Its seems that the Collaborative tapped Horn Project thread has gone out of Control.
Why don't we start a thread on what the important driver parameters, and ratios are for a successful tapped horn.
The Few Nuggets that I have so far.
Driver:
Should have a higher Fs than intended operating range. (By what amount? an octave?)
Should have a relatively Stiff Suspension (Relatively High RMS?)
Cabinet (horn):
Needs to be a certain Percentage longer than the 1/4 wave of intended low cut. (What is that percentage?)
What sort of Compression Ratio is needed with relation to Driver Sd?
Questions:
What is an Optimal expansion from Throat Tap to mouth.
It would appear that a mouth smaller than the Area of the large end of the horn is useful??? Is there a Ratio for the mouth.
Is a tapped horn that Expands and contracts along its path benificial??
Observations:
I have noticed that many of the first attempts at building a tapped horn appeared to be imitating the DTS-20 but it became obvious upon my own early akabak modeling that they were making their internal path to short.
I built my 3 bend model before, the patent photos for the DTS-20 appeared, and got really lucky how close I ended up. There have been a few out there recently that have been longer paths. Of Various design.
I do see some promising HornResp models posted here but haven't seen too many recent results.
To many people have been modeling their Cabs In Eighth space, and getting bloated predicted results. They should be modeled in Half space unless your putting them in a concrete corner.
I hope we can turn this thread into a relatively un-congested resource for useful tapped horn design info. And links to other useful info.
As the other ones seem to be full of more questions about will this be a good driver for a tapped horn, can you model this design that I arbitrarily threw into this dimension.
It would be nice to just have a place to reference the details rather than sift through 1000+ posts
Cheers Boys!!
Antone-
Why don't we start a thread on what the important driver parameters, and ratios are for a successful tapped horn.
The Few Nuggets that I have so far.
Driver:
Should have a higher Fs than intended operating range. (By what amount? an octave?)
Should have a relatively Stiff Suspension (Relatively High RMS?)
Cabinet (horn):
Needs to be a certain Percentage longer than the 1/4 wave of intended low cut. (What is that percentage?)
What sort of Compression Ratio is needed with relation to Driver Sd?
Questions:
What is an Optimal expansion from Throat Tap to mouth.
It would appear that a mouth smaller than the Area of the large end of the horn is useful??? Is there a Ratio for the mouth.
Is a tapped horn that Expands and contracts along its path benificial??
Observations:
I have noticed that many of the first attempts at building a tapped horn appeared to be imitating the DTS-20 but it became obvious upon my own early akabak modeling that they were making their internal path to short.
I built my 3 bend model before, the patent photos for the DTS-20 appeared, and got really lucky how close I ended up. There have been a few out there recently that have been longer paths. Of Various design.
I do see some promising HornResp models posted here but haven't seen too many recent results.
To many people have been modeling their Cabs In Eighth space, and getting bloated predicted results. They should be modeled in Half space unless your putting them in a concrete corner.
I hope we can turn this thread into a relatively un-congested resource for useful tapped horn design info. And links to other useful info.
As the other ones seem to be full of more questions about will this be a good driver for a tapped horn, can you model this design that I arbitrarily threw into this dimension.
It would be nice to just have a place to reference the details rather than sift through 1000+ posts
Cheers Boys!!
Antone-
The problem i see with this is, that the only possible answer to the questions is, that it depends on the driver and the intended response - no specific answer is possible. A horn with little mouth to throat ratio may be a good solution in one case and sub-optimal in another case. Same goes for every system parameter. It all depends on what you have (ie. the driver) and what you want to achieve with it. You can make practically every woofer have a more or less acceptable response in at least one frequency range. But actually this is a good thing! You dont have to find the magic parameter set in horn and woofer, because everything will work if you choose a fitting set of parameters.
100% true MaVo. There are no magic ratios or set rules. It all depends on your application. Besides, an hour tooling around with McBean's Hornresp program is all you need. It's too easy to design a good tapped horn with Hornresp to waste time trying to make up rules that don't always apply.
Rgs, JLH
Rgs, JLH
On that thinking, I'd earlier proposed to Don, to publish his 6 different TH's that's he's drawn up in cad, in a single thread. (maybe a sticky??) Those 6 span from PA to LFE HT subs, so maybe that would be a good starting point. And then, whenever someone comes up with a great addition that's proven out, add that to the list.
