'Topless' U-baffles

scorpion · « **on:** 9 Dec 2006, 10:03 pm »

Ever since MJK answered my question about air cavity resonances between the 12" parallell backwings in his OB-design I have been
thinking over his answer, http://www.audiocircle.com/index.php?topic=32919.110

For the 'regular U-baffle (Box with no back) at least with dimensions covering one or two units, 12" in my case, there is no doubt that you
will have a well accented resonance coinciding with the lengt of the U-baffle. In the same time H-baffles do no have as pronounced rersonance
as U-baffles. But in both cases these are measurable with simple RS Sound Level Meters. I do not think that you will need more sophisticated equipment.

The intention here is to start a discussion about what happens if we allow another degree of freedom, that is to open up the top.
Does this fundamentally change the situation ? Or do we have to calculate carefully with our dimensions ? As I see it if we can have more knowledge
in this respect it could give us important guides to OB-design especially with wings.

I, myself as a layman is not the man to answer these questions, but I think there are members with good theoretical and pratical knowledge that can contribute
a lot in trying to answer these questions.

/Erling

mcgsxr · « **Reply #1 on:** 9 Dec 2006, 10:08 pm »

A good question, and having built an H baffle with neither top nor bottom, strictly for bass, I can tell you that it works fine.

I am sure that a proper H baffle with both top and bottom would produce different bass, but the H I used with my 12 inch fosgate in it worked fine.

I would guess that the baffle is "smaller" effectively speaking, but with careful consideration for overall design, I am sure a workable compromise could be sorted out.

scorpion · « **Reply #2 on:** 9 Dec 2006, 10:15 pm »

Mark,

Yes, this is a good illustration, 'topless' but not 'bottomless'.

/Erling

mcgsxr · « **Reply #3 on:** 9 Dec 2006, 10:41 pm »

Right, floor provides the bottom, correct.

JohninCR · « **Reply #4 on:** 10 Dec 2006, 02:59 am »

Scorpion,

If it's only in the augmenter's section, resonances are easily cured with some stuffing
without ill effects. If you run U's for your main driver, then you have to contend with
1/4 wave as well as standing wave resonances. MJK threw me for a loop when he said
angled sides couldn't prevent resonances, because I've built a number of resonance free
U shaped cabs with main drivers within the cavity. I've also built open backed straight
pipes out of wood for main drivers that required no stuffing to eliminate resonance (pm
me if interested).

To avoid resonances, I never sweep the sides back 90 degrees and I always use slightly
different depth sides, even if they are tapered. I also start wings as close to the driver
cutout as possible. Since I started building winged OB's this way, the only time I have
resonance problems is with the cabs backed up very close to wall, which results in a
hollow resonant sound, apparently due to reflections between the wall and the cavity, but
if that placement is required some stuffing in the cavity works fine.

I have tried some of these U shaped (more V shaped) cabs without any top cap, but
the only time I didn't like the sound better with a top was an OB line array. There's something
about the shortest distance being over the top that is bothersome to my ears, but it's hard
to explain.

johnk... · « **Reply #5 on:** 10 Dec 2006, 05:22 pm »

Not exactly relevant to topless U's but you may find this of interest.

http://www.musicanddesign.com/NaO-II-U-frame.html

It's about how to design a U-frame woofer and how to tune the damping of the 1/4 wave resonance for optimum performance.

Simply removing the top of the U will not only alter the resonance behavior but also significantly impace the low frequency performance.

John k...

scorpion · « **Reply #6 on:** 12 Dec 2006, 01:57 pm »

Thank you John,

I am familiar with your work on U-baffles and have been playing around quite a bit with your and MJK's spreadsheet for
U- and H-baffles.

My interest in this question with regard to bassreproduction is that I would like to have some kind of general dipole-bass
of as small dimensions as possible beeing usable up to 250-300 Hz. But still going down real low into the 30ies and with as few things as possible to correct for.

You wrote in the 'Edge diffraction' thread and published measurements comparing flat and winged baffle rear response.
Bass was quite augmented with the winged baffle but I could not trace any resonance phenomena in your measurements.
So I think there is a case for my question.

/Erling

JohninCR · « **Reply #7 on:** 12 Dec 2006, 04:31 pm »

Erling,

Some brought that Jussi brought up in his OB thread at DIYAudio is that there is a
sonic difference (not just response difference) in the bass region between U's and
dipoles. It relates to placement and the difference in reflections due to the very
different polar response of the two. He mentioned that U's tend to sound better
with placement near the front wall, and that dipoles sound better than U's if placement
is near the side walls. Playing around with placement recently of some of my U's, I'd
have to agree with Jussi.

johnk... · « **Reply #8 on:** 15 Dec 2006, 12:39 am »

Not to contradice Jussi, but the difficulty in comparing U's and conventional dipole woofer is that a dipole woofer is pretty much a dipole woofer, whether it be on a flat baffle, and H frame, W frame, etc. But the performance and polar response of a U is highly dependent on the damping of the rear wave. Either system will have a 6dB increase in SPL if place near a side wall.

Rudolf · « **Reply #9 on:** 24 Dec 2006, 03:07 pm »

Had to get out of the wifes way for some time today. So I started to clean up my harddisk a bit and found this:

It´s a comparison between a H baffle with opened and closed top. Mic position was in the red circle (center of back opening). Black curve is with closed top, red with open top (as shown). Interpretion is up to everybody himself.

scorpion · « **Reply #10 on:** 26 Dec 2006, 07:21 pm »

Rudolf,

I have seen that there has been some dispute over the GBS 515's at diy-Audio. However the measurements are
innovative, it is a pity that SL didn't do the same for his Protos, but we will go further.

