Hi xrk971,
Post #40: "...is it assumed all 4 horns are at the single driver location? Clearly that is not the same as 4 drivers, one at each quadrant of this box..."
Yes, that's how I understand Hornresp simulations. I think it's close enough to get started, because you end up in a single mouth similar to the Hornresp circular mouth.
What might get Circlomanen's FLH even closer would be to add segments 3 and 4 to reflect the front face of the enclosure, but it doesn't seem to make much of a difference when I tried it. I don't know how acurately Hornresp can simulate that.
To some extend I think that putting four horn enclosures together into one is an elegant trick, as people may have a tendency to compare the performance to single driver enclosures without using four single driver enclosures in an appropriate arrangement.
Regards,
Post #40: "...is it assumed all 4 horns are at the single driver location? Clearly that is not the same as 4 drivers, one at each quadrant of this box..."
Yes, that's how I understand Hornresp simulations. I think it's close enough to get started, because you end up in a single mouth similar to the Hornresp circular mouth.
What might get Circlomanen's FLH even closer would be to add segments 3 and 4 to reflect the front face of the enclosure, but it doesn't seem to make much of a difference when I tried it. I don't know how acurately Hornresp can simulate that.
To some extend I think that putting four horn enclosures together into one is an elegant trick, as people may have a tendency to compare the performance to single driver enclosures without using four single driver enclosures in an appropriate arrangement.
Regards,
Thank You Martinsson for very kind words.
I am guessing, but based on the knowledge i have got after many years of designing and discussing (tapped) horns with you. I hope i got it reasonably right.
I would never have come up with this idea without you Martisson! Our discussions are a development platform, and your THAM range of basshorns are the one i want to beat.
Cheers,
Johannes
Hornresp assumes that all the drivers is in the same horn.
I can model one driver in one horn, and it behaves fairly simmilar for a tapped horn. A normal exponetial or hyperbolic horn does not scale inte same way. I could not make a small exponential horn with 4 small 4 inch drivers, but it is simple to build the tapped bandpass hybrid that way. See last page for a 4 inch THBP.
Cheers,
Johannes
I think he meassures with all the drivers parallel with a total driver load of 2 ohm. If i wired them in parallel in hornresp i get 113,6 dB/2,83v/m efficiency to.
It is easier to just simm them in 1 pi instead. Not correct in the lower region of the passband, but much more correct above 50 Hz or soo.
Johannes
This is how i would build the box if i had some large sheets of plywood and 4 B&C 15SW115 lying around.
This is just a guess. I am sure Danley has all kinds of differnt refinements, and a much more elaborate tuning. I don't believe he uses straight pipes like i do, but i belive this is the genral principle behind his new BC-horns.
Cheers,
Johannes
I want to "Out-THAM15" the THAM15 with a smaller box, without going to extremes in power and driver cost. Using two B&C 21IPAL in a small closed box with many kilowatts of power is not very cost efficient, especially so for the DIY comunity.
I think this is a good example of what acan be accomplished with this variation of the tapped horn.
I want to call it THDB12. I don´t have any plans yet, but i could probably draw something if there is some interest. I don´t like CAD, since i have a hard time trying to learn it..
As can be seen, my THDB12 has 2,6 dB more maximum output at 80 Hz. It has a similar overall efficiency, with small variations. It lacks a dB spl-capacity below 60 Hz, but will play 2,6 - 4 dB above 70 Hz. The region between 70 and 120 Hz is very important, as a very substantial amount of the energy in music is centered there. A lack of energy in this region will make the horn sound dull. The exact opposite to what people say about the DSL BC412/415.
This is a smaller box (130,6 L) with a B&C 12NW100 compared to a larger box (167 L) with a B&C 15TBX100.
I have never heard this kind of box yet, so it is only theoretical at the moment. I believe it will sound fantastic with an enormous "kick" and attack, but i don´t know for sure.
Cheers,
Johannes
I think this is a good example of what acan be accomplished with this variation of the tapped horn.
I want to call it THDB12. I don´t have any plans yet, but i could probably draw something if there is some interest. I don´t like CAD, since i have a hard time trying to learn it..
As can be seen, my THDB12 has 2,6 dB more maximum output at 80 Hz. It has a similar overall efficiency, with small variations. It lacks a dB spl-capacity below 60 Hz, but will play 2,6 - 4 dB above 70 Hz. The region between 70 and 120 Hz is very important, as a very substantial amount of the energy in music is centered there. A lack of energy in this region will make the horn sound dull. The exact opposite to what people say about the DSL BC412/415.
This is a smaller box (130,6 L) with a B&C 12NW100 compared to a larger box (167 L) with a B&C 15TBX100.