If someone would take a little time to add some commentary on the how's/why's of each driver/design, it could be a mini how-to for using hornresp, and help those just starting out. It might help someone choose an appropriate driver, of if they already have one, look and see if there is a published design that's a good starting point.
And for those who just want a set of plans, and go with a proven design... it's in one place.
That thread would need to be kept clean, no posts, just plans & documentation.
If someone would take a little time to add some commentary on the how's/why's of each driver/design, it could be a mini how-to for using hornresp, and help those just starting out. It might help someone choose an appropriate driver, of if they already have one, look and see if there is a published design that's a good starting point.
And for those who just want a set of plans, and go with a proven design... it's in one place.
That thread would need to be kept clean, no posts, just plans & documentation.
I think there are some rules of thumb.
There is probably a relationship between Driver (Fs) and other Critical TS Parameters, and Desired TH. High Pass.
I'm sure by exploring successful TH designs, those relationships can be derived I expect that there are trends.
I'm pretty sure that there is a somewhat a optimal or Ideal percentage of path length longer than the Horns useful High pass 1/4 wave length.
I think a formula can be derived, that one could plug a target high pass knee, and get a rough Total path length.
I also think that a rough Expansion rate should be able to be found with some of the other TS parameters, and compression ratio.
To me an optimal model should, be shoot for, flattest response to high pass knee, SPL, and excursion control.
Those criteria, will steady the target a little. I think it should be easily scalable from those constraints.
Antone-
There is probably a relationship between Driver (Fs) and other Critical TS Parameters, and Desired TH. High Pass.
I'm sure by exploring successful TH designs, those relationships can be derived I expect that there are trends.
I'm pretty sure that there is a somewhat a optimal or Ideal percentage of path length longer than the Horns useful High pass 1/4 wave length.
I think a formula can be derived, that one could plug a target high pass knee, and get a rough Total path length.
I also think that a rough Expansion rate should be able to be found with some of the other TS parameters, and compression ratio.
To me an optimal model should, be shoot for, flattest response to high pass knee, SPL, and excursion control.
Those criteria, will steady the target a little. I think it should be easily scalable from those constraints.
Antone-
Antone,
I was interested to see your post, after slogging through about 55 pages of posts on the Collaborative tapped Horn Project last night with very little new info.
I seem to be following some of your (and Tom Danley’s) paths.
Around last January I started checking out the Collaborative tapped Horn Project, and decided to try converting my existing 65” path spiral horns using Lab 12s to a longer path length tapped horn.
The results were disappointing. After reading about your bass cabinets using 2x12” Labs, and after much experimentation, I ended up converting my boxes to dual Lab 12” ported cabinets.
The 2x12” put out more output in the 30-50 range than the 1x12” spiral horns, using 2/3rds the space, only 26.5” x 22.5” x22.5”, weighed the same (100 pounds) cabinets, can do about 130 dB with 2000 watts at one meter half space.
While in the process of that experiment, a friend called up, wanting to add subs to his Peavey system. He is a Peavey dealer, and had 4) low rider 15” that he wanted to use. Since he was using 4) 2x15” cabinets with 2” horns, I doubted that a pair of 2x15” ported cabinets with the Low Riders would keep up, and I suggested he build tapped horns for them.
I have never been a big fan of Peavey, but what the hell, here was a chance for me to get some testing done after someone else made the sawdust!
I designed him a 2x15” tapped horn truck integer cabinet, 32” x30” x 45”, horn path length around 105”, which according to the math should be about a 32 HZ tapped horn. He built a pair, and I went up to assist him in setting up his PA with the new subs out doors for a snowboard competition in Colorado.
Outdoors in the snow, we had little time to do a decent test, but from what I could determine, the 2x15”
at a 20 volt level (50 watts per cone) at one meter, mic on the ground plane, ranged from 116 to 123 dB SPL from 30 to 70 HZ, which averages out to a 99. 25 dB SPL at one meter reading.
The test used sine wave tones at 5 HZ intervals from 30 to 70 HZ. The peak to peak excursion went to 10 mm at 30 hz, only 2.5 at 35 HZ, 6 at 40, 7 at 45 and 50, 6 at 55, 5 at 60 and 1 at 70 HZ.
In Peavey’s literature on this sub, it shows a single Low Rider should do about 94 dB SPL in the range tested in the “Large 4.5 cubic foot cabinet”.
Adding 3 DB expected gain from having two cones to the ported cabinet should show 97 dB SPL .
