This was like you where reading my minds!
This very thing has made my life "more diffucult" (and more time consuming), not only a bad thing, but I can't do anything half-heartedly, and when I get to a point where I don't have the answer... I get stuck.
All of these things you guys talk about is part of the motivation to create this thread. After all, we're dealing with (arguably) the most important part of the frequency range in our hearing. As they say, everything happens in the midrange and there would be no music without it. Everything hinges around it and when it's bad, everything is bad. I love decent bass and clean, extended treble, but you could restrict these ranges and still have good, intelligible sound.
I can't say anything fundamentally bad about 2 way speakers. There are many good ones out there but they're usually not as good as a well designed, true 3 way in many aspects. The one Achilles heal is always going to be the crossover right smack in the middle of the critical midrange. There are a few exceptions to this, ie. when the LF driver is small and goes clean up to about 3 - 4k without too much beaming. That's a hard one and in most cases, if the LF driver can pull that off, it will otherwise be limited in LF extension. So the best thing to do is use it as a dedicated midrange in a 3 way. At the same time, you could easily argue there are very few 2 ways that have LF extension to 40 hz. This is a minimum requirement for me. I won't build or own speakers with less than 40 hz LF extension, unless they have very good mids. In most cases you can just add a good pair of subs (yes, at least 2) and have a decent fullrange capable system.
When I see a cone driver with low 80s dB sensitivity, I don't even consider it, unless it has some magical midrange quality that sets it apart from the vast ocean of pathetic low sensitivity cone drivers. I haven't come across one of these magical drive units yet. Mid 80s dB sensitivity is my cutoff point on drivers with 1" VCs, as they won't be capable of good dynamic range due to limited thermal power handling. There are a few low sensitivity drivers that can be useful. These are the higher end Morel and Dynaudio cone drivers with huge 3" VCs. These usually can handle a good amount of input level to make up for their low sensitivity. The main problem with these is increased odd order HD from the inherently compromised, highly non-linear motor design. I mainly use those for mids if they have decent linearity.
So we're pretty much left with either a decent large dome midrange or an exceptionally good cone midrange. Horn loaded mids can be an exception, but these have their own gremlins and in most cases, they're simply too big to live with in a typical domestic setting. There are obviously some smaller WG equipped mids that can work well in a typical 3 way.
Anyways, back to dome mids...
I can't say anything fundamentally bad about 2 way speakers. There are many good ones out there but they're usually not as good as a well designed, true 3 way in many aspects. The one Achilles heal is always going to be the crossover right smack in the middle of the critical midrange. There are a few exceptions to this, ie. when the LF driver is small and goes clean up to about 3 - 4k without too much beaming. That's a hard one and in most cases, if the LF driver can pull that off, it will otherwise be limited in LF extension. So the best thing to do is use it as a dedicated midrange in a 3 way. At the same time, you could easily argue there are very few 2 ways that have LF extension to 40 hz. This is a minimum requirement for me. I won't build or own speakers with less than 40 hz LF extension, unless they have very good mids. In most cases you can just add a good pair of subs (yes, at least 2) and have a decent fullrange capable system.
When I see a cone driver with low 80s dB sensitivity, I don't even consider it, unless it has some magical midrange quality that sets it apart from the vast ocean of pathetic low sensitivity cone drivers. I haven't come across one of these magical drive units yet. Mid 80s dB sensitivity is my cutoff point on drivers with 1" VCs, as they won't be capable of good dynamic range due to limited thermal power handling. There are a few low sensitivity drivers that can be useful. These are the higher end Morel and Dynaudio cone drivers with huge 3" VCs. These usually can handle a good amount of input level to make up for their low sensitivity. The main problem with these is increased odd order HD from the inherently compromised, highly non-linear motor design. I mainly use those for mids if they have decent linearity.
So we're pretty much left with either a decent large dome midrange or an exceptionally good cone midrange. Horn loaded mids can be an exception, but these have their own gremlins and in most cases, they're simply too big to live with in a typical domestic setting. There are obviously some smaller WG equipped mids that can work well in a typical 3 way.
