As I say, to each their own, but whenever the requirement emphasises the LF above all other criteria, I would tend to go down the route of using drivers that are actualy designed for it, e.g. a couple of 15in HE woofers per channel & crossing outside the telephone band, or at a point of minimal compromise to let the widebander do what it does best.
All helpful guys, thanks.
brsanko- I tried to reply to your PM but it wouldn't work 😕 But thanks.
I dont understand why you get rid of the chamber to plot combined response in hornresp. Either way I plotted combined response for mine and it's terrible.😱 I'm not sure what constitutes a good plot. I'm fairly happy if I can draw a line through the top and then down the hf roll off witout deviation of more than 3dB.
Scottmoose- You're not the first i've heard say fostex is best for simpler passages, but why? What is it that happens when you play the 'wrong' sort of music through them?
I'm pretty sure I want to start my design from scratch but seeing what others have done always helps and inspires😎
brsanko- I tried to reply to your PM but it wouldn't work 😕 But thanks.
I dont understand why you get rid of the chamber to plot combined response in hornresp. Either way I plotted combined response for mine and it's terrible.😱 I'm not sure what constitutes a good plot. I'm fairly happy if I can draw a line through the top and then down the hf roll off witout deviation of more than 3dB.
Scottmoose- You're not the first i've heard say fostex is best for simpler passages, but why? What is it that happens when you play the 'wrong' sort of music through them?
I'm pretty sure I want to start my design from scratch but seeing what others have done always helps and inspires😎
It depends exactly what you're doing with them, so no simple one-size-fits-all answer I'm afraid.
😕
I wasn't really looking for a one size fits all solution, I was just wondering in what way fostex fail when playing more complex pieces.
I wasn't really looking for a one size fits all solution, I was just wondering in what way fostex fail when playing more complex pieces.
As I say, there's no one-size fits all answer to that question. It is context dependant, viz. the driver, the requirements, and what you are doing with it. In some cases it will do better in one area than others, in other cases it will be the reverse & so on & so forth.
Last edited:
So are you saying that the fostex drivers are just more prone to being compromised by their implementation when pushed to their theoretical best performance than other drivers? Or that single driver is just not the way to get high quality wide extension sound at all?
I don't know if compression is the right word to describe it. It'll vary but in general and from a laypersons pov, sounds (insturments, notes, etc), meld together, distort and are awful. Not how you'd experience the piece live or even with a multi-way system and not exclusive of some fostex drivers. Don't get me wrong and think it'll intrude on all pieces, nor do i plan on giving up single full range drivers any time soon.
Hi StoneT,
Any driver will ultimately be limited in terms of SPL, xmax, amount of air displaced (roughly equal to cone area times xmax), thermal limits etc.
When it gets near those limits, it may start to sound "strained," have rising distortion, have thermal compression etc. (I'm no expert but I know it when I hear it.)
If you don't ask the driver to hit its inherent limits, it will continue to sound its best. This is true of any driver, not specifically fullrangers. For example, my car stereo is okay on rock but simply cannot do orchestral -- it severely compresses the dynamic range of the recording.
Fullrangers with ultra-lightweight cones are often "highly resolving," so you can hear little faint details in the recordings that you never heard before. They can also sound really "pure," coherent, realistic. But if you push them too hard, same as any driver, you get distortion or "strain."
Any driver will ultimately be limited in terms of SPL, xmax, amount of air displaced (roughly equal to cone area times xmax), thermal limits etc.
When it gets near those limits, it may start to sound "strained," have rising distortion, have thermal compression etc. (I'm no expert but I know it when I hear it.)
If you don't ask the driver to hit its inherent limits, it will continue to sound its best. This is true of any driver, not specifically fullrangers. For example, my car stereo is okay on rock but simply cannot do orchestral -- it severely compresses the dynamic range of the recording.
Fullrangers with ultra-lightweight cones are often "highly resolving," so you can hear little faint details in the recordings that you never heard before. They can also sound really "pure," coherent, realistic. But if you push them too hard, same as any driver, you get distortion or "strain."
OK guys, with your help I think I'm starting to get there. Stop me If Im way off the mark but the situations where they fall down seem to be playing a wide range of frequencies at the same time.
