Chris A Posted August 27, 2016 Author Share Posted August 27, 2016 "...within 1/4 wavelength at the crossover frequencies...neglecting phase lag due to crossover filters..." As you know, those off-axis ports have to be within 1/4 wavelength or you can't cross over high enough with the lower frequency drivers--there's an acoustic low-pass filter that's quite strong that comes into play. Chris Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted August 27, 2016 Klipsch Employees Share Posted August 27, 2016 4 hours ago, Chris A said: One of the interesting things about the K-402-MEH is that I could "cheat" on the phase compensation and push the group delay back down a bit by delaying the HF channel a little bit, without any apparent issues due to polar lobing (the crossover center frequency is 475 Hz): and the corresponding SPL (on-axis), excess phase (white trace), and total phase (bottom red trace) plot: not sure but from the graph, it looks like the phase could have been "flattened" out more if you had add a bit more delay to lf. also is it possible to open up the spl scale in order to get more detail on the freq response? Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted August 27, 2016 Klipsch Employees Share Posted August 27, 2016 2 minutes ago, Chris A said: "...within 1/4 wavelength at the crossover frequencies...neglecting phase lag due to crossover filters..." i see....etcs are a fine thing. Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted August 27, 2016 Klipsch Employees Share Posted August 27, 2016 14 minutes ago, Chris A said: "...within 1/4 wavelength at the crossover frequencies...neglecting phase lag due to crossover filters..." As you know, if those off-axis ports have to be within 1/4 wavelength or you can't crossover over high enough with the lower frequency drivers--there's an acoustic low-pass filter that's quite strong that comes into play. Chris yes but those ports become the "radiator" and whatever wavelength they are, their apparent acoustic center takes on the "apperance' of the acoustic source and of course depends on the xover freq in question. there are some things about this style of design that "shares" a horn and the ensuing compression of a paper cone woofer and soft cloth surround..... that acoustic filter is because this is essentially a bandpass that is somewhat horn loaded..... Quote Link to comment Share on other sites More sharing options...
Chris A Posted August 27, 2016 Author Share Posted August 27, 2016 How about this... Note that the dip in response at the crossover frequency was due to my choice of too high of a crossover frequency and where I put the off-axis woofer ports in this first prototype. I'll elongate them slightly toward the throat (and the mouth--both directions) when I start up again in a few weeks. It's nice to have A/C in the shop...isn't it? I know that...because I don't currently have it... Chris Quote Link to comment Share on other sites More sharing options...
Klipsch Employees Chief bonehead Posted August 27, 2016 Klipsch Employees Share Posted August 27, 2016 1 minute ago, Chris A said: How about this... Note that the dip in response at the crossover frequency was due to my choice of too high of a crossover frequency and where I put the off-axis woofer ports in this first prototype. I'll elongate them slightly toward the throat (and the mouth--both directions) when I start up again in a few weeks. It's nice to have A/C in the shop...isn't it? I know that...because I don't currently have it... Chris better....and my first question in any shop is....what is the air conditioner thermostat set at... 2 Quote Link to comment Share on other sites More sharing options...
Chris A Posted August 27, 2016 Author Share Posted August 27, 2016 Note that most of the 300-500 Hz dip in response was due to my poor choice of crossing filters and their frequencies, PEQs, and shelving filters...and not really the position of the ports. I was kind of in a hurry when I ran those settings, and I haven't run a sweep since updating the Dx38 settings in that area. I expect to see that gap closed currently...but I need a round tuit first to run another sweep. It sounds pretty good right now. When do I get to see your "secret vault"? I've shown you mine... Quote Link to comment Share on other sites More sharing options...
Chris A Posted August 27, 2016 Author Share Posted August 27, 2016 Hello, hello...where'd the party go? Chris Quote Link to comment Share on other sites More sharing options...
Chris A Posted August 28, 2016 Author Share Posted August 28, 2016 I found that I posted the wrong measurement, above. Here is measurement of SPL/phase after the last iteration on EQ. The dip at 170 Hz was a 1/4 wave boundary bounce cancellation: and the excess group delay plot: Quote Link to comment Share on other sites More sharing options...
Chris A Posted August 28, 2016 Author Share Posted August 28, 2016 Just to give a relative feel for the total phase growth for the K-402-MEH with K-69-A driver versus the Jubilee with TAD TD-4002 driver...the K-402-MEH is the orange trace, while the Jubilee is the purple trace: Chris Quote Link to comment Share on other sites More sharing options...
sunnysal Posted August 28, 2016 Share Posted August 28, 2016 I have posted before about my DEQX based system and how I ended up going back to a analog crossover system. From the scientific standpoint I would never claim it is not possible to improve the performance of a system via these types of manipulation in the digital domain. However, in my case, the added complexity didn't add up to enough value to keep all that equipment in my system. I often mention the difference i discovered between system "performance" and music "reproduction". For me I had better results, for music listening, with my dumbed down system than I did with all the bells and whistles connected. I still wonder about implementing a steeper analog crossover but I have been cured of chasing an improvement in musical enjoyment via digital manipulation in my system. To each his own, your mileage may vary, etc., etc.. Warm regards, from sunny El Salvador, Tony 1 Quote Link to comment Share on other sites More sharing options...