Anyways, back to dome mids...
Last edited:
I would find it interesting if you can comment why you feel it is better to have the main speakers go down to 40Hz (I suppose f3) even if using multiple subs (I believe you meant it this way).
If the requirement is simply to reach 40Hz with the subs, at what frequency do you aim crossing over to the subs?
How can we get so nice low end performance, down to about 400hz, with the M74 with a moving mass of only 2,3 grams?
Regular bass driver with a light cone sucks at the deep end...
Regular bass driver with a light cone sucks at the deep end...
@swak I prefer all thr bass cone from the main L+R speakers, so if they can't go that low (at a decent SPL), a subwoofer needs to pick up the slack. I'd aim to cross at about 80 hz,.depending on room acoustics and placement limitations. Even so, I still try to get some overlap on the low end if the room needs it to blend seamlessly, otherwise I'd still have some flexibility of chosing the most appropriate turnover to the sub. The 40 hz rule implies that (without a sub) the main speakers need to go down cleanly to 40 hz with some decent SPL. I'd of course expect subs to extend the FR further down, otherwise there's little point of having them in the first place. I hope that explains it.
@Jonasz Well, from my POV, I'd choose a higher crossover for the M74A/B/T. 400 hz is pushing it too far and just about smack on the driver's Fs. That causes issues and you'll run into excursion limitations. Even if the mid can do this,its still not advised, just from a reliability standpoint alone. You can get a good midbass to match the distortion performance of the M74. There are plenty of midbass drivers out there which will do that and if it can't be done with a single midbass, then you can run multiple units without issue. Thats due to the WL at crossover being long enough to blend the output of two smaller midbass without any significant off axis destructive interference.
@Jonasz Well, from my POV, I'd choose a higher crossover for the M74A/B/T. 400 hz is pushing it too far and just about smack on the driver's Fs. That causes issues and you'll run into excursion limitations. Even if the mid can do this,its still not advised, just from a reliability standpoint alone. You can get a good midbass to match the distortion performance of the M74. There are plenty of midbass drivers out there which will do that and if it can't be done with a single midbass, then you can run multiple units without issue. Thats due to the WL at crossover being long enough to blend the output of two smaller midbass without any significant off axis destructive interference.
Would you share your method? I do struggle with similar difficulties, sometimes. And you seem to be very successful, seeing your completed project(s)!
(Apologies to profiguy for the off topic!)
Another approach that can sound really good in the mid-range is the small wide range driver with helper woofer. It would not be a generally usable monitor, and usually not play very loud, but can be very captivating with detail for smaller ensembles and "girl and guitar".
I have a pair of 20 year old Fostex Fe127e with Planet10 mods, that can be really charming. With high pass on the Fostex plus 10" and 12" dipole woofers contributing bass and baffle-step compensation, it will play "loud enough" for smaller room use.
Interestingly, I found that moving the woofer down to 28"OC of the full range, that the point source charm of the full/wide range driver can be preserved. Closer spacing ends up sounding like a more normal multi-way speaker.
There is a huge amount of overlap in spectrum between the 10" woofer and the 4.5" Fostex; the 10" still having a contribution at 1000hz. It would be good to understand the physics/psychophysics of why the 28" separation make the difference that it does. I don't remember hearing it as a separate source, or if that was percievable, that it seemed "unnatural".
The 10" Faital 10FE200 has amazing mid range tone for an inexpensive pro driver. It is really more of a mid driver than a bass driver.
I have a pair of 20 year old Fostex Fe127e with Planet10 mods, that can be really charming. With high pass on the Fostex plus 10" and 12" dipole woofers contributing bass and baffle-step compensation, it will play "loud enough" for smaller room use.
Interestingly, I found that moving the woofer down to 28"OC of the full range, that the point source charm of the full/wide range driver can be preserved. Closer spacing ends up sounding like a more normal multi-way speaker.
