I move home often, so can't design speakers for a known room shape or size (and room treatments must usually be modest, too). Headphones and near-field listening have been the answer, but I miss having room-sized speakers. I suspect that horn/waveguide type speakers would have many positives in this situation, and I find MEHs especially interesting for other reasons as well.
I've not gone there because they would need to be large; unfortunately it is the somewhat lower frequencies, say at least down to 500hz or so, that would seem most beneficial for me to direct. Keel's equasion indicates something like an 80cm (32inch) wide horn-mouth for that, depending on what angle is chosen. Though they would probably need to play lower than their directvity, down to around 80-100hz, as I'd want to hand over to subwoofer(s) rather than need an additional bass driver.
However.. it occured to me that a lot of MEHs are quite sensibly sized around the actual drivers, taps and the straighter part of the horn. So, if the seconary flare and any round-overs were easily detachable and dismantlable (for transport, storage and when letting-agents come to inspect), the size might not be such a deal-breaker.
Though I've never built a MEH or even a big waveguide; there seems to be a huge amount to learn and test, especially if I'm on the wrong track.. For those who have experience of such speakers: do my hopes and expectations seem valid and reasonably achievable?
Thanks,
Kev
I've not gone there because they would need to be large; unfortunately it is the somewhat lower frequencies, say at least down to 500hz or so, that would seem most beneficial for me to direct. Keel's equasion indicates something like an 80cm (32inch) wide horn-mouth for that, depending on what angle is chosen. Though they would probably need to play lower than their directvity, down to around 80-100hz, as I'd want to hand over to subwoofer(s) rather than need an additional bass driver.
However.. it occured to me that a lot of MEHs are quite sensibly sized around the actual drivers, taps and the straighter part of the horn. So, if the seconary flare and any round-overs were easily detachable and dismantlable (for transport, storage and when letting-agents come to inspect), the size might not be such a deal-breaker.
Though I've never built a MEH or even a big waveguide; there seems to be a huge amount to learn and test, especially if I'm on the wrong track.. For those who have experience of such speakers: do my hopes and expectations seem valid and reasonably achievable?
Thanks,
Kev
Interesting idea.
I have limited experience, having only built 2 SH50 type.
I've made a fair few other horns and like fabricating things.
I guess it depends on your definition of portable and liftable
What frequency coverage are to aiming for?
3 or 4 or more drivers get heavy when in one package.
The Mini-Meh thread is very interesting.
Amazingly compact.
As soon as you get into birch plywood, even damped chipboard, things get heavy quickly, esp if there is a rear box to provide bass reflex loading.
With portability cones a requirement of toughness.
Got to be a bit like PA rigs, pack up shift, unpack, quick install.
Neutrik Speakon connectors are great for this.
Making the mouth flare removeable would add a bit more weight. Has to be robust enough to take knocks.
Certainly doable. As long as the joins are reasonably smooth.
Are you thinking 3d printable?
I have limited experience, having only built 2 SH50 type.
I've made a fair few other horns and like fabricating things.
I guess it depends on your definition of portable and liftable
What frequency coverage are to aiming for?
3 or 4 or more drivers get heavy when in one package.
The Mini-Meh thread is very interesting.
Amazingly compact.
As soon as you get into birch plywood, even damped chipboard, things get heavy quickly, esp if there is a rear box to provide bass reflex loading.
With portability cones a requirement of toughness.
Got to be a bit like PA rigs, pack up shift, unpack, quick install.
Neutrik Speakon connectors are great for this.
Making the mouth flare removeable would add a bit more weight. Has to be robust enough to take knocks.
Certainly doable. As long as the joins are reasonably smooth.
Are you thinking 3d printable?
Thanks, both. Some good points and lots to read, already.
