If it makes you feel better I labelled my first graph non normalized when I should have said normalized to 0 deg 😀
It certainly does. FIR correcting the comb filtering on my arrays sounds the best. It shouldn't and I shouldn't do it due to positional dependency, steep filters, pre ringing etc. etc. but it sounds better, don't know why but it does.
Sure does.
Noticed the same thing using the four 4” BC 4NDF34-8 mounted on the sides of two EV PA-60 horns, or using the cheap Tymphany TC9FD-18-08 on a conical horn, they just sound better than the best compression drivers I've tested in the 400-1kHz range.
Even at a fraction of a watt level, where the PA-60 have no problem with distortion or response down to 600 Hz, the 4NDF34 (running "full range") sounded so much better that it was no contest raising the crossover point to 950Hz, as high as they could go before their offset horn acoustic band-pass kicked in.
Art
Do you mean the upper mid frequency response ripple that results from combing or its more extreme form in the HF? I would expect audible improvement for the former but don't know what to expect for the latter.
As you know, you can reduce/eliminate combing by frequency tapering but then you get vertical position dependency because the line is no longer uniform or quasi-infinite.
Last edited:
The over 8K dips and peaks. I have tried all sorts of different strategies to try and follow the envelope, psychoacoustic filtering etc. etc. Nothing sounds better than the setting with too much correction for theory. I don't know why, I spent a lot of time trying to make it better because it had to be able to be better, right? I couldn't make it better so I just accepted it as a mystery 🙂 None of that here though just another unsolved audio quirk.
Yep. My only 'theory guess' for the small cones sounding better in the 3-500Hz range and up than CD's made to reach that low, is that the CD's are simply trying to pull too much bandwidth, maybe at the expense of some kind of intermod distortion......????
I'm thinking that because the more I split the 10 octaves into separate slices, the clearer sound has been getting. (split into somewhat equal octave width slices)
Almost startlingly clearer on the current 5-way build (4-way syn + sub)
Good for you fluid 😀
I'm happy to set theory aside, even measurements aside, when it comes to what i hear.
I've been thinking i need to study more about how we hear ....
Mark,
Yes, had been hoping you'd do those IMD measurements we wrote about last April 😉.
Art
Attachments
Haha Art, yes...i totally forgot about that project. (and forgot where the idea that IMD might be the culprit came from....from YOUR measurements 🙂)
I have such a bad memory...
Sometimes, i get a question in my head, google it, start reading some unknown author's post, say to myself "hey this is exactly the info i was looking for"....and then find out i wrote the damn post...😱
Killed a few too many brain cells over the years i guess 😛
You have an awesome memory, thx for continually sharing it.
I have such a bad memory...
Sometimes, i get a question in my head, google it, start reading some unknown author's post, say to myself "hey this is exactly the info i was looking for"....and then find out i wrote the damn post...😱
Killed a few too many brain cells over the years i guess 😛
You have an awesome memory, thx for continually sharing it.
Puzzles
Mark, I've been following this thread from the start, and consequently re-read your Take#7 thread a couple of times.
I'm surprised if you hit the limits, or even "stress", the big co-ax CDs with big horns, in a domestic setting, and with a high-order crossover. My intuition keeps saying "look for another reason". You've ruled out a few hypotheses, and FWIW, I agree.
If I may, I'd like to check that I have mostly understood the observations. I won't link or quote your own words at you, but please let me know where I'm going off-track, if I am.
From the Take #7 thread I see that, with EQ, you could easily hear the kind of "mid" driver used (10PR300s vs. 10FE200). You hypothesized that perhaps the port size is a better match to the 10PR300. I find those points interesting. Could they provide a route to explain the sound change due to the new small mids?
Later in that thread you write that the B&C and BMS coaxes sound similar, and if I'm not mistaken, the differences were mainly at the high end (3.2kHz or higher). I think that's what makes me doubt the "overdriven" CD hypothesis most - it seems unlikely that these two would struggle so similarly that you'd not notice a difference, if one was being stressed more than the other.
My experience limited, but when I've over driven smaller CDs, such as the old 1" BMS ones, below their normal frequency range, the audible distortion has always risen so quickly with level that it's like a hard limit, and there's little doubt about its origin. I've not used such steep crossovers as you, however.
