Mid Bass Horn Design 80-800 Hz

[Steve Mac]

Sorry in advance for the long-winded post. Blame it on bfish.

Filled one more side wall yesterday...two more to go...adhesive is slow to dry...but it's looking good and maybe in less than 2 weeks the side walls will be filled with mounting boards attached and ready.

But I thought to review some comments made in the beginning thread which were interesting but I didn't really respond to. Basically it was a great post by "bfish".

OK, here's the comment by bfish:
Forgot one more item of importance.

The largest (H or W) dimension of the inner unit must be smaller
than the highest WL to be produced by the outer to prevent shadowing.
This rules out using a 511 at 800Hz (don't have to do the math, "been there").

This means the inner horn will likely need to be run right down to its' cutoff,
which is a function of its' dimension. Finding a horn that provides uniform
loading to the desired cutoff, while still maintaining a polar control profile
to match the outer unit may be your biggest challenge, and IMO, is the biggest
flaw of the Engine.

In addition, I was given a link to a JBL pro design of the PD700 series which concerns itself with the same design challenge I'm looking at which is a coaxial approach.

JBL PD700 tech notes

Now, there are several considerations in these tech notes and I directly compared them to this design.
The first problem is reflections off the rear of the compression drivers, back down the throat of the midrange horn, and the aberration in frequency response, and polar response that may result. The second problem is the high-frequency horn acts as obstruction to the proper expansion of the midrange horn and may generate resonances.

Two things they are doing is to accomodate the size of the interior horn in the exterior wall calulation which has been attended in the new design. The second thing they do is to place an acoustic blanket over the inner throat which becomes acoustically absorbent above 700Hz and I figure this accomodates the roll off of the 1.25KHz XO point of the 700 series. They also have a shell around the inner driver which we have plans for in the new design. All in all, except for an acoustic blanket the 700 series and the new coax are very similar in approach.

Now, concerning "shading". There appears a disconnect in at least terminology.
Here's what JBL said in the 700 series tech notes:
Understanding Co-axial ?Shadowing?:
Another design issue in a coaxial system is ?shadowing?. If the percentage of the area of the
midrange horn blocked by the high frequency horn is too large, then shadowing may occur. The effect
causes the midrange horn to behave as two distinct ?cells?, or signal sources. When this occurs, the midrange off-axis response has nulls within the nominal coverage angle. To solve this problem the size of the high frequency horn must be minimized, but must remain large enough to maintain pattern control at the crossover. A delicate design balance must be achieved.

To conclude, JBL maintains that their ratio is 13%(inner/outer dimension), which is 2dB down at various polar locations. The new design is currently lower than this at 10%.


To balance this out with what was described as "shading" I had to research the tech library at JBL until I found this excerpt from part 1 of the Sound System Design Reference Manual:
Diffraction of Sound
Diffraction refers to the bending of sound waves as they move around obstacles. When sound strikes a hard, non-porous obstacle, it may be reflected or diffracted, depending on the size of the obstacle relative to the wavelength. If the obstacle is large compared to the wavelength, it acts as an effective barrier, reflecting most of the sound and casting a substantial ?shadow? behind the object. On the other hand, if it is small compared with the wavelength, sound simply bends around it as if it were not there. This is shown in Figure 1-7

Now, not assuming the above statement is related to the next comment outlined, I decided to get the dimensions and crossover points of the PD 700 series and see how they fit with what was stated anyway:
The largest (H or W) dimension of the inner unit must be smaller than the highest WL to be produced by the outer to prevent shadowing.

15" is the larger dimension of the inner unit of the 700 series. 15"=1.25'::crossover point is 1.25kHz:: u=1130/1250=0.904' at 1.25kHz.
So, 1.25 feet is not smaller than 0.904 feet. Someone needs to reconcile this. I can only think that an aerodynamic chute is not the same as an obstacle, which is why I'm not so sure it is a valid argument. Anyway, calculating the 511 horn yields 565Hz. Utilizing a horn with a superior aspect ratio yields a calculation closer to 700Hz which I believes presents evidence the design is not violating the laws of physics. Even if the laws of physics are wrong.

Anyhow, the link concerning a blanket is interesting. The chute I have designed ends before the driver, which sticks out. I was considering a sock of appropriate material to cover the driver but had not considered covering the 15" speakers at the throat. Thoughts?

<font color="#FFFFFF"><font size="1">[ February 02, 2008 09:21 PM: Message edited by: Steve Mac ]</font></font>

<font color="#FFFFFF" size="1">[ February 02, 2008 09:23 PM: Message edited by: Steve Mac ]</font>

[bfish]

You grokked it well (R. Heinlein).

Never did the math on the JBL. Interesting.

My statement on the inner horn size was merely an application-specific derivation of the diffraction theory. To the best I could see, it was true with my trials in general, only maybe not as cut and dried as the math would indicate. Lots of other stuff going on there too.

The math behind coax horns is way beyond me. Can't hurt to experiment though, and I really think you're on the right track.

Never forget the basic goal...point source. Find a good definition, print it, and paste copies everywhere. Consider how every aspect applies.

I think your inner "shell" should continue rearward to terminate in a point or knife edge. Look inside an old multicell throat.