To answer the questions: I'd like 80hz to say 16khz (or more) frequency range, though I don't expect pattern control below a few to several hundred hertz. I see that 1" compression drivers are frequently said to give better top end, and will go low enough in a home setting. Though even for modest SPLs I've seen a few people still using bigger compression drivers (with bigger diaphragms) in order to eliminate the small midranges, so presumably at least some larger compression drivers must have an acceptable top end; I'm into my 50s now, so can't hear to 20khz at sensible volumes.
I'm not intending to 3d-print this. I'm imagining prototypes would be somewhat crude and made of MDF and filler, but would use better ply and maybe things like fiberglass for the final thing.
EDIT: you're right; weight may also be an issue. I doesn't have to be light, but I'd need to be able to carry each part up/down stairs by myself without it being dangerous. Maybe composite construction (e.g. aluminium and foam or balsa etc) might need to be considered.
Cheers
Kev
To answer the questions: I'd like 80hz to say 16khz (or more) frequency range, though I don't expect pattern control below a few to several hundred hertz. I see that 1" compression drivers are frequently said to give better top end, and will go low enough in a home setting. Though even for modest SPLs I've seen a few people still using bigger compression drivers (with bigger diaphragms) in order to eliminate the small midranges, so presumably at least some larger compression drivers must have an acceptable top end; I'm into my 50s now, so can't hear to 20khz at sensible volumes.
I'm not intending to 3d-print this. I'm imagining prototypes would be somewhat crude and made of MDF and filler, but would use better ply and maybe things like fiberglass for the final thing.
EDIT: you're right; weight may also be an issue. I doesn't have to be light, but I'd need to be able to carry each part up/down stairs by myself without it being dangerous. Maybe composite construction (e.g. aluminium and foam or balsa etc) might need to be considered.
Cheers
Kev
Last edited:
Yes, good call. He posted some pictures here. Very much along the lines that I was considering, albeit tweaked for hi-fi use in small rooms. Very encouraging to see that it works.
I would say you'd be looking at 1 or 2 X 8" drivers to get that low. Maybe 6" at a pinch.
If you don't need much SPL, one bass driver might even do.
I find 1 X 12" plenty loud enough in my hifi room.
Lightest and most compact, I would think is probably a dual concentric compression driver, or dual comp and cone driver.
Would easily reach 16KHz and beyond.
Problem there is developing the horn throat to allow the HF straight out and the for the tapped MF ports further down.
I'm still musing this problem / opportunity, thinking of a SH60 alike at some point.
Good idea on the wood and prototyping - man after my own heart.
Chopping wood with purpose is very therapeutic 🙂
I'd always used high quality birch plywood before this last build, as I had lots of large offcuts.
It's lovely to machine and screw into.
Heavy on the sanding though.
Chipboard takes screws much better than MDF and is nicer to cut and shape / sand.
MDF - the dust the dust!
The P5 chipboard stuff is quite tough and won't melt in the rain😃
I use Evostik urethane expanding wood glue. Wetting the surfaces with a light water spraying and it sets in minutes.
Easily sanded and doesn't crack apart like PVA when knocked hard, plus it's waterproof.
Just don't get it on your skin. It basically has to wear off!
I put flight case handles on the SH50s.
A godsend!
If you don't need much SPL, one bass driver might even do.
I find 1 X 12" plenty loud enough in my hifi room.
Lightest and most compact, I would think is probably a dual concentric compression driver, or dual comp and cone driver.
Would easily reach 16KHz and beyond.
Problem there is developing the horn throat to allow the HF straight out and the for the tapped MF ports further down.
I'm still musing this problem / opportunity, thinking of a SH60 alike at some point.
Good idea on the wood and prototyping - man after my own heart.
Chopping wood with purpose is very therapeutic 🙂
I'd always used high quality birch plywood before this last build, as I had lots of large offcuts.
It's lovely to machine and screw into.
Heavy on the sanding though.
Chipboard takes screws much better than MDF and is nicer to cut and shape / sand.
MDF - the dust the dust!
The P5 chipboard stuff is quite tough and won't melt in the rain😃
I use Evostik urethane expanding wood glue. Wetting the surfaces with a light water spraying and it sets in minutes.