Is this doppler distortion or could it be that the bass vibration disrupts the ability to produce the higher frequencies in other ways also? Eg, if the much of the voicecoil is out of the gap for any time the higher frequency modulations of the magnetic field have less effect on cone motion, loosing detail in the upper frequencies. 
brsanko said that Fs is not an indication of how low a driver will play in a horn but I've just been playing with sims for the 126, 166 and 206. It seems impossible to keep excursion within fostex specified Xmax much below Fs in a reasonably sized enclosure. I know the published Xmax for these drivers is low and they can play low frequencies nicely without distortion at higher excursions, but maybe they can't reproduce anything else cleanly at the same time. Maybe Fostex know what they are talking about.😀
However I think maybe I can make the 206 do what I want if I'm careful with the design and don't play any music through it that has frequencies below 35Hz🙁 Back to simming🙄
Is this doppler distortion or could it be that the bass vibration disrupts the ability to produce the higher frequencies in other ways also? Eg, if the much of the voicecoil is out of the gap for any time the higher frequency modulations of the magnetic field have less effect on cone motion, loosing detail in the upper frequencies. brsanko said that Fs is not an indication of how low a driver will play in a horn but I've just been playing with sims for the 126, 166 and 206. It seems impossible to keep excursion within fostex specified Xmax much below Fs in a reasonably sized enclosure. I know the published Xmax for these drivers is low and they can play low frequencies nicely without distortion at higher excursions, but maybe they can't reproduce anything else cleanly at the same time. Maybe Fostex know what they are talking about.😀
This starts to make a lot of sense🙂
However I think maybe I can make the 206 do what I want if I'm careful with the design and don't play any music through it that has frequencies below 35Hz🙁 Back to simming🙄
Whether it can technically be called doppler is debatable, but AMD / FMD do tend to get lumped under that heading. That's in addition to distortion when the VC exits the gap. Some drivers overload more benignly than others -the Fostex units are relatively forgiving in this respect, from the POV that they rarely crack up, just sound increasingly compressed, and, as you suggested, loosing detail (a euphanism for higher distortion).
Technically that's true insofar as a horn itself isn't especially bothered about the driver's own characteristics, it just requires pressure at the throat to function (although the horn should of course be designed to provide the desired load gain BW with the driver in question). Unfortunately, in most cases certain practical limitations in physical size etc. come into play, and a back-horn is no different to any other cabinet type -you can have two of these three options, but not all of them:
-Small box
-Low bass
-High efficiency.
A compromised horn bass horn uses resonant action to support the LF below the point at which it's impedance matched; it does not provide a purely resistive load for the driver across the whole desired gain BW & as you would expect, the higher the compromise, the less control you have over the driver.
You can get LF out of them all right & many are quite satisfied with the results. It just depends how much you're willing to compromise.
Technically that's true insofar as a horn itself isn't especially bothered about the driver's own characteristics, it just requires pressure at the throat to function (although the horn should of course be designed to provide the desired load gain BW with the driver in question). Unfortunately, in most cases certain practical limitations in physical size etc. come into play, and a back-horn is no different to any other cabinet type -you can have two of these three options, but not all of them:
-Small box
-Low bass
-High efficiency.
A compromised horn bass horn uses resonant action to support the LF below the point at which it's impedance matched; it does not provide a purely resistive load for the driver across the whole desired gain BW & as you would expect, the higher the compromise, the less control you have over the driver.
You can get LF out of them all right & many are quite satisfied with the results. It just depends how much you're willing to compromise.
Last edited:
Why is life always a compromise?🙄
The space I have is about 1850x400x250 or 700x900x300 so not tiny about 200 litres total.
The bass I'm still trying to decide. I think I'll go for 40Hz being not more than 6dB below the rest of the horn loaded frequencies and then attempt to loose some of the upper bass/mids through folding (how do people feel about square spirals?
) and hope for a little help from room gain.Which leaves the efficiency where it ends up but it shouldn't be too bad.
I see this as a compromise of all three factors to a lesser extent.
Late night fettling with hornresp on the 206 (and now keeping an eye on the displacement) seems encouraging.🙂
I'm never going to want 120dB out of these and if the bass is lacking or I feel it's muddying the rest then I'll bite the bullit and cross in a sub somewhere low for that last octave.