There is a huge amount of overlap in spectrum between the 10" woofer and the 4.5" Fostex; the 10" still having a contribution at 1000hz. It would be good to understand the physics/psychophysics of why the 28" separation make the difference that it does. I don't remember hearing it as a separate source, or if that was percievable, that it seemed "unnatural".
The 10" Faital 10FE200 has amazing mid range tone for an inexpensive pro driver. It is really more of a mid driver than a bass driver.
Last edited:
This has been my XO frequency too between the main speakers and the distributed subs. I think it can also be well applied to open baffle speakers (+ monopole distributed subs).
Your thread inspired me to give the D7608 a try (paired with theTW025A20Mg), and I have a pair on the way. I'm still deciding what to use below the midrange, but I'm leaning toward building a 3-way bookshelf speaker without subs. I might end up choosing a woofer for a bass-reflex enclosure— to make something along the lines of a Harbeth 30.1 or one of the classic Snells, but as a 3-way using the D7608s.
ll
I get tense when working, and find that it is helpful to feel where in my body the tension is, and then to repeatedly remind myself to relax the clenching teeth, or some other tightness.
The work I enjoy the most/feel most relaxed in, is that which is of no great consequence. That which is of more consequence, is most rewarding to me when there is a sense of altruism or service or love involved. My hobby work is best done as a relaxing activity, though I do not always achieve such a good-feeling relaxed state. I think much of my tension is conditioned—it is habit. Awareness, and conscious choice to change the tense habit, helps me in the long run.
I think that anxiousness is a human trait that, when appropriate,keeps us moving towards more successful states of achievement. In its most basic instinctive form, it is about survival. It is good to figure out how necessary any particular activity is for our survival, and to impart less emotional tension to those things that are more about happiness and recreation, like hobbies.
Creativity takes some of the moving ahead energy, and channels it into new forms, which I find very satisfying.
Anxiety is hard to trace to its source(s) through mental effort. When I get stuck or down, doing some large muscle activity that is enjoyable is the best therapy. Getting my body energy going through movement can get me unstuck to an amazing degree.
I have seen engineers, who were not builders, do very good work on their homes. I think that it is largely because they are methodical,and don't charge ahead when they are unsure. I approach automotive mechanical problems this way.
I get tense when working, and find that it is helpful to feel where in my body the tension is, and then to repeatedly remind myself to relax the clenching teeth, or some other tightness.
The work I enjoy the most/feel most relaxed in, is that which is of no great consequence. That which is of more consequence, is most rewarding to me when there is a sense of altruism or service or love involved. My hobby work is best done as a relaxing activity, though I do not always achieve such a good-feeling relaxed state. I think much of my tension is conditioned—it is habit. Awareness, and conscious choice to change the tense habit, helps me in the long run.
I think that anxiousness is a human trait that, when appropriate,keeps us moving towards more successful states of achievement. In its most basic instinctive form, it is about survival. It is good to figure out how necessary any particular activity is for our survival, and to impart less emotional tension to those things that are more about happiness and recreation, like hobbies.
Creativity takes some of the moving ahead energy, and channels it into new forms, which I find very satisfying.
Anxiety is hard to trace to its source(s) through mental effort. When I get stuck or down, doing some large muscle activity that is enjoyable is the best therapy. Getting my body energy going through movement can get me unstuck to an amazing degree.
@swak
I'd use 2x SB23NRXS45-8 per side paralleled in a vented cab, crossing at 650 - 700 hz 2nd order, or 600 hz 3rd order. You'll have more flexibility with the midbass using a shaded filter on the woofer furthest from the mid. BSC will automatically be dealt with. The woofer closest to the mid should be run up the the 600 - 700 hz LP limit, depending on the mid's HP filter.