As I write this, I'm wondering if your experience could be due to "hearing" the mids - if the 10" ones sound different after EQ, why shouldn't the the small ones could sound different from either 10"?
When I write "mids" I mean something like "the combination of driver, volume between driver and port, port size, and port location". Perhaps it's related to the Helmholtz resonance/low-pass response of the port, together with the coupling to the horn, and cone resonances in the mids? I don't yet see how the sound "escapes" the EQ however, given the horn size and frequencies involved.
Puzzles.
Ken
Mark, I've been following this thread from the start, and consequently re-read your Take#7 thread a couple of times.
I'm surprised if you hit the limits, or even "stress", the big co-ax CDs with big horns, in a domestic setting, and with a high-order crossover. My intuition keeps saying "look for another reason". You've ruled out a few hypotheses, and FWIW, I agree.
If I may, I'd like to check that I have mostly understood the observations. I won't link or quote your own words at you, but please let me know where I'm going off-track, if I am.
From the Take #7 thread I see that, with EQ, you could easily hear the kind of "mid" driver used (10PR300s vs. 10FE200). You hypothesized that perhaps the port size is a better match to the 10PR300. I find those points interesting. Could they provide a route to explain the sound change due to the new small mids?
Later in that thread you write that the B&C and BMS coaxes sound similar, and if I'm not mistaken, the differences were mainly at the high end (3.2kHz or higher). I think that's what makes me doubt the "overdriven" CD hypothesis most - it seems unlikely that these two would struggle so similarly that you'd not notice a difference, if one was being stressed more than the other.
My experience limited, but when I've over driven smaller CDs, such as the old 1" BMS ones, below their normal frequency range, the audible distortion has always risen so quickly with level that it's like a hard limit, and there's little doubt about its origin. I've not used such steep crossovers as you, however.
As I write this, I'm wondering if your experience could be due to "hearing" the mids - if the 10" ones sound different after EQ, why shouldn't the the small ones could sound different from either 10"?
When I write "mids" I mean something like "the combination of driver, volume between driver and port, port size, and port location". Perhaps it's related to the Helmholtz resonance/low-pass response of the port, together with the coupling to the horn, and cone resonances in the mids? I don't yet see how the sound "escapes" the EQ however, given the horn size and frequencies involved.
Puzzles.
Ken
Hi Ken, thanks for the thought and consideration you've given my projects.
Yeah, i can't begin to fully know what's causing what, with the number or variables in play, and being simultaneously altered.
I stay puzzled, and will add some newfound puzzles after trying to make some replies to your points.
Yeah, i can't begin to fully know what's causing what, with the number or variables in play, and being simultaneously altered.
I stay puzzled, and will add some newfound puzzles after trying to make some replies to your points.
Musing
Mark,
thanks, I'm going to muse on your points for a while.
I read the "Somewhat easier ..." thread, but didn't find anything to help with this puzzle, perhaps I should read it again.
It would be great if you do try the "turn up slowly" test. Perhaps it's simply that you hearing the CDs start to limit - do the co-ax's motors weaken much much before they crash? I don't know that, I couldn't find any directly-useful design information.
When I listen outdoors I'm reminded of the clarity of good headphones, at least in the mid band. I guess that's due to lack of early reflections, except the one that I doubt matters much 🙂
Ken
Mark,
thanks, I'm going to muse on your points for a while.
I read the "Somewhat easier ..." thread, but didn't find anything to help with this puzzle, perhaps I should read it again.
It would be great if you do try the "turn up slowly" test. Perhaps it's simply that you hearing the CDs start to limit - do the co-ax's motors weaken much much before they crash? I don't know that, I couldn't find any directly-useful design information.
When I listen outdoors I'm reminded of the clarity of good headphones, at least in the mid band. I guess that's due to lack of early reflections, except the one that I doubt matters much 🙂
Ken
Hi Mark
Thank you for sharing Syn 9. I was about writing a PM to you asking for details and pictures, after reading about your new horn in the class D thread over on Klipsch-community. Then I thought, check diyaudio, and here it is. Thanks for doing what I have been contemplating. I still have some issues in life that prevent me from getting started to build something.
That makes sense to me somehow.