Easily sanded and doesn't crack apart like PVA when knocked hard, plus it's waterproof.
Just don't get it on your skin. It basically has to wear off!
I put flight case handles on the SH50s.
A godsend!
Last edited:
The 2 way SPL Runt clones are pretty compact. Mouth ca 560 x 320 mm. I can definitely recommend Fane Sovereign 8-225, I use them sealed, but they would work with BR as well. The green horn is 80 cm mouth width with 2x12 and I would definitely not call it portable even without a back box
Attachments
Thanks, I'll add the runt clone to my investigation/reading list!
2x12" would likely be more than I need, so a smaller lighter back box might be possible for me. Though yes you're right: the horn size would still be quite large, and unfortunately I'd be wanting one about the size of that green example. So hopefully making the mouth separate (and maybe even designed for disassembly to a 'flat pack' for storage) would help with that. At least, that is my hope.
2x12" would likely be more than I need, so a smaller lighter back box might be possible for me. Though yes you're right: the horn size would still be quite large, and unfortunately I'd be wanting one about the size of that green example. So hopefully making the mouth separate (and maybe even designed for disassembly to a 'flat pack' for storage) would help with that. At least, that is my hope.
Thanks again. Yes, I think/hope that a couple of 6" or maybe 8" drivers would offer enough bass for my needs,
The concentric compression drivers do seem popular as a way of decreasing crossover frequency and increasing driver spacing. From what I can tell, a valid objective: both for accommodating bigger cone drivers and to place the taps less harmfully near to the compression driver. Though I've also seen it said that a small compression driver teamed with small midranges (as a 3-way) is not only cheaper but also better for the HF and midrange (in hi-fi use) than a dual concentric driver going direct to bass cones (as a 2-way). I'm in no position to judge that, but certainly a 3-way seems easier to cover the octaves needed in band-pass designs.
As this looks promising I'll have to learn hornresp and do some simulations. I can do some bandpass investigations in winISD, but that won't be enough for this project (unless I just winged it with rules of thumb, which seem more like a starting point than a final design).
Last edited:
Yes, all pretty promising.
There must be a reason why Danley uses the BMS 5CN160 comp driver / 5" cone in the SM60F and HRE1 (aimed at hi-end private cinema rooms), rather than the dual concentric compression driver.
I dabbled with BMS 4590ND dual concentrics ages ago.
It looks great on paper 300Hz to 22000!
It's for 2" throat horns, so not so suitable for Synergys I would think.
I never got a sound I was happy with.
A horn that would give good HF comprised the MF and vice versa.
I didn't know about Synergys then.
There are smaller 4959NDs...
ND drivers are the way to go for weight saving. Get your cheque book out though😃
There must be a reason why Danley uses the BMS 5CN160 comp driver / 5" cone in the SM60F and HRE1 (aimed at hi-end private cinema rooms), rather than the dual concentric compression driver.
I dabbled with BMS 4590ND dual concentrics ages ago.
It looks great on paper 300Hz to 22000!
It's for 2" throat horns, so not so suitable for Synergys I would think.
I never got a sound I was happy with.
A horn that would give good HF comprised the MF and vice versa.
I didn't know about Synergys then.
There are smaller 4959NDs...
ND drivers are the way to go for weight saving. Get your cheque book out though😃
I share your objective of as large a horn as reasonably works. And portability has also been one of my main priorities.
The best compromise I've come up with is the build i call syn-10. https://www.diyaudio.com/community/threads/syn-10.383607/
It's a 90x60, 36" wide x 22" tall x 22" deep. Uses two 12"s, four 4"s, and a coax CD.
And weighs only 51 lbs. Part of the catch with weight though, is the horn has no back....the CD and mid drivers are exposed off the back of the horn.
