Don Posted:
"I'm currently working on using two drives to complement each other. Often, the most afficient (highest SPL) )configuration results in a peak at the LF limit, then a dip, then another peak an octave higher. I'm trying to make the second driver "fill in" the dip. Troels Gravesen does something similar with his "doppel horns", where two tapped horns are tuned a fifth apart. It's harder to do it in a single horn."
Hi Don.
You and I are after the same animal. Evil little thing to nail down that magic point where the drivers actually complement each other. You are successfully taking my design apart. All the more power to you. The next lower playing version is to be built some time this week. And I may make the box in such a manner that I can play with the positions of the drivers. I think that this way I can finally get empirical data that matches the theoretical.
I think Bjorno made the comment about highest SPL does not always get you the best response. He is completely correct as are you. The sneaky trade-off is the correct combination. An example would be a broad 3db dip would not be to noticeable but a peak will irritate an acute listener.
"One thing I missed before, your original Hornresp input parameters in post #43 show a large chamber defined between the driver(s) and the "S2" point of the horn. Is that intentional? This may explain the response difference you see. The attached picture shows a Hornresp graph with the chamber (grey line) and without (black line). There is a significant difference. The black line version exactly matches my Akabak model, which does not have the chamber."
All my secrets are getting revealed. Nice work Don.
"I agree, the hourglass or bowtie shape is what you get when you combine chamber (Helmholtz) resonators with 1/4 wave resonators. The aim is to get the best of both in the same box - a true reflex. This is an old use of the term "reflex", from the early days of "steam" radio when tubes / valves were expensive. It was quite common to use one tube to amplify the RF signal, then after conversion to AF the signal would be fed back through the same tube again to amplify the AF (audio) signal. In this case, we're using the same box both as a Helmholtz resonator and as a pipe resonator."
Yes this is basically what an Unhorn is. But the design of the little monster is not that easy. This you know to be sure. The second generation will be put together with more rigor than the first. The first was a try and see. The second is going to be done with some insight and a little less chance and good happenstance.
Mark
"I'm currently working on using two drives to complement each other. Often, the most afficient (highest SPL) )configuration results in a peak at the LF limit, then a dip, then another peak an octave higher. I'm trying to make the second driver "fill in" the dip. Troels Gravesen does something similar with his "doppel horns", where two tapped horns are tuned a fifth apart. It's harder to do it in a single horn."
Hi Don.
You and I are after the same animal. Evil little thing to nail down that magic point where the drivers actually complement each other. You are successfully taking my design apart. All the more power to you. The next lower playing version is to be built some time this week. And I may make the box in such a manner that I can play with the positions of the drivers. I think that this way I can finally get empirical data that matches the theoretical.
I think Bjorno made the comment about highest SPL does not always get you the best response. He is completely correct as are you. The sneaky trade-off is the correct combination. An example would be a broad 3db dip would not be to noticeable but a peak will irritate an acute listener.
"One thing I missed before, your original Hornresp input parameters in post #43 show a large chamber defined between the driver(s) and the "S2" point of the horn. Is that intentional? This may explain the response difference you see. The attached picture shows a Hornresp graph with the chamber (grey line) and without (black line). There is a significant difference. The black line version exactly matches my Akabak model, which does not have the chamber."
All my secrets are getting revealed. Nice work Don.
"I agree, the hourglass or bowtie shape is what you get when you combine chamber (Helmholtz) resonators with 1/4 wave resonators. The aim is to get the best of both in the same box - a true reflex. This is an old use of the term "reflex", from the early days of "steam" radio when tubes / valves were expensive. It was quite common to use one tube to amplify the RF signal, then after conversion to AF the signal would be fed back through the same tube again to amplify the AF (audio) signal. In this case, we're using the same box both as a Helmholtz resonator and as a pipe resonator."
Yes this is basically what an Unhorn is. But the design of the little monster is not that easy. This you know to be sure. The second generation will be put together with more rigor than the first. The first was a try and see. The second is going to be done with some insight and a little less chance and good happenstance.
Mark
It does help the low end response, as the graph showed. I just wondered because I did not see the chamber in the actual box. I have seen tapped horn designs with a chamber, sometimes for its own sake, sometimes as a result of trying to fit two drivers in the same space.
On the subject of multiple drivers, it doesn't seem to be essential to make complicated box structures to try and get the drivers to be "coincident". Arranging the drivers along the length of the horn seems to work just as well.
Akabak allows modelling the different driver arrangements and comparing them.
