Chris A

The following is taken from this source (Bob Katz): https://www.digido.com/portfolio-item/level-practices-part-2/ Farther down the article, it talks about using a "-6 dB" reference for home CDs, etc. as the zero dB monitor reference point (i.e., 77 dB). The article is a good discussion of the entire subject of "how loud" and dynamic range of recordings. While the RIAA-affiliated mixing and mastering shops still have trouble pushing back against the record companies and their "Loudness War" practices, largely, the MPAA (movie) industry practices have set reasonable levels and unofficial standards for mixing and mastering levels. Chris

If you are looking for general guidance on DSP crossovers, in the past I typically looked at the ElectroVoice documentation on the Dx38 or the DC-One for better explanations. The advice in those user manuals doesn't mirror exactly what Xilica provides, but it is useful for those unfamiliar with DSP crossovers. Other than that, you can look at the miniDSP forum and YouTube videos. Chri

Good luck. Chris

Guys, this is all very considerate and all of the OP, but like a lot of online threads, has apparently taken on a life of its own. See: https://blog-fgci.com/2016/08/30/8-solvents-for-cleaning-up-your-next-job/ I've done a fair amount of fiberglassing (boats, acoustic horns, body work, etc.) over the past 50 years and have found acetone to be both useful and effective. I use it just about every time I do resin work. But protection from breathing the fumes is the most important point. Skin contact is way down the list of considerations in my experience. What really happens is that it takes away all the oils in your skin, causing skin cracking and dryness. The solution is to use gloves that don't dissolve in acetone. You can use nitrile or latex gloves, but butyl rubber is the most resistant to acetone. You can typically find these wherever you buy acetone. Chris

The link is posted, above. Chris

That's solder that's been smeared on those nuts. You can see the damage to the plastic standoff on the black plastic piece in the foreground that was caused by someone indiscriminately daubing solder on the terminals and instead melting the plastic with the tip of the soldering iron. You were cutting through solder with your hacksaw--not corrosion. Solder has flux in it (typically) to self-clean the surfaces that are being soldered, and that looks like dark amber or even black when it's been daubed on like that. However much you'd like to blame Klipsch for this, I'd say you need to re-evaluate what has occurred to these crossover networks (the portion that you show here). The seal around the bass bin access door will re-flow slowly over time, so even though you think it wasn't accessed before, if what I'm looking at above is from behind the access door on the bass bin, it's been opened before. Someone probably decided that they didn't ever want to get inside to torque down the connections, and gummed it up with solder, instead. That's a particularly bad job, too since those connections were not designed to be soldered, and Klipsch didn't do that...I can tell you without reservation. Is there a reason why you want to blame Klipsch for this? Chris

I don't believe that they looked like that when they left the factory. Looks like you need a fairly large soldering iron and a solder sucker... Those look like these: Once you can remove most of the solder that was placed on those screw-down terminal blocks, I think you're going to be able to take things apart and remove the leads when you wish to. Chris

Thanks Michael. I didn't receive the email that you did, above, so I was unaware of this communication. I just sent a new email to the company to remind them of my failed AMT-1 driver diaphragm from 29 January 2020, one driver out of the four AMT-1 drivers that I had bought for this purpose--that remains inoperable after its first few minutes of initial operation at 100 dBSPL (1 m on-axis). I was assured via telephone conversation with Rick Caudillo at that time that ESS would replace the failed diaphragm at their cost. I haven't been able to complete my double-stack AMT (stereo) setup to report my own findings here because of this situation, and this project is currently unfinished, awaiting action from ESS to replace the diaphragm. "Hope springs eternal." Chris

It's been a while since I posted an EQ curve in this thread I started in 2015. I just managed to generate a useful demastering EQ curve for the following album (one size fits all for this album): I've enclosed an Audacity demastering EQ curve in XML format here which can be imported under the "Equalize..." command: The Black Knight and Bavarian Highlands Scenes.XML A screen shot: I've owned this disc for 10 years now and I just got my first chance to hear the embedded quality of this recording without the mastering EQ curve that the disc came with--which until today rendered it basically unlistenable (to my ears). Now, I can hear all the voices without that overwhelming midrange/treble sound, the resonance around 100 Hz, and now being able to hear the double basses in balance. It's a pretty good recording that's now enjoyable to hear. If you own this album, give it a try in Audacity using the "Equalize..." command. Chris

Acetone has 1/3 the viscosity of water at room temperature, therefore the Reynolds numbers for similar impingement velocities are 3x higher than water. Open all the windows first, then use a respirator to avoid breathing acetone vapors. (It's the Reynolds numbers attained on the surfaces you're cleaning which do the cleaning action...and acetone is used a lot in fiberglass work to clean stray resin off working surfaces when laying up.) You can also use compressed air for everything that doesn't require wetting action to clean. I'd use compressed air first to knock off anything that's not adhered to the fiberglass surfaces, then acetone for the remainder. Chris