I have been contemplating if the "harshness/clearity-issue" has to do with unproper loading of the CD? That the CD simply can't get traction down low due to lost loading!? If you look at the big horn for your B&C CD it has a throat-section with a diffractionslot to make it load better and ad lenght to the horn to make the horn have a lower cut-off-frequency, the same with the big red horn for the AXI 2050!? Yust a thought, as I am no expert on horns, but maybe some others with experience could comment on this?
I have been wondering for some time, what happens when you do NOT use "filtering out sideband-effects" in the midrange with midrange drivers firing through holes, and instead use a CD as "direct radiator" in the same frequency-range? Maybe you provide some of the ansver in this build?
Well a lot of speculation from my side!? But really exciting to follow your experiments.
Steffen
Thank you for sharing Syn 9. I was about writing a PM to you asking for details and pictures, after reading about your new horn in the class D thread over on Klipsch-community. Then I thought, check diyaudio, and here it is. Thanks for doing what I have been contemplating. I still have some issues in life that prevent me from getting started to build something.
That makes sense to me somehow.
I have been contemplating if the "harshness/clearity-issue" has to do with unproper loading of the CD? That the CD simply can't get traction down low due to lost loading!? If you look at the big horn for your B&C CD it has a throat-section with a diffractionslot to make it load better and ad lenght to the horn to make the horn have a lower cut-off-frequency, the same with the big red horn for the AXI 2050!? Yust a thought, as I am no expert on horns, but maybe some others with experience could comment on this?
I have been wondering for some time, what happens when you do NOT use "filtering out sideband-effects" in the midrange with midrange drivers firing through holes, and instead use a CD as "direct radiator" in the same frequency-range? Maybe you provide some of the ansver in this build?
Well a lot of speculation from my side!? But really exciting to follow your experiments.
Steffen
Hi Steffen, glad to see you here.
I've been close to PMing you to fill you in on new developments, or posting in the Klipsch forum's 'Nearly Full Range' MEH thread Update in progress...
Kept thinking I'd post a follow up over there, but i keep doubting folks will be in sync with what's a clear turn from the prevailing wisdom of a CD straight to large cones. (A formula I've also embraced for two years!)
Anyway, this thread is in a large way, the result of the experiments I was posting over there. Those experiments as you know, were about trying to see if different port shapes and placements for large cones, would improve polars further. (I had turned back to digging deeper with port options, because I felt mouth termination efforts at improving polars had hit the end of practical improvement. Improving polars had been my singular focus.)
The dilemma that ports for a single large cone liked to be close to the throat for raising the cone's response up to the CD, but away from the throat for lowering the cone's response down to the sub became ever more obvious as experimentation continued.
At the same time, the talk of the new DSL Hyperion, with TomD's participation on ASR was going on. Mentions of the SH-50, how it can be (favorably) difficult to aurally locate, kept popping up.
My existing syns sometimes, but more to the rarely side, display that property.
When i added in my sometimes complaint of harshness with increasing SPL....
Well, i said screw it, time to try some small mids...allows better port placement for the large driver, and takes some of the load off the CD's lower end, and who knows...maybe helps with the magic.
I was speculating that using the CD too low is the source of the harness because I've cranked the CDs bigtime in other traditional builds crossed higher, without any trace of harshness.
You may be right about the lack of low end loading on the CD in the syn horns...maybe small signal response curves are simply not giving the full picture.
Anyway again, so i plugged up the test horn's holes that was being used in the Klipsch forum thread, and slapped some pretty cheap drivers on it, just looking for any improvements for any of the three issues. (lack of low end loading, sometimes harsh, and not enough of an aural disappearing act)
All three improved ! And I'm still shaking my head in a bit of disbelief about how much clarity was also gained.
So I built a 60x60 with better drivers. It's even clearer, but i screwed up the low ports by accidentally putting them too far out in the mouth.
The little 10fe200's in the old proto horn have a better bottom end than the 12pr320s in the 60x60.
I also don't like the 60 degree horiz pattern, compared to the 90 degree pattern of the proto.
(i move around alot...gave up sweet spot stereo along time ago, and remain very happy doing so !)
Ok, yet another build in progress.
I need a 60 degree vertical pattern to be able to mount the 12"s on the top and bottom of the horn (huge weight savings vs enclosing the entire horn in a box).
