Placing an object inside a horn

[peterjohnswindley]

Hi there.

I’ve seen horn loaded designs where there is a large mid range front loaded horn and inside the mouth of that horn there is a high frequency horn (loaded with a compression driver).

My question is, how is the mid range horn designed with an object in its mouth?


I know that wavelengths that are relatively large by comparison with the object will diffract around it and relatively small wavelengths will reflect (in free air, at least). And I realise if the crossover between the mid and high freq horns is at a suitable point, the frequencies that would reflect off the high freq horn would be cut out by the low pass filter on the mid.

However, I don’t know what happens when the object is surrounded by a horn.

Is the horn enlarged and/or shaped to compensate for the volume of the HF horn?

Or does the sound just ‘share’ the air in front of the horn and pass through as if it wasn’t there?

Or something else?

Thanks in advance.

 

[Bigun]

"Is the horn enlarged and/or shaped to compensate for the volume of the HF horn?" - I would assume this if the HF horn is large enough, but many mid-bass horns are compromised enough (too short, mouth too small) that it may not matter.

 

[AllenB]

Lower frequencies may be interested in the overall expansion, but diverting the walls that way might not be liked by the higher frequencies.. nor might the obstacle.

 

[weltersys]

My question is, how is the mid range horn designed with an object in its mouth?

Depends on the designer 😉
In my three-way nested Maltese horn system (circa 1992), the HF horn was enclosed in a box with a pyramid rear section which formed the phase plug for the two-part conical mid range horn.

Both horns used an identical expansion profile.

Is the horn enlarged and/or shaped to compensate for the volume of the HF horn?

It may or may not be enlarged.

Looks like EAW's KF700 series mid horn continues to expand from start to finish:

Lots of diffraction effects in that mid horn, it really didn't "come together" until "Gunness Focusing" (temporal equalization using FIR DSP) reduced the artifacts.

Or does the sound just ‘share’ the air in front of the horn and pass through as if it wasn’t there?

The effects of path obstructions on sound waves are frequency dependent, the lower the frequency, the less effect.

Or something else?

Multiple entrant horns reduce most of the diffraction problems inherent with nested horns while reducing the cabinet volume.

Art

 

[newvirus2008]

My question is, how is the mid range horn designed with an object in its mouth? ... wavelengths that are relatively large by comparison with the object will diffract around it ... and if the crossover between the mid and high freq horns is at a suitable point, the frequencies that would reflect off the high freq horn would be cut out by the low pass filter on the mid.

I think you're correct in guessing that the mid-range frequencies possibly do not interfere (much) with the treble horn inside, due to the larger wavelengths.

Is the horn enlarged and/or shaped to compensate for the volume of the HF horn?

I don't think any adjustments are made to compensate for this effect as it seems to be 'allowed' by the laws of physics.

An example would be the JBL 2192/3/4 series that simply 'combines' the separately available 2392/3/4 mids and 2332/3/4 treble horns (using brackets) to get a family of 2-way 'co-axial' horns.

JBL 2192/3/4

 

[peterjohnswindley]

Thanks to all contributors!

Sorry I haven’t replied until now.

It’s good to see a few other designs I wasn’t aware of.

I think the replies cover my question nicely, thank you.


Would it be cynical of me to suggest these ‘nested’ horn systems might perform 95+% as well as a synergy horn? (That is, fulfill the vast majority of their advantages?).

For the record, I love Danley. If I was a PA hire company here in the UK, they would be my first port of call for investments.

 

[weltersys]

Would it be cynical of me to suggest these ‘nested’ horn systems might perform 95+% as well as a synergy horn? (That is, fulfill the vast majority of their advantages?).

I'm not sure what you mean by "cynical", or how you would evaluate performance.
"Synergy" and before that, "Unity" were Tom Danley's trade names for his multiple entry horns (MEH).
There are many different manufacturers now selling MEH, some of them preceded Danley's.
Companies like EAW, EV, and JBL have opted for MEH over nested horns in their current product lines.

Hard to put a percentage on the performance advantage of not having a disruption in the center of the mid horn, but I'd put that as over 5% all by itself.

In terms of cabinet volume utilization, MEH can have ~5-50% (or more) performance advantage over nested horns. The reduction in volume and weight reduce manufacture, transportation, storage, deployment and rigging costs.
In terms of new design tooling costs, MEH can have as much as a 100% performance advantage over nested horns.

Having designed and built both nested horn systems and MEHs, it's obvious to me that the MEH has significant performance advantages.
Had I been aware of the MEH concept in 1992, I certainly would not have used nested horns.