The passive crossovers have the equivalent of 180 degrees of delay in them (higher frequency leads lower frequency), which is not desirable delay: ~1.25 milliseconds at the woofer/midrange crossover (180 degrees at 400 Hz), and ~100 microseconds at the midrange/tweeter crossover (180 degrees at 4.9 kHz). But that's what you get with passive crossovers: The real time alignment issues are mismatches that I quoted earlier/in another thread: 4.8 ms for the bass bin to midrange, and 6.5 ms from the bass bin to the tweeter--mostly due the different lengths of horns/drivers (i.e., an additional 0.7 ms from the midrange to tweeter--about 3.5 wavelengths). That's where you need the DSP crossover to delay the higher frequency channels to time align them to the lower frequency channels (i.e., tri-amping using DSP). This unit has delays available. I use one of these 2x4 HDs for my surround AMT-1/Belle bass bins. If you use REW and a calibrated microphone, you can use the 2x4 HD to EQ out all SPL irregularities. I don't use "room correction software" anymore--even Dirac has issues. I use REW and the DSP crossovers, instead. I linked two other threads at the first post of the thread that help you decide where to put the crossover frequencies from running raw (uncorrected) measurements and overlaying them on each other to see where they naturally cross. This is the best method of setting crossover frequencies, I've found. For a Khorn, those frequencies generally are 400 Hz and 4.9 kHz, but they can vary if you've got in-room acoustic measurements and a DSP crossover. Generally, I recommend first or second order crossover filters. First order filters produce half the (undesirable) phase shifts as second order. Time aligning and EQing the loudspeaker flat using REW measurements really do sound better--even with Khorns. The use of a DSP crossover and multi-amping also opens up the possibility of changing out horns and drivers--something that you really can't do using passive crossovers...unless you redesign the passives completely, which is a difficult task even for experienced loudspeaker engineers. That's the miniDSP 4x10 HD that I referenced in the leading post of this thread. Yes. In the case of certain miniDSP crossovers (the two mentioned here), the quality is good. Xilicas are a little better, but that's also reflected in the price differential. Personal opinion--with experience attached... 😉 But it's your money and time. Chris

I've used a couple of different AVPs--which of course are mainly multichannel DACs with many available codecs and I/O options on the audio portion of the device. The difference that I heard was at the highest frequencies--the "air" and very high frequency clarity of the two. However, if you didn't tell me which AVP I was listening to, I'd be hard pressed to hear any differences between the two. (I don't have use of two-channel-only DACs.) Of course, I pipe everything through downstream DSP crossovers on the output of the AVP, so in theory, I shouldn't hear much difference (if any) at all. But I did.. a very small amount, however. I'd say that it's a bigger difference than one would initially guess. What are you using now (brand, model)? All DSP crossovers I've seen offer adjustable delay--that, and plenty of bi-quads (PEQ filters) per channel. That appears to be an analog active--and it can't offer delay--only all-pass phase delay (less than 360 degrees of phase--which is a tiny amount of delay). I don't recommend using analog active crossovers for horn-loaded loudspeakers (hybrid horn/direct radiator or fully horn loaded), since basically all multi-way/multi-horn loudspeakers need EQ and delay. This is really a DSP crossover FAQ. So you want the cheapest DSP crossover that you can find to drive Khorns? Good luck... I've identified useful DSP crossovers for high efficiency loudspeakers in this thread. I recommend looking at that list and saving your pennies until you can afford one that's useful for your application. I don't recommend cheap DSP crossovers: they will be nothing but headaches. Chris

Did you ever use the PEQ settings that I posted? Note that there is ~22 dB of overall EQ that's required: Chris

Can you forward your Xilica .xdat preset file? Chris

If you've got powered subwoofers, I recommend XLR or Euro/Phoenix connectors from the Xilica to the subwoofers. If not that, then speakON (if the sockets are available on the loudspeaker/subwoofer). If not that, then RCA or bare wire. Chris

I avoid RCA whenever and wherever possible. Neutrik is a manufacturer, and are noted for their XLR and their own unique speakON connectors. The speakON connector is most often used for connection to non-powered loudspeakers: For powered (signal-level) connections, XLR is typically the best to use, since they reject common mode noise (i.e., power-related noise): This sounds like you've got passive subwoofers--not powered. To answer your question, you can go internally and bypass any passive low-pass or high pass filters, using the filters within the Xilica to limit the signal to your separate subwoofer amplifier. Generally I wouldn't bypass an active (i.e., on-board amplifier) subwoofer's circuitry, unless you've got issues with those electronics (i.e., failures). I would simply route the signal-level connection from the Xilica to the active subwoofer--using XLR-->XLR connections, if possible. Chris

I used the prototype without a box, but the effect wasn't very pleasing, in my experience. Bass response actually got more spotty (polars got ugly) and generally weaker. The K-402-MEH really depends on at least half space loading (i.e., the back of the unit is touching a wall), so I wouldn't want to give away the bass extension to space it away from its backstop wall. Chris

