AVS Forum banner

1 - 20 of 444 Posts

·
Registered
Joined
·
11,474 Posts
Discussion Starter #1 (Edited)
I decided to model a horn that would fit in my room. (Because the LHorn and GHorn wouldn't.)

It needed to have several other features:
-Had to support vertical (primarily)
-As few cuts as possible. As simple as possible.
-Throat had allow for vertical corner placement, for gaining extra SPL!
-Driver had to be visible so that I could monitor excursion and so I could remove the driver without destroying the box.
-Had to have decent SPL, extension and linearity
-Had to be based on LMS-18

I call it the HzHorn.
It is 84x25x25
91db @ 18hz, -3db @ 15hz.

Useable up to ~50-80hz (depending on XO and EQ.)

Edit: Version 2.0!

Draft version of HzHorn (v2) in dark, GHorn light:









This is the biggest box I can support without compromising the soundstage!


I've never built a horn before. Did I get the settings correct or ?

Here is the SketchUp file:
https://drive.google.com/file/d/0BwUrIAjh6sJ3ZWhFMndlWmsxZkE/view?usp=sharing
 

Attachments

·
Registered
Joined
·
11,474 Posts
Discussion Starter #2 (Edited)
I wanted to start off with one; and maybe build two or four, one for each corner.

Could handle more power if you used a heavy 20hz HPF (48db/oct or something)... But who would do that???

If you used a HST-18 or UXL-18, you'd have to adjust the power and/or XO accordingly for XMECH limit.
 

·
Registered
Joined
·
1,185 Posts
That's a big old box. But only a 4' square footprint. I likey. :) I know you have a name for it already, but if we're feeling Germanic, we can call it "Das Woofen Coffin." :D


One feature I do like is that you can actually service the driver. I've seen some horn builds that you seal up the driver inside...not a fan of that. Personal preference, of course.
 

·
Registered
Joined
·
21,319 Posts
congratulations on taking a shot at a horn.


if only running one, set the "Ang" to 2.0 pi space, as that seems closest to what folks measure regardless of placement.


the length of the last leg is the distance from the driver to the mouth. in yours, the whole s4 to s5 par 62 row would be all zeroes.


i haven't checked the numbers for precision.
 

·
Registered
Joined
·
11,474 Posts
Discussion Starter #7
the length of the last leg is the distance from the driver to the mouth. in yours, the whole s4 to s5 par 62 row would be all zeroes.
I was thinking that the cement floor and two corner walls would become the mouth. So they would be tune-able and somewhat-decent sized.
I tried setting them to 0 or 1 but hornresp gave me issues.
 

·
Registered
Joined
·
3,799 Posts
Could handle more power if you used a heavy 20hz HPF (48db/oct or something)... But who would do that???

If you used a HST-18 or UXL-18, you'd have to adjust the power and/or XO accordingly for XMECH limit.
That's crazy talk...I certainly don't...:D
 

·
Registered
Joined
·
11,474 Posts
Discussion Starter #9 (Edited)
-Fixed an error with the cut-out diameter.
-Also added absolute-minimum bracing.
Original Post updated with newest.

If I ever get around to making this I'd add lots more bracing, takes too long to model it all.

Here is what it looks like with a UXL-18 mounted on it. Motor hangs out a bit (as would the LMS's). But that's ok by me...


 

Attachments

·
Registered
Joined
·
5,542 Posts
Just from looking at it what you have modeled is in no way representing the cabinet design. The driver loading and throat area is critical. You have an S2 of 299 which seems a bit small judging from the 25" cab width and a large constricting taper which then pinches way tight well after the driver. This is not represented in the sim at all. This pinch point would also cause severe air turbulence issues at war volume I'd imagine. Also you want way more bracing. Trust me you want to go overboard on that if anything. Those are big panels and they will vibrate badly. This ain't a little sealed cab. The forces involved are much higher.

Also simulate at 2.0 or 4.0. I always use 4.0 because it is anechoic and matches well with ground plane. 0.5 or 1/8th space is a best case car cabin type environment. Not related to a big HT at all.
 

·
Registered
Joined
·
11,474 Posts
Discussion Starter #14
Just from looking at it what you have modeled is in no way representing the cabinet design. The driver loading and throat area is critical. You have an S2 of 299 which seems a bit small judging from the 25" cab width and a large constricting taper which then pinches way tight well after the driver.
When I was playing around with the numbers S2 was the most critical, which is the tapered part right in front of the driver.
It seemed that as long as S1 wasn't more than double S1, then the results were smooth.

I had to sacrifice S4 completely, and the accuracy of S3 and L34 is questionable.



Originally I started off with the same parameters as the LHorn but smaller and that results in a horrible peak at 22hz with poor 15hz results.
S2 was too small, for the 25x25 width/depth.
 

Attachments

·
Registered
Joined
·
7,780 Posts
How will you take out the driver of needed? From the pics, it looks to be impossible from the angle of the top of the driver in the cab.
This is a cool build. It would be sweet to build these long ways right into the walls of a new build. That would look awesome.
 

·
Registered
Joined
·
11,474 Posts
Discussion Starter #16 (Edited)
If you extend the backplate and vent out the sides, the performance degrades significantly. (that and... the driver wouldn't fit.)
25hz resonance and 40hz null.



 

Attachments

·
Registered
Joined
·
11,474 Posts
Discussion Starter #17 (Edited)
How will you take out the driver of needed? From the pics, it looks to be impossible from the angle of the top of the driver in the cab.
This is a cool build. It would be sweet to build these long ways right into the walls of a new build. That would look awesome.
Comes out from this side: 19x22


On the 19 side you'll have a 1/4 of clearance on each side of the LMS-18, which is slightly bigger than the UXL-18.
Not sure if the LMS motor will fit in there, it's gonna be close on the one side of the magnet near the throat...

Might have to set the bracing back and maybe make the horn 0.5 to 1inch shorter. (Sim'ed it and that seems to have no impact on performance.)
I wouldn't know until it is being assembled.

The UXL-18 seems to fit nicely though.
 

Attachments

·
Registered
Joined
·
11,474 Posts
Discussion Starter #18 (Edited)
Here is was happens when you change the mouth by 1inch in all directions.

You lose maybe 1db @ 15hz, but gain 1.5 db above that. That's worst-case...


It seems that 60hz performance improves quite a bit by doing that...actually; at the sacrifice of worse 80hz performance.
 

Attachments

·
Registered
Joined
·
11,474 Posts
Discussion Starter #19 (Edited)
You lower it in and then you tilt it.





From what I can tell, there doesn't seem to be any shortage of room to tilt it, there are several inches on both sides.
 

Attachments

·
Registered
Joined
·
5,542 Posts
If I get a chance I will show you what I mean. Your sims look fine but your cab design is not what you are simulating.

S2 is at the center of the radiating area/driver cone. You clearly have way more than 299cm^2 at the center of the driver. Your S3 is the tiny pinch point after the driver. This is probably something like 200cm just by eyeballing it. The length from S2 to the pinch at S3 would be more like 25cm or so. Then your expansion starts towards S4 which is the center of the backside of the radiating surface/driver. This L3 section would be nearly all of the horn length. Your S5 area/ horn mouth is clearly bigger than the 1900cm input. 19x22" is almost 2700cm. Also the final L4 section length is much shorter than the input value and is probably only 30cm or so. This is the distance from the driver cone center to the center of the mouth. Also the vtc and ATC are way off. They should be about 6850 and 1465 for an LMS back mounted on an 18mm baffle.

If you change all of this data to more correctly reflect the cab design as is you will see that it doesn't look so hot anymore.
 
1 - 20 of 444 Posts
Top