Hi entropy455,
Post #380: "... I do not currently own an SPL meter."
Smartphone > free apps > good enough. I talked to a friend about sound pressure level apps, and before I was finished he had downloaded one, and showed it to me. Literally!
Post #380: "... I extend the length the horn to 62 feet, I get a 10 Hz tune in half-space. Is this a complete waste of steel & concrete..."
Yes, I went to 40ft for an example I would consider excessive. But, this is your baby, as they say: "You pays your money, and you makes your choices." 🙂
Prior to making your final choice, take a look here:
http://www.diyaudio.com/forums/subwoofers/119854-hornresp-245.html
Post #2450, and research a little on LeCleac`h horn subwoofers, they are very interesting. Pictures of the indoor version I have found show a platform construction that looks as if it might be a suitable alternative for your project, but levels first.
Regards,
Post #380: "... I do not currently own an SPL meter."
Smartphone > free apps > good enough. I talked to a friend about sound pressure level apps, and before I was finished he had downloaded one, and showed it to me. Literally!
Post #380: "... I extend the length the horn to 62 feet, I get a 10 Hz tune in half-space. Is this a complete waste of steel & concrete..."
Yes, I went to 40ft for an example I would consider excessive. But, this is your baby, as they say: "You pays your money, and you makes your choices." 🙂
Prior to making your final choice, take a look here:
http://www.diyaudio.com/forums/subwoofers/119854-hornresp-245.html
Post #2450, and research a little on LeCleac`h horn subwoofers, they are very interesting. Pictures of the indoor version I have found show a platform construction that looks as if it might be a suitable alternative for your project, but levels first.
Regards,
Attachments
Last edited:
1. Buy and SPL meter and find out how loud is loud.
2. "Listen" to your music sources on a real-time analyzer and see what you can find, if anything, south of 30 Hz. Or 35 Hz. Another test is to use digital processing to filter out below. say, 35 Hz, and see if you can tell (esp. when somebody else is pressing the test button... piece of cake to set up with a Behringer DCX2496.. which you'll be buying soon enough anyway).
My HiFi is blasting away now "Historic Organs of France" fun recording if you love woody reeds and real big diapason stops. dBC (slow) hits about 90 when the room is alive at more than real-world organ levels.
Ben
Last edited:
Oliver,
1) "Good enough" is a phrase I'm quite used to, but in the case of a Smartphone > free apps, since the electret mic pre-amp voltage is very low ( around volts IIRC) by comparison to the 48 volt used in "pro" applications (or the 130 volt B&K mics I use for "serious" testing), it clips around 110 dB, resulting in grossly inflated SPL readings. The Smartphone dB/RTA/etc. apps are OK as long as used far enough away from the source that the mic hears under the 110 SPL clip limit. Same applies to a number of low-voltage electret mics commonly used for testing, they just are not up to the "real world" of loud stuff.
2) For anything but 8Hz pipe organ, or attempts to communicate with elephants or submarines, I agree.
3)The mouth round-over of the Le`Cleach horn makes perfect sense for a mid/HF horn to reduce diffraction, but for a bass horn, a 90 degree "barn door" boundary will give considerably more forward output for the materials used.
Art
Last edited:
Here's my homework assignment:
The lowest fundamental “large” organ note (C-1) is 8.18 Hz. This is not going to a happen out of my horns – let’s just forget about it. The lowest fundamental "normal" organ note is around 16 Hz. I’ve actually experienced this at a church in Virginia – very cool. The organ bellows in the basement were powered by a giant electric motor. Not sure what the horsepower rating was (dude wouldn’t actually let me walk into pump room, because the large belts & pulleys were completely unguarded – and I was only 12 at the time.) The lowest fundamental frequency of a Tuba is 16 Hz. Honestly, I could care less about a Tuba. In middle school band we poured water into some girl’s Tuba, and it sounded like a wet fart. She cried, and I felt bad. Really - - - I did. . . . The lowest fundamental frequency of an Imperial Grand Piano is 16.35 Hz (ok - what’s an imperial grand piano?) The lowest fundamental frequency of 5 string bass guitar is 30.868 Hz (now we’re talking) The lowest fundamental frequency of a standard 88 key Piano is 32.70 Hz (now we’re talking)
The above being said - I will openly acknowledge that it’s quite possible that the sound I'm “seeking” to achieve with my horn system is not necessarily the ability to reproduce down to 20 Hz, but rather the ability to achieve clean & effortless sound reproduction of the bottom octaves in general. As just-a-guy has accurately pointed out, when you push drivers hard, they don’t seem to sound as good.