Chris A Posted August 28, 2016 Author Share Posted August 28, 2016 I believe that we've talked about the digital crossover that you used might have been a factor in you not staying with your tri-amp setup. I've found that there is a definite threshold in digital crossover quality, i.e., I presently don't recommend the Behringer DCX2496 or entry-level DEQX model. The miniDSP units may or may not be good choices, but their models been identified as being susceptible to USB noise from poorly implemented USB power supplies from computers. It remains to be found whether or not miniDSP can be implemented such that it doesn't create sound quality issues. In all instances where there are sound quality issues with digital crossovers, it's been my experience that all the problems actually can be tied to the analog electronics used...or the analog architecture used (in the case of the Behringer). The newest digital crossovers that are coming onto the market all seem to be addressing the issues of the first generation of units including the above-mentioned cases, especially a move to 24/96 high fidelity processing. The only company that I see that seems to not be responding to customer complaints of sound quality and insufficient power supplies seems to be Behringer--which has for years been the "cost leader" in PA--and live music hardware for starving musicians. Chris Quote Link to comment Share on other sites More sharing options...
sunnysal Posted August 28, 2016 Share Posted August 28, 2016 When I was looking at digital crossovers and room correction, DEQX was the undisputed technology leader, I bought the top of the line at the time and looking at current secs for the HDP 4 and 5 it is hard to see how the systems you are referring to could be any better. Behringer is another level all together from the DEQX products, perhaps you are not familiar with them or their products? As I said, I don't doubt those who enjoy thier systems but for me, with the best stuff available at the time, it didn't float my boat. Regards, T Quote Link to comment Share on other sites More sharing options...
Chris A Posted August 28, 2016 Author Share Posted August 28, 2016 On a side note, something that Greg Oshiro mentioned to me candidly was that he could hear the limitations of Khorn midrange and tweeter horns/drivers once the obfuscating and masking effects of time misalignment, first-order crossover filters, and loudspeaker-room equalization issues were corrected. That's also true in my experience with tri-amping a Belle. Even though the timbre of the loudspeaker was corrected so that it integrated between the Jubilees, it revealed its limitations in tweeter and midrange horn performance. So some of the limitations of tri-amping for sound quality is wrapped up in the performance of the Heritage loudspeaker horns/drivers themselves. I've had other forum members state this same observation to me. So I upgraded my Belle to a K-510/K-69-A-based JuBelle configuration and those issues almost completely disappeared. Indeed, many would class JuBelle performance today as state-of-the-art design. I can definitely tell you that the Klipsch Jubilee is not limited by its horn design, nor some DIY improvements to other Heritage upgrades, such as the JubScala or JuBelle and perhaps one or two other third-party designs reusing basic Heritage designs. I can also think of a few other yet-unrealized loudspeaker configurations that also would not be limited by horn/driver designs like present stock Klipsch Heritage configurations. Chris Quote Link to comment Share on other sites More sharing options...
Deang Posted August 28, 2016 Share Posted August 28, 2016 "One of the conversations in another thread discussing AA networks, I commented on the observation that if you carefully re-equalize your loudspeakers ... that you can't tell the difference in sound from before the parts change." - specifically capacitor types. I don't see anything in this thread that validates the above statement, which I thought was the point of the new thread. Since capacitors with very similar ESR numbers sound completely different from one another - I'm rejecting the premise. Compare a Type A built with Sonicap (a very good measuring capacitor) to the same network built with Auricap, and then again with AudioCap Theta. While that network only has two capacitors - the sound changes. Quote Link to comment Share on other sites More sharing options...
Deang Posted August 28, 2016 Share Posted August 28, 2016 If you don't like the "obfuscating" effects and limitations of the horn/drivers, you can buy a nice Tractrix horn from Dave Harris and some de120s from Crites. Quote Link to comment Share on other sites More sharing options...
Deang Posted August 28, 2016 Share Posted August 28, 2016 The output of the driver/horn falls like the proverbial rock at 5.8kHz, and when combined with the first order electrical filter - falls even faster - the acoustic behavior is not first order. Quote Link to comment Share on other sites More sharing options...
Chris A Posted August 28, 2016 Author Share Posted August 28, 2016 Dean, I can respect that you disagree but I can't respect the basis of your argument which is apparently "argument from ignorance" (argumentum ad ignorantiam) . I prefer other and better non-fallacious methods in order to discuss or argue productively. Otherwise it's a waste of both of our time. Because of that observation, your arguments currently appear to me to have purely inflammatory intent. If you're willing to do some controlled experiments using something in addition to your assertion of "just knowing", or having "golden ears", then I'd probably respond more directly to your premise--i.e., capacitor materials have a certain "signature sound" independent of equalization and time/phase compensation. I've done those experiments and I have strong evidence that significantly counters what you are saying. What I've found are, for me, facts. I don't argue facts. I've found that's a waste of time. I believe that Roy also mentioned some reasons to question the conclusions you put forward based on other factors. I agree with him based on your answers to my follow-up questions in the other thread. If there are other personal issues that I should be aware of then I believe that you should convey those, but privately, please. Life can be hard and I certainly understand that there are many things that get the better of us at times. Sincerely, Chris 1 Quote Link to comment Share on other sites More sharing options...
Deang Posted August 28, 2016 Share Posted August 28, 2016 You were the one who made the claim, so why is the burden on me? Please show how two passives built with different parts can be equalized to sound the same. Personal issues? I'm not building right now because I'm in pain, that doesn't mean I'm not lucid. I have no issues with you. Quote Link to comment Share on other sites More sharing options...
Chris A Posted August 28, 2016 Author Share Posted August 28, 2016 I believe that you missed it already. The early testing of the Belle was with passive crossover and old caps, carefully EQed. After the caps were replaced, I carefully re-EQed and found the sound was indistinguishable. Then I removed the passive and direct coupled the drivers to the Dx38, setting the EQ to match the passive. No difference in sound. Then I corrected the time/phase misalignments--big difference in sound and in measured results. Chris Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.