Time to get back to the forum again to sound out some ideas and get some input from all the gurus out there! 🙂
Waveguides... A shallow "horn" used to controll the directivity and off-axis response of tweeters as exemplified by SEAS in their DXT tweeter H1499-06 27TBCD/GB-DXT and various experimental applications such as: Zaph|Audio
Obviously great stuff for matching directivity with mids/ midwoofers.
Apart from that, I believe a controlled, or "narrow" dispersion is most usefull managing unwanted reflections in a less than optimum, i.e. most domestic environments, listening rooms.
Just bought a pair of Fountek FE85 on sale at my local DIY pusher, and that got me thinking.....
The small diameter of these fullranges will translate in to a very wide dispersion well up in frequency, (compared to say a 6,5" midwoofer) but at some point it will start beaming and end up with a dispersion which is probably narrower than a dome tweeter in the treble range.
From my initial reasoning, the narrow high frequency dispersion is not necessarily an evil in it self, on the contrary, but the wide dispersion in the lower frequencies will be a disadvantage.. varying dispersion pattern and a lot of room reflections at lower frequencies..
Now, how about concidering that little FE85 as a slightly large tweeter with a very low resonance frequency?
Not entirely unlike a 2" horn-driver?
Ok, placing a 2-300Hz cut-off horn is an obvious solution. but what about a large wave guide that could be effective down to 2-300 Hz rather than 1-2000Hz for a tweeter?
EQ'ed bakc flat again, that should give a fullrange source with controlled directivity covering the entire critical midrange area with no x-overs ot tweeter.
So, where is the flaw in my thinking??
Bring it on!! 🙂
Waveguides... A shallow "horn" used to controll the directivity and off-axis response of tweeters as exemplified by SEAS in their DXT tweeter H1499-06 27TBCD/GB-DXT and various experimental applications such as: Zaph|Audio
Obviously great stuff for matching directivity with mids/ midwoofers.
Apart from that, I believe a controlled, or "narrow" dispersion is most usefull managing unwanted reflections in a less than optimum, i.e. most domestic environments, listening rooms.
Just bought a pair of Fountek FE85 on sale at my local DIY pusher, and that got me thinking.....
The small diameter of these fullranges will translate in to a very wide dispersion well up in frequency, (compared to say a 6,5" midwoofer) but at some point it will start beaming and end up with a dispersion which is probably narrower than a dome tweeter in the treble range.
From my initial reasoning, the narrow high frequency dispersion is not necessarily an evil in it self, on the contrary, but the wide dispersion in the lower frequencies will be a disadvantage.. varying dispersion pattern and a lot of room reflections at lower frequencies..
Now, how about concidering that little FE85 as a slightly large tweeter with a very low resonance frequency?
Not entirely unlike a 2" horn-driver?
Ok, placing a 2-300Hz cut-off horn is an obvious solution. but what about a large wave guide that could be effective down to 2-300 Hz rather than 1-2000Hz for a tweeter?
EQ'ed bakc flat again, that should give a fullrange source with controlled directivity covering the entire critical midrange area with no x-overs ot tweeter.
So, where is the flaw in my thinking??
Bring it on!! 🙂
One of the things I like about full range drivers is the narrowing directivity. Though you'll change that about it if you use it as a compression driver and build a phase plug for it, making it no different to other compression drivers.
Hi,
Its doable but requires an extremely large / wide baffle.
Its fairly obvious 200Hz needs to be 10 times bigger than 2KHz,
and that is 100 times the frontal baflle area, 100 times bigger.
rgds, sreten.
Its been done with larger drivers, these are waveguides not horns :
Its doable but requires an extremely large / wide baffle.
Its fairly obvious 200Hz needs to be 10 times bigger than 2KHz,
and that is 100 times the frontal baflle area, 100 times bigger.
rgds, sreten.
Its been done with larger drivers, these are waveguides not horns :
Last edited:
Yes, well, perhaps my ramblings were not too clear.. if I wanted a compression driver and horn, I'd go for that in the first place.. 🙂
What i was thinking about was simply using a wave-guide on the FE85 as has been done with dome-tweeters
All in the hope of narrowing the lower frequency dispersion down to better match the inherently narrow high frequency dispersion of such a driver.