I love the bass-punch of small drivers. The problem is you need a giant wall of them. Bass-reflex enclosures help, but most reflex enclosures unload long before hitting 20 Hz (sometimes long before hitting 30 Hz). I have experimented with bass reflex enclosures that tuned low (between 20 Hz and 25 Hz.) Every time I’ve been disappointed. The sound is either rubbery, lacking sharpness, lacking fidelity, or lacking "something". . . . I think it’s because the cones are just too heavy, and the BL values are just too low – it’s an inherent problem of low Fs drivers (in my opinion). And yes I achieved a proper alignment, and yes I fed them plenty of clean power. . . I’ve been building systems for 25 years now, and I still haven’t found that “sound” that I’m looking for - which I can only best describe as a higher-end reference type audio system, but with the lower octaves properly adjusted for the Maff curve – so that the sound is more true to the concert - aka live sound.
This is why I want to build a straight exponential horn using 12” drivers – specifically with a high BL, low Qts, and low MMD drivers. These type of speakers (outside of a horn) are excellent mid-bass drivers at best. And I don’t mean that as a putdown. They are genuinely wonderful sounding mid-bass units – tight, punchy, sharp, and clean. My hope is to extend this mid-bass fidelity into the bottom octaves, using the outstanding impedance-matching properties of the exponential bass-horn. Or said another way, achieving high SPL is not the driving factor for this project. My goal is system fidelity. I’m building the horn outside, because I don’t want to deal with room nodes. I’m building the horn out of concrete, because I want it to last for 50 years. If my end result is only 105 dB in my listening area, yet I achieve that “sound” that I’m looking for, then the project is a complete success. I’m not ready to accept that going louder is the solution to my problem – as my wife’s stereo is plenty loud, plenty efficient, and feed with plenty of clean power. It sounds good, but it's still not the sound I’m looking for. . . .
And if everything goes to crap, there's always the bulldozer. . . . .
Last edited:
Oliver, as mentioned in point # 3 of post 374:
Tom Danley's "Paraline" HF horn coupling in a small format horn loaded line array, and the Maltese horns, which are the cleanest sounding horns I have experienced, regardless of designer, I can say with no reservations that the Paraline sounds better in multiples than the narrow, "non overlapping" approach (That would be something like multiple SH-25). That said, if you placed the system near the listening position, a pair of 3" diaphragm drivers on a horn of the proper dispersion to cover the area will sound better than any multiple driver system from a distance, unless you are after the reverberation such a distance will afford.
Multiple SH-25 would be literally exactly like twice as wide dispersion Maltese horns, which I demonstrated to Tom many years before the SH-25 was in his wheelhouse.
The Maltese horns have twice the DI of the SH-25, but still would still not "keep up" with Eric's horn design other than in the "hot tub" zone.
A virtual single source point multiple driver horn like the SynTripP would be preferable to separate mid/high horns for many reasons, but the clean output is limited to the DI "Q" of the HF driver potential.
Most anyone would know all this stuff out after designing dozens of systems, and listening to live reinforcement and testing hundreds of those systems over four decades, if they lived to tell the tale.
Thanks to David McBean, anyone able to operate a Windows PC can design a large bass horn that will perform virtually the same as the simulation, if the enclosure is reasonably close to the simulated cross-sectional area throughout the horn expansion, and the enclosure is rigid.
