hi Paul and mainframe99,
Regarding the 'compression distortion': briskly correctly explained the problem above. The concern is that the excursion of the driver will modulate the box volume and thus cause a nonlinear air compliance. However, this nonlinear spring will simply add to the drivers own typically nonlinear suspension and the net effect will still typically be domniated by the drivers typically progressive suspension. Such Kms(x) disortion mostly causes harmonic disortion down towards fs and below (where it is not much audible) and very little IMD higher up (see our blog post Low frequency harmonic distortion is almost inaudible. So what’s the point of low distortion drivers? - PURIFI). Again, this mechanism comes from changing the box volume and has nothing to do woth the drriver design.
A related issue is that the driver surround deforms in a non linear way due to the high back pressure in a small selaed box - essentially a nonlinear modulation of the the drivers effective radiating area Sd. This causes a substantial modulaiton of the air spring stiffness (stiff ness of the piston is proportional to Sd squared).
For this back pressued induced Sd modulation, our Sd neuttral surround is much more robust to back pressure than a normal half roll so in that way the PTT6.5X is more suited for small sealed boxes than most other drivers.
In the end, everyone can measure the distortion coming out of a driver in a sealed box and that is what counts. Mark Bushinsky has published numerous such measurements in this forum including IMD and the PTT6.5X.
Cheers,
Lars
Regarding the 'compression distortion': briskly correctly explained the problem above. The concern is that the excursion of the driver will modulate the box volume and thus cause a nonlinear air compliance. However, this nonlinear spring will simply add to the drivers own typically nonlinear suspension and the net effect will still typically be domniated by the drivers typically progressive suspension. Such Kms(x) disortion mostly causes harmonic disortion down towards fs and below (where it is not much audible) and very little IMD higher up (see our blog post Low frequency harmonic distortion is almost inaudible. So what’s the point of low distortion drivers? - PURIFI). Again, this mechanism comes from changing the box volume and has nothing to do woth the drriver design.
A related issue is that the driver surround deforms in a non linear way due to the high back pressure in a small selaed box - essentially a nonlinear modulation of the the drivers effective radiating area Sd. This causes a substantial modulaiton of the air spring stiffness (stiff ness of the piston is proportional to Sd squared).
For this back pressued induced Sd modulation, our Sd neuttral surround is much more robust to back pressure than a normal half roll so in that way the PTT6.5X is more suited for small sealed boxes than most other drivers.
In the end, everyone can measure the distortion coming out of a driver in a sealed box and that is what counts. Mark Bushinsky has published numerous such measurements in this forum including IMD and the PTT6.5X.
Cheers,
Lars
Is there any difference in size of a woofer, when considering the acoustics impedance? I was of the thought that a smaller woofer has a very difficult job to push the air in the same manner as a bigger driver. I'm a bit in doubt whether exursion can compensate for radiation area.
yes, the acoutic impedance changes with woofer size. However, the point is that the far field on axis SPL is only a function of the air volume acceleration produced by the piston/cone. This means that it does not matter whether we use a large short stroke woofer or a smaller long stroke woofer as long as they give the same volume displacement/acceleration. The issue with long stroke has previously been that IMD was high. The PTT/PURIFI woofers have drastically reduced this IMD. The advantage of a smaller woofer is then that dispersion is better than for a large woofer and so is the midrange response. Cheers /Lars
Thank you Lars 
In my design, I aim for a slow roll-off below around 80hz to make an easier mix with my multi-sub system, so that might save me some space in the main speakers. But - what happens when I cross my woofer at around 400hz to my midrange?
Maybe this is why 8" woofer are so popular - because they can both go high enough to reach a smaller midrange, but still play low enough without audible distortion. hmmm... still lots of things to try out
In my design, I aim for a slow roll-off below around 80hz to make an easier mix with my multi-sub system, so that might save me some space in the main speakers. But - what happens when I cross my woofer at around 400hz to my midrange?
Maybe this is why 8" woofer are so popular - because they can both go high enough to reach a smaller midrange, but still play low enough without audible distortion. hmmm... still lots of things to try out
I think there’s lot of solutions; not a unique one.
For instance, the multiple subwoofer approach is aimed at getting an in-room bass response that is less variable of seating position eg. Home theatre. But as Todd Welti of Harman explained, this is completely unnecessary if you’re optimising for just one person/one seat- you only need 1 subwoofer if you want to tune for your own listening position!
But if you have speakers that can reach down to, say, 50Hz, or 40Hz; which the Purifi 6.5X is capable of doing; as Erin Harrison has shown (extended (40Hz to 3200Hz) multi tone compression tests):
100dB-
Purifi Audio PTT6.5W04-01A 6.5 Inch Woofer Review
then it’s already playing down into the sub-bass. You can thus try it as them as part of your multi-sub solution. Let them play full range, and try crossing your subs to them.
Trying to crossover at a fixed 80Hz to blend with your subwoofers may pose a challenge in itself; but forcing a constraint on yourself.
I think being flexible in your approach and trying frequencies between 50-120Hz is a good idea.
I personally find DSP solutions eg Dirac or Accourate invaluable for smoothing out frequencies below the Schroeder frequency, but manual solutions also work well if you’ve got basic measurement equipment.
As for the question; what’s the good woofer for 80Hz to 400Hz? again I think that depends on how loud or cleanly you want to play. Plenty of 5” “subwoofers” can do 80-400Hz; but so can 10” pro midbasses. I don’t think there’s anything special about the ubiquity of 8” midwoofers; apart from it’s an easily band for them to do; with only very basic motor and soft parts design.
