Hello,
I'm a fan of PS Audio's Ask Paul section, which gets uploaded to yt almost every day. This thread is based on one of his answers to the question - what midrange driver to use, where he advised to utilize something fast.
Based on your experience, how important the speed of a driver would be to acquire a real like (natural) tonal reproduction and what does the speed depend on?
As far as I'm into the subject, I see a few theories:
1. Mounting the driver on an open baffle - removing the back wave and back pressure from interfering
2. Finding a driver with a good ratio of moving mass to force factor BL (magnet strength). More magnet power for moving the lighter membrane, hence some acceleration term, as I saw in some manufacturer's specs.
3. Adding a second driver to work with in conjunction, splitting the woofer workload in two, hence splitting the driver excursion
4. Using a higher end woofer?
For example, investing in several of these and putting them in an MTM.
Midrange
By doing all that and something more, would there ever be a way to compare a midrange membrane driver to a ribbon, AMT or electrostats?
I'm a fan of PS Audio's Ask Paul section, which gets uploaded to yt almost every day. This thread is based on one of his answers to the question - what midrange driver to use, where he advised to utilize something fast.
Based on your experience, how important the speed of a driver would be to acquire a real like (natural) tonal reproduction and what does the speed depend on?
As far as I'm into the subject, I see a few theories:
1. Mounting the driver on an open baffle - removing the back wave and back pressure from interfering
2. Finding a driver with a good ratio of moving mass to force factor BL (magnet strength). More magnet power for moving the lighter membrane, hence some acceleration term, as I saw in some manufacturer's specs.
3. Adding a second driver to work with in conjunction, splitting the woofer workload in two, hence splitting the driver excursion
4. Using a higher end woofer?
For example, investing in several of these and putting them in an MTM.
Midrange
By doing all that and something more, would there ever be a way to compare a midrange membrane driver to a ribbon, AMT or electrostats?
Well, I wouldn't know. Either way, I'm trying to find a driver that has neutral/natural sound signature and realize what exactly to look for.
I've heard a lot of conventional ported speakers, some of them with a high price tag, but for most of them, you kind of always know that a voice is originating from a speaker. One exception was the mark Daniel with an AMT tweeter, which was a different experience, voices sounded very real.
I've heard a lot of conventional ported speakers, some of them with a high price tag, but for most of them, you kind of always know that a voice is originating from a speaker. One exception was the mark Daniel with an AMT tweeter, which was a different experience, voices sounded very real.
Dan Wiggins has shown us that all fast means is that a driver has good high frequency response. So in other words you want a midTweteer not a midrange.
I suspect what you mean is that you would like a driver that starts and stops quickly to more closely follow the input signature
I am not a big fan of most tweeters, i believe that in an MTM one should XO such that teh XO frequency is less than a quarter wavelength of the centre-to-centre of the 2 Ms.
I have my favourites.
dave
I suspect what you mean is that you would like a driver that starts and stops quickly to more closely follow the input signature
I am not a big fan of most tweeters, i believe that in an MTM one should XO such that teh XO frequency is less than a quarter wavelength of the centre-to-centre of the 2 Ms.
I have my favourites.
dave
I'd say efficiency is the #1 parameter to a subjectively "fast" result: most particularly a dynamically fast on-set.
Still, you can have a "fast" result and lack decay - and that tends to sound a little "zippy" and "dry": providing emphasis only to the dynamic attack from the signal.
It's tough to get a "fast" result with good or excellent decay in a dynamic driver.
That AT driver would likely be a good match with a ribbon tweeter like the RAAL drivers (or Fountek ribbons).
IF however you really want a similar sound (to a ribbon, AMT, or planar), then something like this would be more suitable:
GRS PT5010-8 10" Planar Mid/Tweeter 8 Ohm
Still, you can have a "fast" result and lack decay - and that tends to sound a little "zippy" and "dry": providing emphasis only to the dynamic attack from the signal.