The LF portion of a sound system is important, but considering most instrument and vocal fundamentals are above 80 Hz, and hearing acuity is greatest in the 800-4kHz range, "top" enclosures are by far the defining elements of a system.
Folks with enough money to consider hundreds of hours of design and construction time, purchasing tons of concrete, and thousands of dollars of drivers might consider consulting with someone familiar with outdoor system integration before the cement trucks arrive 😉.
Art
Entropy455 -
I think if you follow my two suggestions for how to establish your basic criteria empirically, you'd be a lot better of than taking bandwidth with your speculative views on the world of sound.
You also need to take care in digesting advice from the world of commercial sound. Do you want a music system handling like a Ford pick-up truck or a Lotus?
B.
I think if you follow my two suggestions for how to establish your basic criteria empirically, you'd be a lot better of than taking bandwidth with your speculative views on the world of sound.
You also need to take care in digesting advice from the world of commercial sound. Do you want a music system handling like a Ford pick-up truck or a Lotus?
B.
Last edited:
Eric,
Yes, the Cat D9, the ultimate EQ.
An "imperial" is longer than a grand, which is longer than a baby grand, which is slightly longer in string length than most upright pianos.
The lowest fundamental frequency of a standard 88 key Piano is not 32.70 Hz (now we’re talking..) it is 27 Hz, only 2 Hz lower than the Fb of the bass horns that I am ready to build after finishing my sail rig and home theater subs.
Take a look at the 8 foot long plywood an Outback can haul in one load- that's material for six 131 dB SPL (141 by the usual manufacturer spec.) 29 Hz Fb horn subs and about 10 top cabinets. They would do around 141 dB (or 151) at two meters, enough to effing blow you out of your tub at two meters 😛
Art
Attachments
Last edited:
Folks with enough common sense would realize that OP's goal of 105 db from midrange and up doesn't require anything special. Even my antique Klipsch Heresy that I found in the garbage and restored to working order can do that (just as an example, not a recommendation), especially since OP is considering moving the listening position a lot closer to the speakers.
As long as the mains are reasonably close to the sub to avoid lobing issues at the crossover frequency and can play 105 db at the listening position there are no "outdoor system integration" issues to consider, I still can't believe you are trying to profit off this.
Incidentally OP's 120 db bass, 105 db mains target is incredibly close to THX spec, and while I haven't ever heard of a MAF curve it sounds a lot like a Fltecher Munsen curve. So OP doesn't like flat, he want's a bass bump. No problem with that.
By the way, since you keep talking about barn doors, do you have any idea how large they are going to need to be to get appreciable forward directivity (gain) at the bottom of the passband? A couple of small barn doors like the ones you did for your 35 or so hz subs isn't going to do much of anything at 20 hz. You need a boundary big enough to reflect (no allow wrap around) at the lowest frequency of interest, the bigger the better, and at 20 hz that barn door is literally going to have to be the size of a barn or it's only going to have significant effect up at the top of the passband where it's not needed since the horn is already very directional up there.
This is getting good. I can't wait for you to say it. Say it please. ESL panels. You know you want to.
As far as your "establishing basic criteria emipirically", how you do propose OP should measure the THD at his preferred listening levels, since THD is at least as important as spl in this application? You don't seem to realize that reaching 90 db with ultra high distortion (due to your antique subwoofer's tiny xmax and tiny sealed box) in a tiny concrete walled senior's apartment is a bit different than hitting 90 db with close to zero distortion outside.
JAG,
1) I missed the OP's goal of 105 dB, reference the post. To produce 105 dB at 24 meters outdoor ain't going to happen with antique Klipsch Heresy "garbage", as you call it.
2) Yes, I do.
Since the doubling of frontal area in my "barn door" frontal boundary experiments resulted in a uniform 3dB gain at the low corner frequency in subs ranging from 60 Hz to 35 Hz, a 20 Hz horn would require the same doubling of area to achieve the same 3 dB forward gain.