The tapped horn would appear to be only 2.25 dB more efficient (sensitive?) using a 25 cubic foot cabinet compared to a 9 cubic foot ported cabinet. Not good at all in terms of real estate, unless the TH was way less distorted. Unfortunately I did not test that parameter and had never listened to the Low riders in ported cabinets, so I can’t offer any subjective difference, though I would say that the Low Riders seemed quite “clean” in the tapped horn from what I heard.
That said, the tapped horn subs seemed to “keep up” with 8) 15” above them, which would indicate subjectively that they may have been more quite a bit more efficient.
Bottom line, it would be very interesting to compare real world measured response of specific cones in “optimum” bass reflex cabinets compared to the same real world measurements using tapped horns of the same volume.
I finally have the SMAART program, and excellent test mics, so anyone in the Albuquerque/Santa Fe area that wants to do some real world “before and after” tests could be accommodated.
Without a ported vs tapped horn cabinet size consideration, which for the most part seems lacking in the Collaborative tapped Horn Project, it is hard to make valid comparisons.
Art
I was interested to see your post, after slogging through about 55 pages of posts on the Collaborative tapped Horn Project last night with very little new info.
I seem to be following some of your (and Tom Danley’s) paths.
Around last January I started checking out the Collaborative tapped Horn Project, and decided to try converting my existing 65” path spiral horns using Lab 12s to a longer path length tapped horn.
The results were disappointing. After reading about your bass cabinets using 2x12” Labs, and after much experimentation, I ended up converting my boxes to dual Lab 12” ported cabinets.
The 2x12” put out more output in the 30-50 range than the 1x12” spiral horns, using 2/3rds the space, only 26.5” x 22.5” x22.5”, weighed the same (100 pounds) cabinets, can do about 130 dB with 2000 watts at one meter half space.
While in the process of that experiment, a friend called up, wanting to add subs to his Peavey system. He is a Peavey dealer, and had 4) low rider 15” that he wanted to use. Since he was using 4) 2x15” cabinets with 2” horns, I doubted that a pair of 2x15” ported cabinets with the Low Riders would keep up, and I suggested he build tapped horns for them.
I have never been a big fan of Peavey, but what the hell, here was a chance for me to get some testing done after someone else made the sawdust!
I designed him a 2x15” tapped horn truck integer cabinet, 32” x30” x 45”, horn path length around 105”, which according to the math should be about a 32 HZ tapped horn. He built a pair, and I went up to assist him in setting up his PA with the new subs out doors for a snowboard competition in Colorado.
Outdoors in the snow, we had little time to do a decent test, but from what I could determine, the 2x15”
at a 20 volt level (50 watts per cone) at one meter, mic on the ground plane, ranged from 116 to 123 dB SPL from 30 to 70 HZ, which averages out to a 99. 25 dB SPL at one meter reading.
The test used sine wave tones at 5 HZ intervals from 30 to 70 HZ. The peak to peak excursion went to 10 mm at 30 hz, only 2.5 at 35 HZ, 6 at 40, 7 at 45 and 50, 6 at 55, 5 at 60 and 1 at 70 HZ.
In Peavey’s literature on this sub, it shows a single Low Rider should do about 94 dB SPL in the range tested in the “Large 4.5 cubic foot cabinet”.
Adding 3 DB expected gain from having two cones to the ported cabinet should show 97 dB SPL .
The tapped horn would appear to be only 2.25 dB more efficient (sensitive?) using a 25 cubic foot cabinet compared to a 9 cubic foot ported cabinet. Not good at all in terms of real estate, unless the TH was way less distorted. Unfortunately I did not test that parameter and had never listened to the Low riders in ported cabinets, so I can’t offer any subjective difference, though I would say that the Low Riders seemed quite “clean” in the tapped horn from what I heard.
That said, the tapped horn subs seemed to “keep up” with 8) 15” above them, which would indicate subjectively that they may have been more quite a bit more efficient.
Bottom line, it would be very interesting to compare real world measured response of specific cones in “optimum” bass reflex cabinets compared to the same real world measurements using tapped horns of the same volume.
I finally have the SMAART program, and excellent test mics, so anyone in the Albuquerque/Santa Fe area that wants to do some real world “before and after” tests could be accommodated.
Without a ported vs tapped horn cabinet size consideration, which for the most part seems lacking in the Collaborative tapped Horn Project, it is hard to make valid comparisons.
