Since you're not using the same horn, driver, and off-axis port locations, those bass bin response peaks will be different for you. I recommend a calibrated microphone and REW to measure them. You're going to need it anyway to dial in the MEH assembly.
"Qualitative" means the opposite of quantitative. It means that you can see areas on sim plots where the frequency response will differ, but it will not be measurable differences, only visual indications that the response will differ and approximately where those frequencies will be.
Chris
"Qualitative" means the opposite of quantitative. It means that you can see areas on sim plots where the frequency response will differ, but it will not be measurable differences, only visual indications that the response will differ and approximately where those frequencies will be.
Chris
Seems misleading to me too.
Never heard anyone else rename the ports, 'off-axis ports'.
What's the rationale?
Confused about notch filters
Would someone please explain why we want to notch out some frequencies? I understand the basic principle of the Unity/Synergy mids and the (I supposed) need to crossover to "tweeter" before the first notch (1/2 wave reflection). If this (and multiples) are already minima, what is the point of a notch filter? What am I missing in this concept? Is someone trying to bend the "rules" (physics!) and get more HF output of of the mids???
Would someone please explain why we want to notch out some frequencies? I understand the basic principle of the Unity/Synergy mids and the (I supposed) need to crossover to "tweeter" before the first notch (1/2 wave reflection). If this (and multiples) are already minima, what is the point of a notch filter? What am I missing in this concept? Is someone trying to bend the "rules" (physics!) and get more HF output of of the mids???
When you look at the raw response of the woofers on both the SH-50 and the K-402-MEH, you can see response peaks above the first notch frequency:
The idea is to provide a steep crossover slope without the accompanying phase shifts inherent in typical IIR crossover filters (i.e., Butterworth, Bessel, Linkwitz-Riley, Chebyshev, etc.). All the woofers need is a 90 degree phase lag at their ~500 Hz crossover frequency to match the phase of the next higher driver (the midranges in the SH-50 and the compression driver in the K-402-MEH). All you need to do with either woofer channel is to attenuate the peaks above the first notch frequency, i.e., 1320 Hz and 2600 Hz in the K-402-MEH. Once you do that, the resulting low-pass crossover slope of the woofer is about 10 dB/octave--by itself without added electrical crossover filters.
Once you attenuate those response peaks above first notch frequency, then you get almost flat phase response out of the entire loudspeaker, as seen in the phase plot that I posted earlier:
The "secret sauce" of the Danley Synergy is this almost-flat phase response across the entire audible bands without the use of FIR filtering. Once you hear the difference from a full-range MEH that can control its polars down to below 200 Hz (SH-50) or 100 Hz (K-402-MEH), you'll understand why the Synergy series horns are held in such high regard.
Chris
Last edited:
When you look at the raw response of the woofers on both the SH-50 and the K-402-MEH, you can see response peaks above the first notch frequency:
Chris
Thanks for clearing that up. Really nice graph. I can also quite clearly see you 475hz "acoustical crossover" in action. Is this how you get there? Model the horn till you see the curve becoming something like this? Starting from the 1/4th WL distace, and seeing where it starts dropping of, as to change the off axis distance till it matches is?
Awesome, I'll play with it some and send you the graph for you to confirm. Also here my message I sent you in mail. Maybe more relevant here:
"I love playing with the software, but it’s good to know it’s really accurate for predictions like the one I’m trying to make. How did you figure you needed 2x 15” instead of 2x 12”? I also noticed there is almost 0 horn loading visible in hornresp. As you already told me not to look to hard at it, I didn’t. But still.. I know from you it cant be 0 hornloading right?
I will definitely start with the Em Kappa 15C I have, so modelling is not that important for now, but for later it maybe is. In the end I want to match the BMS 4594 down to at least 100hz. What would be the best way to go about this? I’m figuring I need 130dB for 1 woofer (including the hornloading) to get there (142dB) with 4 woofers, or 127dB with 6 woofers (x6 12” maybe?).
Could I just look at their performance in a similary sized sealed box to guess their performance in the MEH? "
"I love playing with the software, but it’s good to know it’s really accurate for predictions like the one I’m trying to make. How did you figure you needed 2x 15” instead of 2x 12”? I also noticed there is almost 0 horn loading visible in hornresp. As you already told me not to look to hard at it, I didn’t. But still.. I know from you it cant be 0 hornloading right?
I will definitely start with the Em Kappa 15C I have, so modelling is not that important for now, but for later it maybe is. In the end I want to match the BMS 4594 down to at least 100hz. What would be the best way to go about this? I’m figuring I need 130dB for 1 woofer (including the hornloading) to get there (142dB) with 4 woofers, or 127dB with 6 woofers (x6 12” maybe?).