It's tough to get a "fast" result with good or excellent decay in a dynamic driver.
That AT driver would likely be a good match with a ribbon tweeter like the RAAL drivers (or Fountek ribbons).
IF however you really want a similar sound (to a ribbon, AMT, or planar), then something like this would be more suitable:
GRS PT5010-8 10" Planar Mid/Tweeter 8 Ohm
Yeah, I searched for it as well.. and came up with squat. 😱
I think the replies were at least as far back as 2007.. 😱 😀
Basically Dan was trying to debunk the whole "fast" bass thing as it didn't really work for the products he was producing.
Obviously it's a transient character that people are describing, not the ability of the driver to produce higher freq.s. (..or actually oscillate more rapidly).
What I've found of Dan Wiggins' writing is an article about woofer speed. It does answer my question to some part, where an inductance is playing the key part in woofer going and stopping process. Isn't that the whole point of an ideal woofer, one that follows the input as close as possible?
Higher inductance did alter transient response, response to a signal was weaker and reaction to a signal stopping was delayed. What is left unexplained to me is why did the higher inductance leave a different result on frequency response graph only above 1 kHz. It was interesting as well to see that added mass only affected the low frequency sensitivity. A lot of new information to think about, that's for sure.
However, I found nothing about driver speed having to do just with high frequency range. Did you draw that conclusion from the graph in his writing? If so, why would it make any difference if I had a midtweeter or midrange, provided that both have the same inductance?
Why do you say that AT midrange mentioned above would go well with a ribbon?
I'm having a problem with room acoustics, to the point that I want something radically different, like dipoles. Been reading a lot about Linkwitz's speakers design and I'm thinking about DIY something with an open baffle with consistent horizontal response. So far I've seen some interesting midrange of 5 inch (AT from above or Satori), the idea is to use them from around 200 to 4000 Hz, so one driver to cover most where the music mostly is. Any critics and advice are most welcome.
Higher inductance did alter transient response, response to a signal was weaker and reaction to a signal stopping was delayed. What is left unexplained to me is why did the higher inductance leave a different result on frequency response graph only above 1 kHz. It was interesting as well to see that added mass only affected the low frequency sensitivity. A lot of new information to think about, that's for sure.
However, I found nothing about driver speed having to do just with high frequency range. Did you draw that conclusion from the graph in his writing? If so, why would it make any difference if I had a midtweeter or midrange, provided that both have the same inductance?
Why do you say that AT midrange mentioned above would go well with a ribbon?
I'm having a problem with room acoustics, to the point that I want something radically different, like dipoles. Been reading a lot about Linkwitz's speakers design and I'm thinking about DIY something with an open baffle with consistent horizontal response. So far I've seen some interesting midrange of 5 inch (AT from above or Satori), the idea is to use them from around 200 to 4000 Hz, so one driver to cover most where the music mostly is. Any critics and advice are most welcome.
Last edited:
Hello
the word fast is hard to understand in the context : any driver will be fast enough if the amp permits it. It's not really the driver alone but the couple amp/driver that matters. So if one is concerned enough by this, he should choose an active system and a good designed amp for each driver (understand in relation to what the driver must do, i.e. frequency range)
As you pointed out, the excursion will depend of the Sd at iso Spl output and the frequencies involved. The "little" driver with more Xmax will go faster (than the big) to reach the frequency needed at the needed spl. But the thermal compression can collapse that dynamic behavior (and the power response level waited)
Imho the only area where one can talk about fastness is the frequency range. I find the T&S datas of a tweeter harder to undrstand at my limited understanding level than a mid.
The 3) item of your list makes sense and the choice of the length of the voice coil as well I will add. Plus you want certainly avoid thermal conpression (higher efficienty helps as the design of the voice coil, motor, cooling, etc) . So it's a whole, you can have a sealed with the right Qtc 0.55 with a well damped rear wave.