Art
weltersys
Like the beginning of this thread said. Not all drivers are good in TH's.
And did you design the enclosure with akAbak or HR?
I have a particular 15 that dosent work so
well in a TH but in a normal horn it bangs. (which I believe is due to the driver having a "loose" cone with allot of Xmax. IE it needs to rear to be enclosed for spring pressure.)
I think the big plus to THs is for small enclosures.
Not hard getting something 4-500 Lt to do good freq/spl #'s BUT to get something say less than 50 to do it is another story. I have an 8" design (Tang Band W8-740P) that is 47.5 Lt's that is 126 db 50-125 with -10 Db being 37 and 166 Hz. (a bit of EQ to get flat above 125 though on roughly 170 watts). I am looking for a different driver for this design. Something as cheap with a higher SPL. (as this one is 84db)
I have built 3 TH's so far.
The first was a test box for the 15 I referred to
in the part 1. The other 2 are single 12" in roughly 102 Lt box (box total is roughly 133). 130 Db 50-275 Hz. -10 is 40 Hz. I only run them to 200 Hz. That is on 430 ish watts. I have used these a few times and love them.
One thing I have found from looking at
other peoples designs are that there are 2 distinctions. Home Vs PA. Home (HT) wants a super flat response graph going down into the teens. Which is pretty simple to do given that the enclosure wont be being moved and also room response. Where as for PA you want SPL then flat response. I design for max SPL at 1 Pi / 2 Pi and not 1w@1m at .5 Pi . (yes those #s above are at 1 Pi)
You might want to look at ScreamUSA's design.
Like the beginning of this thread said. Not all drivers are good in TH's.
And did you design the enclosure with akAbak or HR?
I have a particular 15 that dosent work so
well in a TH but in a normal horn it bangs. (which I believe is due to the driver having a "loose" cone with allot of Xmax. IE it needs to rear to be enclosed for spring pressure.)
I think the big plus to THs is for small enclosures.
Not hard getting something 4-500 Lt to do good freq/spl #'s BUT to get something say less than 50 to do it is another story. I have an 8" design (Tang Band W8-740P) that is 47.5 Lt's that is 126 db 50-125 with -10 Db being 37 and 166 Hz. (a bit of EQ to get flat above 125 though on roughly 170 watts). I am looking for a different driver for this design. Something as cheap with a higher SPL. (as this one is 84db)
I have built 3 TH's so far.
The first was a test box for the 15 I referred to
in the part 1. The other 2 are single 12" in roughly 102 Lt box (box total is roughly 133). 130 Db 50-275 Hz. -10 is 40 Hz. I only run them to 200 Hz. That is on 430 ish watts. I have used these a few times and love them.
One thing I have found from looking at
other peoples designs are that there are 2 distinctions. Home Vs PA. Home (HT) wants a super flat response graph going down into the teens. Which is pretty simple to do given that the enclosure wont be being moved and also room response. Where as for PA you want SPL then flat response. I design for max SPL at 1 Pi / 2 Pi and not 1w@1m at .5 Pi . (yes those #s above are at 1 Pi)
You might want to look at ScreamUSA's design.
Hi FlipC
I am using a Mac, and unfortunately there seems to be no modeling programs that work on the Mac platforms.
I like the fact that various modeling programs can prevent cutting up wood in vain, but there are many things that can only be learned by actually building and measuring the results.
The huge peaks and dips the modeling programs predict in the upper ranges of the TH that actually don’t happen in the real world are a case in point.
Being an empirical type of designer
^), I like “rules of thumb”, to get in the ballpark before I butcher the wood, regardless of what a computer prediction says.
You said:
“there are 2 distinctions. Home Vs PA. Home (HT) wants a super flat response graph going down into the teens. Which is pretty simple to do given that the enclosure wont be being moved and also room response. Where as for PA you want SPL then flat response. I design for max SPL at 1 Pi / 2 Pi and not 1w@1m at .5 Pi . (yes those #s above are at 1 Pi)
I am not sure of what you mean by 1 Pi / 2 Pi and .5 Pi.
I prefer measurements and predictive programs in 1/2 space, as most of us locate our subs on the ground plane, while the distance to walls varies significantly.
Also, -10 dB SPL response sounds half as loud, while your ear hears progressively less in the low end. In the low frequency range, attention to the -3 dB low corner should be paid close attention in a TH cabinets.
