FFT, IFFT is not limited to MP. The impulse response defines the system, be it MP or not. FFT-IFFT is simple a tool used to go between frequency and time domain. FFT, IFFT also apply to responses that are the sum or product of different responses, as the sum of multiple reflections. I discuss this here about 1/2 way down the page, starting with equation 9. You don't have to understand the math, just read the text. Dipole_modesA with conclusion, "Thus we see that the impulse response at any point in the room is the convolution
of the impulse response of the source with an impulse train that is the sum of the impulse responses of the transfer functions between source and listening position for each mode." [I.e. and impulse train that is the sum of the impulses from each reflection, or of each mode.]
Ha! That does look familiar.
Your project inspired me to get a miniDSP, having a lot of fun with that processor.
It's been a while since I've touched FFT - you are right, of course, and the analysis you did should be a must read for anyone really trying to understand acoustics in a room !
Particularly interesting was the part about which type of source has potentially the best in-room transient response when we know the response is not MP.
Here, you name two choices:
- a source exciting the fewest modes, when one ignores other considerations (like power requirements), that is: a dipole aligned to the room's main axis. This intuitively copes well with Linkwitz's (unidimensional) simulations showing that reaching steady-state for a modal peak takes a lot longer than between modes
- a source in the immediate proximity of the listener
Now there are at least two lines of thought:
1. the "who cares" camp, as advocated by Dr. Geddes on the argument that transient response at LF is inaudible anyway once you uniformly "fill" the room with multiple sources (exciting as many modes as possible) and you only ever hear steady-state due to ear's integration time.
*It would be interesting to analytically look further into the case of multiple sources. Your Eq. 14 becomes a sum of convolutions between each source's impulse response and its respective train of impulses associated to the transfer functions for each mode. It's hard to draw conclusions, but one can actually imagine that by EQ-ing and delaying each of the sources one can ultimately make impulse contributions from multiple sources into a certain mode cancel each other (isn't that what actually happens also for a dipole ? like the impulse contributions of the 2 monopoles along the non-longitudinal modes cancel each other). If that was the case, then multiple sources _might_ also achieve superior transient response.
2. the "better safe than sorry" camp, which would sleep better knowing they have a theoretical optimal in-room transient response.
For camp 2, maybe a combination of the two kind of "good transient" sources might be a solution: dipoles aligned to room's axis + woofer monopole(s) under/to the sides of the listening couch. Of course, this would make it a "multiple sub" approach in the end, but a particularly interesting one.
bzfcocon,
What about a double bass array? In the best case, it should achieve even better response than a dipole. Dipole's bass wave will still bounce around the room, if less than in case of a monopole. DBA's wave will hit the listener exactly once and will be absorbed completely behind him.
I have heard neither, mind
Just trying to understand transient response better.A DBA or SBA (at the end of the day also a multiple source approach) is definitely an interesting option if you can afford the cost and space (and have an understanding wife and/or landlord). It could be more difficult to integrate with the mains though, especially if the mains are dipoles (for controlled directivity in the mid-bass)
One can also consider a pair of bass dipoles in front + a monopole wall in the back working on the same principle as a DBA.
One can also consider a pair of bass dipoles in front + a monopole wall in the back working on the same principle as a DBA.
Well, there will a backwave from dipoles reflecting from a front wall. Maybe it's better to have a cardioid bass in front (possibly by combining a monopole woofer and dipole mids) and a sealed sub in the rear flush to the wall?
Some people report good results from just having two subs, one in front, one in the back.
Apparently I am the only one here who has actually heard the LX minis.
I spent a nice afternoon with a pair and would have to pronounce them as
fairly spectacular.
Seems unlikely NP as all of the reasons so far posted that they could not possibly work!
Unknowing noob that I am I ordered the plans before I realized they were fundamentally flawed.Seems unlikely NP as all of the reasons so far posted that they could not possibly work!
SL could probably save himself much time if he just consulted with the members here before embarking on a new speaker design.
I have my LXmini setup up/running now and can agree with Nelson's evaluation.
Dave.
The LXmin has a transition from rather monopole response below 700Hz, through a brief cardiod region and then into the dipole response above 1.5kHz (at least to the best of that driver's abilities). This seems to go against the belief that even power response and directivity is important, since these will be significantly different above and below the LXmini crossover point. Was that sacrificed in order to provide a small 2-driver system or is this some shift in the paradigm?
I'm curious about this because I am currently designing the next iteration of my "medium width open baffle" system that I hope to bring to the NorCal DIY audio meeting in October. The currently design is a 3-way system, designed with the woofer in a medium sized open baffle, the midrange operating nude, and the tweeter in/on a separate small baffle. It should be able to provide a dipole-like response from below 100Hz to 20kHz based on my driver measurements and some ongoing modeling. If a dipole response pattern below 500Hz is not so important I could change the woofer from OB to sealed omni and reduce the form factor considerably. For this reason I am interested in the tradeoff that the LXmini seems to represent, or the latest thinking on response patterns at various frequencies.
I'm curious about this because I am currently designing the next iteration of my "medium width open baffle" system that I hope to bring to the NorCal DIY audio meeting in October. The currently design is a 3-way system, designed with the woofer in a medium sized open baffle, the midrange operating nude, and the tweeter in/on a separate small baffle. It should be able to provide a dipole-like response from below 100Hz to 20kHz based on my driver measurements and some ongoing modeling. If a dipole response pattern below 500Hz is not so important I could change the woofer from OB to sealed omni and reduce the form factor considerably. For this reason I am interested in the tradeoff that the LXmini seems to represent, or the latest thinking on response patterns at various frequencies.
The LXmini and the LX521 are both relying on drivers with cones made of ... woven polypropylene with excellent internal damping ... (as on the SEAS product pdf). Can anyone comment on these drivers, and on this specific membrane material, e.g. in terms of stored energy? SL has done some beautyful research about driver distortion and stored energy some years ago, so I would be very interested to know how these new drivers compare to his own work as in Midrange distortion test
Nelson,
what do you mean with "fairly spectacular"?
I'm owning a kind of Plutos. Would the LXmini be a better speaker than Pluto or not?
Maybe you experienced the totally free from the speakers effect i had on the bigger LX.
The SEAS unit is certainly a bending wave transducer.
It is just a bit more ridgid then the Manger and breaks up at a higher frequency.
Because there is not much amplitude with typical music the bending wave action has low enough distortion.
A perfect piston has no distortion as such ( with a perfect motor and suspension ) but we humans are superbly deaf for that kind of distortion.
The Coclea is a single ended bending wave sensor in the oposite direction.
Both bending wave generator and bending wave sensor are not reziprocal or we whould hear a never ending echo.
The SEAS unit is certainly a bending wave transducer.
It is just a bit more ridgid then the Manger and breaks up at a higher frequency.
Because there is not much amplitude with typical music the bending wave action has low enough distortion.
A perfect piston has no distortion as such ( with a perfect motor and suspension ) but we humans are superbly deaf for that kind of distortion.
The Coclea is a single ended bending wave sensor in the oposite direction.
Both bending wave generator and bending wave sensor are not reziprocal or we whould hear a never ending echo.
SL could probably save himself much time if he just consulted with the members here before embarking on a new speaker design.
I have my LXmini setup up/running now and can agree with Nelson's evaluation.
Dave.
Hi,
No subjective lacke of treble ?... And mid-bas impact & transcient (due to the lacke of trebles) ?
Apparently I am the only one here who has actually heard the LX minis.
I spent a nice afternoon with a pair and would have to pronounce them as
fairly spectacular.
I will call this the PASS SMASH haha!
If it's good enough for him then they are good enough for me!
Lebiu,
I've only heard the Plutos at Burning Amp a few years back, and the mini's not at all But the choice should be easy enough - the Plutos need to be well away from walls on all sides, the Mini's can be located closer to the front wall. Pick the one that suits the room.
Bill
In audio there is often lots of petty jealousy and put downs, Kinda like the saying that "the fights in academia are so vicious because there's so little at stake".
I think it's laudable that people such as Joachim Gerhard and Nelson Pass are willing to say such complimentary things about another persons work re the LX 521c
http://www.diyaudio.com/forums/vend...aker-inspired-linkwitz-lx521.html#post3842016
and the LXmini speaker series here.
I think it's laudable that people such as Joachim Gerhard and Nelson Pass are willing to say such complimentary things about another persons work re the LX 521c
http://www.diyaudio.com/forums/vend...aker-inspired-linkwitz-lx521.html#post3842016
and the LXmini speaker series here.
SL considers the Pluto obsolete at this point, so a person embarking on a new project should choose the LXmini. It sounds better top to bottom than the Pluto and seems somewhat less position-sensitive (as SL advertises)...at least in my room. The choice is fairly clear....IMHO.
Cheers,
Dave.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.