A few thoughts:
Uneven response was raised as in issue with Synergy/Unity speakers, or at least with the current commercial versions. Adjusting the crossover networks, as Tom mentioned, can solve this issue. But, other than for an expert, that's (fiendishly?) difficult. I started my whole current DIY journey with the intent of building speakers, but waiting for my Unity's to arrive from Nick, I built a few amps, then other stuff. More than 10 years later I have yet to design and build a crossover - principally because I didn't want to spend the time to learn the software tools, and to buy or make bunches of inductors and other parts. What I did to is buy a DEQX. Though the learning curve is not trivial, it's in my estimation much easier and faster than diddling with passive components. My system has four sets of drivers - Tad 2001, Misco Mids, Lambda 15" mid bass, and Lambda 12's for subs. Each driver has its own amp. SE tube amp for the CD, FET transformer coupled amp for the miss (a version of Susan Parker's Zeus), SS amps for the bass and sub (FC Mihai's FC100). I've used others too, with equally good results. BTW the sub amps are driven by MiniDSP modules mounted in the DEQX chassis and powered by the DEQX power supply (feeding filters and regulators then the MiniDSP)
I can get perfectly flat 1M response and adjust in room response as well (especially useful for the bottom end). I can change amps or drivers and within 1/2 hour make any necessary adjustments. And the drives are time aligned in software. Adding subs and accounting for spatial offset is easy too.
Perturbations due to the mid entry holes is another issue mentioned. One way to minimize this is simply to use smaller entry holes (with appropriate shaping of the fustrum). This will lower the potential output capability, but for home use Danley's speaker output is way more than needed. I certainly have had no issues doing this for my mids, and some adjustments are feasible for the mid low drivers on the Synergys too.
I could perceive some horn character (HOM's?), but I bought some foam from
Earl, and that completely eliminated the horn character. I did not need to fill the entire waveguide in order to achieve this either.
Mention was made of the possibility of the mids (and by extension the bass drivers in the Synergy) perturbing the compression driver. Interesting point, but my guess is that this in minimal. On has to take into consideration that the compression ratio is likely to work in reverse, in this regard. The CD entry will work as a bandpass too.
The single inherent drawback is size.
In short, it works very well, with a natural open sound. The quality or perception of realism is entirely dependent on the recording itself.
Sheldon
Uneven response was raised as in issue with Synergy/Unity speakers, or at least with the current commercial versions. Adjusting the crossover networks, as Tom mentioned, can solve this issue. But, other than for an expert, that's (fiendishly?) difficult. I started my whole current DIY journey with the intent of building speakers, but waiting for my Unity's to arrive from Nick, I built a few amps, then other stuff. More than 10 years later I have yet to design and build a crossover - principally because I didn't want to spend the time to learn the software tools, and to buy or make bunches of inductors and other parts. What I did to is buy a DEQX. Though the learning curve is not trivial, it's in my estimation much easier and faster than diddling with passive components. My system has four sets of drivers - Tad 2001, Misco Mids, Lambda 15" mid bass, and Lambda 12's for subs. Each driver has its own amp. SE tube amp for the CD, FET transformer coupled amp for the miss (a version of Susan Parker's Zeus), SS amps for the bass and sub (FC Mihai's FC100). I've used others too, with equally good results. BTW the sub amps are driven by MiniDSP modules mounted in the DEQX chassis and powered by the DEQX power supply (feeding filters and regulators then the MiniDSP)
I can get perfectly flat 1M response and adjust in room response as well (especially useful for the bottom end). I can change amps or drivers and within 1/2 hour make any necessary adjustments. And the drives are time aligned in software. Adding subs and accounting for spatial offset is easy too.
Perturbations due to the mid entry holes is another issue mentioned. One way to minimize this is simply to use smaller entry holes (with appropriate shaping of the fustrum). This will lower the potential output capability, but for home use Danley's speaker output is way more than needed. I certainly have had no issues doing this for my mids, and some adjustments are feasible for the mid low drivers on the Synergys too.
I could perceive some horn character (HOM's?), but I bought some foam from
Earl, and that completely eliminated the horn character. I did not need to fill the entire waveguide in order to achieve this either.
Mention was made of the possibility of the mids (and by extension the bass drivers in the Synergy) perturbing the compression driver. Interesting point, but my guess is that this in minimal. On has to take into consideration that the compression ratio is likely to work in reverse, in this regard. The CD entry will work as a bandpass too.
The single inherent drawback is size.
In short, it works very well, with a natural open sound. The quality or perception of realism is entirely dependent on the recording itself.
