I got diverted in sorting out crossovers etc needed to test this. But now with just a couple of crude notch filters on the hf108 drivers, I've been listening to them crossed over at anything above 600hz and finding them impressive even with no waveguide or doughnut.
Unfortunately that impresiveness extends to exposing poor recordings, too. But with good recordings I'm, so far, very happy. Their dynamics (probably words like 'attack' here too) make things sound very real, especially vocals and live recordings. Though it has taken a few days to get accustomed to, as my previous speaker's inability to resolve so much does result in a smoother and more relaxed sound.
So, I shall be progressing with this idea.
Though... I'm surprised how well they sound just hung there, raw drivers as they are. Maybe the edge diffraction at the corners of the throat is so close to the pistonic output that one cannot distinguish. Similarly, there is a case for using a tiny baffle, e.g. the driver's diameter and no more (they of course have a sealed back, so no rear waves). It does make me wonder if a large doughnut is worth the extra desk space.
Unfortunately that impresiveness extends to exposing poor recordings, too. But with good recordings I'm, so far, very happy. Their dynamics (probably words like 'attack' here too) make things sound very real, especially vocals and live recordings. Though it has taken a few days to get accustomed to, as my previous speaker's inability to resolve so much does result in a smoother and more relaxed sound.
So, I shall be progressing with this idea.
Though... I'm surprised how well they sound just hung there, raw drivers as they are. Maybe the edge diffraction at the corners of the throat is so close to the pistonic output that one cannot distinguish. Similarly, there is a case for using a tiny baffle, e.g. the driver's diameter and no more (they of course have a sealed back, so no rear waves). It does make me wonder if a large doughnut is worth the extra desk space.
https://www.hifinews.com/content/duevel-bella-luna-diamante-loudspeaker is said to sound pretty good with its compression driver and diffusor that avoid classic horn. Result should depend about the room though. As it's 360° diffusion, we maybe call that decontrolled directivity !
I wonder about k-tube lower limit as well ? https://www.diyaudio.com/community/...for-1-compression-drivers.344571/post-5959017.
Maybe ask @freddi ?
I wonder about k-tube lower limit as well ? https://www.diyaudio.com/community/...for-1-compression-drivers.344571/post-5959017.
Maybe ask @freddi ?
Thanks for the info. I don't know much about k-tubes so that is an interesting idea, and it looks like they have some interesting properties. Although, I found a post by freddi suggesting a 1" tube would be best above about 1.6khz, and I'm looking to go much lower (the HF108's impedance peak is at about 700hz, which might be a sensible lower limit for it).
If I understand correctly, a bigger diameter (and longer) k-tube could be used at lower frequency, but then it wouldn't match the 1" exit, which some people seem to think is important. It might however influence my choice of driver for the bigger room-sized speakers I'm intending to make in the future. A 1.4" or even 2" exit for those might be a reasonable choice.
If I understand correctly, a bigger diameter (and longer) k-tube could be used at lower frequency, but then it wouldn't match the 1" exit, which some people seem to think is important. It might however influence my choice of driver for the bigger room-sized speakers I'm intending to make in the future. A 1.4" or even 2" exit for those might be a reasonable choice.
The "diffusor" of the Duevel speakers is actually a horn with a special radiaton pattern.
Regards
Charles
Regards
Charles
Yes it is. At least less intrusive as far as the zize is the blocking factor that motivated the op.
It is an interesting idea. It is very true that I don't have the space for waveguides big enough to control directivity down to <1khz, but for near-field listening I don't think that matters, as room effects are minimal at these frequencies. In fact a few early reflections can even make things sound more real (less anechoic) than listening purely to the direct sound.
That diffusor seems to take things a step further, by purposely aiming for wide, omnidirectional patterns - such as Linkwitz liked IIRC. I've mixed feelings about that on a desk, though; it might be even better, but it might be going too far. My concern would be that with PC monitors and other stuff in near proximity, it could be quite challenging to achieve a decent placement for something that fires equally in all (horizontal) directions. Not sure if that is a valid concern or not.
