Speakerplans.com Homepage
Forum Home Forum Home > Plans > New Projects Forum
  New Posts New Posts RSS Feed - 15" Horn to go over X1s
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

15" Horn to go over X1s

 Post Reply Post Reply
Author
Message
JulianDA View Drop Down
Registered User
Registered User
Avatar

Joined: 29 May 2018
Location: Germany
Status: Offline
Points: 17
Post Options Post Options   Thanks (0) Thanks(0)   Quote JulianDA Quote  Post ReplyReply Direct Link To This Post Topic: 15" Horn to go over X1s
    Posted: 15 January 2019 at 12:46pm
Hello fellow Speakerheads,

in an attempt to try another Kickbin over X1s to maybe replace our HD15s, i simulated and modelled a small FLH. The Plan was to build a speaker that is half the size of an X1 Letterbox (56,9x50,8x69cm (WxHxD) ), so four Horns will stack neatly onto one to give a nice height for our Tops. Another criteria was a decent frequency response down to 60Hz, so we can use a but24 HPF at roughly 80Hz to meet the X1. 
After some time with Hornresp i came up with the following simulation for the existing Oberton 15B450 Speaker. (I simulated in eightspace, because the Horn will mainly be used in a stack of four on the ground or on just one X1 (so roughly on the ground))

   
This looked promising, so i spend some more time in Autodesk Inventor and put this simulated Horn in a wooden box :P My first design was this slotloaded one:
The problem with this is that there is no way to put the driver in, not enough place for the magnet in the backchamber, a lot of wasted space, and to much crazy angles on small pieces of wood.....so i came up with another idea in the following drawing:
Its more simple and doesnt look to bad. The driver can be mounted from the right side, has probably enough room (maybe after some minor changes) and is easily accessible. The only problem is now that i dont know how to put this correctly into Hornresp. I dont think this will change to much of the frequency response because the significant areas are nearly the same and the small loss in lenght should be well compensated by the virtual hornextension in front of the hornmouth.

My questions to you are now:
- how can i simulate this kind of hornthroat more accurate?
- is there another way of folding this horn better in the given space?
- can i get an extended frequency response on the high end in comparison to the HD15 even though there are two bends in the horn (we run the HD15 currently to 151Hz and i would like to get to 180Hz without the typical "honkyness"

Thanks in advance,
Julian
Back to Top
DMorison View Drop Down
Young Croc
Young Croc


Joined: 14 March 2007
Location: Aberdeen
Status: Offline
Points: 1155
Post Options Post Options   Thanks (0) Thanks(0)   Quote DMorison Quote  Post ReplyReply Direct Link To This Post Posted: 15 January 2019 at 8:46pm
Originally posted by JulianDA JulianDA wrote:

The problem with this is that there is no way to put the driver in, not enough place for the magnet in the backchamber, a lot of wasted space, and to much crazy angles on small pieces of wood.....so i came up with another idea in the following drawing:
Its more simple and doesnt look to bad. The driver can be mounted from the right side, has probably enough room (maybe after some minor changes) and is easily accessible. The only problem is now that i dont know how to put this correctly into Hornresp. I dont think this will change to much of the frequency response because the significant areas are nearly the same and the small loss in lenght should be well compensated by the virtual hornextension in front of the hornmouth.

My questions to you are now:
- how can i simulate this kind of hornthroat more accurate?
Thanks in advance,
Julian

Use the Offset Driver option in the number of drivers field. You then set S1 to near enough zero (I think 0.1 is the lowest value it accepts), representing the very thinnest end of the "wedge" in front of the driver, with S2 being the sectional area in front of the middle of the driver.
Not sure about your other questions, sorry.
HTH,
David.
Back to Top
JulianDA View Drop Down
Registered User
Registered User
Avatar