It has a handle on the top, and I can get it thru doors, up/down stairs, etc easily enough. That said, it is a bit awkward just because it's a big triangle/pyramid, but if this 71 yr old can move it around, can't be too bad
If I had to move it to a difficult location, the CD comes off with 2 screws, which knocks 6+ lbs and 4" of depth off the tab.
When it come to the smallest, lightest, version I've built that i think worked....it was another 90x60 31" wide x 19" high.
It used two 10"s.
I really don't think that getting below 30" wide is going to allow anything other than small lows, and like you said, doesn't get to 500Hz pattern control.
I've also built versions that had 36" wide primaries, and detachable flares that expanded them to 48-49" wide.
Detachables work, but are a PIA to make, imo. Art Welter might have them best solution there.
I've had trouble with resonances at cranking SPL.....bass energy from subs want's to turn the flares into energy grabbers i think.
Anyway, just some random thoughts/experiences...hope they help...
The best compromise I've come up with is the build i call syn-10. https://www.diyaudio.com/community/threads/syn-10.383607/
It's a 90x60, 36" wide x 22" tall x 22" deep. Uses two 12"s, four 4"s, and a coax CD.
And weighs only 51 lbs. Part of the catch with weight though, is the horn has no back....the CD and mid drivers are exposed off the back of the horn.
It has a handle on the top, and I can get it thru doors, up/down stairs, etc easily enough. That said, it is a bit awkward just because it's a big triangle/pyramid, but if this 71 yr old can move it around, can't be too bad
If I had to move it to a difficult location, the CD comes off with 2 screws, which knocks 6+ lbs and 4" of depth off the tab.
When it come to the smallest, lightest, version I've built that i think worked....it was another 90x60 31" wide x 19" high.
It used two 10"s.
I really don't think that getting below 30" wide is going to allow anything other than small lows, and like you said, doesn't get to 500Hz pattern control.
I've also built versions that had 36" wide primaries, and detachable flares that expanded them to 48-49" wide.
Detachables work, but are a PIA to make, imo. Art Welter might have them best solution there.
I've had trouble with resonances at cranking SPL.....bass energy from subs want's to turn the flares into energy grabbers i think.
Anyway, just some random thoughts/experiences...hope they help...
They do indeed; thank you! I'd also read some of your other threads with interest, but haven't seen the v10 that you linked to, so that is helpful too.
I think/hope that my modest home hifi level requirements makes the task a little easier. In particular the pressure near the mouth should be fairly low, which might allow lightweight or thin extensions. Though I can see how bass drivers could cause problems through shaking/vibrating the structure generally; I've even had that occur with quite small tweeter waveguides in the past. A very good point! Possible handing off to subwoofers at around 80hz may help, but there is still a fair bit of bass energy above there. Definitely food for more thought. I wonder if, to some degree, opposing drivers might be linked together more strongly to better cancel their forces; an angular if less effective version of some other cabinet and subwoofer designs.
Thanks,
Kev
I think/hope that my modest home hifi level requirements makes the task a little easier. In particular the pressure near the mouth should be fairly low, which might allow lightweight or thin extensions. Though I can see how bass drivers could cause problems through shaking/vibrating the structure generally; I've even had that occur with quite small tweeter waveguides in the past. A very good point! Possible handing off to subwoofers at around 80hz may help, but there is still a fair bit of bass energy above there. Definitely food for more thought. I wonder if, to some degree, opposing drivers might be linked together more strongly to better cancel their forces; an angular if less effective version of some other cabinet and subwoofer designs.
Thanks,
Kev
Last edited:
The SynTripP horn extensions are just one example of detachable horn sections.
Community light and Sound had fiberglass/balsa wood core bass horn extensions that could be stacked like Dixie cups:
No problems at over 130dB SPL at one meter, using 1/2" ply for "wave guides" in front of 10 x8", 5 x3" diaphragm HF drivers and an 18" tapped horn :
In the 1970s I made 2x15" exponential bass horns from two pieces of laminated 1/4" plywood.