I have never heard this kind of box yet, so it is only theoretical at the moment. I believe it will sound fantastic with an enormous "kick" and attack, but i don´t know for sure.
Cheers,
Johannes
Last edited:
http://www.diyaudio.com/forums/subwoofers/106558-another-tapped-horn.html
I have had this "itch" in the back of my mind, that i have seen something similar to my idea here, and i belive Tom Danley was involved somehow..
I found the above thread...
Quotes from Tom Danley below:
I don't know if the last quote from Tom Danley is about the tapped horn or from the tapped horn with a front quarterwave resonator.
It would not be far fetched to imagine Tom Danley using this idea in his latest BC415....
This forum is a goldmine of information!
Thanks everyone who contributes to this forum...
Cheers,
Johannes
I have had this "itch" in the back of my mind, that i have seen something similar to my idea here, and i belive Tom Danley was involved somehow..
I found the above thread...
Quotes from Tom Danley below:
I don't know if the last quote from Tom Danley is about the tapped horn or from the tapped horn with a front quarterwave resonator.
It would not be far fetched to imagine Tom Danley using this idea in his latest BC415....
This forum is a goldmine of information!
Thanks everyone who contributes to this forum...
Cheers,
Johannes
Oliver,
Hornresp seems to ignore the effect of a segment going out to a larger flat boundary, while I have measured a 3 dB increase in LF response doubling frontal area.
Just another one of those things that keeps me making sawdust rather than spending too much time simulating.
Art
Indeed! His DTS20 and its later, lower tuned, dual driver DTS10, among others that made him the TH 'leader of the pack'..........
The latter..........
For sure, though don't know if he did or not.
Indeed it is! You're welcome!
GM
Horns are not efficient att producing bass!!!
If one wants to maximize bass output from any given size box, the quarter wave resonator is the most efficient. Much more efficient then a horn.
Tom Danley knows this, and this is part of why i believe this tapped horn(pipe) - quarter wave resonator hybrid is the way the BC415 works.
The problem, as is very obvious in the response comparison below, is the very narrow bandwidth of the quarter wave resonator. It has a very deep depression between the the quarterwave resonances. By using one extra quarter wave resonator in the deep depression, its easy to get a extreme efficiency over a narrow but useful bandwidth.
The spl charts below is a 60 liter, one meter long QW pipe (not tapped) and an one meter long hyperbolic horn (not tapped) with 1788 cm2 mouth. Both 60 liters.
This is part of why i can get more spl and the same low cut of with a 12 inch 10 mm xmax driver in a smaller box, compared with THAM15, with a 15 inch driver with 11 mm xmax in a larger box.
Cheers,
Johannes
If one wants to maximize bass output from any given size box, the quarter wave resonator is the most efficient. Much more efficient then a horn.
Tom Danley knows this, and this is part of why i believe this tapped horn(pipe) - quarter wave resonator hybrid is the way the BC415 works.
The problem, as is very obvious in the response comparison below, is the very narrow bandwidth of the quarter wave resonator. It has a very deep depression between the the quarterwave resonances. By using one extra quarter wave resonator in the deep depression, its easy to get a extreme efficiency over a narrow but useful bandwidth.
The spl charts below is a 60 liter, one meter long QW pipe (not tapped) and an one meter long hyperbolic horn (not tapped) with 1788 cm2 mouth. Both 60 liters.
This is part of why i can get more spl and the same low cut of with a 12 inch 10 mm xmax driver in a smaller box, compared with THAM15, with a 15 inch driver with 11 mm xmax in a larger box.
Cheers,
Johannes
Last edited:
I think a tapped horn has higher SPL per unit power than a plain 1/4-wave resonator that is not a horn. If you also consider a broad flat response, the FLH is probably better than a TH. I am not sure what you are trying to show above but it is not clear to me. As an example, a good TH can typically achieve 110dB sensitivity at 1w and 1m from 50Hz to 100Hz without too much trouble. Horns are 1/4 wave resonators so the distinction must be regarding expansion to have a horn like mouth.
Last edited:
Hi Art,
Post #50: "Hornresp seems to ignore the effect of a segment going out to a larger flat boundary, while I have measured a 3 dB increase in LF response doubling frontal area."
David McBean indicates the possible lack of accuracy in his "Caution" when using section w/ unusual aspect ratios.
You are definitely in good company when you indicate, that building and measuring trumps simulations.
Regards,
Post #50: "Hornresp seems to ignore the effect of a segment going out to a larger flat boundary, while I have measured a 3 dB increase in LF response doubling frontal area."
David McBean indicates the possible lack of accuracy in his "Caution" when using section w/ unusual aspect ratios.