Sheldon
Last edited:
I can't imagine why we are still messing around with passive crossovers. Save for people being amp tweak crazy. Digital , provided you keep to the rule of no extra conversions simply walks all over passive. Great amps are dirt cheap, especially when you don't need much amp. I get that there us a limit somewhere with filter correction, but I highly doubt that these efficient designs with minor ripple are anywhere near being beyond what you can correct 100% with the transfer. I suppose two wires to a speaker is a plus, but when they are all low latency wireless with an ip address that won't matter either.(-:
Last edited:
One of the first things that occurs to everyone that messes with one of these, is "why not just put the drivers onto the walls of the horn", similar to this:
An externally hosted image should be here but it was not working when we last tested it.
instead of this:
There's a number of reasons why the second solution works better:
1) The first reason is pathlength. To get the high frequencies to blend with midrange frequencies, we want the two radiators to be within about one third of a wavelength at the crossover point. With a crossover point of 1khz (34cm) that means that our maximum gap between the two drivers is about 11.3cm (4.5"). That would be impossible if the drivers were bolted to the walls of the horn; you couldn't get them close enough.
2) The air under the cone forms a low pass filter. Same idea as a bandpass sub. It's like a crossover filter, but it's acoustic, not electronic. JLH posted some measurements over on the main thread that showed near flat response... with no electrical or electronic xover filter at all! Just the acoustic filter formed by clever juggling of all the variables!
3) An acoustic filter reduces harmonic distortion
4) One of the 'neatest' parts about the acoustic filter is that it can clean up the phase response of the entire speaker. Here's an example of how this works:
Every one of us has heard a bandpass sub that didn't sound 'right.' We have heard bandpass subs that sounded 'slow' or out of 'sync' with the rest of the music. IMHO, the reason that this happens is that bandpass subs are resonant in their passband. For instance, if you have a bandpass sub with a bandwidth of 40hz to 80hz, you'll typically find that the helmholtz resonance is right in the middle of the passband. Due to this, there's a delay that's introduced at this frequency. You can see the delay in a number of measurements. You can see the delay in the CSD of the sub, and you can see the delay in the impedance curve of the sub.
Now bandpass subs aren't the only subs that do this; the thing that's rather unique about bandpass subs is that the delay is right in the middle of their response. A sealed sub has that phase shift too, but in a sealed sub the phase shift happens very gradually, and the phase shift also gets higher and higher as we go further and further out of band.
All of that make sense?
OK, so that's one of the 'odd' things about the various drivers in the Unity horn. They're delayed in their passband, but that delay actually has the net effect of moving their perceived location *backwards.* It makes the midranges "seem" to radiate from the throat, and it makes the woofers do likewise. (IE, the delay from the midrange coupling chamber is only a few centimeters, because the frequency is something like 500hz. At the lower frequencies, the delay is much greater, because the frequency is much lower.)
An externally hosted image should be here but it was not working when we last tested it.
My favorite analogy is that the drivers are 'handing off' to each other. In most loudspeakers, we WANT to extract a lot of bandwidth out of the drivers. It's the reason that two-way speakers are the norm. But in a Unity horn, we actually get BETTER phase response with MORE drivers. Because each driver is covering a narrow bandwidth, and that 'handoff' is part of the design.
Check out the phase measurements from Danley Sound Labs, and note how the phase is flattest in the six octaves from 200hz to 13500hz. In the pic above, I've shown the phase and frequency response of the SH50.
^^^ all of that is off the top of my head, and I'd *strongly* encourage anyone to investigate this further using a copy of hornresp or Akabak. There are literally 4-6 people on the forum that seem to understand how all the phase relationships in a Synergy horn work, and my understanding of all of this just scratches the surface. Danley, Bill Waslo, John Haskins, and John Sheerin have posted invaluable data on this over the last decade or so. Not just here but also on audioasylum and the bass list.
Last edited:
As Earl has also mentioned, some linearity issues can't be resolved via equalization.
When I had asked for higher res. (anechoic) polar plot it was specifically with regard to looking at this issue - that cannot currently be appreciated with the polar's I've seen on the Synergy.
I specifically mentioned "Uniformity".
This isn't a matter of trying to achieve "constant directivity" (which in the context of most loudspeakers is nothing but a fanatasy), but rather about how uniform the response is from each near axis.
As I've mentioned before - we hear the entire polar, and are particularly sensitive to about +/- 15 degrees (..from any axis that is referenced to the listener as generally 0 degrees), more so horizontally than vertically. Also, we can (over time) easily hear +/- 1.5 db deviations from "flat" as long as the deviation is a broader band deviation (..or a lower/wider "Q") - particularly at higher freq.s and especially in the range of 2-6 kHz.