Does that sound sensible?🙂
Sorry I got booted for a couple days for fighting but I'm back. To answer your question about the chamber; Vrc and Lrc are for the back chamber in a front loaded horn or the front chamber in a compound horn, so in a simple rear loaded horn their is no chamber in front of the driver and the rear chamber is specified with Vtc and Atc. In leu of not playing music with really low bass I'd put a simple high pass filter on the speakers or use an amp with one, that way you don't lose the dynamics of the bass within it's pass band. but honestly there really isn't that much material below 40hz inmost recordings so I wouldn't worry about it unless it actually becomes an issue when listening. I really don't understand what people are talking about saying that Fostex speakers aren't good for dynamic passages from my experience that is what they excell at provided you design the cabinet right and don't push them too hard. One thing you'll notice when you continue to experiment with hornresp is that typically the bigger the speaker is the larger the horn will have to be to reach certain frequencies. But if you have room for a 200L cabinet then I'd say go ahead and shoot for the FE206En it will give you phenominal bass down to 40hz and you won't sacrifice anything on the top end. To my ears the 206 has some of the best treble I've heard. Let us know what you come up with and if you have any more questions. You should be able to PM me now that I'm out of the ol' penalty box.
Hmmm, not sure I want to consort with a troublemaker like you...😉You say you like the highs of the 206, may I respectfully ask your age? My high hearing is pretty good (evidence of an unspent youth) and I have three younguns that like to listen too. I dont want to be adding a supertweeter. Actually I'm finding it easier to get 40Hz and low displacement out of the 206 than the 166 (in terms of a smaller cab) so if you can attest to the highs I might go this way.
I've seen you like to keep an eye on the impulse response of the design, but everything I do to improve it results in increased mids through the horn. Is this something you would address with cabinet layout / stuffing? I suspect that folding to alter frequency response would also change displacement, impedance and maybe impulse characteristics. Is there a reliable way of predicting this? Or even any 'rules of thumb'?
I'm 35, My hearing is still quite good although I know there are those who can hear things I can't. I know I can hear at least 18K quite well.
Honestly especially when it comes to realism in the bass the impulse response is probably the single most important information you can get from hornresp.
As far as the rising mid range, I'd say live with it, once you've heard how beautiful the midrange is from Fostex speakers you'll probably wish you had more. The ultra light paper cones and strong magnets make the details in the midrange just shine through and give a realism you never thought possible. If it's too much, turn the bass up a bit. I have had silk dome tweeters from the likes of Vifa, Boston, B&W, and many more and I didn't hear anything from them that I don't hear from the 206s.
Honestly especially when it comes to realism in the bass the impulse response is probably the single most important information you can get from hornresp.
As far as the rising mid range, I'd say live with it, once you've heard how beautiful the midrange is from Fostex speakers you'll probably wish you had more. The ultra light paper cones and strong magnets make the details in the midrange just shine through and give a realism you never thought possible. If it's too much, turn the bass up a bit. I have had silk dome tweeters from the likes of Vifa, Boston, B&W, and many more and I didn't hear anything from them that I don't hear from the 206s.
Oh yes and keeping the chamber small does indeed improve impulse response but it does allow more midrange through the rear horn (not a good thing) and this can definitely be addressed with stuffing. You may want to consult with some of the transmission line fans on stuffing tips, as I have no software to simulate it's effects and have not done extensive experimentation in this area.
For best results try to keep rear horn response well attenuated above 200hz as not doing so can cause imaging issues and lack of detail in the midrange, as well as peaky response. A comprimise between this and best impulse is something to shoot for. The easiest way to experiment with hornresp is to brake your horn up into 4 conical sections so that you can use the loudspeaker wizard in the tools menu. It's convenient sliders make modeling 100s of times quicker and easier. Then when you reach the optimum comprimise that way you can try changing the segments to exp to see if it improves things.
Yep, it's the loudspeaker wizard that I've been playing with. If only all the plots could be on one fullscreen window that would be perfect
My current response seems to roll off starting at 200Hz at ~12dB/octave is that ok? It still means that at 200Hz the horn and driver response are both near max. I'm working on the impulse response but I'm not sure what is 'good'. I can see improvement but for all I know I may hav improved it to just 'Really bad' from 'painful'😕
:
My current response seems to roll off starting at 200Hz at ~12dB/octave is that ok? It still means that at 200Hz the horn and driver response are both near max. I'm working on the impulse response but I'm not sure what is 'good'. I can see improvement but for all I know I may hav improved it to just 'Really bad' from 'painful'😕
Yeah that's about what you'd want to shoot for. If when you do combined response your midrange doesn't get really peaky and ugly you did it right. And of course since you are bending your horn and hornresp models it straight you'll lose some more midrange than it shows too, especially if you put some stuffing in there. Hornresp is also going to show a big dip in the combined response where the two responses meet, it won't really be there, or at least not anywhere near as bad as it shows.
- Status
- Not open for further replies.