I'd use 2x SB23NRXS45-8 per side paralleled in a vented cab, crossing at 650 - 700 hz 2nd order, or 600 hz 3rd order. You'll have more flexibility with the midbass using a shaded filter on the woofer furthest from the mid. BSC will automatically be dealt with. The woofer closest to the mid should be run up the the 600 - 700 hz LP limit, depending on the mid's HP filter.
Is the shading on the woofer what is sometimes called a .5 system, as in "3.5"way? This is where the low-pass on the lower woofer is at a very low frequency, in order to extend the bass, and like you indicated not have to apply as much (other) BSC?
Hello !
Here are my speakers.
There are SEAS Excel W22NY003 speakers that are close to the technical specifications of the Triangle speakers, which you'll most likely only find in France.
I'm not claiming anything, but they do the job!
Sensitivity : 89 dB/2 V/ meter >> 4 Ohm
As we know that most amplifiers handle low-impedance electrical loads well, sometimes doubling the power divisible by two to impedance. With 91 dB/2.83 V/meter, we can say that most mid-woofers drivers easily reach the treble level.
However, my tweeters had to be attenuated; they produce 96 dB/2.83 V/meter >> 8 Ohms.
Best regards!
But let's get back to the point!
French expression.
Sincerely,
Laurent
Ps : I plan to use it from 500 to 5000 Hertz in a 3-way passive radiator configuration.
French expression.
Sincerely,
Laurent
Ps : I plan to use it from 500 to 5000 Hertz in a 3-way passive radiator configuration.
Last edited:
I forgot to comment on this post. There's alot of wisdom offered by LO. He's considered by many to be the most practically educated person on speaker design. I studied alot of his theory on individual driver analysis. His resonance theories, how to deal with them, and audibility of specific driver idiosyncrasies are for the most part spot on. I started evaluating new drivers via his method of listening to them on large baffles with full range music and pink noise. I also do this with drivers in free air, specifically LF drivers, to gauge their motor noises and sound coming off the chassis at varying angles and frequencies.
Lastly (and most importantly), he explains that static sine waves reproduced by individual chassis reveal much more than momentary bursts or sweeps typically used by measuring gear. This is a big one for me, as it tells you pretty much everything you'll need to know about successfully implementing a driver in a design. The other big one is resonance filtering and audibility of resonances after the fact.
There are alot of guys who will say resonance peaks are usually linear in nature and can be successfully eliminated without noticeable distortion by the listener. I disagree with this. You can however, sucessfully notch out specific out of band breakup modes if the driver pass band doesn't include them. This is critical with metal dome tweeters and I'll only use them in conjunction with an LCR.
Lower Q, shallow breakup modes are easier to deal with and can sometimes be within the driver's operating range without being annoying to the ears. Again, measurements don't always show the whole picture here.
How does this apply to dome mids? LO states he doesn't particularly like dome mids. That's where I apply the other wisdom I've accumulated to proving this otherwise. By far, the most critical design element of an exceptional sounding dome mid is the rear cavity and its role as part of the driver suspension. If the diaphragm loading isn't near optimal, the distortion will be significant, hurting lower mid performance. The compliance of the air in the chamber has to dominate that of the mechanical suspension (surround).
The rear chamber air compliance also keeps the dome from rocking considerably close to Fs and increases mechanical power handling. The surround is another important element in a mid dome and needs to be narrow enough to suppress rocking, radial and axial resonant modes, but compliant enough to allow reasonably linear dome movement.
All this is why I place so much emphasis and importance on the rear chamber of a dome mid (and tweeter). If this element isn't right, the driver won't perform at its full potential. It will also suffer from other issues indirectly related. This literally makes or breaks the entire midrange driver and it's also the reason why many designers don't use dome mids anymore. They're not easy to implement and there aren't that many excellent examples of midrange domes. The ones that are exceptional need alot more care to incorporate in a system than most cone mids.
When a dome mid is done properly though, it opens up the midrange and exposes all the little details (bad and good) in a recording that normally aren't heard with most other types of mid drivers or fullranges. The mids from a dome can even be better than some planars and electrostats.