I've debated how far to take the horizontal pattern towards 90 degrees, because the 60x60 is a little clearer.
So under construction is a split-the-difference 75x60, using the better drivers.
I know much of this post is repeat of prior info in this thread (other than new current project), but i felt i should give you a clearer sense of why i moved to the small mids.
My biggest takeaway so far, besides being delighted at having found a significant SQ improvement..is to let go of agonizing over polars.
It appears there are bigger fish to catch than super polars;
fish i didn't even know were swimming in the pond 😀
I've been close to PMing you to fill you in on new developments, or posting in the Klipsch forum's 'Nearly Full Range' MEH thread Update in progress...
Kept thinking I'd post a follow up over there, but i keep doubting folks will be in sync with what's a clear turn from the prevailing wisdom of a CD straight to large cones. (A formula I've also embraced for two years!)
Anyway, this thread is in a large way, the result of the experiments I was posting over there. Those experiments as you know, were about trying to see if different port shapes and placements for large cones, would improve polars further. (I had turned back to digging deeper with port options, because I felt mouth termination efforts at improving polars had hit the end of practical improvement. Improving polars had been my singular focus.)
The dilemma that ports for a single large cone liked to be close to the throat for raising the cone's response up to the CD, but away from the throat for lowering the cone's response down to the sub became ever more obvious as experimentation continued.
At the same time, the talk of the new DSL Hyperion, with TomD's participation on ASR was going on. Mentions of the SH-50, how it can be (favorably) difficult to aurally locate, kept popping up.
My existing syns sometimes, but more to the rarely side, display that property.
When i added in my sometimes complaint of harshness with increasing SPL....
Well, i said screw it, time to try some small mids...allows better port placement for the large driver, and takes some of the load off the CD's lower end, and who knows...maybe helps with the magic.
I was speculating that using the CD too low is the source of the harness because I've cranked the CDs bigtime in other traditional builds crossed higher, without any trace of harshness.
You may be right about the lack of low end loading on the CD in the syn horns...maybe small signal response curves are simply not giving the full picture.
Anyway again, so i plugged up the test horn's holes that was being used in the Klipsch forum thread, and slapped some pretty cheap drivers on it, just looking for any improvements for any of the three issues. (lack of low end loading, sometimes harsh, and not enough of an aural disappearing act)
All three improved ! And I'm still shaking my head in a bit of disbelief about how much clarity was also gained.
So I built a 60x60 with better drivers. It's even clearer, but i screwed up the low ports by accidentally putting them too far out in the mouth.
The little 10fe200's in the old proto horn have a better bottom end than the 12pr320s in the 60x60.
I also don't like the 60 degree horiz pattern, compared to the 90 degree pattern of the proto.
(i move around alot...gave up sweet spot stereo along time ago, and remain very happy doing so !)
Ok, yet another build in progress.
I need a 60 degree vertical pattern to be able to mount the 12"s on the top and bottom of the horn (huge weight savings vs enclosing the entire horn in a box).
I've debated how far to take the horizontal pattern towards 90 degrees, because the 60x60 is a little clearer.
So under construction is a split-the-difference 75x60, using the better drivers.
I know much of this post is repeat of prior info in this thread (other than new current project), but i felt i should give you a clearer sense of why i moved to the small mids.
My biggest takeaway so far, besides being delighted at having found a significant SQ improvement..is to let go of agonizing over polars.
It appears there are bigger fish to catch than super polars;
fish i didn't even know were swimming in the pond 😀
Thanks Mark
I have been drawing some sketches and started some kind of log-book with ideas and decisions I make along the way.
I had a stress-breakdown this spring, and my brain almost stopped working! So some projects in my house that I need to finish before building horns got on full stop. I have only recently taken up small projects again in small steps. One of the projects is to clear a room in my basement to become my horn-workshop. That implies some plumbing and electrical work as it is the washing room I am moving upstairs.
I walk a lot and mostly have synergies on my brain. I´m pretty visual, so I can build in my mind while walking and can turn the horn around to tinker with how to do. When I have had a good idea, I write it down in my book for later. Thinking on paper is a good thing too!
