[Rudolf]

MJKs project for an OB and H Frame speaker system http://www.quarter-wave.com/Project08/Project08.html encourages me to show my own system with a comparable topology:


It still is a test bed so looks don´t have priority yet.
The H frame has been discussed here: http://www.audiocircle.com/index.php?topic=39094.0
For the system description it should be sufficient to show the basic response diagram of the H frame:


Black line is the dipole near field response. Red curve is near field with an added 6dB/oct LP filter at 200Hz. At 1 m distance and with an added 300 Hz LP it "levels out" to the purple response. Since the measurements are not gated or smoothed, they already show much room influence at 1 m distance.

With its Fs~40 Hz and Qts=0.5 the driver does not reach down as far as the Goldwood in MJKs project. Even with EQ he misses 20 Hz by more than one octave – a strong argument for MJKs strategy to employ drivers with Qts>1.

The other drivers are the Monacor SPH 176 http://www.monacor.de/typo3/index.php?id=84&L=1&artid=4084&spr=EN&typ=full and two pairs of Monacor DT-25N http://www.monacor.de/typo3/index.php?id=84&L=1&artid=3574&spr=EN&typ=full



Active crossover and EQ are done with a Behringer DCX 2496 feeding into the 6 power amps of an Onkyo TX-SR505E AV-Receiver (seen behind the left speaker). XO frequencies are 250 Hz and just below 2 kHz.

In the first implementation midrange driver and tweeter had been mounted on the mid axis of a 40 cm wide OB. A simulation with BoxSim  resulted in this non-satisfying radiation pattern from  0.8-3 kHz:


Cutting off 10 cm off the baffle at the right side resulted in a much better response to the left – and a worse (?) to the right:


The main contribution to a smooth response was NOT the offset of the driver, but the reduction of the baffle width to nearly double the driver diameter (as discussed by JohnK in http://www.musicanddesign.com/Dipoles_and_open_baffles.html)
In many cases such a small OB is in conflict with the required SPL at the lower end of the frequency range, but since Linkwitz could do it that way in his Orion it should be within reach.

Doubling the tweeter is no must, but has resulted in an obviously better 3D representation. I first added some other-make tweeter to the back that I had lying around. But only when I used the same tweeter for front and back the result became convincing. I will have to put the rear tweeter another 2.5 cm to the back to get it symmetrical to the midrange voice coil, like SL did. It never stops …

My stereo basis is 2 m wide and I listen at 2 m distance to the baffles. They are toed-in so the driver axis crosses 40 cm in front of my ears. In the back the driver axis aims into the room corner (this seems the same as in MJKs room). At the listening position I get this response for the left speaker:



Measurement is not calibrated, so all SPL is relative. Don´t look at anything above 8 kHz - my mic does not work there. The dip at 250 Hz is the floor reflection, any dips below are room dependant.

I hope this will lead some people to enter the world of OB and H frame too. Welcome! 

Rudolf

[Russell Dawkins]

Congratulations, Rudolf.

Does it sound as nice as it measures?

This is just laziness on my part, but before I look deeper into it for myself, could you briefly describe the advantage of H baffle over plain open baffle of the same frontal dimension. I am presuming it results in lowering the frequency at which front-to-back cancellation occurs and is beneficial in that it is inherently more stable physically, being deeper - but are there downsides, apart from cavity resonances if crossed at too high a frequency?

[Rudolf]

With the H frame bass is really clear and effortless. Music like F.U.B.B. from Wishbone Ash or "Eight Miles High" live from the Byrds "Untitled" album show off every note the bass player does. I love it!
Talking about more contemporary music: In "Confessions.. " Madonna is constantly changing the width of the acoustical panorama. This has never been more convincingly conveyed to me than with the rearward tweeters. Its like someone stretching the sound to the sides of the room in one single moment.

Some time ago I did a comparison between different types of dipoles:http://www.audiocircle.com/index.php?topic=49940.0
The first diagram in that thread compares a H frame which is as deep as it is wide (and high) with an OB of the same frontal dimension. Its dipole roll-off frequency is half that of the OB. So in the dipole roll-off area the H frame is at least 6 dB more efficient than the OB.
One should limit the (total) depth of the H frame to its width/height. This way the cavity resonance - if it ever shows up - can easily be hidden in the Xover to the midrange or fullrande driver. The H frame retains the figure-8 radiation pattern all along its usable range - which U- or W frames don´t.
If we are talking about below 200-250 Hz I really don´t see any downsides apart from aesthetics: A plain open baffle may look more elegant than a cube with an upright plank on top. 

[scorpion]

We are certainly on the same track when considering economic low bass performance.
Then  we will see a lot of different mid-high combinations like yours to perform more or less
to our expectations. 

/Erling

[Rudolf]

... we will see a lot of different mid-high combinations ...

For sure!
While the choice of suitable mid- and high-Q bass drivers is somewhat limited, there is a lot of possibilities above. How about a pyramid on top of the H frame with a midrange driver on the bottom and a tweeter in the peak? Or lets revive the Gradient 1 with a H frame as the basis.

[gainphile2]

Looks nice albeit bulky (in the photo).

Have you considered the acoustic alignment between the drivers, or at least delay the mid/tweet using allpas delay circuit ?

[Rudolf]

Looks nice albeit bulky (in the photo).

Well it is bulky in reality too. And that far out in the room there is no chance to hide it.

Have you considered the acoustic alignment between the drivers, or at least delay the mid/tweet using allpas delay circuit ?

I tried and measured it from the listening position. Could see (and hear) no difference. The distance between midrange baffle and H frame baffle is just a bit more than 1/8 lambda at the x-over frequency. I have asked MJK for his opinion regarding the baffle position: http://www.audiocircle.com/index.php?topic=59619.msg530336#msg530336

Rudolf

[weidok]

How about a pyramid on top of the H frame with a midrange driver on the bottom and a tweeter in the peak?

I build a pyramyd U frame...

[Rudolf]

Hi weidok,

I see that you utilize a SLS 12. How loud does it get in your U frame? Sufficient for a 22 m² room (music only and neighbour-compatible SPL)? Would you prefer a second SLS per side?

Rudolf

[weidok]

Hi weidok,

I see that you utilize a SLS 12. How loud does it get in your U frame? Sufficient for a 22 m² room (music only and neighbour-compatible SPL)? Would you prefer a second SLS per side?

Rudolf

Hi Rudolf One SLS 12 preforms very well in our living room (40m2)
I drive them with teh Omnes Audio 2.1 II  4 channel T amp with
the Class T PSU powersupply and a 15V AC toroidal transformer.


link power supply http://www.audiodigit.com/index.php?section=89