I wait with bated breath... (Why does everyone spell that "baited"? Don't they teach English at school any more? )
Have a fancy vanity cabinet to assemble then there is watching paint dry. I cut it all out today and did some of the rough assembly. The life of a cabinet maker. That gives me time to work. Four small easy boxes to make then the two horns tapped and Unhorn. The weird thing is the latest work I'm doing on a circa 20hz car horn for 2 SDX7 is functioning more and more like a tapped horn. My current best is 44 liters and 22hz. If this pans out I may be happy to have a very small box go down low.
Mark
The UNHORN 2 Dual 8 inch version is almost ready. It is an almost exact replica of the simulation so you guys will find out how it operates just after I do.
Mark
Mark
I now know what pain in th rear end means
This iteration is a real pain to fold correctly. I have worked it through twice yesterday and today it dawned on me that the useless leftover area that I could do nothing with in the folds would allow me to take it down a little lower if I repositioned the folding method. In plain englrish I'm making one part longer to make the whole thing fold better into a rectangle.
But it is not easy! AHHHH
98% of the way. Whoever said perseverance and invention go together should pay me a visit. I folded unhorn 1 in 20 minutes. This son of a motherless dog will end up taking the better part of a day. Here is a tidbit to tease you.
This is with 2 pcs MCM 8 inch. It is much flatter with dual Trio8. That is the driver the enclosure is optimised for. But in the real world the 3db dip is not going to mean a hill of beans amidst all the standing waves and enigmodes we get in real rooms or automobiles. The driver compliment will get you where you want to go.
How big is the box? At this point 28" x 10" x 19 1/2" Not small but pretty much unbeatable for the size and output. With the TRIO8's and 300 watts you will peak out around 125db. Try modeling the drivers in a reflex cabinet and see what you get!
Again this can only be done using the TRIO8's you will get a paltry 114 with the MCM drivers. Not bad either way. When I have a box tomorrow I will post some pics.
Mark
This iteration is a real pain to fold correctly. I have worked it through twice yesterday and today it dawned on me that the useless leftover area that I could do nothing with in the folds would allow me to take it down a little lower if I repositioned the folding method. In plain englrish I'm making one part longer to make the whole thing fold better into a rectangle.
But it is not easy! AHHHH
98% of the way. Whoever said perseverance and invention go together should pay me a visit. I folded unhorn 1 in 20 minutes. This son of a motherless dog will end up taking the better part of a day. Here is a tidbit to tease you.
An externally hosted image should be here but it was not working when we last tested it.
This is with 2 pcs MCM 8 inch. It is much flatter with dual Trio8. That is the driver the enclosure is optimised for. But in the real world the 3db dip is not going to mean a hill of beans amidst all the standing waves and enigmodes we get in real rooms or automobiles. The driver compliment will get you where you want to go.
How big is the box? At this point 28" x 10" x 19 1/2" Not small but pretty much unbeatable for the size and output. With the TRIO8's and 300 watts you will peak out around 125db. Try modeling the drivers in a reflex cabinet and see what you get!
An externally hosted image should be here but it was not working when we last tested it.
Again this can only be done using the TRIO8's you will get a paltry 114 with the MCM drivers. Not bad either way. When I have a box tomorrow I will post some pics.
Mark
Are we there yet???
Still a bit left over. Had a son to pickup and entertain this weekend unexpectedly. Life in the fast lane. All fooling around stops to be a good father. But tomorrow I still have the day off. There are some benefits to being self employed.
Mark🙄
Still a bit left over. Had a son to pickup and entertain this weekend unexpectedly. Life in the fast lane. All fooling around stops to be a good father. But tomorrow I still have the day off. There are some benefits to being self employed.
Mark🙄
Parts made but not woofing yet
In the ever present desire to better an already better design....
I have all the pieces for a great many boxes but none assembled. Work rears it's ugly head once again. The good thing is the weekend is coming fast and assembly generally is quite quick. I really did design the horn one more time. I really want it to fit into a more or less rectangular shape. So tossing and turning the fold and I get a more or less useful rectangular shape. The greatest struggle is keeping the drivers at the location simulated and the compression chamber created in an area that will still let the horn function. Nasty but accomplished.
Here is some more info:
Not the largest box at under 2 cubic feet ( 28 liters /cubic ft ) We'll see how it performs. I'm thinking that this one in a car will be lethal. In a house it should be a room pounder that is if the simulations hold true again.