Could I just look at their performance in a similary sized sealed box to guess their performance in the MEH? "
Hmmm....i believe that is a misleading and inaccurate conclusion.
Whether FIR filtering is used or not is immaterial in evaluating the Danley Synergy crossover design.
It is a passive IIR x-over design, which means it can be fully replicated with traditional IIR active processing....with no FIR is sight...
The use of passive x-overs is much more about business strategy than anything else, imo.
FIR could/can be used to improve the sonic performance.
I've had the pleasure of conversations with Tom, Mike, Ivan, and a number of other fine Danley folks this week at infocomm.
I asked opinions about the use of active FIR processing in place of the passives. I learned that some HT enthusiasts have done that, and were very pleased with the results.
There's no 'secret sauce' in the crossovers imo, just a lot of great balanced innovative engineering.
I tried to express a huge appreciation on behalf of our DIY community to those guys for their continued willingness to share information (that doesn't jeopardize their business interest. Really cool, down to earth folks!!!!
If you're willing to do what it takes to do FIR filtering, you will experience even lower phase response wander. What I've found, however, is that FIR filtering is typically not needed if you stick to the precepts that Tom Danley put out there: stay within 90 degrees of flat phase and you're going to get a lot more than 90% of the benefit in terms of subjective listening performance.
To the OP: there is a secret sauce, and I've heard it--in spades. You can also talk to user: Delicious2 and user: MisterVego on the K-forum about their recent crossover filter upgrades using K-402-sized horns that control their polars down to ~100 Hz. In my experience, it's a quantum jump in subjective listening performance while using full-range directivity controlled (fully horn-loaded) loudspeakers.
I believe that the issue comes when you lose polar directivity control in-room well above ~200 Hz (as I believe the quoted individual has exclusively experienced, but hasn't mentioned this little tidbit yet), then you lose a great deal of the effects of phase flattening across the board, like Danley gets with its passive crossovers and full-range directivity loudspeakers. I've heard it in the K-402-MEH and the Jubilees, vs. using typical "named" crossover filters that introduce phase distortions, and I can tell you that the effect is dramatic, but difficult to verbalize exactly how it sounds different (see https://community.klipsch.com/index.php?/topic/182419-subconscious-auditory-effects-of-quasi-linear-phase-loudspeakers/&do=findComment&comment=2379562...about 3/4 way down the page where it starts with "Well, I learned a lot over the past two days...").
Chris
Last edited:
Synergy crossovers do come together quite readily. It's the nature of the beast, nothing untoward or 'special' going on. The notch doesn't prevent phase changes, and passive is suitable.
If by special, you mean it sounds good, OK but this is not unique to synergies. There are other ways to account for the regions mentioned.
If by special, you mean it sounds good, OK but this is not unique to synergies. There are other ways to account for the regions mentioned.
I think this is another case of tying variables together in a way that doesn't flow logically.....and further misdirection....
Polar directivity in a Synergy is a function of the horn size and geometry, waay more than any particular crossover design? That's why there are different size Synergy's, eh?
Crossover design in a Synergy is more about getting the acoustic sources to fully work together, than any enhancement in pattern control.
That's a big part of really understanding the Synergy concept imo...Synergy an acoustical concept more than anything else.
Crossovers are just the always necessary component in any multi-band design.
Measurements and crossover designs for speakers like Synergy's are undoubtedly made in anechoic or near anechoic environments.
What does the loss of indoor directivity well above 200Hz have to do with crossovers (or anything really)?
Besides, the best way to evaluate directivity (and crossover frequency/phase response) is outdoors or as reflection free as possible.
Meyer had, maybe still has, a series of video tutorials on crossover design that included charts for the major xover types, that showed response and phase by order and polarity.
The tutorials went on to show how a UPA-1P (circa late 1990's) was optimally tuned using asymmetric, different type, different order, different polarity, different x-over freq, 'named' crossovers.
That kind of methodology, that Danley is now doing passively has been widely known for a long time.
Crossover design certainly does effect the polars of typical two ways or mutiways with acoustic centers that are separated perpendicularly to the listener.
But again the whole point of a Synergy is to load the acoustic centers into a common source, and rid that x-over effect on polars.
Again, no secret sauce as has been claimed, imo.
Passive totally works, active IIR probably works even better, and FIR probably works best....
If there is a secret sauce , I would call it an excellent design with which ever path is taken...
Maarten (the OP), I think I'll again restrict my participation to the K-forum and email directly as we've used over the past couple of years.
To others still interested in this subject: if you want to talk more about the subject of full-range MEH design/test or Danley-style crossovers, I recommend the K-forum (where there's actual moderation). My handle back there is "Chris A". I find it a very interesting subject that I think will continue to gain considerable momentum over time.