Crap, I have not added anything worthwhile to Antone’s initial thread.
I am using a Mac, and unfortunately there seems to be no modeling programs that work on the Mac platforms.
I like the fact that various modeling programs can prevent cutting up wood in vain, but there are many things that can only be learned by actually building and measuring the results.
The huge peaks and dips the modeling programs predict in the upper ranges of the TH that actually don’t happen in the real world are a case in point.
Being an empirical type of designer
^), I like “rules of thumb”, to get in the ballpark before I butcher the wood, regardless of what a computer prediction says.You said:
“there are 2 distinctions. Home Vs PA. Home (HT) wants a super flat response graph going down into the teens. Which is pretty simple to do given that the enclosure wont be being moved and also room response. Where as for PA you want SPL then flat response. I design for max SPL at 1 Pi / 2 Pi and not 1w@1m at .5 Pi . (yes those #s above are at 1 Pi)
I am not sure of what you mean by 1 Pi / 2 Pi and .5 Pi.
I prefer measurements and predictive programs in 1/2 space, as most of us locate our subs on the ground plane, while the distance to walls varies significantly.
Also, -10 dB SPL response sounds half as loud, while your ear hears progressively less in the low end. In the low frequency range, attention to the -3 dB low corner should be paid close attention in a TH cabinets.
Crap, I have not added anything worthwhile to Antone’s initial thread.
GM
Yes, there is a thread on it. I cant recall if it was in the Collab thread or not. Even with my builds I can attest to it.
weltersys
0.5 Pi = Corner loading =1/8 space
1 Pi = Floor & Wall =1/4 space
2 Pi = Ground only =1/2 space
4 Pi = Full space =Full space
Most people use 1/8 space.
Especially HT environments. IE room gain etc. I use 1/4 space as I will be running more than 1 sub. Easiest way for me to accommodate a db gain for TH's in HR.
As far as the -10db goes.
This is to show the drop off rate. Which is sharp in my case. At 120DB that is still a usable level for a PA though.
I am attempting to install OSx86
for testing purposes. But since OSX is FreeBSD re wrapped. And I believe HornResp works on linux. You should be able to run in Mac.
If anything under emulation. Definately worth the time to figure it out. A great program. Just need someone to build a folding program to use its info.
Now back to original topic.
Since manufacturer's do not release any info on the amount of compression a particular cone can handle. It is a general rule that a TH likes a stiff coned driver. I am going off of what I have modeled and built. This is a priority over Xmax. IE the 15" inch driver I have with 15+ mm of Xmax and loose cone vs the 12" with 8mm (or is it 6..?) with a super tight cone. The 12 outperforms. (other values being roughly the same) Though the 15 in a big box (400-500 Lt) will go into the sub sonic region easily - IE sub 20 Hz)
Power rating on a driver isnt as crucial
as say in a sealed or ported enclosure. My models have shown that you will reach Xmax before peak power. Now this isnt always the case. In one I had the compression ratio high and it was showing it could takeover 3 times it's rated power before it reached xmax.
I still say you would have to have 2 categories
for "rules of thumb" for TH's. Home Theater (TH) usage and Pro Sound (PA). My 15 is much better suited for HT while the 12 is for PA.
Yes, there is a thread on it. I cant recall if it was in the Collab thread or not. Even with my builds I can attest to it.
weltersys
0.5 Pi = Corner loading =1/8 space
1 Pi = Floor & Wall =1/4 space
2 Pi = Ground only =1/2 space
4 Pi = Full space =Full space
Most people use 1/8 space.
Especially HT environments. IE room gain etc. I use 1/4 space as I will be running more than 1 sub. Easiest way for me to accommodate a db gain for TH's in HR.
As far as the -10db goes.
This is to show the drop off rate. Which is sharp in my case. At 120DB that is still a usable level for a PA though.
I am attempting to install OSx86
for testing purposes. But since OSX is FreeBSD re wrapped. And I believe HornResp works on linux. You should be able to run in Mac.
If anything under emulation. Definately worth the time to figure it out. A great program. Just need someone to build a folding program to use its info.
Now back to original topic.