For all you guys thinking about the box in a car here it is at 1oo watts. 1/8th space. If you look closely at the SPL chart there is the ghost of a 1 watt input. The excursion plot is X-max + 25% or 10 mm.
Any comments?
Mark
In the ever present desire to better an already better design....
I have all the pieces for a great many boxes but none assembled. Work rears it's ugly head once again. The good thing is the weekend is coming fast and assembly generally is quite quick. I really did design the horn one more time. I really want it to fit into a more or less rectangular shape. So tossing and turning the fold and I get a more or less useful rectangular shape. The greatest struggle is keeping the drivers at the location simulated and the compression chamber created in an area that will still let the horn function. Nasty but accomplished.
Here is some more info:
Not the largest box at under 2 cubic feet ( 28 liters /cubic ft ) We'll see how it performs. I'm thinking that this one in a car will be lethal. In a house it should be a room pounder that is if the simulations hold true again.
For all you guys thinking about the box in a car here it is at 1oo watts. 1/8th space. If you look closely at the SPL chart there is the ghost of a 1 watt input. The excursion plot is X-max + 25% or 10 mm.
Any comments?
Mark
Last edited:
Hello.
Would love a Tutorial from keying in the driver parameters to finished design. Would like to try various drivers and designs myself. But, I'm really enjoying what I'm reading, so thank you, just for the read, but a tutorial would be nice.
Regards
Fuzzymuff 🙂
Would love a Tutorial from keying in the driver parameters to finished design. Would like to try various drivers and designs myself. But, I'm really enjoying what I'm reading, so thank you, just for the read, but a tutorial would be nice.
Regards
Fuzzymuff 🙂
yeah a tutorial starting from hornresp to drawing of the enclosure would be nice for not so technical person as myself. =)
Hi,FYI, Not a tutorial but: See my pictures comparing the last presented 'Unhorn' with simple ported and closed boxes.
b
Attachments
Hello Bjorno
Thank you for your comments and the time you took to create your simulations. I have always found your comments constructive and kind and I thank you for your findings.
One thing to clarify is that this box is designed for in home use. The posts that have been made as of # 205 and down have been simulated in a room environment. Or walls floor and ceiling. If you put it into a corner bump up the efficiency by 3 db. If you put it in a basement corner where there is a concrete floor and two corner walls bump it up a bit more. If you are putting it in a car you get 6 db more efficiency.
The only bit of argument that I have with your findings is relatively minor. I simulated in 1 pi environment. Or a room with a ceiling and walls. Your one simulation at least, is in a corner loaded environment. ( .5 pi ) That is not exactly the same.
It is interesting that you point out the need for both a high pass and low pass filter on the UNHORN.
Any Woofer playing the low end of music will need a low pass filter. That is a moot point that you already know this to be true.
But any horn system by it's very nature needs a high pass as well. This you already know. That being said any tuned port sysytem when played below it's system resonant frequency is libel to self destruct as will a horn. We have all seen the madly flapping cones in a vented enclosure that signal the quick end of the drivers voice coil. The unfortunate thing with a horn is you cannot see it ( cone flapping ) happening. But you can hear it and as with all sound systems there are limitations to this design.
In order of performance these drivers have been simulated in this enclosure.
SDX7 flattest response and lowest response.
TRIO 8 decently flat response and greater efficiency at maximum power due to greater cone area and greater X-max.
MCM 8 inch woofer not so flat but not to bad. SPL is nothing to sneeze at either for 1 watt or maximum power.
What am I putting into the enclosure? The test box is going into my car with dual TRIO 8's. I have a full range design worked through to use 2 SDX7 and two EL70's and this is the real reason for the design of this box in the first place. A forum member was the seed idea for this application to be mentioned in the first place. He will get an opportunity to try it out. If he likes it I have encouraged him to post.
The UNHORN already has a bit of efficiency advantage in your simulation. But in the apples to apples simulation the advantage in not a mere 3db better ( twice as efficient ) but actually 6 db ( or 4 times as efficient ). That is not small potatoes! The trade off for a double tapped double resonant cavity enclosure is box complexity and some size disadvantage. But the gain is not insignificant is terms of system efficiency. The real thing I was after in the first place was greatest efficiency for the smallest enclosure. I'm not an expert in designing tapped horns. But I can consistently design a smaller UNHORN than a tapped horn. And the response is almost the same between the two different designs. The size difference is almost 50% smaller for the UNHORNS. The group delay is a bit better for the UNHORN as well when compared to a Tapped horn. So all in all if the latest design effort pans out I may be on to something a little bit interesting.