What i was thinking about was simply using a wave-guide on the FE85 as has been done with dome-tweeters
All in the hope of narrowing the lower frequency dispersion down to better match the inherently narrow high frequency dispersion of such a driver.
Sreten,
you are in deed right, I would hardly anticipate such a wave guide to be small! 🙂
looking at "conventional" horns, a horn with a cut-off of, say 200 Hz will have a large-ish mouth area and be relatively long.. but what about a wave guide?
Wave guides for tweeters are undeniably shallower/shorter than a theoretically correct "conventional" horn...
I guess the question really is how does one determine the dimensions/ profile of a waveguide vs a horn??
you are in deed right, I would hardly anticipate such a wave guide to be small! 🙂
looking at "conventional" horns, a horn with a cut-off of, say 200 Hz will have a large-ish mouth area and be relatively long.. but what about a wave guide?
Wave guides for tweeters are undeniably shallower/shorter than a theoretically correct "conventional" horn...
I guess the question really is how does one determine the dimensions/ profile of a waveguide vs a horn??
A low horn cut-off does not mean it has pattern control to that frequency.
“Waveguides” are usually conical other than the throat and mouth transition areas, the clear devices sreten posted in #3 look like horns to me.
Without getting caught up in the semantics of “horns” vs “waveguides” a simple conical expansion (in which the sidewall angles determine the dispersion angle up to the point where cone piston diameter takes over) requires a mouth about 1 wavelength wide to provide pattern control to that wavelength.
Speed of sound is 1130 ft per second, 1130 /300 Hz =3.77 feet.
Big.
The transition is not on/off at that frequency, even a conical “waveguide” of 1 foot diameter will provide some control much lower, but will raise sensitivity above that point, requiring compensating EQ if running the HF driver lower than around a 1130 Hz crossover point.
A 2” exit driver also beams like your little FE85 drivers unless the throat is pinched to a narrow width.
I saw the advertisement on a magazine some years ago , and they cost a lot...
much ,much more than a normal person would like to spend on some passive device . I guess that , if you have a big piece of thick vinyl and a big oven , you can put a load in the center ...et voilà ! you can have your homemade waveguide . Also ,some unused / broken big woofers might be used for the purpose . It's just a matter of sensibility between the ways of a loudspeaker system ,so if a woofer has ,say ,92 dB sens. and you want to use a little 2" in a sort of FAST arrangement , using the WG will raise its sensitivity .
Just plot through the frequency and deciBel graphic table at which frequencies the wg+fr combo is gaining respect to the fr alone .
Hi,
Typically a waveguide is much shorter than a horn and does not
go as low as the mouth circumference would indicate the cut-off
frequency for proper horn loading. Also the efficiency boost of
a waveguide is a lot less than that given by full horn loading.
Its semantics as to whether its horn loading or a waveguide,
as waveguides untilise "horn loading", but it seems clear to me.
This is a waveguide :
This is a (long throw) horn :
At some point the two blur.
rgds, sreten.
Considering it further the picture I posted #3 does not have a typical
waveguides profile / expansion rate so I guess you could call it a horn.
"All waveguides are horns, but not all horns are waveguides" (Earl Geddes)
Typically a waveguide is much shorter than a horn and does not
go as low as the mouth circumference would indicate the cut-off
frequency for proper horn loading. Also the efficiency boost of
a waveguide is a lot less than that given by full horn loading.
Its semantics as to whether its horn loading or a waveguide,
as waveguides untilise "horn loading", but it seems clear to me.
This is a waveguide :
This is a (long throw) horn :
An externally hosted image should be here but it was not working when we last tested it.
At some point the two blur.
rgds, sreten.
Considering it further the picture I posted #3 does not have a typical
waveguides profile / expansion rate so I guess you could call it a horn.
"All waveguides are horns, but not all horns are waveguides" (Earl Geddes)
Last edited:
Interresting... I was sort of assuming that waveguides would need to have some sort of curvature but if it is only a matter of smoothing out the transition to the baffle.. hmmm.. that would require some sort of profile/ radius? And what would determine that transitional profile??
