Thank you. Just checking. Lol
I haven't pushed the 8PE21 that hard. Maybe I should try my red neck IMD test on it.
I haven't pushed the 8PE21 that hard. Maybe I should try my red neck IMD test on it.
Peter Morris design should be considered, it seems to do what you are looking for:
https://soundforums.net/community/threads/new-diy-mid-high-90deg-aka-pm90.11601/page-9
It is a long thread, Hornresp models, actual measurements, polar response, DSP settings, just don't ask me on what page each are located ;^)
Art
Last edited:
I was in the same dilema as you but ended up using a ready available horn structure. If you're located in Europe I would suggest buying a pair of "Audac HS212" end eviscerate the little coaxial horn. If you're in US then you could buy those in photo which are similar if not identical, available on eBay US, too expensive to ship overseas. The horn is 51cmx51cm at mouth and 45cm in lenght (70cm incuding backchamber) and uses 12inch speaker. The coaxial horn&driver has phaseplug role, but for 600Hz you don't need it (it's necessary if crossing at 1.5kHz as in Audacs). The output is quite smooth from 100Hz up, very rigid and clean, completely different of a folded variant I was used before (shrink down "Show horn" by Bruce Edgar).
Attachments
I'm entertaining the idea of using 2 x 12MH32 per horn (same mouth area (Am) to shorten the length a little and improve low end loading a little. Is this a fair assumption? Looking at a given expansion rate, doubling the throat surface (At) by means of doubling drivers in theory should be able to maintain the same expansion rate, just starting it further down at the point of a larger surface area in the horn.
I dont think folding the horn will get me the lowest distortion performance I'm looking for, but if need be, I'm open to internally curving the beginning throat area 90 deg to get a slightly longer horn to adhere closer to the optimal flare geometry. I've done this once by layering kerf cut pieces of wood strips and backing them with thin plywood to stiffen it all up. That at least provides a smooth taper to the critical begining part of the flare into the rest of the horn and keeping distortion low while enabling a higher cutoff limited by the driver itself.
Its been a long time since I built a larger horn. There wasn't much available info on them to the average speaker hobbyist back when I was doing it in the 80s, way before the internet. I just wish it wasn't still such a guarded thing, despite today's availability of shared knowledge (or misinformation). Both Inlow and Crowe want money for plans, which may not even be their own designs and could possibly be the wrong design for my purposes. It just can't be that hard to design a midbass horn from scratch if I'm willing to calculate it out. I just don't have th patience for spreadsheets. I'd rather write my own programs in basic on my lame little Sharp pocket PC with all of its 4kB of ram - Yes, I'm old lol. It would be nice if I could use an Android device for this purpose and not be locked into using a PC.
Does anyone know where I can find the plans for the Edgar System 100 horn? I'm also looking for the later 80 Hz version and any other similar style proven hypex midbass horn design. Subbing in a smaller driver means a smaller throat and consequently a longer horn, but if my theory in doubling the driver throat area is correct it may be possible to use a similar flare expansion curve.
I dont think folding the horn will get me the lowest distortion performance I'm looking for, but if need be, I'm open to internally curving the beginning throat area 90 deg to get a slightly longer horn to adhere closer to the optimal flare geometry. I've done this once by layering kerf cut pieces of wood strips and backing them with thin plywood to stiffen it all up. That at least provides a smooth taper to the critical begining part of the flare into the rest of the horn and keeping distortion low while enabling a higher cutoff limited by the driver itself.
Its been a long time since I built a larger horn. There wasn't much available info on them to the average speaker hobbyist back when I was doing it in the 80s, way before the internet. I just wish it wasn't still such a guarded thing, despite today's availability of shared knowledge (or misinformation). Both Inlow and Crowe want money for plans, which may not even be their own designs and could possibly be the wrong design for my purposes. It just can't be that hard to design a midbass horn from scratch if I'm willing to calculate it out. I just don't have th patience for spreadsheets. I'd rather write my own programs in basic on my lame little Sharp pocket PC with all of its 4kB of ram - Yes, I'm old lol. It would be nice if I could use an Android device for this purpose and not be locked into using a PC.
Does anyone know where I can find the plans for the Edgar System 100 horn? I'm also looking for the later 80 Hz version and any other similar style proven hypex midbass horn design. Subbing in a smaller driver means a smaller throat and consequently a longer horn, but if my theory in doubling the driver throat area is correct it may be possible to use a similar flare expansion curve.
hi profiguy, read this paper: http://xlrtechs.com/dbkeele.com/PDF/Keele (1977-05 AES Preprint) - LF Horn Design Using TS Paras.pdf I don't think there is any need for pre-made plans if you have a grasp of basic geometry.
I suppose you're right about that. I've read that paper and did some preliminary math. I'm just looking for that bit extra, something I may not know about, to give me an edge. Building a horn and realizing you're screwed it up is not a productive or economically wise thing.