Since manufacturer's do not release any info on the amount of compression a particular cone can handle. It is a general rule that a TH likes a stiff coned driver. I am going off of what I have modeled and built. This is a priority over Xmax. IE the 15" inch driver I have with 15+ mm of Xmax and loose cone vs the 12" with 8mm (or is it 6..?) with a super tight cone. The 12 outperforms. (other values being roughly the same) Though the 15 in a big box (400-500 Lt) will go into the sub sonic region easily - IE sub 20 Hz)
Power rating on a driver isnt as crucial
as say in a sealed or ported enclosure. My models have shown that you will reach Xmax before peak power. Now this isnt always the case. In one I had the compression ratio high and it was showing it could takeover 3 times it's rated power before it reached xmax.
I still say you would have to have 2 categories
for "rules of thumb" for TH's. Home Theater (TH) usage and Pro Sound (PA). My 15 is much better suited for HT while the 12 is for PA.
I also would love to see some distilled recipies for tapped horn construction. Even rough formulas for S1 and S4 area based on driver sd, what T/S parameters to target, etc would be very helpful.
Putting random drivers into hornresp and playing with the wizard passes the time, but it doesn't seem the most sane route for designing to particular goals.
Likewise it might be nice to start gathering up finished tapped horn designs and their measurements in a thread easier to scan than the monster thread.
Lastly, I've gotten both hornresp and akabak to run under parallels on OSX.
I'd be very interested in writing an open source cross platform tool like akabak if someone could help fill in my math deficiencies.
Putting random drivers into hornresp and playing with the wizard passes the time, but it doesn't seem the most sane route for designing to particular goals.
Likewise it might be nice to start gathering up finished tapped horn designs and their measurements in a thread easier to scan than the monster thread.
Lastly, I've gotten both hornresp and akabak to run under parallels on OSX.
I'd be very interested in writing an open source cross platform tool like akabak if someone could help fill in my math deficiencies.
GM said:
Were the cases, that too low measurement resolution made the high q peaks look smaller than they are and that room modes masked them, ruled out? As long as this isnt the case, i prefer to think that the simulations are showing reality. Since, once i start doubting the sims, i may as well build a horn without them and end up in the very position i wanted to criticise with my initial post.
I am still not sure, what one could gain from rules of thump. Once one has a clear vision of what the horn has to do, ie upper and lower corner frequency, max spl, etc, you just dial around in hornresp until those demands are satisfied. Where do the rules of thump grip in this scenario?
For example: I want a horn for pro sound. It shall be able to do x spl, above the frequency y. I choose a driver, which is fitting for my intended use (in this case a pro sound woofer) with power handling and xmax, so that my X spl demand can be satisfied. Then i type in some random horn params and press ctrl+h. Then i adjust horn length so that lower corner is satisfied. Then i play with the area sliders (and maybe front chamber and added inductance, which are often not needed) until i get a flat response. Then i do a max spl check in hornresp and see if my driver choice was ok, or if i need to buy a more expensive driver.
Now you can do some customisation. For example, if you have truck space availabe but no money for additional drivers, you can make the hornmouth bigger for some additional spl. Or you could decide to use several small horns instead of one big to buy several cheap drivers instead of one big. But this is just a matter of taste.
As a last step, post your design along with a 3d pic of the folding and ask if others think its good. There are allways small details that could be made better.
No rules of thump used in this - or maybe, this is a rule of thump?
For example: I want a horn for pro sound. It shall be able to do x spl, above the frequency y. I choose a driver, which is fitting for my intended use (in this case a pro sound woofer) with power handling and xmax, so that my X spl demand can be satisfied. Then i type in some random horn params and press ctrl+h. Then i adjust horn length so that lower corner is satisfied. Then i play with the area sliders (and maybe front chamber and added inductance, which are often not needed) until i get a flat response. Then i do a max spl check in hornresp and see if my driver choice was ok, or if i need to buy a more expensive driver.
Now you can do some customisation. For example, if you have truck space availabe but no money for additional drivers, you can make the hornmouth bigger for some additional spl. Or you could decide to use several small horns instead of one big to buy several cheap drivers instead of one big. But this is just a matter of taste.
As a last step, post your design along with a 3d pic of the folding and ask if others think its good. There are allways small details that could be made better.
No rules of thump used in this - or maybe, this is a rule of thump?