Lars’ 8” midwoofer with 10mm x-max and 90dB/2.83V sensitivity will be one to be looking towards; because there’s not many equivalents…
For instance, the multiple subwoofer approach is aimed at getting an in-room bass response that is less variable of seating position eg. Home theatre. But as Todd Welti of Harman explained, this is completely unnecessary if you’re optimising for just one person/one seat- you only need 1 subwoofer if you want to tune for your own listening position!
But if you have speakers that can reach down to, say, 50Hz, or 40Hz; which the Purifi 6.5X is capable of doing; as Erin Harrison has shown (extended (40Hz to 3200Hz) multi tone compression tests):
100dB-
Purifi Audio PTT6.5W04-01A 6.5 Inch Woofer Review
then it’s already playing down into the sub-bass. You can thus try it as them as part of your multi-sub solution. Let them play full range, and try crossing your subs to them.
Trying to crossover at a fixed 80Hz to blend with your subwoofers may pose a challenge in itself; but forcing a constraint on yourself.
I think being flexible in your approach and trying frequencies between 50-120Hz is a good idea.
I personally find DSP solutions eg Dirac or Accourate invaluable for smoothing out frequencies below the Schroeder frequency, but manual solutions also work well if you’ve got basic measurement equipment.
As for the question; what’s the good woofer for 80Hz to 400Hz? again I think that depends on how loud or cleanly you want to play. Plenty of 5” “subwoofers” can do 80-400Hz; but so can 10” pro midbasses. I don’t think there’s anything special about the ubiquity of 8” midwoofers; apart from it’s an easily band for them to do; with only very basic motor and soft parts design.
Lars’ 8” midwoofer with 10mm x-max and 90dB/2.83V sensitivity will be one to be looking towards; because there’s not many equivalents…
Last edited:
I have a 10 channel preamp with build in dsp for all channels and 6 x 200W and 4 x 500w - all in 8 ohm (2 ohm capable ). I also have software, calibrated microphone and pro soundcard.
To me, it's also nice to be able to walk around my apartment and have smooth and even bass all over. Distributed bass also lowers the amount of stress on each subwoofer a lot.
Still.... I would need 2 purifi woofers per side to keep up sensitivity.
How audible is distortion at around 150hz in a 4 way system?
Last edited:
@piotr z, now that’s an idea, nice 2cuft floor standers with two Purify 8” 90dB/W/m woofers doing 30hz to 750hz, bliesma 3” dome (choose your flavour) up to 3khz and then a T25 (flavour to match mid), all low distortion and a high efficiency design, what more could you want but will not be cheap even using the al mid dome and al tweeter my guess is the drivers will cost you $1500+ per speaker.
Since everyone else seems to be asking Lars Risbo for the transducers they want for Christmas, I'd like to know if there are any plans for midrange drivers with higher efficiency, more in the range of PA.
Why not just move the midrange corner down a bit? The Bliesma midrange seems suited to a 400-500Hz transition, depending on the particular x-over and SPL target.
Cut this out as there is no need/sense for that in such design and You'll end up with what i said before.
Last edited:
The textreme has an advantage of lesser 2nd order harmonics - compared to getting the purifi to 400 - 500 hz.
Wow! You're leaving a lot of critical music out with that lineup! Add a super tweeter and at least a 12" and 15". I think you'll be missing the range in which feeling and intuition live, which is so crucial to the music.
imho, ymmv
Hmm... apparently IMD and THD is meaningless. So many ways to look at it:
Geddes on Distortion perception
Geddes on Distortion perception
I find that thread to be quite misleading. It is a combination of what seems to be correct information about perception, combined with some attempt to make driver design less important. We have heard the story many times, "we can just EQ the same frequency response, and nobody will hear the difference".
Some attempts has been made to lower distortion by several manufacturers. The method was to make drivers more linear by modifying motor and suspension to cancel each other out, and end up with the lowest measured THD, only to discover that IMD (in those rare cases where it was actually measured) was made worse, and the perceived sound did not improve. One of the pitfalls of this method is that the suspension linearity hardly affects higher frequencies at all, but the motor still has the same non linearity. But since the higher THD normally occurs at low frequencies, it seems to work well. Another problem is that the suspension is only a part of the compliance. The box design will affect the frequency range where the suspension plays an important role, and how far the excursion goes as a function of power and frequency, and since the non linearity of the motor is a function of both excursion, current and polarity at the same time, the result will only be valid for the driver used with the box it was mesured in to begin with.
It is of great importance to be able to improve specific parameters on drivers individually before drawing any conclusions that "this or that does not matter". It is not always so easy to know what part of the measured data you are actually reacting to when listening.
Some attempts has been made to lower distortion by several manufacturers. The method was to make drivers more linear by modifying motor and suspension to cancel each other out, and end up with the lowest measured THD, only to discover that IMD (in those rare cases where it was actually measured) was made worse, and the perceived sound did not improve. One of the pitfalls of this method is that the suspension linearity hardly affects higher frequencies at all, but the motor still has the same non linearity. But since the higher THD normally occurs at low frequencies, it seems to work well. Another problem is that the suspension is only a part of the compliance. The box design will affect the frequency range where the suspension plays an important role, and how far the excursion goes as a function of power and frequency, and since the non linearity of the motor is a function of both excursion, current and polarity at the same time, the result will only be valid for the driver used with the box it was mesured in to begin with.
It is of great importance to be able to improve specific parameters on drivers individually before drawing any conclusions that "this or that does not matter". It is not always so easy to know what part of the measured data you are actually reacting to when listening.
how low is low distortion and at what level will distortion become a problem to you?