That diffusor seems to take things a step further, by purposely aiming for wide, omnidirectional patterns - such as Linkwitz liked IIRC. I've mixed feelings about that on a desk, though; it might be even better, but it might be going too far. My concern would be that with PC monitors and other stuff in near proximity, it could be quite challenging to achieve a decent placement for something that fires equally in all (horizontal) directions. Not sure if that is a valid concern or not.
Assuming you mean "compression driver & horn combination" vs dome tweeter directly in baffle, which would be the usual case, the causes have been plenty identified:
https://jblpro.com/de/site_elements/tech-note-characteristics-of-high-frequency-compression-drivers
https://www.klippel.de/fileadmin/_migrated/content_uploads/Loudspeaker_Nonlinearities–Causes_Parameters_Symptoms_01.pdf
Direct radiating dome tweeters of course also have a completely different radiation pattern than a compression driver in a horn, which leads to different room interaction. But, the sharpness and hardness that listeners dislike about compression drivers, that dome tweeters don't have, is a natural and unavoidable result of the compression chamber and horn loading. It inherently increases both sensitivity (good) and nonlinear distortion (bad).
”the sharpness and hardness that listeners dislike about compression drivers, that dome tweeters don't have”
Thats simply very far from true.
There are really bad regular metal and soft dome tweeters and there are bad compression drivers and there are bad horns. Out there.
And there are really bad speaker builders and really, really bad crossover builders and bad speaker tuners. Out there.
And there are also REALLY really good and good sounding examples of ALL of the above. 🎺🙂🎸
Thats simply very far from true.
There are really bad regular metal and soft dome tweeters and there are bad compression drivers and there are bad horns. Out there.
And there are really bad speaker builders and really, really bad crossover builders and bad speaker tuners. Out there.
And there are also REALLY really good and good sounding examples of ALL of the above. 🎺🙂🎸
I was concerned about such things too, so have gone into this with an open mind on whether it would work or fail. Listening first hand appears to be the only way, since opinions around the forum are so varied and often completely opposed.
In the event, I've been pleasantly surprised by how well the compression drivers work. One could always argue over subtleties, but there is really nothing at all stark or objectionable - at least once you objectively match their SPL with the woofers. At worst, they just seem quite revealing, which can go both ways depending on the quality of the source or recording. I'm certainly enjoying the clarity and realism of things like voices, strings and piano, drums etc.
I wonder if perhaps they work nicely because I am using them at 'much' lower SPLs than they would normally be required to produce; less air compression etc. The lack of a horn also means no 'horn sound', that people have cited as detrimental with CDs. The most immediate issue is that they're too sensitive, so prone to revealing amplifier noise and the amp's volume knob covers all the desirable SPL range in a very tiny amount of movement. But I can pad them to get over that, just hadn't expected to need it in an active system.
I'm currently using them down to 600hz. I wouldn't happily put a hifi tweeter to that task; it would either be incapable, full of distortion or too expensive to risk. So, in my particular application it might make more sense to compare the compression drivers to a small wide-range 'mid-tweeter' driver, such as a markaudio alpair 5.3ms. I would say the compression drivers have better dynamics (and maybe better top end), and certainly better power handling. But the 5.3ms can sensibly go much lower than my CDs, so a lot more flexibility on XO.
In the event, I've been pleasantly surprised by how well the compression drivers work. One could always argue over subtleties, but there is really nothing at all stark or objectionable - at least once you objectively match their SPL with the woofers. At worst, they just seem quite revealing, which can go both ways depending on the quality of the source or recording. I'm certainly enjoying the clarity and realism of things like voices, strings and piano, drums etc.