The most troubling part is when the loudspeaker's response doesn't maintain that uniformity horizontally at higher freq.s where we are most attuned to hearing interaural differences in intensity.
Basically what this means as that even if the loudspeakers "match-up" for a single axis - displaying the same freq. response, that they do not fully "match-up" as an average across a broader horizontal axial response. i.e. within +/- 15 degrees the sound is NOT the same from Left and Right loudspeaker.
Below is an older version of Earl's Summa that has the resolution in the polar necessary to see this sort of behavior.
Attachments
Last edited:
That applies to any speaker. Acoustic issues such as reflections and refraction cannot be corrected, except acoustically. I doubt that the Unity/Synergy suffer in this comparison with other designs. As for "constant" directivity; these designs seem to me to be very good. This is supported by the fact that you don't have to have your head in a vise to get a consistent image and frequency response. I certainly don't perceive any frequency response anomalies when moving around the room - except in bass response. I plan to improve that by using working out a multiple sub approach.
Sheldon
What I do know, is that it (active) is a heck of a lot easier - especially for the diy'er. If you are making production speakers, the cost per speaker to do a complex passive is spread thin. And, as I mentioned, making changes in drivers, speaker position, amps, rooms, is trivial with the programmable active setup. One issue in all active, is that a good grounding layout is essential to banish hum. But that's a pretty simple logic problem.
Sheldon
Very true, but some speakers are better in this respect, and more than a little. (..that Revel loudspeaker horizontal plot that I provided awhile ago was considerably more uniform except for that very narrow-band dip at 3.4 kHz.):
http://www.diyaudio.com/forums/mult...rns-no-drawbacks-no-issues-9.html#post3624547
Speaking a "constant dirctivity" here is waveguide/driver combo (without any eq.) measured by augerpro that is (all else equal) also more uniform:
https://5e8772ee-a-62cb3a1a-s-sites...Yt1Bbq3D_pgiLl1--Q5tgsZt0mdw==&attredirects=0
Hmm, maybe you can get to it from this page:
https://sites.google.com/site/drive...aveguides/qsc-pl-000446-gp/frequency-response
Pete,
I like you studied the design for a very long time...and I recently committed to designing myself a version for home use, and I'm in the process of finishing them up. (I'm listening to them right now using Sound Easy's Digital Filter function...the passive parts are on a UPS truck bound for my house right now.)
Any of the positive descriptors of sound quality associated with these speakers are pretty much what I'm hearing. I do not find them to place the image in your lap...but they do have stunning pinpoint imaging. I could see where some might interpret it that way.
As far as frequency response? I've got mine easily +/-2dB except for Tom's aforementioned HF tilt downward. They tend to need more equalizing elements than a standard tower 3 way speaker (I've built plenty)...but that's not that big of a deal.
Someone posted regarding the shape of the midrange and woofer ports. The midrange ports are critical, and I can't tell you how many shaping variations I went through. Three major horn designs and more midrange porting experiments than I care to count. In one post I vaguely remember that Tom mentioned at some point you have to build a few to figure out the things you can and can't change...I agree wholeheartedly. If you have them shaped wrong you can get odd resonances that store a lot of energy. Placed wrong and you don't get the frequency response flatness or extension you want.
Tom has been very kind with his time and knowledge on these lists. I cannot thank him enough. As a result I have the best sounding speakers I've ever heard. I'm sure there are things I did wrong, but they must be relatively minor. I've been aware of his designs since the mid 90's when I was the admin for the old bass-list mailing list at U.T. and to have a version in my living room now...that's pretty dang cool.
Scott
Scott, that is a 90 db range... not much can be deciphered from that. I agree that uniformity and smoothness is most important. Horns and CDs always suffer from that compared to direct radiators. As long as there are no drastic changes off-axis, the synergy should be in the clear.
Note that he also has it "zoomed"..
I personally think both can be reasonably well deciphered (perhaps it's something evolved from looking at Stereophile plots). 😛
BUT it's not nearly as clear as something like Earl's older Summa plot. In fact I don't think I've seen a plot as good as that - even his newer ones on his website aren't nearly as easy to "read". (..and even then the axis scale is a bit to "squashed".)
Last edited:
Your guessing here because there is no data to support this position. If I were to guess, as you just have, I would "guess" the opposite.
Why can't we get "good" data and settle this issue and stop just "guessing".
PS. Scott's polar map is at least seven years old and I have improved the systems significantly since then. Look at the polar maps for the active Abbey on my website). It is exceptional and the data shows that.
Let's see, the speakers that you designed and built are "the best sounding speakers I've ever heard." Wow, that's the first time I have ever heard someone make that claim! 🙄
Yes, *older* design..