Also the feeling of speed may depends of the filter slope and physical layout of the drivers in the loudspeakers and the tweeter "quality". If fast word is taken for transcient.
I may be wrong but one should find the good trade offs for the mid drivers, i.e. T&S with low Le, Low Qts, Low Qes numbers and a good motor with enough efficienty not to suffer of thermal compression and a not too long voice coil length...
Then , if the amp is good, the design is good (filter and all), the driver (or the loudspeaker as it's a mid that needs a tweeter) will play fast enough i.e. plays what is on the disc. Maybe the better way to arrive at such a result is to play Full range around the mid (i.e. a mid that plays low and high enough )
Short anwer : an ESL63 or ESL57 🙂
short middle answer : focus on the amp quality.
But hey a mid alone is not enough... a loudspeaker is a whole and full of trade offs and you already knowing that !
My very two cents, I'm also involved by such feeling as I consider transcient and timing an important of the music. But notice anyway the music is coming to your ears at the same speed (more or less 344 m/s ). I dunno if that helps though.
Near ESL, I experience electo magnet speaker have something on them about subjective "fastness" -better motion control ?- but as ESL are much more complicated and expensive, and at the end the power supply also matters a lot... which is also called an "amp" for a classic driver !
the word fast is hard to understand in the context : any driver will be fast enough if the amp permits it. It's not really the driver alone but the couple amp/driver that matters. So if one is concerned enough by this, he should choose an active system and a good designed amp for each driver (understand in relation to what the driver must do, i.e. frequency range)
As you pointed out, the excursion will depend of the Sd at iso Spl output and the frequencies involved. The "little" driver with more Xmax will go faster (than the big) to reach the frequency needed at the needed spl. But the thermal compression can collapse that dynamic behavior (and the power response level waited)
Imho the only area where one can talk about fastness is the frequency range. I find the T&S datas of a tweeter harder to undrstand at my limited understanding level than a mid.
The 3) item of your list makes sense and the choice of the length of the voice coil as well I will add. Plus you want certainly avoid thermal conpression (higher efficienty helps as the design of the voice coil, motor, cooling, etc) . So it's a whole, you can have a sealed with the right Qtc 0.55 with a well damped rear wave.
Also the feeling of speed may depends of the filter slope and physical layout of the drivers in the loudspeakers and the tweeter "quality". If fast word is taken for transcient.
I may be wrong but one should find the good trade offs for the mid drivers, i.e. T&S with low Le, Low Qts, Low Qes numbers and a good motor with enough efficienty not to suffer of thermal compression and a not too long voice coil length...
Then , if the amp is good, the design is good (filter and all), the driver (or the loudspeaker as it's a mid that needs a tweeter) will play fast enough i.e. plays what is on the disc. Maybe the better way to arrive at such a result is to play Full range around the mid (i.e. a mid that plays low and high enough )
Short anwer : an ESL63 or ESL57 🙂
short middle answer : focus on the amp quality.
But hey a mid alone is not enough... a loudspeaker is a whole and full of trade offs and you already knowing that !
My very two cents, I'm also involved by such feeling as I consider transcient and timing an important of the music. But notice anyway the music is coming to your ears at the same speed (more or less 344 m/s ). I dunno if that helps though.
Near ESL, I experience electo magnet speaker have something on them about subjective "fastness" -better motion control ?- but as ESL are much more complicated and expensive, and at the end the power supply also matters a lot... which is also called an "amp" for a classic driver !
Last edited:
Hi Allen,
Could you please inputt about that subject in relation to the concept of phase, group delay, filters all that stranges things, please ? 🙂
Could you please inputt about that subject in relation to the concept of phase, group delay, filters all that stranges things, please ? 🙂
Ordinarily, frequency response is derived from transient response, phase is derived from frequency response, group delay is derived from phase. They're different ways of saying the same thing.