Hi, profiguy
I was at Dr. Bruce Edgar's home about 7 yrs, ago, basically asking the same questions. Unfortunately, all the research he did says NO. You can't screw with physics !! You mentioned a FLH design that "cleanly" reproduces 100hz. to 600hz. There are some "crafty" designs out there ( Art mentioned one) buy not really made for home use and NOT a FLH design. The mid bass horn is the most difficult to properly build and intergrate with the rest of the system. This is what it took me to get close to what you are looking for.
https://www.youtube.com/results?search_query=meet+joe+and+his+amazing+diy+horn+speakers
Best of luck,
Joe
I was at Dr. Bruce Edgar's home about 7 yrs, ago, basically asking the same questions. Unfortunately, all the research he did says NO. You can't screw with physics !! You mentioned a FLH design that "cleanly" reproduces 100hz. to 600hz. There are some "crafty" designs out there ( Art mentioned one) buy not really made for home use and NOT a FLH design. The mid bass horn is the most difficult to properly build and intergrate with the rest of the system. This is what it took me to get close to what you are looking for.
https://www.youtube.com/results?search_query=meet+joe+and+his+amazing+diy+horn+speakers
Best of luck,
Joe
once you have a basic design you can deviate from the 'ideal' in Hornresp to see what you can get away with. Just be aware that effects like acoustic cancelation accross the cone of the driver are not modeled so this needs to be taken into account.
If your looking for more detail this book is very good:
https://hornspeakersystems.info/
In terms of tricks to reduce size I can only think of porting the back chamber which is used in some prosound horn subs, this results in a shelved down response in the LF but also the driver will take considerable power around tune as excursion is minimised, it might also be a bit of a pain to crossover to the subs due to phase shifts. https://martin-audio.com/products/subwoofers/mlx
Thanks for thst link. Once again, good suggestions regarding the design. I've been trying to look at other areas to compensate for the physical size demands and didn't want to end up with some weird phase shifts out of band. That would really complicate integration on the crossover side. Its always a challenge to use a bandaid approach that enables physically "liveable" dimensions. Luckily I don't need every ounce of efficiency, so that's one area I can wiggle on.
I've always been a fan of designing the back chamber area with the flexibility of varying the volume by means of moving or adding a dividing wall (like Edgar's shoehorn has). The chamber volume somehow always ends up being smaller than calculated for reactance annulling.
Alot of problems with weird resonances can be traced back to poor dampening of the back chamber or geometry issues causing standing waves and reflections that make it back through the cone, being that most drivers optimized for horn loading have relatively thin cones to start with. I dont think thats a major concern on a midbass horn with limited bandwidth as with a design playing into the upper mids. I also see alot of apprehension to sufficiently dampen the rear chamber vs a potential loss in low end efficiency. That can really fix many of those residual streaks and blips in the CSD and clean up the impulse response, as does adding enough bracing to the flare itself. I remember how bad those older Altec VOTTs were in that regard and how much it helped to add more bracing to them. At least I don't need to worry about extra weight in that regard.
Maybe someone can tell me how critical it is to use continuous bends in the flare vs multiple straight wall sections for a horn that won't be playing that relatively high. While it looks more eye appealing, IMO it just complicates the design and makes it harder to achieve sufficient wall thickness.
That brings me to a similar topic. The last horn i built with curved plywood panels ended up using sand filled walls to deaden the flare. While not that practical from a weight and portability perspective, I found that sandwiching thinner panels with sand works very well to kill most panel resonances, even when the panels were braced. I got that idea from some older Leak speakers I owned which sounded very clean in the mids, so I decided to experiment using this construction method with other builds. From my personal experiences, for a midbass horn you typically only need 3/16 - 1/4" plywood walls inside and out filled with roughly a 3/8" layer of sand. I cant emphasize enough how well this works and the first time I built an enclosure with sand filled walls was a big eye opener for me. The midrange was incredibly inert and open sounding without any hint of the typical box resonances you'd get in the mid 100s hz area. Its even more effective with curved panels, so it would be a given to build a horn this way.
Some tips when using sand filled walls - You need to pay attention to sealing any tiny pin holes in the wood and glue joints which will allow sand to leak out over time with vibration. To avoid this, I usually coat the inside walls of the panels with a layer of watered down wood glue to fully seal it. This also allows you to bend the panels without as much fear of splintering or cracking the outer panel radius surface and it adds a little more rigidity to the wood itself. * VERY IMPORTANT * fill walls only using FULLY DRIED and STERILIZED SAND. This keeps mold and mildew from growing inside the sand and walls. I also mix in a little bit of silca gel desiccant with the sand for extra measure (use approved respirator when doing this). The tiniest bit of moisture in the sand can wreak havoc here, so its important to be thorough here. You'll also need to carefully drill fill holes for adding the sand into the top corners of the panel sections. Be patient when adding sand. It may take a while to get it settled down enough and into all the inner edges. I like to use a small palm sander with a rag attached to the sanding surface to vibrate the sand into place. Its not that big of a deal if you don't get the tiniest last bit of air pockets filled. There will be some more settling anyways over time when you play the speakers. Its not going to be a problem. When sufficiently filled with sand, plug holes with wood glue and wooden dowels, then sand flush.