So, assuming a conical wave-guide, with a mouth area given by the wavelenght of the cut-off, the depth would then be a function of the cone-angle desired for directivity-controll, i.e. the dispersion at the frequency the driver starts becomming directional.. well, this all makes sense.
But if it makes sense from a practical and hi-fi point of view is perhaps another issue... 🙂
So, assuming a conical wave-guide, with a mouth area given by the wavelenght of the cut-off, the depth would then be a function of the cone-angle desired for directivity-controll, i.e. the dispersion at the frequency the driver starts becomming directional.. well, this all makes sense.
But if it makes sense from a practical and hi-fi point of view is perhaps another issue... 🙂
Obviously, a true horn is best if increased efficiency is a criteria, but for my case, I'm solely interrested in the directivity controll.
Winder if there are other means of narrowing dispersion at lower frequencies, e.g. absorbing "baffles" on the sides of the driver?
Winder if there are other means of narrowing dispersion at lower frequencies, e.g. absorbing "baffles" on the sides of the driver?
I was going to write that if you make a wg too large ,then ,at the mouth , the large surfaces would become 'boundaries ' !!
But that's the opposite of what a ( well projected ! ) wg does , as it makes the speaker less dependent of room reflections , as also from speaker's edge diffractions ( or tweeter's flange ) .
For dimensioning a waveguide you can put some parameters
D - Diameter of the mouth ( m )
A - Angle of the cone ( ° )
F min ( Hz ) = 25000 / D / A
But that's the opposite of what a ( well projected ! ) wg does , as it makes the speaker less dependent of room reflections , as also from speaker's edge diffractions ( or tweeter's flange ) .
For dimensioning a waveguide you can put some parameters
D - Diameter of the mouth ( m )
A - Angle of the cone ( ° )
F min ( Hz ) = 25000 / D / A
Low frequencies are basically omnidirectional, and are hard to absorb, anechoic chambers use about 3 foot wedges of rock wool to be good to around 60 Hz.
Soffit mounting your speakers limits radiation to 180 degrees.
In a relatively small room like most domestic listening environments, pattern control to 500 Hz is pretty much "good enough", and won't require a huge waveguide to match the dispersion of a 15" woofer at that frequency.
Who said 15 inches !?! 
For the matters of dispersion and directivity , it would be far better
to use two littler cones ( faster ! ) , say , two eight inches .
For the matters of dispersion and directivity , it would be far better
to use two littler cones ( faster ! ) , say , two eight inches .
Hi,
i don't know where this thread is wandering off to. I''d say a big waveguide
on say a 3" driver will not have much effect above ~ 2.5kHz, and would suit
a high efficiency 3" with limited excursion and low Qts, which typically has
an inexorably rising response, a waveguide could flatten the response,
down to a point where IMO you'd need stereo subs to take over.
You could define wave guides as no narrowing of the driver diameter.
Its not that simple but it excludes compression drivers. The example
I posted in #3 would then be a wave guide, which I still think it is.
I'd say big waveguides have to be matched to a (suitable) driver, and
x/o to stereo bass/sub units to keep the driver in its effective range.
rgds, sreten.
i don't know where this thread is wandering off to. I''d say a big waveguide
on say a 3" driver will not have much effect above ~ 2.5kHz, and would suit
a high efficiency 3" with limited excursion and low Qts, which typically has
an inexorably rising response, a waveguide could flatten the response,
down to a point where IMO you'd need stereo subs to take over.
You could define wave guides as no narrowing of the driver diameter.
Its not that simple but it excludes compression drivers. The example
I posted in #3 would then be a wave guide, which I still think it is.
I'd say big waveguides have to be matched to a (suitable) driver, and
x/o to stereo bass/sub units to keep the driver in its effective range.
rgds, sreten.
I remember the price on the magazine was over 10 K$ for two speakers , stands , and polycarbonate concave discs .
Since I have two dead woofers😱 that have a large enough voice coil diameter to accomodate a 2,5 " , I was making some research today looking for budget FR , specially those with a phase plug in the center to eliminate some reflections and turbulences of air at the beginning of the throat of the wg .