MaVo said:
Obviously possible, but haven't done any proofs-of-concept, so have to rely on others input. The proof is in the listening IMO and a number of folks have commented that it's not the case, though in theory the sims where the driver is facing out at the extreme end of the mouth would have the greatest amount, so no doubt in my mind that since most drivers must be mounted frame out that this damps it some, especially if its bulk isn't accounted for. Also, most TH pics posted show a cutout in the side of the horn Vs the straight mouth HR assumes (don't know about Akabak) which in theory should affect it too. With the side mouths, it seems that damping the walls around the driver might be worthwhile if the sims are accurate. After all, the DTS20's driver is way up the horn where it's acoustically better damped and TD still felt the need to add what appears to be a fair amount of damping around it.
GM
in my case,
which is PA use, I run a digital crossover with a 48 db cut at 200 Hz so
even if the cabinet would create peaks that the sim shows (between 200-400 Hz) I am not reproducing it.
Mavo hit it on the nail.
But lets get back to the OG questions of the thread that sumsound asked:
1.) We know that the driver have a higher Fs than intended operating range. To what amount?
Hmm, Looking at the sims and builds I would say it needs to be close to your intended Fs. Definitely not a full octave. My rule of thumb has become no more than 10 Hz no less than 5 within my intended Fs.
2.) Should have a relatively Stiff Suspension (Relatively High RMS?)
Yes, as I have tested this. 15" driver with 15 mm Xmax - 3.3 RMS Vs 12 with 8mm Xmax - 7.9 RMS. The 12 outperforms.
3.)Needs to be a certain Percentage longer than the 1/4 wave of intended low cut. (What is that percentage?)
Hmm, I haven't looked into this. The first TH I did was by paper and built around a final size. The second 2 were built using HR first around a given size.
4.)What sort of Compression Ratio is needed with relation to Driver Sd?
Would there be a rule for this. I would think nothing more than the general rules that already apply for horns.
5.) What is an Optimal expansion from Throat Tap to mouth?
HR already gives you this. I make the first and last correct and give the middle an expansion off the first. IE I don't sweat it to be 100%
I am using EQ after all.
6.) It would appear that a mouth smaller than the Area of the large end of the horn is useful??? Is there a Ratio for the mouth.
I don't understand the question.
7.)s a tapped horn that Expands and contracts along its path beneficial?
I think questions 5-7 are answered by HR. It gives you the optimum layout already. Beyond the initial compression ratio I don't think it would be of any benefit to re compress the waves.
which is PA use, I run a digital crossover with a 48 db cut at 200 Hz so
even if the cabinet would create peaks that the sim shows (between 200-400 Hz) I am not reproducing it.
Mavo hit it on the nail.
But lets get back to the OG questions of the thread that sumsound asked:
1.) We know that the driver have a higher Fs than intended operating range. To what amount?
Hmm, Looking at the sims and builds I would say it needs to be close to your intended Fs. Definitely not a full octave. My rule of thumb has become no more than 10 Hz no less than 5 within my intended Fs.
2.) Should have a relatively Stiff Suspension (Relatively High RMS?)
Yes, as I have tested this. 15" driver with 15 mm Xmax - 3.3 RMS Vs 12 with 8mm Xmax - 7.9 RMS. The 12 outperforms.
3.)Needs to be a certain Percentage longer than the 1/4 wave of intended low cut. (What is that percentage?)
Hmm, I haven't looked into this. The first TH I did was by paper and built around a final size. The second 2 were built using HR first around a given size.
4.)What sort of Compression Ratio is needed with relation to Driver Sd?
Would there be a rule for this. I would think nothing more than the general rules that already apply for horns.
5.) What is an Optimal expansion from Throat Tap to mouth?
HR already gives you this. I make the first and last correct and give the middle an expansion off the first. IE I don't sweat it to be 100%
I am using EQ after all.
6.) It would appear that a mouth smaller than the Area of the large end of the horn is useful??? Is there a Ratio for the mouth.
I don't understand the question.
7.)s a tapped horn that Expands and contracts along its path beneficial?
I think questions 5-7 are answered by HR. It gives you the optimum layout already. Beyond the initial compression ratio I don't think it would be of any benefit to re compress the waves.
GM said:
I haven't had the chance to do any proper measurements on my TH build, but running a frequency sweep shows there is a lot of HF hash. Even though I have the sub crossed at 80hz (sorry I don't know the slope of my AVP's crossover) I was getting a 200hz signal that, at a guess, was 0db down from the bandpass signal.
Now I'm using DSP on the sub channel for eq and to attenuate the signal above the band pass.
When a/b ing the DSP in and out of the signal path I don't notice a huge difference when it's eliminating the hash, but it does sound a bit better. My sub is between my mains which helps.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.