I wonder if perhaps they work nicely because I am using them at 'much' lower SPLs than they would normally be required to produce; less air compression etc. The lack of a horn also means no 'horn sound', that people have cited as detrimental with CDs. The most immediate issue is that they're too sensitive, so prone to revealing amplifier noise and the amp's volume knob covers all the desirable SPL range in a very tiny amount of movement. But I can pad them to get over that, just hadn't expected to need it in an active system.
I'm currently using them down to 600hz. I wouldn't happily put a hifi tweeter to that task; it would either be incapable, full of distortion or too expensive to risk. So, in my particular application it might make more sense to compare the compression drivers to a small wide-range 'mid-tweeter' driver, such as a markaudio alpair 5.3ms. I would say the compression drivers have better dynamics (and maybe better top end), and certainly better power handling. But the 5.3ms can sensibly go much lower than my CDs, so a lot more flexibility on XO.
Compression tweeters can be used to give no audible problems. A common issue is it's up to you to provide the acoustic conditions and the EQ. If you don't do this as is needed, you may not like the result.
By this point I'm convinced enough with this use of compression drivers to think about designing a baffle for them. I had it in mind to make a toroid or donut shape with a radius big enough to be effective in minimising edge diffraction, both as sound exits the driver throat and as it wraps around the baffle edges.
However, listening to them with no baffle, I'm often surprised by how well they already image. My theory/guess is that diffraction at the corners of their throat is effectively still coming from the same 1" diameter hole, albeit the edges, so might not be distinguishable as a separate source of radiation.
So I'm now wondering about putting them in a sphere rather than a toroid, since that would be both easier and more compact (as befits near-field speakers). I don't have a way to easily measure diffraction though, so can't really verify this. I suppose that I might make some crude mock-ups to test it subjectively. But I'd also be interested in hearing if people think a toroid would be worth doing, to begin with?
Thanks!
However, listening to them with no baffle, I'm often surprised by how well they already image. My theory/guess is that diffraction at the corners of their throat is effectively still coming from the same 1" diameter hole, albeit the edges, so might not be distinguishable as a separate source of radiation.
So I'm now wondering about putting them in a sphere rather than a toroid, since that would be both easier and more compact (as befits near-field speakers). I don't have a way to easily measure diffraction though, so can't really verify this. I suppose that I might make some crude mock-ups to test it subjectively. But I'd also be interested in hearing if people think a toroid would be worth doing, to begin with?
Thanks!
I would go with the sphere.
Anything else starts the path towards a waveguide. At that point, you might as well use optimized waveguides.
I'm also surpised how good a compression driver can sound without a horn or baffle. I've used them quite often just sitting next to a midrange driver.
Anything else starts the path towards a waveguide. At that point, you might as well use optimized waveguides.
I'm also surpised how good a compression driver can sound without a horn or baffle. I've used them quite often just sitting next to a midrange driver.
Thanks for the thoughts. Similarly, I'm testing them just sat there in front of a midbass driver, and find it surprising that they sound so good. Hopefully that bodes well for when they're housed more thoughtfully, and also corrected more finely for frequency response.
Yes, I've no great need of a waveguide in this application, and couldn't afford the space for a decent one anyway. So at most I would do something to help diffraction. But maybe that could gain fairly little in the area immediately at the throat/exit.
Yes, I've no great need of a waveguide in this application, and couldn't afford the space for a decent one anyway. So at most I would do something to help diffraction. But maybe that could gain fairly little in the area immediately at the throat/exit.
Perception of diffraction is spl dependent. Not as much of an issue at lower spl's and close listening distances.
A generous roundover at the edge of the driver, sphere, teardrop, or a couple of inches of damping material will be enough to put your mind at ease on this aspect.
The room will have a bigger impact in this case but it sounds like you are already having success as is.
A generous roundover at the edge of the driver, sphere, teardrop, or a couple of inches of damping material will be enough to put your mind at ease on this aspect.
The room will have a bigger impact in this case but it sounds like you are already having success as is.