That's the song of electricity 😀.. pardon : of the wavelengths !
... stranger and stranger ! I wisch I was a Jedi too !
... stranger and stranger ! I wisch I was a Jedi too !
Last edited:
All of this makes me rethink the whole room acoustics problem I'm trying to avoid and makes me buy random boxes and be done with it 😀
Active crossovers are a no go, since adding several expensive DACs and perhaps a tube preamp for each driver is a costly endeavour. On the other hand I am not an expert in a field of amp electronics, so I'll just buy a decent amplifier, maybe two for added power (but feeding from the samepreamp and DAC). Mini dsp with its analogue to digital to analogue conversion makes my hair rise.
It is a compromise, so I'm trying to achieve a good result in 50-4000Hz area. Hence the thread for finding a decent midrange/midtweeter. What you mentioned building around a full range, it's a good proposition. I thought of 4 inch mid with 10 inch woofer for sub 300-400 range.
If moving mass isn't as important in transient behaviour ("speed"), and if the voice coil inductance has an effect only on upper range mids (2k Hz +), why does everyone brag about ribbon midranges?
I do recall hearing an AMT from Mark Daniels (crossed about 900hz), that had such an real sound to it. Maybe I just never heard a 4-5 inch midtweeter with decent specs. Either way, what makes a driver neutral and 'real' sounding?
I've been listening to magnetoplanar mid and tweeter for 7 years now, but never had the feeling they are fast. As well paper/composite/aluminium woofers and mids sound just the same.
This is in diy 3-4 way speakers. But in two-ways the bad habits of drivers come apparent.
This is in diy 3-4 way speakers. But in two-ways the bad habits of drivers come apparent.
Fast / Speed?
When I see these and similar terms used to describe an asset a given driver has, it is an understandable use of these words.
I think a better, more accurate term is resolution. The one thing that all really good speakers do/share along with tonal balance. And the better the speaker, the wider the high resolution bandwidth. Magniplaner mid bass comes to mind. Set up correctly, the mid bass is loaded with sonic-spatial details and correct tactile weight. You get the feeling you are hearing all there is to hear from a given recording. When you experience that feeling from say 40 Hz up to and beyond 10Khz you have a special speaker system. Just my $0.02 worth.
When I see these and similar terms used to describe an asset a given driver has, it is an understandable use of these words.
I think a better, more accurate term is resolution. The one thing that all really good speakers do/share along with tonal balance. And the better the speaker, the wider the high resolution bandwidth. Magniplaner mid bass comes to mind. Set up correctly, the mid bass is loaded with sonic-spatial details and correct tactile weight. You get the feeling you are hearing all there is to hear from a given recording. When you experience that feeling from say 40 Hz up to and beyond 10Khz you have a special speaker system. Just my $0.02 worth.
I use Allen Wright’s term DDR. The ability of a device to claerly reproduce the small stuff even in the precense of the big stuff.
This is like looking at a body of water. FR is the surface, we would like it mrror flat, vut what is happening at the bottom?
dave
Not 'everyone' does. I don't like most ribbons for example. Some exceptions. And those that do are not necessarily talking about the somewhat fallacious concept of 'speed'. There are many reasons why some people like some drive units. Many may be real, some only in their own heads.
Look at it this way. Assuming moving coil drive units operating within their piston bandwidth, then if it is capable of accurately producing, say, a 2KHz signal, then by definition it can move quickly enough to reproduce all frequencies below that, which have a longer wavelength (i.e. are 'slower'). That applies down to the mass corner frequency (take as 2Fs/effective Qts where you account for any series R in circuit). Below that, you have a system response of driver & any relevant box load, which will affect the transient response in that BW.
Caveat to this all are the piston bandwidths, and the assumption there are no resonant modes at all through that region through suspension issues, reflections etc. It breaks down, or at least becomes more complex when you have a drive unit that uses TL modes to extend the upper end of its usable BW.
Last edited: