I like it.... and yes a simple LM3886 or LM4780 is plenty to drive them.
In terms of distortion numbers and the various merits of different cone materials I don't have much to say. The CS125 is one of the clearest most natural midranges I've used though. I think one of the key features that makes it so good is the fact that it just doesn't compress like a typical driver. When you turn it up and use more power it doesn't sound strained. If you use a ported box & have significant information below port tuning you can bottom the driver and cause distortion but if you run it right... it just get's louder without breaking up. I think at lower volumes the dynamic peaks in the program material are portrayed much better. You don't seem get momentary higher distortion products due to the millisecond dynamic peaks. The driver just sounds more natural. That is my hypothosis at least. Not based on anything other than a hunch....
In terms of distortion numbers and the various merits of different cone materials I don't have much to say. The CS125 is one of the clearest most natural midranges I've used though. I think one of the key features that makes it so good is the fact that it just doesn't compress like a typical driver. When you turn it up and use more power it doesn't sound strained. If you use a ported box & have significant information below port tuning you can bottom the driver and cause distortion but if you run it right... it just get's louder without breaking up. I think at lower volumes the dynamic peaks in the program material are portrayed much better. You don't seem get momentary higher distortion products due to the millisecond dynamic peaks. The driver just sounds more natural. That is my hypothosis at least. Not based on anything other than a hunch....
Joules said:
I can´t say which one is better since I´ve never compared in same application. However Accuton usually have better cone behaviour and Seas lower motor distortion so the application (as always) decides which one is best. Accuton midrange units have lower distortion than Seas though so it is primarly as bass units or lower mids Excel may have a better performance.
But as always it depends on..
So for standard speakers I could use any of them for a two way and for a three way I would likely use a Accuton mid and Seas bass. In a four way Seas as upper bass/lower mid and Accuton as upper mid. I would typically use higher order slopes with the Seas due to the often more severe break up while Accuton can work with more shallow slopes and a resonance trap.. which should be used with a Excel driver as well. These traps need to be designed for the exact driver you have, one for each driver in the stereo pair since the break ups never match exactly.
/Peter
Note that nonlinearities in the Cms, BL, and Le - with respect to excursion - will excite the breakup peak. No amount of linear prefiltering can eliminate this. The crossover will NOT eliminate these issues (unless it's an IIR-based DSP crossover or uses extensive feedback from the acoustic output of the driver).Pan said:
This is why cones with high Q breakup modes - like raw metal or ceramic units - always sound harsh when pushed, yet cones with low Q breakup modes can be pushed significantly further before they sound harsh.
Sure, you can limit their bandwidth, but then you need to have multiple drivers in the system, which introduces its own set of acoustic problems, which can often be much worse than just living with a bit of low Q breakup modes from a softer, better damped cone.
Dan Wiggins
Adire Audio
Yes, and that´s why I recommend to lowpass below the frequencies that are most likely to have harmonic products (from Bl, inductive, and mechanical non linearities) that would be most likely to excite the break up.
Also if the driver has low distortion overall (like the Excel drivers) and are operated at a sane level or are highpassed, then there is no problem IMO. A threeway or more is necessary anyway IMO to achieve certain qualities.
A driver with inherent low distortion will always excite the breakup less than a driver with higher dist.
And sure, even if we lowpass a driver with 6k break up peak below 2k (which has its 3rd harmonic coincident with break up) there will be higher order harmonics from lower frequencies that will coincide with breakup, only these will (under good cirumstances) be very low. For example the 6th order harmonic at 1k will excite the breakup, however a good driver (W22) has 0.05% 2nd and 0.09% 3rd harmonics at 1k, then it´s very likely that the 6th harmonic will be a magnitude down at 0.01% or so (or less), and with the gain from the breakup will only increase to about 0.1%.
I don´t see this as a problem and in the end we could see it as a trade, a low distortion overall with minor peaks at a narrow range instead of a higher overall level of distortion which I believe is more destructive.
Or am I missing something?