Recently I made a real life experiment, building a wood-frame to see if a 80x60 horn 59"x43"would fit into my room, up against the sidewalls. Well, I and my wife looked at it a while, and decided to shrink it 10% to be only 135 cm (53") wide 😀. Maybe I can manage to upload a picture of the mockup.
I have to go now.
Steffen
I have been drawing some sketches and started some kind of log-book with ideas and decisions I make along the way.
I had a stress-breakdown this spring, and my brain almost stopped working! So some projects in my house that I need to finish before building horns got on full stop. I have only recently taken up small projects again in small steps. One of the projects is to clear a room in my basement to become my horn-workshop. That implies some plumbing and electrical work as it is the washing room I am moving upstairs.
I walk a lot and mostly have synergies on my brain. I´m pretty visual, so I can build in my mind while walking and can turn the horn around to tinker with how to do. When I have had a good idea, I write it down in my book for later. Thinking on paper is a good thing too!
Recently I made a real life experiment, building a wood-frame to see if a 80x60 horn 59"x43"would fit into my room, up against the sidewalls. Well, I and my wife looked at it a while, and decided to shrink it 10% to be only 135 cm (53") wide 😀. Maybe I can manage to upload a picture of the mockup.
I have to go now.
Steffen
Attachments
Yikes. Glad you bounced back and are going strong again 🙂
Your room and speakers look great. WOW...that looks like a HUGE horn frame 😀
Your room and speakers look great. WOW...that looks like a HUGE horn frame 😀
Hi Mark
Have you seen this, lots of measurements:
Danley Sound Labs SH-50 Professional Audio Speaker Review
A test of the SH50.
Steffen
Have you seen this, lots of measurements:
Danley Sound Labs SH-50 Professional Audio Speaker Review
A test of the SH50.
Steffen
Hi Mark
I have "finished" some sketches for my 80 by 60 degrees MEH/SYN, filling in your midrange-drivers and 15" woofers that I have 2 pairs of. That could be a starting point for me. The port-hole-placements are completely arbitrarily, just to have a picture and have something to point at. I hope it´s OK I share the sketches in tis thread?
When I look at the sketches it really becomes clear, that it should/could be some sort of an advantage to have smaller holes near the throat and bigger ones further out, minimizing disturbance of the CD!? That was actually the idea that I tried to convey over in the other forum. Small mid-ports do what they are good at and big woofer-ports do what they are good at further out!?
I am really only intuitive, as I haven´t build any horns yet or measured anything, but it kind of makes sense.
In my first sketches I planed to use 4 times 10PR320 16 ohm woofers, to get close enough to the throat and have the port-holes in the corners. With midranges, I could start experimenting with my 15" woofers! But I feel tempted to use 4 times 12PR320 16 ohm woofers instead! Then I could have a little brother to the SH96 in each room-corner! 😀
An idea for you to try maybe? In my sketch I made two smaller woofer-ports for the 15" woofers, instead of one bigger one, placed so that they kind of divide the area of the sound-bubble i four equal spaces, seen from the front. I have some idea/hunch that this will pressurize the sound-bubble most symmetrically!? It could also be an alternative to placing the portholes in the corners, where 1/4 wave-lenght distance (horizontally) begins to restrict, how far out the ports can be placed!? Just another thought.
I really must say, that your midrange-thing has inspired me, thanks.
Steffen
PS: I have scanned my sketches (PDF)and try to upload them, could not manege to rotate the horizontal one. Hope it works.
I have "finished" some sketches for my 80 by 60 degrees MEH/SYN, filling in your midrange-drivers and 15" woofers that I have 2 pairs of. That could be a starting point for me. The port-hole-placements are completely arbitrarily, just to have a picture and have something to point at. I hope it´s OK I share the sketches in tis thread?
When I look at the sketches it really becomes clear, that it should/could be some sort of an advantage to have smaller holes near the throat and bigger ones further out, minimizing disturbance of the CD!? That was actually the idea that I tried to convey over in the other forum. Small mid-ports do what they are good at and big woofer-ports do what they are good at further out!?
I am really only intuitive, as I haven´t build any horns yet or measured anything, but it kind of makes sense.