If your expansion yields a spherical wave, the wave will be perpendicular to the outer wall at any given point. You're blowing a bubble with the big horn. Now you're gonna stick an object in the middle of that bubble that's gonna blow its' own bubble, and those bubbles have to merge coherently.

<font color="#FFFFFF" size="1">[ February 02, 2008 11:33 PM: Message edited by: bfish ]</font>

[bfish]

Being unable to do the math, I have to work from mental images...

So, I'm imagining a 5x3 multicell. Now I remove all the cells but the center one, also leaving the peripheral walls. Now I adjust for the voids between the cells I removed, and I should have the theoretical equivalent of what you're shooting for.

Can ya dig it?

[Steve Mac]

Originally posted by bfish:
Being unable to do the math, I have to work from mental images...

So, I'm imagining a 5x3 multicell. Now I remove all the cells but the center one, also leaving the peripheral walls. Now I adjust for the voids between the cells I removed, and I should have the theoretical equivalent of what you're shooting for.

Can ya dig it?

<font size="2" face="Verdana, Helvetica, sans-serif">Close. When I made accomodations in the expansion of the wedges, which contain all of the "information", I alloted for the outside dimension of the mouth of the 511 back 17"...which is pretty close to covering the whole shebang, horn plus driver.
This picture shows the cab that is planned inset which shows it stopping short of covering the rear driver. The desire was to get easy access to the rear driver.
Inset cab for 500hz horn
The plan was to plaster the metal horn with epoxy or "floor patch? " and slide it into the chute. The rear driver would then be exposed at the back of the chute and could be covered or wrapped with a sock/blanket.

When I described a horn with a better aspect ratio, I was really referring to the 500hz rectagular edgar horn. I grafted this pic from Boyd's 815 cab pics on the late Mr. Eckle's site. It is uncanny how similarly shaped this horn is to this design:

Just imagine the 815's turned sideways.

[bfish]

For some reason, my weak brain thinks having the mid horn mouth behind the big horn mouth is asking for problems. The only commercial example I can think of is the Urei duplex, and the smoother transition it yielded, due reportedly to the extended horn and to a lesser degree, the edge treatment.

On the other hand, the farther forward you bring it, the greater the source offset...

At some point the theory becomes academic, and you've just gotta prototype (or punt). I've spent weeks pondering the theory, but learned more in 3-4 days of tinkering and chirping.

<font color="#FFFFFF" size="1">[ February 03, 2008 11:52 AM: Message edited by: bfish ]</font>

[bfish]

Originally posted by Steve Mac:
A large point source with the imaging of the 15" duplex would surely be something to behold...

<font size="2" face="Verdana, Helvetica, sans-serif">Conversely, a tiny point source the size of a pinhead would have the least issues...

The real challenge ahead will be obtaining any semblance of similarity in polar patterns. While you're not looking for uniform coverage of a theater audience, it would be nice to not have noticeable artifacts as you move around the room.

You're at the point where I chunked my cardboard models in the fire and went fishing...

Forge on.

[bfish]

Here's an idea of one of my earlier attempts;



It had a few things going for it, the near proximty of the rear of the shell minimized reflections. The mouth aspect ratios and percent of area complied with the "rules" you noted from the JBL ref. The discontinuity between the driver cutout and the beginning of the horn was reduced.

On the other hand, the narrow BW, already compromised 828 horn turned to garbage. GM said it would, and it did. Here's where you have a shot, having modified the outer horn to accomodate an inner.

[GM]

Originally posted by Steve Mac:

The thing is, if you give me gallons...then give me pounds, and tell me gross or net please.

<font size="2" face="Verdana, Helvetica, sans-serif">Sheesh! I know better than to post when I'm tired, but sometimes do it anyway and didn't notice the big, bold type at the top that said it was for brass.

Anyway, I agree, I didn't understand the odd mix of units.

GM

[GM]

Originally posted by bfish:
For some reason, my weak brain thinks having the mid horn mouth behind the big horn mouth is asking for problems.

<font size="2" face="Verdana, Helvetica, sans-serif">You're right, the mouth's acoustic impedance will roll off the HF horn's output and be a poor match acoustically to the larger horn and without a pressure wave simulator, then AFAIK it's a trial n' error solution as to the best fore/aft position.

GM

[GM]

Originally posted by Steve Mac:

A large point source with the imaging of the 15" duplex would surely be something to behold.

Because the mouth dimension is increased by ~250 in^2........

<font size="2" face="Verdana, Helvetica, sans-serif">Tom Danley achieved it years ago with his original two way Unity and pushed the directivity to a lower frequency with the three way, which combined with his largest tapped horn is the first system I've ever heard that can reproduce a lightning strike accurately enough that I can't discern the difference and why I'll never waste any more time building another critical listening system for myself that's not of at least the same type/performance.

Hmm, I can tell from the pics you didn't change the flaring to account for the 511, so there will be some anomalies in its response even with the extra mouth area and the 511 in an aero pod, though don't have a clue how audible it will be.

Note that I believe the 511's polar response is a better fit to the big horn's response if turned vertically, so I recommend trying it both ways. It was due to all these issues that I suggested the three flat sided, horizontal woofer layout horn. A pity they wouldn't fit the room space as it would appear as a large point source at ~ the same distance as a M19.

GM