Mark
P.S.
sixgen and Fuzzymuff
What you guys are asking is bigger than everything I have posted to this thread! But there is quite a bit of info already in this thread as to what matters and what does not. The best bet is to download the program punch in the numbers and play with the horn box section of the input screen. Pretty soon when you toggle over to the schematic diagram area you will see what the numbers you have punched in look like in an enclosure. You can press calculate and see what your machinations have turned into. The basic rule of thumb is that the longer a horn enclosure the lower it will play. The UNHORN is an attempt to make the most efficient box in the smallest volume box. Now there is a true oxymoron.
Thank you for your comments and the time you took to create your simulations. I have always found your comments constructive and kind and I thank you for your findings.
One thing to clarify is that this box is designed for in home use. The posts that have been made as of # 205 and down have been simulated in a room environment. Or walls floor and ceiling. If you put it into a corner bump up the efficiency by 3 db. If you put it in a basement corner where there is a concrete floor and two corner walls bump it up a bit more. If you are putting it in a car you get 6 db more efficiency.
The only bit of argument that I have with your findings is relatively minor. I simulated in 1 pi environment. Or a room with a ceiling and walls. Your one simulation at least, is in a corner loaded environment. ( .5 pi ) That is not exactly the same.
It is interesting that you point out the need for both a high pass and low pass filter on the UNHORN.
Any Woofer playing the low end of music will need a low pass filter. That is a moot point that you already know this to be true.
But any horn system by it's very nature needs a high pass as well. This you already know. That being said any tuned port sysytem when played below it's system resonant frequency is libel to self destruct as will a horn. We have all seen the madly flapping cones in a vented enclosure that signal the quick end of the drivers voice coil. The unfortunate thing with a horn is you cannot see it ( cone flapping ) happening. But you can hear it and as with all sound systems there are limitations to this design.
In order of performance these drivers have been simulated in this enclosure.
SDX7 flattest response and lowest response.
TRIO 8 decently flat response and greater efficiency at maximum power due to greater cone area and greater X-max.
MCM 8 inch woofer not so flat but not to bad. SPL is nothing to sneeze at either for 1 watt or maximum power.
What am I putting into the enclosure? The test box is going into my car with dual TRIO 8's. I have a full range design worked through to use 2 SDX7 and two EL70's and this is the real reason for the design of this box in the first place. A forum member was the seed idea for this application to be mentioned in the first place. He will get an opportunity to try it out. If he likes it I have encouraged him to post.
The UNHORN already has a bit of efficiency advantage in your simulation. But in the apples to apples simulation the advantage in not a mere 3db better ( twice as efficient ) but actually 6 db ( or 4 times as efficient ). That is not small potatoes! The trade off for a double tapped double resonant cavity enclosure is box complexity and some size disadvantage. But the gain is not insignificant is terms of system efficiency. The real thing I was after in the first place was greatest efficiency for the smallest enclosure. I'm not an expert in designing tapped horns. But I can consistently design a smaller UNHORN than a tapped horn. And the response is almost the same between the two different designs. The size difference is almost 50% smaller for the UNHORNS. The group delay is a bit better for the UNHORN as well when compared to a Tapped horn. So all in all if the latest design effort pans out I may be on to something a little bit interesting.
Mark
P.S.
sixgen and Fuzzymuff
What you guys are asking is bigger than everything I have posted to this thread! But there is quite a bit of info already in this thread as to what matters and what does not. The best bet is to download the program punch in the numbers and play with the horn box section of the input screen. Pretty soon when you toggle over to the schematic diagram area you will see what the numbers you have punched in look like in an enclosure. You can press calculate and see what your machinations have turned into. The basic rule of thumb is that the longer a horn enclosure the lower it will play. The UNHORN is an attempt to make the most efficient box in the smallest volume box. Now there is a true oxymoron.
A bit more fooling around.
As for simulating 2 MCM drivers in a vented box this is the best I can get.
Note the box is 60 liters. The input power is 300 watts. The UNHORN can do this in a smaller enclosure and with less than 50% of the power.
Mark
As for simulating 2 MCM drivers in a vented box this is the best I can get.
An externally hosted image should be here but it was not working when we last tested it.
Note the box is 60 liters. The input power is 300 watts. The UNHORN can do this in a smaller enclosure and with less than 50% of the power.
Mark
Last edited:
Hi bjorno,
I notice that your ported and closed box examples include very short 3-segment cylindrical horns. Note that they are not really necessary - for direct radiator systems, all segment values can be set to zero. Notes 12 and 13 on page 18 of the Hornresp Help file refer.
