top of the corner horns, though, spilling off the sides of the bass unit by about 3" each side, and bringing the whole stack to a height of just a few inches shy of 6'.

I bought them originally to do the top end on a pair of University Classic bass horns I built. They are stunning together, and the size fits perfectly, even if the pairing is visually formidable.

What I would like to do, is, simply EQ the bass horn a bit in the 300-400 Hz region, to boosting it to catch up with the HR90, which CAN be integrated easily into the cabinetry. What I don't know, is, how well does the bass horn respond to EQ'ing this range? Anyone done it?

I agree that a possible change in driver could help a bit, but, you know, I have listened to these horns with the original WS1504's that Speakerlab put in them, and I've also put a pair of EV15L's in them, and I must say, the K33E's sound WAY better than either of those. So, I am inclined to stay with the K33E's and try to EQ these into correspondence with the HR90s.

-Mike

[Edgar]

What I would like to do, is, simply EQ the bass horn a bit in the 300-400 Hz region, to boosting it to catch up with the HR90, which CAN be integrated easily into the cabinetry.  What I don't know, is, how well does the bass horn respond to EQ'ing this range?  Anyone done it? 

I have no experience with this.

Didn't Bruce Edgar mention that adding reflectors at the internal horn bends improved the bass cabinet's HF performance?

I agree that a possible change in driver could help a bit, but, you know, I have listened to these horns with the original WS1504's that Speakerlab put in them, and I've also put a pair of EV15L's in them, and I must say, the K33E's sound WAY better than either of those.

As they say, the proof of the pudding is in the eating. No arguments here.

Greg

[DrWho]

Just curious...do you have or are you interested at all in measuring your mods?

I
wonder if building yourself some false corners wouldn't help boost the
HF extension a little bit. Relying on a lossy seal with drywall isn't
exactly a recipe for maximized performance from the horn...

Also,
the Khorn LF is a bit squirrely in its midband performance (like around
200Hz). You'll note that the modern Klipsch balancing networks have EQ
integrated into them to notch this down quite a bit. By reducing the
level in this region, you're in effect comparitively boosting the
output at the higher frequencies in addition to the lows (which are
rolling off slowly).

Just keep in mind that as you widen the bandwidth of a driver, that
the distortion goes up exponentially. There is a reason PWK put the
K400 on the Khorn instead of many other alternatives...

[Horatio]

Yes, I've considered making the measurements...I've done a quick glance at the links in your last message and see you've been on the track, too. I'm kind of moving broadly in the direction of complete system equalization with respect to the in room response. Far from getting there, at this point, but, this is where I think I'm headed.

Agree on the false corners part: in my case, the corners aren't exactly perfect: I have a house with alot of windows, even though it was built in the 1920's, and the front windows are floor to near ceiling, multipane swing out types. The horns sit in corners that do not allow good sealing, as the sides are against the window borders, etc. The room I'm playing them in is pretty large, relatively speaking. I suspect that sealing these up a bit better would improve matters. Lots of compromise here, to be sure, and it undoubtedly influences the sound I get.

I did not know that the later models of the Khorn used networks to EQ the response a bit more- I lost track after Type AA.... I figured to simply boost and attenuate my way to smoother response viz, HR90 integration. Right now, I do it by ear, listening to lots of different source material. Yes, it's a bit of an exercise in relativism, but, there's a certain self consistency that eventually results.

I think I'll see if I can get hold of a decent parametric EQ unit and see what develops.

Appreciate the comment on the problems of opening up driver bandwidth- one of the surest ways to screw up the sound, especially if you're dealing with marginal drivers to begin with, is, push them outside their intended range. The converse is also true; you can take some pretty cheap drivers, and, provided you narrow the range of use, can make them sound pretty decent. I figure if quality pro sound compression drivers can handle, say, 40 watts continuous at 400 hz, and the response curves go lower, they oughta be pretty good on a couple of watts peak, a bit lower. Yep, the distortion goes up, but right now, what I am hearing in the lower midrange in terms of amplitude response is washing any increase in distortion, at least right now. Qualitative statement, to be sure!