/Erling

johnk... · « **Reply #11 on:** 26 Dec 2006, 10:44 pm »

I don't know why I posted this in another thread, but I'll repeat it here.

"I was reading the thread on Open top U-frames and made some quick measurements comparing the rear radiation from a flat baffle dipole (11" wide) and one with 8.5" deep side wings, straight back. These are pretty deep wings, but the point is that once you start adding wings things become more complex, for better or worse. Green is rear w/o wings, red with wings."

.

DYI, the baffle and wings are about 40" tall.

JohninCR · « **Reply #12 on:** 27 Dec 2006, 02:27 am »

John,

What's causing the 10db higher SPL in the rear radiation with wings below 300hz?
Are they very nearfield measurements? That's the kind of graph I would expect
on the front side.

Davey · « **Reply #13 on:** 27 Dec 2006, 05:49 am »

Seems pretty obvious to me. It's because "D" is larger.

The front and back responses will look somewhat different, but the "winged" baffle will show more SPL because the cancellation from front/back is less effective.

Cheers,

Davey.

JohninCR · « **Reply #14 on:** 27 Dec 2006, 07:41 am »

Davey,

That was my knee-jerk reaction too, until I read JohnK's post again and realized
he said measured in the rear. I'm hoping John actually measured in front, otherwise
I am confused. I like those measurements if they are from the front, because they
would really make the case for wings on bass augmenters (+10db essentially for free
up to 300hz is about equal to 3 times as many drivers).

Measuring from the backside 90 degree wings don't have any effect on D. That's
why a perfect U-Baffle has the null to the rear, since the front and rear wave travel
distances are closest to equal along the rear axis. Wings do act as a waveguide
containing the rear radiation for their length, which would distort very nearfield
measurements.

Davey · « **Reply #15 on:** 27 Dec 2006, 04:20 pm »

No, I didn't knee-jerk. I read it a couple of times.

I am assuming these measurements are from John's NaO Mini system with and without the wings. I also believe the measurements were not very close or nearfield because of the roll-off in both measurements which is approximately 6db/octave.

Whether the measurements are taken in front or back (back in this case) "D" is larger for either measurement when the wings are attached and thus the SPL output will be larger and the "Fequal" point changed.

I'm sure John will elaborate the next time he stops by this forum.

Cheers,

Davey.

johnk... · « **Reply #16 on:** 27 Dec 2006, 11:20 pm »

The only point of the measurement was to see if the wings changes the radiation impedance seen by the back side of the driver. The winged measurements were taken in the exit plane of the wings. I forget where the unwinged measurement was taken. But the point here is that there is an obvious resonance at about 550 Hz in the winged rear SPL which is absent from the unwinged response. The conclusion is that there must be a change in radiation impedance due to the wings even though the top is opened. I'll try to look at it again and maybe with different length wings, but right now I'm rather busy with other projects.

JohninCR · « **Reply #17 on:** 28 Dec 2006, 03:47 am »

Davey,

Nearfield explains the increased bass SPL primarily because the output has
been constrained, so there's naturally more pressure. "D" is the difference
between the front and rear waves' travel distance to the listening position.
eg You can cut the front of a typical H frame baffle all the way back to
the driver mounting baffle, and D is unchanged. The radiation pattern
changes from dipole toward cardiod.

wrt The link to SL's discussion of D that you gave someone at DIYaudio,
one of SL's diagrams is misleading, the one for an H baffle. While in a
symmetrical H baffle D happens to equal the depth, the proper measure of D
is from the back the driver around the rear edge and back forward to the
point even with the baffle plane. That single diagram has prevented most
from understanding how folded baffles affect bass. That's why H baffles
are common and U's aren't, yet H's are much more compromised, starting
with double the floor space in a wasted manner.

Davey · « **Reply #18 on:** 28 Dec 2006, 07:27 am »

John,

With all due respect, I think you need to read the information on Siegfried's and John K's pages more carefully.

And start learning to take some measurements to confirm your theories. SoundEasy (one option) is a really nice program with extensive testing suite and John K. is a wizard with it so you will have expert help.

Cheers,

Davey.

Rudolf · « **Reply #19 on:** 28 Dec 2006, 02:56 pm »

Davey,
I am not sure where you found John to need reading SL more carefully. I believe John is right in regarding SLs H baffle diagram as misleading:

In part "a" and "b" the distance "D" is following the cone positions (source of "+" and "-"). SL should have followed this in "c" too (like the red pathway I drew in his diagram).

Instead he connected "D" with the front and back of the H baffle (and transfering "+" and "-" to those positions too). And this, I believe, has been a source of some misinterpretation.

John is quite right in his remark "You can cut the front of a typical H frame baffle all the way back to
the driver mounting baffle, and D is unchanged. The radiation pattern changes from dipole toward cardiod." This implies, of course, that you are only looking at the front of the baffle.

Where I can´t follow John is his comment "That's why H baffles are common and U's aren't, yet H's are much more compromised, starting with double the floor space in a wasted manner." Cardioids and dipoles are different kettle of fish and it is no use to plainly regard one better than the other. It strictly depends on the individual design goals.

Rudolf