If however you also want to examine directivity characteristics, then a single short segment will be required.
Kind regards,
David
Attachments
Hi mwmkravchenko and everybody else,
Just stopping in to say a few things: Your results have inspired me to move beyond simulating closed and ported boxes, this UNHORN is incredible! I downloaded HornRESP and have been learning it in my spare time.
Also, I was getting ready to build one of these boxes but have held off because of the larger 2xSDX7 horn that goes deeper that has also been discussed -- I listen as well to some big classical recordings and organ music, should I continue waiting or just build a small UNHORN?
Thanks and regards,
mista2
PS. I second what Fuzzymuff said, if nothing else this has been an exceptional read!
Just stopping in to say a few things: Your results have inspired me to move beyond simulating closed and ported boxes, this UNHORN is incredible! I downloaded HornRESP and have been learning it in my spare time.
Also, I was getting ready to build one of these boxes but have held off because of the larger 2xSDX7 horn that goes deeper that has also been discussed -- I listen as well to some big classical recordings and organ music, should I continue waiting or just build a small UNHORN?
Thanks and regards,
mista2
PS. I second what Fuzzymuff said, if nothing else this has been an exceptional read!
yes i have been experimenting on the program. i just would like to thank you guys for this thread. sorry about my request 🙂 hope to learn more from this thread! God Bless!
Mista
Wait for the finished UNHORN2 measurement results. I have a couple of days this week that I have booked off to get some of this work done. I promise a plenty pictures and graphs.
David McBean is the guy who deserves the credit. I have thought about how to shorten a horn for quite a while. But thinking what may work and figuring what does work is only possible when you have an accurate method to simulate the ideas. Without Hornresp I would have to go back to the plunking in numbers in a formula and fighting with my TI calculator. Many thankx again David.
Mark
Wait for the finished UNHORN2 measurement results. I have a couple of days this week that I have booked off to get some of this work done. I promise a plenty pictures and graphs.
David McBean is the guy who deserves the credit. I have thought about how to shorten a horn for quite a while. But thinking what may work and figuring what does work is only possible when you have an accurate method to simulate the ideas. Without Hornresp I would have to go back to the plunking in numbers in a formula and fighting with my TI calculator. Many thankx again David.
Mark
Thanks for the quick reply!
I'll go ahead and wait for the new design on your suggestion. Very excited to see the results of UNHORN2! I'll run through plenty of pictures myself upon starting the build pending your posts.
mista2
I'll go ahead and wait for the new design on your suggestion. Very excited to see the results of UNHORN2! I'll run through plenty of pictures myself upon starting the build pending your posts.
mista2
Thank you Mark!
But,Sorry, I'm not going to use valuable thread space answering your IMO odd reaction to the message I brought with my posted pictures but here are some newer ones pointing at the information I used as an input. See pictures a and b.
Why don't you post the input HR screen addressing the two pictures in posting # 207 I scrutinized to be reflecting a very misaligned design?
This could save you if I've done severe mistakes when comparing your Unhorn to my closed resp. ported simulations. I would immediately confess If I'm wrong.
b
Attachments
Not a tutorial but:
Hi David,
😱 Thank you for noticing my misuse of your eminent HR program.
I simply randomly copied an older closed box template I've used to simulate an closed box many HR revisions ago.
I understand the importance to be correct when publishing plots that could fall back on you as there could be a lot of HR newbies that could be unnecessarily misled by my posted plots.
🙂 So I've made a new attempt to rectify and sorting this problem out:
See picture c and d
b
Hi David,
😱 Thank you for noticing my misuse of your eminent HR program.
I simply randomly copied an older closed box template I've used to simulate an closed box many HR revisions ago.
I understand the importance to be correct when publishing plots that could fall back on you as there could be a lot of HR newbies that could be unnecessarily misled by my posted plots.
🙂 So I've made a new attempt to rectify and sorting this problem out:
See picture c and d
b
Attachments
Hi bjorno,
I thought that was most probably the case 🙂.
Exactly - that was my concern. It seems that Hornresp can be confusing enough to users even at the best of times 🙂.
Many thanks, much appreciated.
Kind regards,
David
I thought that was most probably the case 🙂.
Exactly - that was my concern. It seems that Hornresp can be confusing enough to users even at the best of times 🙂.
Many thanks, much appreciated.
Kind regards,
David
- Status
- Not open for further replies.