Anyways, just thought posting this may end up helping someone in the future.
I've always been a fan of designing the back chamber area with the flexibility of varying the volume by means of moving or adding a dividing wall (like Edgar's shoehorn has). The chamber volume somehow always ends up being smaller than calculated for reactance annulling.
Alot of problems with weird resonances can be traced back to poor dampening of the back chamber or geometry issues causing standing waves and reflections that make it back through the cone, being that most drivers optimized for horn loading have relatively thin cones to start with. I dont think thats a major concern on a midbass horn with limited bandwidth as with a design playing into the upper mids. I also see alot of apprehension to sufficiently dampen the rear chamber vs a potential loss in low end efficiency. That can really fix many of those residual streaks and blips in the CSD and clean up the impulse response, as does adding enough bracing to the flare itself. I remember how bad those older Altec VOTTs were in that regard and how much it helped to add more bracing to them. At least I don't need to worry about extra weight in that regard.
Maybe someone can tell me how critical it is to use continuous bends in the flare vs multiple straight wall sections for a horn that won't be playing that relatively high. While it looks more eye appealing, IMO it just complicates the design and makes it harder to achieve sufficient wall thickness.
That brings me to a similar topic. The last horn i built with curved plywood panels ended up using sand filled walls to deaden the flare. While not that practical from a weight and portability perspective, I found that sandwiching thinner panels with sand works very well to kill most panel resonances, even when the panels were braced. I got that idea from some older Leak speakers I owned which sounded very clean in the mids, so I decided to experiment using this construction method with other builds. From my personal experiences, for a midbass horn you typically only need 3/16 - 1/4" plywood walls inside and out filled with roughly a 3/8" layer of sand. I cant emphasize enough how well this works and the first time I built an enclosure with sand filled walls was a big eye opener for me. The midrange was incredibly inert and open sounding without any hint of the typical box resonances you'd get in the mid 100s hz area. Its even more effective with curved panels, so it would be a given to build a horn this way.
Some tips when using sand filled walls - You need to pay attention to sealing any tiny pin holes in the wood and glue joints which will allow sand to leak out over time with vibration. To avoid this, I usually coat the inside walls of the panels with a layer of watered down wood glue to fully seal it. This also allows you to bend the panels without as much fear of splintering or cracking the outer panel radius surface and it adds a little more rigidity to the wood itself. * VERY IMPORTANT * fill walls only using FULLY DRIED and STERILIZED SAND. This keeps mold and mildew from growing inside the sand and walls. I also mix in a little bit of silca gel desiccant with the sand for extra measure (use approved respirator when doing this). The tiniest bit of moisture in the sand can wreak havoc here, so its important to be thorough here. You'll also need to carefully drill fill holes for adding the sand into the top corners of the panel sections. Be patient when adding sand. It may take a while to get it settled down enough and into all the inner edges. I like to use a small palm sander with a rag attached to the sanding surface to vibrate the sand into place. Its not that big of a deal if you don't get the tiniest last bit of air pockets filled. There will be some more settling anyways over time when you play the speakers. Its not going to be a problem. When sufficiently filled with sand, plug holes with wood glue and wooden dowels, then sand flush.
Anyways, just thought posting this may end up helping someone in the future.
..or even want to use it that low. Horns that load tend to have interesting group delay issues near their cutoff.
The 100Hz region is a challenge, you fight for every bit of bandwidth when designing a horn here. If you are going to use a multiple sub system then you would be able to set a lower limit on this horn that isn't so low.
the front loaded VoT A5 and A7 made that just fine with a 15/16" Altec (80 hz to 800 hz with a sota dynamic) ? But I do not know if we can call that horn... was certainly more a wave guide to match the directivity of the multi-cells horn above. Not sure it is has good as air impedance load from a horn that really load the driver ?
Sligthy off topic, maybe a baffle less 12" to 15" could make that 100 hz to 600 hz and mimic the cleaness and articulation of a front loaded horn but without the ultimate spl ? A la Gradient as Juhazi member in the Aino Gradient thread https://www.diyaudio.com/community/...aborative-speaker-project.231353/post-7003847 ? Forgett it if too much off topic, I apologize in advance for that.
Sligthy off topic, maybe a baffle less 12" to 15" could make that 100 hz to 600 hz and mimic the cleaness and articulation of a front loaded horn but without the ultimate spl ? A la Gradient as Juhazi member in the Aino Gradient thread https://www.diyaudio.com/community/...aborative-speaker-project.231353/post-7003847 ? Forgett it if too much off topic, I apologize in advance for that.
Thank you for that info. Much appreciated and welcome. Anything helps to inspire me. I dont need alot of gain, just good fidelity with reasonable spl capability to match the A290 in character and tonality.
The front flare on the A7 is well engineered IMO. The actual genius is how its integrated to work with the reflex enclosure.
I'll look at that design a little closer. Something to learn from vintage designs.
I'll look at that design a little closer. Something to learn from vintage designs.