Actually, the distortion question is a good one on another point: the EV CD horns, the early ones, used a hypex expansion in the throat, and so, one might expect more distortion as a result. The interesting qualitative consequence of this design is that the wide, stable radiation pattern and smooth frequency response seems to trump any distortion concern right away; compared to a straight shot exponential, these are really really nice to listen to.

Thanks,

Mike

[DrWho]

Yea, it's always about compromise. The part I find most interesting about this hobby is how people prioritize and deal with the compromises.

Btw, what are you using for amplification and your signal processing? Are you familiar with the Crown XTi? They're decent amps with a killer DSP section built into them. For the price of a quality analog parametric EQ, you could probably get 3 XTi's and move to a fully tri-amped configuration with time-alignment and very powerful signal processing. I have found that any tradeoffs made in the quality of the amplification are far offset by the ability to finer tune the speaker system.

If you're interested, you should look up bhendrix...he's managed to score a killer forum deal over at full compass.

[Horatio]

What I'm using: I'm kind of old school, in that I am using a Hafler preamp, which has two line outputs, one of which is driving BK (U.K.) Electronics MF100 Mosfet amps (these are mono modules, and I have six of them for triamp purposes, but only use a pair in this configuration, for the bass section) feeding the inputs of Type A crossovers, of which only the bass and the tweeter sections are active. The other output is driving a Rane AC23 crossover, of which I am using only the midrange, with CD equalization for the HR9040/HR90 horns. The upper cutoff is set to feather into the Type A's cross to the tweeters. The midrange output of the Rane is fed to a Sonic Impact Tripath amp (the obnoxiously cheap one that is still rediculously nice sounding), which is what, maybe 8 watts? This is a hackneyed setup to be sure, but it sounds pretty dogone nice, believe it or not. My intent is to crib something up that gets into the right neighborhood (proof of concept, really), and then, make a final pass with the right set of DSP gear to tie it down well. I know alot of guys don't like Behringer's 2496 series, but, it seems if one's careful to stablize the clocks and then to pay attention to the outputs to the amps, it oughta be reasonable to stay digital all the way to the point where the amps appear, and do with with reasonable quality. CD -> SRC -> DQX -> DCX -> amps.

I have been considering doing a complete PC-based setup, looking for a stable sound card (uh, yeah) that I can set up for use on my system (e.g., using the 5.1 or whatever outputs as independently programmed channels, so that I can set eq, delays, etc., etc. Flexibility is the driver- the basic hardware and compute horsepower's there- what's needed is better software). Not all that easy to do, still, with decent quality, it seems. Much further ahead to use outboard commercial DSP-based stuff, at least at this point. Have not looked into the Crown model you're speaking of. Will do.

[DrWho]

Going digital all the way through the Behringer is a great idea. There's a guy on the forum (sfogg) that has modified his Behringer to have 3 SPDIF outputs...basically by grabbing the digital bit stream before it dumps into the D/A's on the Behringer. He's then feeding this directly into a Panny receiver that has a digital to PWM conversion for the switching amps on the inside...I think he's reworking it to work for 3 channel bi-amping, but then that would mean it will work for 2 channel tri-amping too. Whether or not you consider PWM to be digital or not, this approach essentially keeps it digital all the way up to the speakers...

[Horatio]

A direct feed of SPDIF to the PWM inputs is close to digital nirvana. I'll have to look into that-thanks!

Checked out the Crown amps- pretty slick product. I think these speakers are woefully overpowered by an amp of that size, though, don't you? I had that worry with the MF100 100 watt amps I'm using; can't imagine what something larger'd do on these speakers.

Thanks for the heads-up on this stuff-

Mike

[Deang]

"I have found that any tradeoffs made in the quality of the amplification are far offset by the ability to finer tune the speaker system."

Yup.

[CECAA850]

Dean, you have PM.