They can be seen in this 1978 photo behind plastic:
The horns mouths were 48", the throat 7" wide, top and bottom flats attached with rods and turnbuckles.
Standard plywood was bent over curved struts, similar to airplane wing construction.
Using "bendy ply", much tighter curves are possible.
Took a while to explain to stagehands, slowed setup time, but reduced the horn transportation volume to about 25% of the assembled size.
Toured arenas across the country using them, as long as the tension was correct, no problems with shaking or vibrations.
Art
Attachments
Very interesting indeed - thank you, Art!
Packing down to around 25% is very impressive. It is also useful to hear that a few different methods have done what I'd hoped (and in much more demanding situations, too). So this can definitely be made to work, then; it'll likely hinge on my inventiveness in making it successful for my particular situation. I'm guessing that won't be too huge a hurdle, in a relative sense - not compared to all the learning and testing required to make the main MEH successful to begin with.
So far I've done a fair bit of reading around the forum on the subject and think I understand, at least in concept, some of the principles and trade-offs that need to be juggled. So my next step will be to try some simulations with actual driver and horn parameters, to see where that gets me. So I'll first look at getting to grips with hornresp... though the initial challenge will be trying to remember my microsoft windows password!
Thanks again,
Kev
Packing down to around 25% is very impressive. It is also useful to hear that a few different methods have done what I'd hoped (and in much more demanding situations, too). So this can definitely be made to work, then; it'll likely hinge on my inventiveness in making it successful for my particular situation. I'm guessing that won't be too huge a hurdle, in a relative sense - not compared to all the learning and testing required to make the main MEH successful to begin with.
So far I've done a fair bit of reading around the forum on the subject and think I understand, at least in concept, some of the principles and trade-offs that need to be juggled. So my next step will be to try some simulations with actual driver and horn parameters, to see where that gets me. So I'll first look at getting to grips with hornresp... though the initial challenge will be trying to remember my microsoft windows password!
Thanks again,
Kev
Thanks again for this; I read your Runt thread and it does look very promising as a basis for my application too. The lack of midranges could be quite helpful in reducing size and complexity. It depends on how low the crossover needs to be, but I may not need to upgrade my existing 1" compression drivers; time will tell. So anyway, I think this is going to be well worth taking further.The 2 way SPL Runt clones are pretty compact. Mouth ca 560 x 320 mm. I can definitely recommend Fane Sovereign 8-225, I use them sealed, but they would work with BR as well. The green horn is 80 cm mouth width with 2x12 and I would definitely not call it portable even without a back box
Just for interest, I had a play around with some dimensions. It doesn't mean a lot at this stage since nothing has been simulated for driver positions or choice of specific drivers etc. but as a sanity check I looked at how a pair of 8" woofers might be arranged and also 4x 6" woofers (which are perhaps similar in total) on a 90x60 degree cone. The 6" ones would cost a lot more and weigh more but (if they'll go down to 80hz) they would also allow more choice of aspect ratios and slightly smaller size for the outer casing. Though there is less size difference than I had anticipated, to warrant the extra weight and cost.
Just a few example possibilities:
(click to enlarge)
So, I might be able to pack a pair of these into a 15" cube for storage/transport etc, which would be good. Though obviously the add-on extensions to the cone and flare would take up additional space.
Thanks,
Kev
Last edited:
Thanks. Thats good to know.
I see that 15" midbass drivers have been used successfully in 2-way MEHs, but they are fairly huge. They take up all the flat part of even quite big horns, and the driver's thickness means that removing the flair wouldn't make the box much less wide/tall either. Thankfully, for my purposes and modest SPL it does look like 8" drivers would suffice, so at least the flare and probably also part of the horn stick out beyond the drivers, so could be usefully detachable.
Predictably I've forgotten my windows password, so installing hornresp isn't as simple as it should be. I have just set up a new virtual machine for windows on my PC, but it then occured to me that reinstalling an old laptop might bet better; later on I'll be wanting to plug in a measurement microphone and probably work outside to measure things and test crossovers, DSP etc.