Cheers,
Art
Having actually driven a twin turbo V8 Lotus, I will take the Lotus please. . . However I am failing to make the proper inference towards your comment. Are you implying that commercial sound is the Ford pickup truck, or is it the Lotus???
I have heard some Pretty fantastic live sound systems - probably the best to date was the Eagles Hell Freezes Over concert in Columbus Ohio 1995. I have also heard some pretty fantastic high-end master reference audio systems - one in Columbus Ohio, and a similar system in Silverdale Washington.
Again, I am looking to achieve a "hybrid" sound - in that the high-end systems sound amazing, but are not properly adjusted for the Maff curve (doing so is considered a bastardization within the world of high-end audio) - whereas the commercial sound systems make very apparent non-linear adjustments to the response curve - for a more dynamic "live" sound.
Question: the majority of time when you personally listen to music, what is your estimated SPL in the listening area? How much higher will your system go, before it starts to sound bad (i.e. should you throw the occasional party. . . .? Or if you just want to crank it up every now and then. . . .?
The Fltecher Munsen curve is exactly what I was referring to.
I have no idea why I was referring to it as the Maff curve. . . You guys are setting me strait again!
I have no idea why I was referring to it as the Maff curve. . . You guys are setting me strait again!
Post 370 - http://www.diyaudio.com/forums/subw...bass-horn-design-question-37.html#post4872012
My intuition (which I openly admit is rather novice in this realm) is that my listening area needs about 120 dB in the bottom 3 octaves, and about 105 dB in the mid/high octaves - for desired backyard get-togethers / parties. And for easy listening while soaking in the hot tub with the wife, I could probably drop those values by 15 dB across-the-board (i.e. turn the volume down).]
Also see post 367 where OP contemplates moving the listening spot to within 30 feet of the horn mouth - http://www.diyaudio.com/forums/subw...bass-horn-design-question-37.html#post4871913
Perhaps I need to reevaluate my planned listening distance. The 24 meter setback was assuming a pair of large stereo bass horns, spaced 60 feet apart at the mouth.
Now that I'm leaning towards building just one bass horn, I see no reason why I cannot converge the listening area closer to the horn.
If spaced my mids & highs about 30 feet apart, the sweet spot would be about 30 feet in front of the bass horn mouth. I'd still cover a decent listening footprint for throwing the occasional party, and my overall SPL (source) requirements would drop. . .
Am I the only one paying attention here? If OP moved the listening position to within 4 meters of the horn mouth the Klipsch Heresys absolutely could do 104 db at the listening position (if the 116 db max spl in the marketing brochure is correct). It wouldn't be a low distortion solution but I just mentioned it because it's a very pedestrian speaker that can be had for about $400 used and it does meet the 105 db goal if the listening spot isn't too far away. Obviously OP will want a better speaker than the Klipsch but I'm pointing out how ridiculous the mains suggestions are getting.
Are you saying you kind of like slightly crappy sound so long as it is bright?
For sure, you've heard rock bands with fuzz boxes sounding just great on a good home HiFi. But what do you think the Brandenberg Concertos would sound like in Wrigley Field? Sure, fantastic systems for rock concerts, but not fantastic for sound quality. You've also got to distinguish when a sound system IS the instrument and when it is just trying to faithfully reproduce the instrument. Likewise, a muscle car is fantastic for going in a straight line fast; but my Lotus Elan Plus2, was a whole different set of automotive (and sometimes other kinds of more interpersonal) joys.
In these parts, what you are calling adjustment for the Fletcher-Munson curve is also known as "loudness" compensation. Since it is all but impossible to implement well, it is out of favour. Some fraught efforts to make it fly with a miniDSP. If you like (and I kind of do like it for late night listening), just look for a pre-amp with a dual 4-gang volume control with loudness circuit.