The data and information that I've presented here really represents the tip of the iceberg from that which I've learned benchmarking with the SH-50, the K-402-MEH work, and Danley-style crossover filter work applied to Jubilees, MEHs, and other loudspeaker configurations. It's made a world of difference in my enjoyment/involvement level of late. This is something that's easy to share (albeit elsewhere).
To the two guys having particular issues (i.e., not the OP): good luck.
Regards and best wishes,
Chris
To others still interested in this subject: if you want to talk more about the subject of full-range MEH design/test or Danley-style crossovers, I recommend the K-forum (where there's actual moderation). My handle back there is "Chris A". I find it a very interesting subject that I think will continue to gain considerable momentum over time.
The data and information that I've presented here really represents the tip of the iceberg from that which I've learned benchmarking with the SH-50, the K-402-MEH work, and Danley-style crossover filter work applied to Jubilees, MEHs, and other loudspeaker configurations. It's made a world of difference in my enjoyment/involvement level of late. This is something that's easy to share (albeit elsewhere).
To the two guys having particular issues (i.e., not the OP): good luck.
Regards and best wishes,
Chris
I would say one of if not the best advantage of designing with synergy/unity is that the crossover can be done without concern with off-axis interference lobes from the drivers. Everything is within a quarter wavelength and the waveguide sets the directivity. Basically behaves like a single driver horn but with the benefits that come with dividing the frequency range into drivers that can handle them best. So, get it right on-axis and use a good waveguide (set up so the lower freq ports don't mess up higher frequencies though) and the off-axis takes care of itself.
I share that assessment..... thinking you mean that the crossover on a synergy/unity is put in place without concern for off-axis response.
That is the beautiful property of a synergy/unity imo...stemming from its acoustic center collocation.
That makes the typical 2-way, multi-way, crossover lobing problems melt away, and need to be largely discarded with a synergy/unity/MEH (whatever the name
It’s a real shame reading this Chris but as I have studied the ins n outs and whys of the successful creation of the synergy design and hats off to danley n co
And your meh creation based upon the findings of danley n co and well documented contribution are a great feat. Some real gems n nuggets of info.
Without stating the obvious but named crossovers by Tom dick n Harry are worlds apart
As for describing the port entry’s regardless of semantics yes they are off axis always will be off axis the only port on axis is the compression drivers port. Throat
Now some may think reading this it’s tosh but guess you ain’t built n tried one
Neither have I ....but the compression drivers I have tried on several synergy style horns blows away ten different highly regarded waveguide/horns here that include jbl 2344 jbl 2380 altec 511b and seos
Any way to test if you made your port big enough? I just set up my synergy, and everything seems to work as it should, but I would like to know if the ports are large enough to allow enough output.
To not cause to much high frequency distortion,ho much place could a port take in on a give level. Could this be expressed in a percentage? My horn is pretty large so I figure a port of a certain size will cause a smaller distortion than a similar port in a smaller horn.
To not cause to much high frequency distortion,ho much place could a port take in on a give level. Could this be expressed in a percentage? My horn is pretty large so I figure a port of a certain size will cause a smaller distortion than a similar port in a smaller horn.
I agree with what you posted, but I'd add one more thing:
I think it's possible that the varying depths of the drivers is playing a part too.
I've listened to a SH50, a Gedlee Summa, and a Lambda Unity Horn, all in the same day.
Probably the "weirdest" part of the SH50 sound was how it wasn't apparent where the *depth* of the soundstage began or ended.
And I thing a big part of that may be because the phase response is excellent, but the *depth* of the various drivers varies quite a bit. The woofer taps are over a *foot* closer to the listener than the tweeter is!
Patrick, What were your observations on how those three speakers compared in terms of sound and what each one was good at and possibly specific things that stood out or we're unpleasant. And finally if you were forced to listen for 10+ hours of music at medium to high levels, which speaker do you think would be most tolerable and most pleasant in the long term listening?
For my money, Bill Waslo's speakers are tough to beat. They combine the great imaging of the Danley Synergy horns with a treble that's a little bit better.
You can see this in the measurements at Danley Sound Labs; the midrange taps are definitely having an impact on the overall frequency response.
Bill's speakers reduce the size of the taps, which seems to clean up the treble, at the expense of maximum output.
Both the Danley speakers and the Waslo speakers are great, but the Waslo speakers sacrifice maximum output for smoother response. Plus, they cost way less.
Possibly the defining characteristic of the Gedlee speakers is that they can play loud while sounding laid back. I work at home, and back when I was single I would frequently have them playing at semi-deafening levels for HOURS.
I don't make meth, I write software, but I used to laugh when I watched Breaking Bad and thought "gee, that looks like something I'd put in my living room."
I doubt anyone got the joke, but in some of my YouTube videos I've played this track, which is from Breaking Bad.
Last edited:
