[pedroskova]

Rudolf,

I have a large U-frame speakers and have been playing with rear damping. When it is not stuffed (e.g. lightly filled) the response is simular to typical U-frame response with peaks and dips. When it is fully stuffed (packed and compressed) the response is basically the same as a monopole.

The interesing finding is that when it is partially stuffed, at higher frequencies the response is like the monopole and at lower frequencies the response is like dipole. This is of course due to the LP filter from the damping material. This is not unexpected.

I have the opening to my U-frames covered with 2" compressed rock wool, and it does indeed form a very efficient low pass filter, especially in regards to the U-frame's pipe resonance... at this point, my current notch filter is probably overkill.  It's still a dipole in its passband since the rock wool is more or less transparent in the lower frequencies.

[JeffB]

Rudolf posted this link above: http://www.linkwitzlab.com/models.htm#A
Section H - Psycho-acoustic 3 kHz dip
is quite interesting.
This is the first time I have ever seen an argument that a completely flat response might not be optimal.
I have often heard the claim that pro-audio studio monitors don't sound good because they strive for very flat response,
whereas, home speakers strive to be more pleasant sounding.  I always thought this was nonsense, and that everybody strived for flat response, except for maybe a few makes of British speakers.

[gainphile2]

Correct, but notice that in his latest design (Orion) SL abandoned that idea, there is no such notch in Orion.

Btw. Does anyone know the Frequency, Q, and depth of Orion's midrange notch filter?

I found it interesting that SL never mention to equalise the dipole peak, only the "driver basket resonator" such as in Phoenix. This is very different from NaO where clearly John is targeting the dipole peak.

[panomaniac]

The Gradient is a slick looking design.  But I've always wondered about the side firing woofer.  I've tried it some and have had a hard time with the crossover.  The dirvers is 90 degs out, after all.

Maybe it depends on how low you cross...

[panomaniac]

Section H - Psycho-acoustic 3 kHz dip

Funny, I was EQing yesterday and thought to try the famous "BBC Dip".  Wanted to know exactly what is it, I ran across this from the Harbeth site:

"According to Harbeth's founder, who worked at the BBC during the time that this psychoacoustic effect was being explored, the primary benefit this little dip gave was in masking of defects in the early plastic cone drive units available in the 1960's. A spin-off benefit was that it appeared to move the sound stage backwards away from the studio manager who was sitting rather closer to the speakers in the cramped control room than he would ideally wish for."

See here: http://www.harbeth.co.uk/faq/#13

True? I don't know....

[HiFiNutNut]

There are quite a number of implementations of the BBC dip. Different people (and companies) could tell you different things. Some say at 2.5kHz, some say 3.5kHz, some say starting at 1kHz with a 3dB dip at 2.8kHz and ending at 6kHz, etc.

It really depends on the type of music and how the recordings were made. I have done numerous experiments on my speakers changing the depth and width of the BBC dip.

For classical music, I found a response profile close to very flat is preferable, especially with opera. Any dips (and of course, peaks) would colour the sound of a soprano's voice.

For an audiophile sound that is easy on the ears, a mild narrow band BBC dip is preferable. You may find that the background is quieter, the sound is smoother and cleaner, the music is less busier, it is more airy, and the background is pushed further.

Overdoing it, the music notes lose all outlines, music becomes too soft, attacks are missing, and musicians would say that your system is rather inaccurate.

For me, I do BBC dip based on the response of my drivers. I prefer a narrow 1.5dB dip centred at 2.8kHz, only starting to drop down at 1.8kHz, and ends at 4kHz. This is way less than the recommendations from the old school BBC dip, but sounds a lot more balanced.

Regards,
Bill

[panomaniac]

Thanks Bill, I think I'll try it!

[Rudolf]

Funny, I was EQing yesterday and thought to try the famous "BBC Dip".  Wanted to know exactly what is it, I ran across this from the Harbeth site: ...

True? I don't know....

Funny how the explanation of Harbeth is almost the contrary of what Blauert found about directional bands:
http://www.uni-koeln.de/phil-fak/muwi/fricke/103blauert.pdf
Look for Fig.2 . I have added the same diagram in another presentation that looks more readable to me:

Green is to the front, red to the back, blue to the top. Y-axis is probability of detection of the sound source direction.

[weidok]

Rudolf,

Commenting also on your latest answer. I think that combining bass H-baffles with smaller mid/treble-baffles, as we have discussed earlier, is a very
promising way to go.

/Erling

I am playing with paramyd U baffle's  and  i am very happy with the sound 



I call it the LOBO NoBox
woofer  Peerless SLS 315
FR Fostex FE 167 E
i am planning to test a Tangband  W5-1611 SA

here is the drawing

[panomaniac]

Hey, I know that woofer!   Same one I'm using now.

Nice looking rig, how are you doing the crossover?

[hum4god]

another example for a good looking narrow baffle, the lyngdorf dp-1



it kind of inverts the idea of pyramid (upside down )
low freq. are handled by two subs.

[Telstar]

Very interesting thread.
I was also considering a triangular (more or less rounded) shape, but my goal was just to extend bass response having the bass drivers (AE OB15s) close to the floor, and with wings.

Could you run a simulation of a 12" driver in a 40cm wide baffle? That would be my midrange. I'm not sure how wide should be my baffle in that point.

[panomaniac]

The Lyngdorf is a good sounding rig.  Heard them at RMAF.  Very small baffle.  Corner subs.

[Rudolf]

Regarding Lyngdorf and Lobo:
At the driver position baffle width should be no larger than the wavelength of the highest frequency you want to transmit. I know that this is very hard to do with a tweeter and 10-20 kHz, but baffles which are much wider than 2" at the tweeter position are far from optimal. In that case one should at least position the tweeter as near to the top edge of the baffle as possible. I know that you can´t achieve this with a FR driver and it´s whizzer cone - and its ragged frequency response off axis proves my argument. But regarding the tweeter the Lyngdorf is a clear example of "form (very much) over function"!!

[Rudolf]

Could you run a simulation of a 12" driver in a 40cm wide baffle? That would be my midrange. I'm not sure how wide should be my baffle in that point.

This is a 12" driver in a 110 cm tall baffle, horizontally centered at 80 cm height:



You see that
- wider baffles give more bass
- smaller baffles reduce the "peakyness" of the first dipole peak
- a 12" driver´s off axis response isn´t much good above 1 kHz regardless of baffle size

[Telstar]

You see that
- wider baffles give more bass
- smaller baffles reduce the "peakyness" of the first dipole peak
- a 12" driver´s off axis response isn´t much good above 1 kHz regardless of baffle size

Thank you, Rudolf. I tried to use the software, but i dont speak german and it's hard to guess.
The difference between 30 and 40cm baffle width is not much and things start to get bad at 50cm.

To accommodate easier room requirements, I think to go for 30-35 cm

Could you also run a simulation for 35cm? The 12" driver is centered at 62,5cm not at 80cm btw.
If you can also set the driver specs in this software, those are Qt=0.6 and Fs=50hz with no crossover.

Cheers

[JeffB]

Hi Rudolf,

Questions:
1) Is the main cause of the "peakyness" related to reflections off the baffle or related to the backwave wrapping around the baffle?
It has been my assumption that it is reflections off the baffle.  Hence, it seems to me, that the problem would exist identically on closed boxes.
You indicated earlier that the effect is much less on closed boxes.  The simple fact that the effect gets worse on wider baffle has me virtually convinced that it is the reflections on the baffle, because it is now harder for the backwave to get around the edge to interfere. 

2) Having to have a narrow 1" or 2" baffle for the highest frequencies seems to ultimately lead to a rather complicated multi-way design of 4-way or more.
Thus, I wonder if there is way to cheat this "peakyness" effect using waveguides?