No electrostatic system ever reaches super high loudness. For most music, conveniently almost all my present components are crying "uncle" by at most 100dBC. For getting my loud jollies, 80dBC is just fine. Pieces need to be played at appropriate volumes, although not clear what page of an engineering manual that would appear on.
Perhaps you should be more careful in distinguishing which among your earnestly held beliefs are based on substantiated information and which are based merely on what seems to you on the spur of the moment to be compelling logical principles.
Ben
Apparently you don't. This is a simple baffle diffraction issue and baffle diffraction is super simple to sim. Any forward gain is based purely on the size and dimensions of the baffle (barn door) boundary.
So here's a quick sim. Check my math. The original horn mouth was 235422 sq cm. I used my handy "convert" program to tell me that's equivalent to 36490.48 sq inches. Then I figured out the diameter of a circle with that area. I used 200 inch diameter, which is a bit smaller, but since the horn mouth is going to be square and probably smaller than the original plan this is fine. A 200 inch diameter horn mouth will be 31400 sq inches area.
So then it's off to the diffraction simulator. First a sim just to show the diffraction of the horn mouth itself (assuming it's round and with just pieces of wood to fill in the corners to make it a square shape). I pulled the virtual mic back 30 feet to reflect the actual listening position, because if the mic is closer you don't see much gains, with the mic this far back you see all the diffraction (forward directivity) gains.
An externally hosted image should be here but it was not working when we last tested it.
As you can see, the horn mouth size alone is enough to provide some pretty significant diffraction gains even down near 20 hz. So how big do we need to go with barn doors to get that 20 hz 3 db more diffraction gain? You might be surprised.
Assuming OP doesn't want his barn doors any higher than the horn mouth (the mouth is almost 17 feet high already if it's round) the only direction you can go with the wall is horizontal. Here's a wall with a total length of 400 inches (over 33 feet wide with the horn mouth centered in the middle).
An externally hosted image should be here but it was not working when we last tested it.
So what did we gain by extending the wall to 33 feet? Looks like about 1 db more at 20 hz. Hardly worth the effort.
If you want to get paid you need to seriously step up your game. If OP builds a cement or masonry wall it's going to cost maybe $10000 and gain about 1 db at 20 hz. This isn't a joke, this is tens of thousands of dollars we're talking about, uneducated guesses are not going to cut it here, especially since you think you are worth getting paid. Diffraction simulators are free and readily available, you should know better than to throw out wild guesses, especially when you are off by so much.
Uh, no. . . I’m saying I like slightly crappy sound, as long as it smells like flowers. There’s a huge difference. . . .
This is fascinating and very informative - which of my earnestly held beliefs are based merely on my perceived logical and compelling - yet spur-of-the-moment principles? I will carefully incorporate this information into building a better set of horns. . . .
FWIW, a popular modification is to put a 347 CID turbocharged small-block ford into an early 90s Miata. With wide profile tires, this combination will run 10s in the quarter, and also pull mid 0.9s on the skid-pad. You want to talk about earnestly held beliefs being upended? You should see the faces of Ferrari, Lotus, and Lamborghini owners when they get their butts handed to them by a Mazda Miata, with less than 10-grand under the hood. . . . It’s priceless. . . . Do you own a Lotus, Ferrari, or Lamborghini Ben?
So weltersys, how did you end up so wrong on this? You did a single empirical test. You started with very small speakers, way too small to have any significant diffraction gains at the bottom of the passband. Then you tripled (or more) the frontal area of the speakers with barn doors, this was enough to gain some diffraction gains near the bottom of the passband because the passband was pretty high, nowhere near 20 hz.
This is entirely different. We are dealing with a passband that extends an octave lower than your empirical test speakers, and the horn mouth in this case is already large enough to provide some significant diffraction gains at the bottom of the passband. So making the wall significantly larger in physical terms doesn't do a whole lot in frequency wavelength terms. You would have to make the wall several times larger than shown to gain 3db in diffraction gains at 20 hz, and at 17 feet high and 33 feet wide it's already extremely large.