-but the older directivity plot itself is still overall better than anyone else's that I've seen (including your newer ones, though they have some advantages as well that the older one doesn't, like the improved detail on the Y axis and the reference to preferred listening axis.)
It will be nice when REW (Room Eq Wizard) has their's done. Hopefully it will incorporate the 3D mapping feature found in their spectrogram:
Attachments
Earl,
Sometimes the DIY "amateurs" making the 🙄 claims are jaded former hifi professionals with 20 years of technical experience, multiple patents, skill and ears. Sometimes there are other people who are competent too.
If you're ever in Austin, let me know, come on over...we'll drink some wine, you can have a listen.
Scott
Attachments
Check out the phase measurements from Danley Sound Labs, and note how the phase is flattest in the six octaves from 200hz to 13500hz. In the pic above, I've shown the phase and frequency response of the SH50.
Not to denigrate Tom's particularly nice phase response in the measurement shown, but the phase response shown in that graph isn't really flat over that range. Bear in mind that the horizontal axis there is LOG of frequency. A flat phase response (or constant delay) would have a constant slope (dPhase/dFrequency) when graphed in a Linear frequency plot. A flat slope in a LOG frequency plot like that (dPhase/dLog(frequency)) would only be constant delay in the special cases where phase is the same everywhere. In a log(F) plot, a constant delay or linear-phase curve with an added delay would actually slope steeper and steeper as frequency increases, not be straight like that.
Yeah, just being picky...
Scott,
Similarly the diy Synergies I have built are also the best sounding I've heard- and therange does include all that I designed myself. But then again, I've never heard real Danley Synergies, Unitys, or any of Earl's speakers, so a grain of salt is called for, of course.
Care to share any details you've found about shapes? I have found the magic of using "frustrums", but still don't know if it's better to have round holes (more area away from the corners) or oval holes (covers more of the horn length). I did find that you can significantly reduce the effect of the ports on higher frequencies by covering the holes with thin (say, T-Shirt) cotton. It of course reduces the sensitivity of the drivers feeding out through the ports, too, but there is output SPL to spare in a living room with these, anyway. The cotton also seems to smooth the response shape from the drivers feeding through it.
The effects I see from the ports (on frequencies sourced from closer to the horn apex) is most significant when measured at the "0 degree" axis. At other angles they tend to smooth out pretty well. Fortunately, as recommended with Earl's speakers, these are arranged well toed-in so the 0degree axis isn't so important.
The Dunlavey SC/IV, which is in fact a multi-way (3 way actually), speaker's phase and impedance plot below..
The design's got some other serious flaws (horizontal axial response particularly), but impedance (at least with respect to being resistive) and phase isn't one of them. The impedance is quite low though, requiring a fair bit of current from the amplifier.
Attachments
Last edited:
Aha Scott, There you have hit upon the other possible elephant in the room. Those speakers and in fact many similar, *are* 3 way speakers. They are in fact, the speakers that are being used to master the recordings we are judging the other speakers on! There are very few speakers used for mastering other than large 3 ways like my "sort of hated" 801's and the Dunlavys Dynaudios etc.One thing they all have in common is they are all 3 way, have a dedicated midrange in that critical area. The last standard 2 way speaker, which was largely abandoned BTW when the B&W's became poplular was in fact the Altec 604 and the Urie variants of it, obviously these are coaxial point source speakers with the point source and dynamic range charm but the FR discrepancies can still be heard on the recordings that have been mixed/monitored on them. This is at it's heart why I started this thread. A (3 way?) coherent speaker with extreme dynamic range and flat FR is now, at least I feel, much needed for mastering chores, not to mention that dynamic range hath it's charm just for listening (-: Fact: We are now dealing with Hip Hop as THE standard pop music, typically listened to at levels that would drive Misters Fletcher and Munson insane. A dynamic mastering speaker like this that also has midrange finesse would be a great tool.
The last time I read that statement before this thread though, was......... 😀
In aviation, we had this DIY guy named Burt Rutan. Guy was always making silly claims for those home built airplanes. Thing is they looked different than those other home built airplanes. Some say they were better, but most people were kind of scared of them because they looked different. sometimes it is hard to get the point across when a new improved product pretty much looks the same. At least different looking speakers can't kill you.
In aviation, we had this DIY guy named Burt Rutan. Guy was always making silly claims for those home built airplanes. Thing is they looked different than those other home built airplanes. Some say they were better, but most people were kind of scared of them because they looked different. sometimes it is hard to get the point across when a new improved product pretty much looks the same. At least different looking speakers can't kill you.
Last edited:
- Status
- Not open for further replies.