Art

 

[chromenuts]

Interesting that this topic popped up.

I’ve been experimenting with a front horn/open baffle hybrid project for a while.

More recently I added a supertweeter I had to see if it would give me some desired clarity and detail I felt was lacking.

Because of the size of the horn it was difficult to find a good location for the supertweeter. I ended up just propping them at the edge of the horn on top of the OB bass section and angled them as best I could.

I was thinking this was a lousy solution and wondering how I could make it better by getting the supertweeters closer or at least along the same axis as the drivers in the horns.

I was brainstorming about adding some sort of low profile wedge shaped clamp on shelf for them somewhere along the center line of the horn near the mouth.

Roughly located as in this pic:

I wasn’t sure if this would be better than what I was currently doing with them or if they would have a negative effect on the performance of the horns.

 

[AllenB]

chromenuts, you’ll get lobing either way. This is less critical with supertweeters than tweeters as your cross is higher than the more sensitive 3-4kHz range. As a result, it can be worth focusing on power instead, by running polars on your current cross and adjusting it accordingly.

 

[chromenuts]

Hi Allen

I’m still playing in the kiddy pool when it comes to these projects and especially the technical testing part. Generally, computers hate me…and the feeling is mutual.

This was a bit of a “what if” idea. The horns were a gift and I didn’t have proper throat adapters for a compression driver, so I just tried the Fostex 108ez on the 3” throats…and was surprised by the quality of the sound they threw into space. Depending on the material, and equipment driving them, sometimes I felt like they were a little lacking in detail.

The OB bass panel is obviously derived from Nelson’s original SLOB article.

I’m afraid I scaled the baffle back a bit too much and have been struggling with getting the response and output I had hoped for from it.

I recently added some wings in the back which gave slight improvement. I think the buyout Peerless drivers I used just don’t have the Xmax needed. Changing them is a hassle since I made the chambers barely big enough for them in a push/pull configuration. Everything else I find with more Xmax has a bigger frame and magnet assembly.

The most I’ve done is some sweeps from the listening position with REW and my UMIK-1 to get an idea if the adjustments I’m making with the B5 make any sense.

I don’t know how one would take polar measurements. Does REW do that?

Regarding the supertweeter…I’m not too sure how I feel about firing them directionally towards the listening position in general. As you said…I’ll get lobing anyway…not to mention timing issues and etc.

If there’s no real way to cure those issues…as I’m not interested in DSP…I was actually considering trying a tweeter mounted firing upwards with a diffuser (omni-directional?)

 

[AllenB]

Being above the sensitive range reduces our ability to hear timing differences, phase, diffraction. Lobing is not bad, necessarily. Here I used it to indicate a certain range of variation you can expect from small changes in position. Also, what looks right and what sounds right are not always the same thing.

So I think you're on the right track.

Supertweeter design should be acoustic (DSP won't cure those issues) and there are a couple of more common ways of approaching it. One, like a regular tweeter and the other involves a more diffuse and pseudo-random treble.

 

[weltersys]

I was thinking this was a lousy solution and wondering how I could make it better by getting the supertweeters closer or at least along the same axis as the drivers in the horns.

I agree side mounting the tweeter is a lousy solution, and vertically above or below is still a big separation jump.

I wasn’t sure if this would be better than what I was currently doing with them or if they would have a negative effect on the performance of the horns.

I think any deleterious affect from co-axially mounting the tweeter in front of the midrange would be far offset by the more uniform off-axis response the virtual point source will provide.
The tweeter could also reduce the mid driver's "beaming".

Using a "nose cone" (semi-phase plug..) behind the tweeter would make the transition smoother.
A toy football cut in half would take care of both tweeters, and would actually lower the mid horn's Fc slightly.
Covering the nose cone with felt or absorptive foam would reduce diffraction effects around the tweeter.

The UREI time aligned coax used foam around the tweeter.

If your crossover is in the 5kHz range, there may be 3 or 4 points along the mid horn's central axis that will phase align the tweeter/mid response. A temporary wire support (arrow pointing to the curved support bridge) could allow sliding the tweeter in and out to find the optimum position for frequency and phase response.
Once you find the sweet spot, a more elegant mount could be made.

Art

 

[wchang]

One might fire super-tweeter through a horn entry-tap bouncing off a very small 45° periscope convex "spoon-head" so as to time-align with the main driver and optimize dispersion. Wool-lined edges and surfaces as needed.

My PrimeRadiant Axia started out as ceramic-dome car tweeter centered inside 15" driver cone with "plast-tape" whizzer-waveguide. Not time-aligned coherent until I flipped it over to bounce off the convex dustcap etc.