In my first sketches I planed to use 4 times 10PR320 16 ohm woofers, to get close enough to the throat and have the port-holes in the corners. With midranges, I could start experimenting with my 15" woofers! But I feel tempted to use 4 times 12PR320 16 ohm woofers instead! Then I could have a little brother to the SH96 in each room-corner! 😀
An idea for you to try maybe? In my sketch I made two smaller woofer-ports for the 15" woofers, instead of one bigger one, placed so that they kind of divide the area of the sound-bubble i four equal spaces, seen from the front. I have some idea/hunch that this will pressurize the sound-bubble most symmetrically!? It could also be an alternative to placing the portholes in the corners, where 1/4 wave-lenght distance (horizontally) begins to restrict, how far out the ports can be placed!? Just another thought.
I really must say, that your midrange-thing has inspired me, thanks.
Steffen
PS: I have scanned my sketches (PDF)and try to upload them, could not manege to rotate the horizontal one. Hope it works.
Hi Steffen, for sure, pls feel free to post about your project here.
(I think threads belong to all of us, not just the OP).
Always glad when we mutually inspire each other...
Yes, splitting the mid and woofer ports and drivers, seems to let them do their job better than a single larger port and driver for both.
One thing that still puzzles me though, vs measurements, is how much of a negative effect on the CD's response, do the ports actually have?
I think maybe a lot less than commonly assumed.
You really are going after a monster! 😀
I'd truly like to make some that big, but portability is a must requirement for me. I simply love outdoor listening too much....
Interesting you bring that idea up 🙂
Here's the back side of the top and bottom horn flares which mount the 12"s, that i made yesterday for the 75x60 underway..
I must admit though, I'm only trying it to see....I quit thinking about what kinda wave comes out of what a while back...simply seems too theoretical, or rather too far deep below real world signal to noise....
(I think threads belong to all of us, not just the OP).
Always glad when we mutually inspire each other...
Yes, splitting the mid and woofer ports and drivers, seems to let them do their job better than a single larger port and driver for both.
One thing that still puzzles me though, vs measurements, is how much of a negative effect on the CD's response, do the ports actually have?
I think maybe a lot less than commonly assumed.
You really are going after a monster! 😀
I'd truly like to make some that big, but portability is a must requirement for me. I simply love outdoor listening too much....
Interesting you bring that idea up 🙂
Here's the back side of the top and bottom horn flares which mount the 12"s, that i made yesterday for the 75x60 underway..
I must admit though, I'm only trying it to see....I quit thinking about what kinda wave comes out of what a while back...simply seems too theoretical, or rather too far deep below real world signal to noise....
Hi Mark
Well 4 times 12PR320´s in each side is maybe slightly overkill.😀
I have been looking at the homepage of B&C, and they don´t show a 16 ohm version of the B&C 4NDF34?! Is it discontinued? I would prefer to have four 16 ohm units in parallel.
Can you think of other suitable 4" or 5" drivers to use as midranges?
And yes, they are going to get heavy! I plan to make things a bit modular. I actually have an idea for some sort of crane to lift up the horns into the corners!
Interesting, that you have had a similar idea regarding the woofer-ports.🙂
Also interesting to notice, how tolerant the synergy-concept seems to be to port-placement, as your experiments show.
Happy building, looking forward to the next prototype of yours. 🙂
Steffen
Well 4 times 12PR320´s in each side is maybe slightly overkill.😀
I have been looking at the homepage of B&C, and they don´t show a 16 ohm version of the B&C 4NDF34?! Is it discontinued? I would prefer to have four 16 ohm units in parallel.
Can you think of other suitable 4" or 5" drivers to use as midranges?
And yes, they are going to get heavy! I plan to make things a bit modular. I actually have an idea for some sort of crane to lift up the horns into the corners!
Interesting, that you have had a similar idea regarding the woofer-ports.🙂
Also interesting to notice, how tolerant the synergy-concept seems to be to port-placement, as your experiments show.
Happy building, looking forward to the next prototype of yours. 🙂
Steffen
Yep, B&C currently only shows a 16 ohm version of the 5" 5NDF38; no 16 ohm 4".
I didn't know the 4NDF34-16ohm existed until i stumbled onto it at TLHP.
Maybe they've been discontinued...???
Their site said twenty something of them in stock when i ordered my eight.
And another yep for getting to a 4 ohm load....i think every amp i own is optimized for 4 ohms.
A little more work done on 12" ports for flow...now ready to reassemble horn 🙂