Joined: 29 May 2018
Location: Germany
Status: Offline
Points: 17
Post Options Post Options   Thanks (0) Thanks(0)   Quote JulianDA Quote  Post ReplyReply Direct Link To This Post Posted: 17 January 2019 at 10:28am
Thank you David, I totally forgot about this feature in Hornresp :D
After some changes to the backchamber to make the driver fit inside, I simulated the horn with an offset driver. Here is the newish design and the simulation results i got:



The thing with this simulation is that i am not sure about the compression ratio of my hornthroat. I didnt want to go over 2:1, because i dont think the driver will like it that much. But with the input for the offset driver i am now at 2,76:1. My gutfeeling tells me, that the actual ratio will be lower in this case because of the volume inside the drivers cone, but i am not sure about that.
Another problem might be the asymmetric loading of the cone because of the shallow baffle angle. 

So first of all, is the Hornresp input for S2 correct like i did it? 
And can somebody with more experience than me tell me if the loading of the cone is problematic in this case? 


Edited by JulianDA - 17 January 2019 at 12:53pm
Back to Top
JulianDA View Drop Down
Registered User
Registered User
Avatar

Joined: 29 May 2018
Location: Germany
Status: Offline
Points: 17
Post Options Post Options   Thanks (0) Thanks(0)   Quote JulianDA Quote  Post ReplyReply Direct Link To This Post Posted: 21 January 2019 at 2:24pm
Because I did not know if i can trust my simulation, my mate an i just build the first prototype yesterday in his garage with the help of some nice techno music Big smile





It is just a quick and dirty OSB board prototype, but i think it turned out good enough :)


So today I did the first frequency response measurement in an open field. 
It is a groundplane measurement taken 4m away from the horn using the Beyerdynamics MM1 microphone and the Steinberg UR12 Soundcard.


Compared to the simulation of just one Horn in 2pi, so standing on the ground, it looks relatively decent.


But as you can see, my simulation shows the frequency dips caused by the driver offset in much lower frequencys. To me this looks like good evidence for my theory that due to the very narrow start of the horn before the driver cutout, and the mounting of the driver on the baffle (so reducing the space even more), the actual offset is just the radius of the cone. 
Another difference is the predicted low end of the horn vs the measured response. I think i measured more low end because of the virtual hornextension in front of the hornmouth. 
The last difference is the response dip at roughly 550Hz. This frequency corresponds to a quarter wavelength of around 16cm, so roughly the radius of the cone. I think this will be a cancelation because of the phase difference of the sound generated at both ends of the driver. 

So by changing the driver offset to the cone radius and adding half the height of the hornmouth to the length of the last segment, i get the following simulation, that really does look like the measurement:


I think this design looks promising and the simulation seems to be spot on.
But maybe i draw the wrong conclusions and got the right simulation just by pure luck. Maybe somebody with experience in horn design could tell me if this looks right?

And i got one more question:
In the measurement you can see a small "ripple" around 133Hz that isnt in the simulation. To me this looks like a resonance caused by a reflection inside the Horn. I concluded that this might be a reflection from the mouth of the horn to the backwall because the quarter wavelength of 133Hz is just under 65cm and my box is 67,5cm deep. Does this explanation sound right? And am i right by thinking this goes away by using more horns in a stack, due to the bigger moutharea and therefore better impedance matching to the sorrounding air?
Back to Top
luthier View Drop Down
Registered User
Registered User
Avatar

Joined: 21 July 2013
Location: UK - Canterbury
Status: Offline
Points: 240
Post Options Post Options   Thanks (0) Thanks(0)   Quote luthier Quote  Post ReplyReply Direct Link To This Post Posted: 21 January 2019 at 5:52pm
Afraid I'm not experienced enough to really help you out with any of your questions. However, I can say your design looks great, very well thought out and efficient! 

Looking at your measurement at first I thought the "ripple" at 133hz would be caused my the microphone, but the manufacturers measurements say it's near perfectly flat. From my understanding peaks in the response will be greatly smoothed out when the horn is used in multiples. The  last hornresp sim you posted does show a very small bump around 100hz. Maybe you could manipulate this to match your measurement and work back?