/Peter
Also if the driver has low distortion overall (like the Excel drivers) and are operated at a sane level or are highpassed, then there is no problem IMO. A threeway or more is necessary anyway IMO to achieve certain qualities.
A driver with inherent low distortion will always excite the breakup less than a driver with higher dist.
And sure, even if we lowpass a driver with 6k break up peak below 2k (which has its 3rd harmonic coincident with break up) there will be higher order harmonics from lower frequencies that will coincide with breakup, only these will (under good cirumstances) be very low. For example the 6th order harmonic at 1k will excite the breakup, however a good driver (W22) has 0.05% 2nd and 0.09% 3rd harmonics at 1k, then it´s very likely that the 6th harmonic will be a magnitude down at 0.01% or so (or less), and with the gain from the breakup will only increase to about 0.1%.
I don´t see this as a problem and in the end we could see it as a trade, a low distortion overall with minor peaks at a narrow range instead of a higher overall level of distortion which I believe is more destructive.
Or am I missing something?
/Peter
rick57 said:
Hi Rick,
did you hear Orions in Melbourne ? I'd be very interested to have a listen.
BTW, I have a pair of speakers that use Excel mids too (VAF i66) and they sound very neutral and detailed. "Ringing" is well supressed there as waterfall would suggest
http://www.vaf.com.au/new_home.htm?...g/sig_title.htm&catalog/products/i66_spec.htm
(W11 breakup is at ~10kHz)
It could be an interesting comparison.
Bratislav
Take a look at this comparison of a metal W15 Seas Excel and the equivalent polycone (M15). While the W15 has slightly superior linear distortion numbers (probably inaudible) the M15 is generally superior in terms of non-linear (stored energy) performance above 1kHz.
Personally, I think a small & light poly or paper cone is the best material for the midrange (voice) region. Metal cones are probably better suited to the lower frequency range (<500Hz).
http://206.13.113.199/ncdiyaudio/mark/Midrange test group/midrange_test_data.htm
Personally, I think a small & light poly or paper cone is the best material for the midrange (voice) region. Metal cones are probably better suited to the lower frequency range (<500Hz).
http://206.13.113.199/ncdiyaudio/mark/Midrange test group/midrange_test_data.htm
David Gatti said:
Perhaps, but a number of speaker designers who had choice to use whatever they liked (and I'm not talking cranks, but people with very solid engineering background like Linkwitz and Phil Vafiadis), picked Excels for midrange duty. It can't be just a coincidence.
Even from LDSG site :
"This has been recommended as arguably the best midrange or mid/bass of its size" (W15)
All said, I have to admit that I prefer voicing of ATC-75S's in friend's ATC 100's but get this - for piano, violins and like ! For actual voices (especially female sopranos) and horns I prefer my i66 ( two W11 as mids). Go figure.
Bratislav
I haven't seen anything related to it in the DIY world, but the B&W FST drivers sound gorgeous. I've never heard them except in a B&W speaker though. Aluminum cones tend to be a little more consitent. There are some horrible paper cones, far worse than any aluminum cone, there are some good paper cones out there too.
David Gatti said:
All the way up to 1.4k W15 is competitive with the other two slightly better performers in the "linear" dist. test. and about the best below.
In the linear test W15 has in general 10dB reduction in distortion compared to the other drivers... that is not a small difference and is absolutely audible. -50dB is about 0.3% while -40dB is about 1%. One driver had -30dB which is 3%.
/Peter
Rick,
In my experience having used Scanspeak carbons; Seas Excel magnesiums; Accuton Ceramics; Yamaha Berylliums; the Heil AMT; Focals and a myriad of other cones & domes - the ultimate midrange unit for intelligible voicing in AV use would have to be the humble Fostex FE206E (from Madisound @ $84ea !!).
I use my Fostex 206E in a 20L BR box modelled on WINISD to roll off around 85 Hz, in combination with a subwoofer. The electrostatic-like imaging and clarity of speech (and other TV sounds) is amazing on the Fostex's ...you will hear a mouse fart at 100 yards on these speakers!