This is entirely different. We are dealing with a passband that extends an octave lower than your empirical test speakers, and the horn mouth in this case is already large enough to provide some significant diffraction gains at the bottom of the passband. So making the wall significantly larger in physical terms doesn't do a whole lot in frequency wavelength terms. You would have to make the wall several times larger than shown to gain 3db in diffraction gains at 20 hz, and at 17 feet high and 33 feet wide it's already extremely large.
Here we go, all you need is a barn door 33 feet high by 33 feet wide to get your baffle step hump down to 20 hz and get full diffraction gain across the whole passband. I'm sure the OP won't mind a 33 foot high 1100 sq foot wall in his backyard. That's about what it's going to take to get your extra 3 db of gain at the bottom of the passband.
(If you move the virtual mic back to 100 feet the whole graph moves up so the flat part at higher frequencies rides the 6 db line, so you have to stand back a bit to get the full diffraction effect.)
(If you move the virtual mic back to 100 feet the whole graph moves up so the flat part at higher frequencies rides the 6 db line, so you have to stand back a bit to get the full diffraction effect.)
An externally hosted image should be here but it was not working when we last tested it.
If memory serves correctly, home owners in Kitsap County (residential property) may only build structures 35 feet tall - without a permit (I own the airspace up to 35 feet, no questions asked - interestingly, I have trees that are taller than 35 feet. . . .) You can however get permits to go higher. And then even higher, federal rules kick in - requiring warning lights for aircraft. That being said - my horn wall will be a maximum of about 3 feet above the top of the horn's mouth (a safety rail - to keep people from falling down from the hillside above (aka the backfill). The rail could be solid (aka part of the wall) if it would provide acoustic value - otherwise, I'll probably make the safety rail out of steel pipe - to lower the profile of the wall.
Bentoronto, I get it – you are a reference audio system kind of guy. You want to play a recording of a string orchestra, and have it sound like you’re sitting in front of a string orchestra.
At this particular point in my life, that’s not what I’m looking for -
I want to play a recording of Chili Peppers, and have it sound like I’m listening to Chili Peppers (without necessarily having to hit concert-level SPL). On a reference audio system, I’d blow out my ears with midrange before the bass drum starts to “kick.” My wife’s stereo came close to the sound I’m looking for – but it’s still not quite there. Part of the problem I’m sure is room nodes – which is why my next system will be outdoors.
But look at the bright side Ben, simulations indicate my horn will be pretty darn flat down to 20 Hz. This means I “could” one day level-off the SPL curve, and make the system suitable for listening to string orchestra. . .
Bentoronto, I get it – you are a reference audio system kind of guy. You want to play a recording of a string orchestra, and have it sound like you’re sitting in front of a string orchestra.
At this particular point in my life, that’s not what I’m looking for -
I want to play a recording of Chili Peppers, and have it sound like I’m listening to Chili Peppers (without necessarily having to hit concert-level SPL). On a reference audio system, I’d blow out my ears with midrange before the bass drum starts to “kick.” My wife’s stereo came close to the sound I’m looking for – but it’s still not quite there. Part of the problem I’m sure is room nodes – which is why my next system will be outdoors.
But look at the bright side Ben, simulations indicate my horn will be pretty darn flat down to 20 Hz. This means I “could” one day level-off the SPL curve, and make the system suitable for listening to string orchestra. . .
Last edited:
The dbx 4800 or 4820 has a programable bass curve called Auto Warmth and it is magic! This is not the "subsonic synthesizer" thing that's in all the driveracks that sounds like poop.
Entro, I usually run a Noise Level Analysis program when we have get togethers but it had been nearly a year since I have had the big rig out and I don't recall how loud it plays. My mic's flat line at 122 dB SPL and it will top that.
Barry.
To put this in perspective, a spherical sound source would need to radiate more than 142.47 megawatts of acoustic power into free space to produce a sound pressure level of 190.7 dB at a point 1 metre from the centre of the source 🙂.