Then Sreten talked about higher sensitivity drivers ...🙄
Yes, wehere is this thread wandering off to... 🙂 Guess I'm just looking for some good ideas for doing something interesting with these FE85 drivers I just picked up at random.. 🙂
Weltersys has a good point.. absorbing low frequencies do require quite some volume of absorbing material.. best case it would only end up as a waveguide made from glassfiber matt, but without the benefit of increased sensitivity i guess.
Acheiving directional controll down to 500 Hz would certainly be a big improvement over what my current 2-way speakers do.
I used to have a pair of Coral flat 8's and H-40 horn tweeters crossed high. Allthough they were probably all over the place in terms of frequency response, they had very clear sound and good imaging, I suspect this had somethign to do with the beaming of the 8 and the 40 or so degree dispersion angle of the tweeter horn.
Weltersys has a good point.. absorbing low frequencies do require quite some volume of absorbing material.. best case it would only end up as a waveguide made from glassfiber matt, but without the benefit of increased sensitivity i guess.
Acheiving directional controll down to 500 Hz would certainly be a big improvement over what my current 2-way speakers do.
I used to have a pair of Coral flat 8's and H-40 horn tweeters crossed high. Allthough they were probably all over the place in terms of frequency response, they had very clear sound and good imaging, I suspect this had somethign to do with the beaming of the 8 and the 40 or so degree dispersion angle of the tweeter horn.
Hi,
I guess you could go for a FAST (full range assisted) in a fairly standard box
with a 90dB or so bass unit x/o 1st order series at the baffle step frequency.
Another alternative is a version of MJK's :
http://www.quarter-wave.com/OBs/OB_Design.pdf
Possibly with 2 of these each side ($12 each)
GRS 12PF-8 12" Paper Cone Foam Surround Woofer 292-412
A pair a side will take the sensitivity from 87dB / 8ohm to 93dB / 4 ohm
which is then in the right +7dB ball park for the 86dB Fountek FE85.
You'd need to adapt the bass x/o in the article to 4 ohms.
rgds, sreten.
Last edited:
Yes, adding a woofer to assist this little fulrange has crossed my mind.. and for the very low frequencies, I have a pair of 15" active subwoofers x'ed over at 70 Hz 24 dB/Oct.
On the same sale, I bought a pair of SEAS H1316-08 CA26RE4X 10" woofers. Tey were 12 Ohm versions which should match the efficiency of the FE85's reasonably well. (got the pair for just over 100$ as well, cheap! )
since I have subwoofers taking care of the lowest octaves, perhaps an OB design could be something?
Very interresting article you linked to, will read now! 🙂
On the same sale, I bought a pair of SEAS H1316-08 CA26RE4X 10" woofers. Tey were 12 Ohm versions which should match the efficiency of the FE85's reasonably well. (got the pair for just over 100$ as well, cheap! )
since I have subwoofers taking care of the lowest octaves, perhaps an OB design could be something?
Very interresting article you linked to, will read now! 🙂
Hmmm... just had a creative moment here.... I can't let go of the thought of trying something "interresting" with those fountek FE85's...
Then I came across this on parts express:
Dayton Audio H12RW 12" Round Waveguide 1" Threaded 270-312
Not very expensive either, the modest outlay could be justifiable for experimental purposes...
What about cutting the throats (no pun intended) back to 3" and adding the founteks?? Should at least provide some directional controll quite a bit down in frequency.. Then I could play around with those 10" SEAS woofers and see if they would x over at some point.. obviously the wave-guided founteks would need some equalizing back to flat, but some op-amp based filters should take care of that...
The woofers would not be very directional at a low-ish x over frequency, but perhaps an open baffle solution could change that??
Then I came across this on parts express:
Dayton Audio H12RW 12" Round Waveguide 1" Threaded 270-312
Not very expensive either, the modest outlay could be justifiable for experimental purposes...
What about cutting the throats (no pun intended) back to 3" and adding the founteks?? Should at least provide some directional controll quite a bit down in frequency.. Then I could play around with those 10" SEAS woofers and see if they would x over at some point.. obviously the wave-guided founteks would need some equalizing back to flat, but some op-amp based filters should take care of that...
The woofers would not be very directional at a low-ish x over frequency, but perhaps an open baffle solution could change that??
- Status
- Not open for further replies.