3) And if a waveguide would work is there any hope of using a coaxial where the midrange would act as a waveguide for the tweeter?
Do you have any thoughts on how something like a P Audio
http://www.loudspeakersplus.com/html/paudiocoax.html
BM8 CXA or
BM12 CXHA
would work.

[Rudolf]

Thank you, Rudolf. I tried to use the software, but i dont speak german and it's hard to guess.

Telstar,
that´s why I don´t recommend to download Boxsim for non-German speakers. It is packed with settings windows where you need to understand the German terms. But the baffle algorithms are exactly the same as in EDGE! I only use Boxsim because I get the horizontal and vertical 30° and 60° response angle in a single run too.

The difference between 30 and 40cm baffle width is not much and things start to get bad at 50cm.

That´s why I keep telling, that you get best results with an effective baffle width that is no larger than twice the cone diameter.

Could you also run a simulation for 35cm? The 12" driver is centered at 62,5cm not at 80cm btw.

This looks VERY nice to me!!

If you can also set the driver specs in this software, those are Qt=0.6 and Fs=50hz with no crossover.

This simulation is with an ideal driver (linear response from 0 to infinity). I can do simulations for specific drivers if I have the specific parameter set for Boxsim (which includes more than the TSP). It will be easier for you to estimate  the Q induced drop off than for me to build even a minimised parameter set. Took me about three hours to get the "ideal driver" model running correctly.

[JeffB]

Well, I just found graphs of the P Audio drivers, they look a little scary to me in the upper frequencies.
http://www.p-audio.co.uk/products/db_product_1_6_bm-8cxa.htm
and
http://www.p-audio.co.uk/products/db_product_1_6_bm-12cxha.htm

[Rudolf]

1) Is the main cause of the "peakyness" related to reflections off the baffle or related to the backwave wrapping around the baffle?
It has been my assumption that it is reflections off the baffle.  Hence, it seems to me, that the problem would exist identically on closed boxes.
You indicated earlier that the effect is much less on closed boxes. The simple fact that the effect gets worse on wider baffle has me virtually convinced that it is the reflections on the baffle, because it is now harder for the backwave to get around the edge to interfere. 

Try to see it this way:
There is NO reflection from any point on the baffle plane. It is exclusively from the baffle (or other) edges. This baffle edge diffraction happens to every closed box baffle. For an OB you have to include the edge diffraction from the second source on the rear side. Both diffractions neutralise in the baffle plane and add up on the driver axis. So all effects double in intensity for an OB compared to CB. You can see this effect very easily in EDGE when checking the "Open baffle" box.

The simple fact that the effect gets worse on wider baffle has me virtually convinced that it is the reflections on the baffle, because it is now harder for the backwave to get around the edge to interfere.

All dimensional effects on a baffle MUST be seen in relation to the wavelength involved. A large baffle which may be large for 1 kHz is small for the 10 Hz wave. So you did not want to say "the effect gets worse on wider baffle", but you really intended to say "the effect gets worse for higher frequencies".
And it gets worse if the baffle is wide in relation to the cone diameter. You can simulate this effect very easily in EDGE too.

2) Having to have a narrow 1" or 2" baffle for the highest frequencies seems to ultimately lead to a rather complicated multi-way design of 4-way or more.Thus, I wonder if there is way to cheat this "peakyness" effect using waveguides?

That is true. But fortunately our ear is not very sensitive at the highest frequencies - both WRT SPL level and radiation pattern.

3) And if a waveguide would work is there any hope of using a coaxial where the midrange would act as a waveguide for the tweeter?

There should always be hope in life.
But a midrange cone is a bad waveguide and a good waveguide can´t act as a midrange cone. The only acceptable combination (concluded from measurements I have seen) was the Thiel SCS. And that only 30° off axis. On axis response is as ragged as with any better fullrange whizzer cone.