  
Back to Top
gen0me View Drop Down
Young Croc
Young Croc
Avatar

Joined: 20 February 2016
Location: UK
Status: Offline
Points: 904
Post Options Post Options   Thanks (0) Thanks(0)   Quote gen0me Quote  Post ReplyReply Direct Link To This Post Posted: 21 January 2019 at 6:07pm
Originally posted by luthier luthier wrote:

at 133hz 
Does it sound muddy, resonative, hard to eq?
Try taking the speaker out, than knocking into enclosure. Any similarities?
I appreciate every like :)) https//www.facebook.com/genomesoundsystems
Mixes: https://www.mixcloud.com/gen-ome/
Back to Top
JulianDA View Drop Down
Registered User
Registered User
Avatar

Joined: 29 May 2018
Location: Germany
Status: Offline
Points: 17
Post Options Post Options   Thanks (0) Thanks(0)   Quote JulianDA Quote  Post ReplyReply Direct Link To This Post Posted: 23 January 2019 at 12:05pm
Originally posted by luthier luthier wrote:


However, I can say your design looks great, very well thought out and efficient! 


Thank you! But this is just my first try at designing a horn myself, so I think i will be running in some problems with it that i did not think of :D

Originally posted by gen0me gen0me wrote:


Does it sound muddy, resonative, hard to eq?
Try taking the speaker out, than knocking into enclosure. Any similarities?


To be honest, i can not hear it at all, but i am very bad at hearing this kind of stuff in the lower frequencys LOL
So I fear the knocking-trick will not give me clear results (or just results biased by what i expect/want to hear)
But the Idea of Resonances of the Box itself is a good one, thanks! Next weekend i will try to put in more braces (especially in front of the driver) and measure the response again.

Originally posted by luthier luthier wrote:


The  last hornresp sim you posted does show a very small bump around 100hz. Maybe you could manipulate this to match your measurement and work back?


I think the bump at 100Hz is there (what i traced in blue), but its "masked" by the resonance (1)
But maybe i see just what i would like to see there Confused

The ripples 2 and 3 look almost exactly like in the simulation (a good sign i think)

And what i traced in Orange and circled in yellow is what i believe to be the effect of the angled driver (the phase differences of the sound coming from the opposite sites of the cone, as seen from the perspective of the Horn)

Sadly i can just work on the Horn on weekends, so now the only thing i can do is speculate, ask here for help, and do simulations Big smile
Back to Top
luthier View Drop Down
Registered User
Registered User
Avatar

Joined: 21 July 2013
Location: UK - Canterbury
Status: Offline
Points: 240
Post Options Post Options   Thanks (0) Thanks(0)   Quote luthier Quote  Post ReplyReply Direct Link To This Post Posted: 26 January 2019 at 11:34pm
Interested to know if you've got any further with this? 

Edited by luthier - 26 January 2019 at 11:34pm
Back to Top
JulianDA View Drop Down
Registered User
Registered User
Avatar

Joined: 29 May 2018
Location: Germany
Status: Offline
Points: 17
Post Options Post Options   Thanks (0) Thanks(0)   Quote JulianDA Quote  Post ReplyReply Direct Link To This Post Posted: 27 January 2019 at 9:47am
On Friday i did some more braces infront of the driver and also looked for more resonating panels. But due to the shitty weather here in germany i cant measure if anything changed Confused
I also worked a little bit on our new midtop design containing an 18S 15nmb420, and an 18S nsd1480n on the xt1464 Horn. But this also needs a good measuring Session, so there was not to much progress.

In the next two weeks i am busy studying and writing my bachelors degree, so all the projects will have to be paused.  After that the work can continue ans i will give you an update of anything interesting that happens :)
Back to Top
 Post Reply Post Reply
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.01
Copyright ©2001-2018 Web Wiz Ltd.

This page was generated in 0.094 seconds.