Regards,
Steve M.
In my experience having used Scanspeak carbons; Seas Excel magnesiums; Accuton Ceramics; Yamaha Berylliums; the Heil AMT; Focals and a myriad of other cones & domes - the ultimate midrange unit for intelligible voicing in AV use would have to be the humble Fostex FE206E (from Madisound @ $84ea !!).
I use my Fostex 206E in a 20L BR box modelled on WINISD to roll off around 85 Hz, in combination with a subwoofer. The electrostatic-like imaging and clarity of speech (and other TV sounds) is amazing on the Fostex's ...you will hear a mouse fart at 100 yards on these speakers!
Regards,
Steve M.
Pan said:
I tend towards speaker sthat i call mostly full-range... a single driver from 80-250 Hz reaching up to at least 6-10k. Help on the bottom or top if necessary. Active on the bottom, maybe as high as 3rd order low pass, typically 1st order or nothing on the FR at the top and the bottom, 1st or 2nd order on the top... depends on the system (i have quite a few going thru, and things are constantly evolving). Of course those XOs only count if there is a helper top or bottom.
dave
posted by David Gatti :
I think a small & light poly or paper cone is the best material for the midrange (voice) region. Metal cones are probably better suited to the lower frequency range (<500 Hz).
I think you’re right. While male voice fundamentals average 130 Hz (female 210 Hz), and some people average about 90 Hz, it’s 500 – 3000 Hz that’s important for intelligibility.
For music a lot of energy is often in 100 – 500, so for that I’d go metal cones.
Steve
I think the Fostex 206E would be excellent. Being 8 inch (vs 4.5 in) it would start to beam much lower, but do you know at what frequency the whizzer cuts in?
It also needs a bigger box, but though Vas is 55 litres, you only use 20 litres?
Maybe Xmax is low (1.5 mm) in part as a trade-off for the nice efficiency (96 dB)?
Tempting. You’ve heard a lot, though none mentioned seemed to be other top quality paper mid cones.
It seems you get more for your money with the Fostex.
- The CSS with XBL2 has triple Xmax, but with higher Sd of Fostex, Vd is close to equal.
- Although it’s a crude indicator, while BL of the Fostex is about 50% higher, the Mms of the Fostex is 15 g, triple that of the CSS. So I suspect the CSS may have more detail.
- The Fostex is 10 dB more efficient!
I think I’d go for the Fostex for music (much higher efficiency) and AV including sound effects, but probably still stick with the CSS for speech (much lower Mms) and a more compact box.
Dave
how would you compare the 206E to the CSS?
I think a small & light poly or paper cone is the best material for the midrange (voice) region. Metal cones are probably better suited to the lower frequency range (<500 Hz).
I think you’re right. While male voice fundamentals average 130 Hz (female 210 Hz), and some people average about 90 Hz, it’s 500 – 3000 Hz that’s important for intelligibility.
For music a lot of energy is often in 100 – 500, so for that I’d go metal cones.
Steve
I think the Fostex 206E would be excellent. Being 8 inch (vs 4.5 in) it would start to beam much lower, but do you know at what frequency the whizzer cuts in?
It also needs a bigger box, but though Vas is 55 litres, you only use 20 litres?
Maybe Xmax is low (1.5 mm) in part as a trade-off for the nice efficiency (96 dB)?
Tempting. You’ve heard a lot, though none mentioned seemed to be other top quality paper mid cones.
It seems you get more for your money with the Fostex.
- The CSS with XBL2 has triple Xmax, but with higher Sd of Fostex, Vd is close to equal.
- Although it’s a crude indicator, while BL of the Fostex is about 50% higher, the Mms of the Fostex is 15 g, triple that of the CSS. So I suspect the CSS may have more detail.
- The Fostex is 10 dB more efficient!
I think I’d go for the Fostex for music (much higher efficiency) and AV including sound effects, but probably still stick with the CSS for speech (much lower Mms) and a more compact box.
Dave
how would you compare the 206E to the CSS?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.