You could use the K55 midrange driver, but then you'd have to use a separate tweeter--like the K-77 or other tweeter, since the K-55 dies at about 6 kHz. If you instead use the K-510-clone and a full range 2" compression driver (just like the Jubilee design), then you're converting to two way and eliminating the separate tweeter with all the attendant issues of time alignment and acoustic polar lobing of having the tweeter and midrange mouths more than 1/4 wavelength from each other at the crossover frequency (nominally 4.9 kHz). If you really got to have a three-way design, then use the BMS 4592ND, which has two diaphragms internally, and you tri-amp the bass bin, and the two internal diaphragms of the 4592ND. I'd got the PEQs posted on using the 4592ND on a K-402 horn posted on this forum. I recommend the Celestion Axi2050 or Radian 950BePB if going two-way. This eliminates the lobing problems and all the harshness associated with the stock Khorn design. The K-510-clone horn eliminates the loss of polar control problem until you get down to ~500-600 Hz, which is a big elimination of the issues requiring placing the absorption material around the top and front of the loudspeaker. Yes. This is the same technology that's used in the K-402 horn that's the top horn on the Jubilee. It's a straight-sided horn with a tractrix mouth roll-out. I also use the K-402-MEH by itself in the K-402-MEH design. Chris

Someone that knows the AK3 wiring should chime in here. The old PWK-era style of wiring had the two leads from the woofer coming directly out of the bass bin cabinet--and those are what you're looking to attach to. Those older crossovers allow the bypassing of the bass bin choke (coil) directly at the connector strip in the top hat. Chris

Yes--to get the electrical resistance and reactance of the passive crossover networks out of the circuit (which causes your amplifiers to work that much harder, and introduce additional phase shifts). The best situation is direct-coupling the amplifier output to drivers. Chris

Then add the 4.8 ms delay to the midrange/tweeter channel. Not in the least. PWK showed this conclusively 50+ years ago when he showed the effect of adding Dynamat-like damping material to the outside of the K-400 horn--which did almost nothing at all. However, when the midrange horn mouth is clamped to the front baffle, all the ringing ceased (just like you'd get if you clamped the rim of a bell and tried to ring it). If you really want to avoid the problems of midrange horn polar control loss and harsh K-77 tweeter sound, I'd recommend using a K-510 clone horn and a good "full range" 2 inch compression driver (e.g., FaitalPRO HF200, HF20AT or HF2000, BMS 4592ND dual diaphragm, Radian 950PB or 950BePB, or Celestion Axi2050, etc.) crossing over at 450 Hz to the bass bin using a DSP crossover. The problems with harshness, time alignment and all other related acoustics issues will disappear (after you dial them in with a DSP crossover and some measurement app like REW). If you can find K-402 horns, it will sound even better. I'd recommend the Celestion Axi2050 or BMS 4592ND for the K-402 horn. Chris

If you're using a DSP crossover (i.e., not an active analog crossover), try inserting ~6.5 ms of delay on the tweeter channels, and ~4.8 ms delay on the midrange channels (assuming tri-amping). Then you'll correct the timbre shifts due to time misalignments of the horn-drivers, which are part of your 'bright and harsh' sound. Alternatively, if you release the tweeter from its position inside the top hat and bring it out to rest on top of the top hat, you can also time align it to the midrange by placing it at the back of the top hat--with the joint of the tweeter/horn approximately right over the top of the midrange driver/horn joint--with the tweeter...and perhaps a little farther back from the front of the loudspeaker to correct the phase shift due to the electrical crossover network itself. When you get the tweeter aligned with the midrange (usually this is within ±1/4 inch from exact time alignment), you'll hear the soundstage of the loudspeakers open up wide and produce a much more realistic feeling of the original recording space (acoustic-only instrumentation - like an orchestra or solo acoustic guitar within a reverberant venue are the best recordings for hearing this--sometimes really good vocals that were not multitrack recorded and stacked together on a mixing board). Additionally, if you put at least 1" thick absorption material on top of the loudspeaker, you'll absorb a fair amount of the re-radiating midrange energy between 400--2000 Hz, where the midrange horn is spilling its vertical polar energy into your floor, ceiling, and on the top of the loudspeaker. If you let the absorption material hang out a little over the top of the top hat, it will be more effective at controlling ceiling bounce. If you place some of the same thickness of absorbent material on the front face of the bass bin, it will also catch more of that re-radiating midrange energy that adds to your bright/harsh sound. Chris

Outside of North America there apparently is a problem with availability of K-402 horns, as has been reported by all that have tried to obtain them. For those that have found a way to get them to their country, the import duties tend to be ferocious (like doubling the cost, etc.). It's very unfortunate. Chris

"Not much"...is the answer. Throat reactance annulling isn't really an issue with the K-402-MEH design using DSP crossovers, I've found. I have done half volume and double volume back chamber trades within Hornresp and found so little difference in power response at low frequencies that I've ignored the effects of the acoustic reactance at the lowest frequencies. Chris