You are definitely in good company when you indicate, that building and measuring trumps simulations.
Regards,
I think AkAbak may be able to handle the horn mouth coming out of a large flat boundary as the size of the baffle is specified as well as the size and position of the horn mouth radiator. The larger your box is the closer it approaches infinite baffle or 2pi (assuming far from floor). What AkAbak will also handle is the diffraction at the edge of the boundary. However that is probably not an issue at these low frequencies.
What I am curious about is what happens when you put two FLH's mouth-to-mouth and have the radiator coming out perpendicularly. Is the effective path length now double the single horn so that there may be a half wave mode somewhere for double the length in addition to the usual quarter wave mode for the single length?
What I am curious about is what happens when you put two FLH's mouth-to-mouth and have the radiator coming out perpendicularly. Is the effective path length now double the single horn so that there may be a half wave mode somewhere for double the length in addition to the usual quarter wave mode for the single length?
Last edited:
Hornresp seems to contradict your statement (see simulated comparison above in prevous post, between QW pipe - non expanding - and an expanding horn).
I try to stick to verifiable facts as much as possible. I have hard time building 20 boxes a day to measure and verify Hornresp to within a dB or so. Hornresp is at least consistent enough in any deviation from reality, that i can ignore these deviations for these simple comparisons.
Yes, the horn has a broad flat response compared to a straight non expanding quarterwave resonator. I define a quarter wave pipe as a non expanding (with 1% or so), straight pipe, anything expanding is a horn, even though it might be compromized in size.
The pure non expanding quarter-wave resonator will have a deeper response with more spl, compared to a horn of the same size (once again see the hornresp simulations i provided above). Instead of contradicting me based on "think" and "believe", please show me with a Hornresp/Akabak sim (like i try to do for everything i claim).
It is much easier to have a meaningful discussion that leads to a greater understanding and hopefully a development if there is at least some reapetable verifiable facts like a Hornresp simulation. It would be best to build and measure everything, but since no one wants to build a lot of boxes each day for a simple discussion, we should settle for a acceptable level of accuracy - Hornresp or Akabak..
Cheers,
Johannes
I have done these simulations in AkAbak many, many times - I just don't have them with me as I use my phone for most of my posts. I thought it was a basic tenet, that a TH has more specific efficiency than a straight pipe TL. Otherwise we would all build straight pipe TL's instead of folded horns. Certainly, a straight pipe is easier to build. I will get some sims to show you but if as I recall, the TH will get about a 10dB boost over a straight TL. It only makes sense if you think how the TH uses both sides of the driver: that is 3dB. An expanding horn gets another 6dB over a straight pipe easily. I am not trying to be a contrarian - it is just well known. Perhaps others with modeling experience can chime in. Btw, I totally agree that sims in either HR or AkAbak, etc are fine and good for this. I never said you have to build it to prove it.
So let me understand the constraints: both boxes have to be of equivalent internal volume, and is there any requirement in width of the passband and how flat it needs to be? Certainly, anyone can make a very high-Q resonator that has a high peak and no bandwidth.
So let me understand the constraints: both boxes have to be of equivalent internal volume, and is there any requirement in width of the passband and how flat it needs to be? Certainly, anyone can make a very high-Q resonator that has a high peak and no bandwidth.
Last edited:
Quarter wave vs Tapped Horn
Quarterwave vs Tapped Pipe (non expanding tapped horn)
I made some simulations. I tried to keep the total volume the same, but the expanding tapped horn got 0,7 liters more. Same driver (B&C12NW100) same power and same width of the passband (within reason). I tried to keep the lower corner as close as possible, and no extreme high Q peaks...
I can´t say it is a clear cut win for the expanding TH, especially considering the extreme compression i had to use to get the size down to the same 36 liters.
Cheers,
Johannes
Last edited:
I would never build such an expanding tapped horn. That extreme compression will lead to very big losses and extreme amounts of distortion. The simple non tapped QW seems very efficient, and usable. The tapped pipe has a very slight edge at the low corner, but nothing worth caring about.
It seems i was wrong about the efficiency of the QW, but considering the problems with using this extremely powerful driver in a as small box as possible, i would prefere the tapped pipe.
This was my starting point for this thread. I belive you can get a much higher power density (in real life with real drivers) from simple non expanding tapped horns or QW-pipes.
Cheers,
Johannes
It seems i was wrong about the efficiency of the QW, but considering the problems with using this extremely powerful driver in a as small box as possible, i would prefere the tapped pipe.
This was my starting point for this thread. I belive you can get a much higher power density (in real life with real drivers) from simple non expanding tapped horns or QW-pipes.
Cheers,
Johannes
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.