I see that 15" midbass drivers have been used successfully in 2-way MEHs, but they are fairly huge. They take up all the flat part of even quite big horns, and the driver's thickness means that removing the flair wouldn't make the box much less wide/tall either. Thankfully, for my purposes and modest SPL it does look like 8" drivers would suffice, so at least the flare and probably also part of the horn stick out beyond the drivers, so could be usefully detachable.
Predictably I've forgotten my windows password, so installing hornresp isn't as simple as it should be. I have just set up a new virtual machine for windows on my PC, but it then occured to me that reinstalling an old laptop might bet better; later on I'll be wanting to plug in a measurement microphone and probably work outside to measure things and test crossovers, DSP etc.
8" woofers + 1" driver is definitely a good combination - but it is on the edge of practicality - not every combination works. BMS4550 is the original driver (and successfully used with Fane Sovereign 8-225 in this clone), modified D.A.S. M-50 works pretty fine, too, BC DE500 was so-so. A pair or quad of smaller woofers would allow higher crossover, 1 kHz is the limit with 8" woofers for my builds, currently mine are set to something like 960 Hz using the natural rolloff of the woofers as a part of the acoustic crossover. A pair of (well chosen) 6" woofers should be more than enough for home volumes, a quad definitely. The added cost and weight for the quad might be actually worth it. I am heading towards to monstrous round 3way MEHs at the moment, but the Runt clones will definitely stay in my home office. I cannot fit anything better there.
Thanks, that is all very helpful; real-world experience with things very close to what I'm aiming for. My current compression drivers are a pair of 1" Faital HF108; they'll get down well below 1khz sweetly and with dynamics to spare at my normal listening levels. However, at very high (home) SPLs they can start to sound a bit pushed (without a waveguide that loads highly, anyway). So yes, I tend to agree that this might be hovering around the threshold of suitability.
I shall need to be more decisive on the scope for this project. For the current use, the HF108s would be great to below 1khz. However, on reflection it would be prudent to allow for future eventualities, needing more SPL - such as for Home Theater use, or bigger rooms than the current one. That could mean crossing higher (to smaller/more woofers), but I think there are still other reasons why crossing low remains desirable.
In that latter case, a compression driver with bigger diaphragm could make sense. Realistically that would mean stepping up in throat size too, but some of the modern ones apparently don't beam as much as their exit diameter might indicate, and my hearing age is old enough to not care about reaching the full 20khz either. So something like a 1.4" with a clever phase plug might be a better compromise in my case. I know some people like the dual ring types, but I'm not sure they're for me, or necessary for my application.
Anyway we'll see if I can breathe life back into my windows laptop and do some simulations. If I can get to grips with hornresp, anyway. My ideas are going to be vague until that point but IMO still important to consider the general format needed. It would be upsetting to expend all this time and money, to then realise my target wasn't what I wanted.
Good luck with the big 3-way!
I shall need to be more decisive on the scope for this project. For the current use, the HF108s would be great to below 1khz. However, on reflection it would be prudent to allow for future eventualities, needing more SPL - such as for Home Theater use, or bigger rooms than the current one. That could mean crossing higher (to smaller/more woofers), but I think there are still other reasons why crossing low remains desirable.
In that latter case, a compression driver with bigger diaphragm could make sense. Realistically that would mean stepping up in throat size too, but some of the modern ones apparently don't beam as much as their exit diameter might indicate, and my hearing age is old enough to not care about reaching the full 20khz either. So something like a 1.4" with a clever phase plug might be a better compromise in my case. I know some people like the dual ring types, but I'm not sure they're for me, or necessary for my application.
Anyway we'll see if I can breathe life back into my windows laptop and do some simulations. If I can get to grips with hornresp, anyway. My ideas are going to be vague until that point but IMO still important to consider the general format needed. It would be upsetting to expend all this time and money, to then realise my target wasn't what I wanted.
Good luck with the big 3-way!