AndrewT said:
With a point source driver, ~three octaves of actual horn loading due to compression ratio (CR) limitations. Horns are normally designed to be a combination of compression plus rising on axis output though, so can cover many octaves if a very narrow 'sweet spot' is acceptable and a proper phase plug is used.
GM
MikeHunt79 said:
Do this only if you are prepared to start thinking about room treatment. The little notch due to unperfect addition around the xo will be swamped by numbers of much bigger peaks and notches from the room modes. I would say, if it sounds good, let it be, since you probably cant see it anyways.
As an example i added a picture of the inroom response of a loudspeaker, which would be flat from 20-300hz without the modes.
Attachments
My 40Hz tapped horns
I finally decided to pull the pictures of my 40Hz tapped horns off the digital camera. Below are a few shots during construction. The outer dimensions are 8.5” deep X 10” wide X 54” tall. So, for 40Hz performance it is quite small.
First picture is of the folding without the braces in.
The below picture shows the access door and where the woofer mounts.
This next picture is the horn with the braces in place. The braces are ½” birch ply and the rest of the horn is ¾” birch ply. This is right before the last side goes on and it is sealed up forever.
Here the B&C 8PE21-16 is going into the horn. It is a tight fit, but worth it because the horn is very small compared to my old ones.
Lastly, the woofer nestled into its final place. The bass output from a pair of these is very impressive. Especially considering their rated Xmax is 1mm and Xmech is 4.5mm.
I would encourage anyone to model and build tapped horns. They perform great, they are easy to build, and are the real deal.
Rgs, JLH
I finally decided to pull the pictures of my 40Hz tapped horns off the digital camera. Below are a few shots during construction. The outer dimensions are 8.5” deep X 10” wide X 54” tall. So, for 40Hz performance it is quite small.
First picture is of the folding without the braces in.
The below picture shows the access door and where the woofer mounts.
This next picture is the horn with the braces in place. The braces are ½” birch ply and the rest of the horn is ¾” birch ply. This is right before the last side goes on and it is sealed up forever.
Here the B&C 8PE21-16 is going into the horn. It is a tight fit, but worth it because the horn is very small compared to my old ones.
Lastly, the woofer nestled into its final place. The bass output from a pair of these is very impressive. Especially considering their rated Xmax is 1mm and Xmech is 4.5mm.
I would encourage anyone to model and build tapped horns. They perform great, they are easy to build, and are the real deal.
Rgs, JLH
I have read the whole thread so far. I have some questions that I have not been able to answer by the info in this thread.
- Driver selection:
I have found that there seems to be three schools regarding the Mmd/Bl proportions. If using a ratio of Mmd/Bl you find: 1) A ratio of approx. maximum 3 where there are preferably lighter cones, say below 100 grams. 2) Heavier cones than in the first group, no apparent upper limit in Mmd, Bl said to be high. Both of these first group of drivers are mostly PA-drivers. 3) Car drivers, with very heavy cones and lower Bl than in the first two groups. Is a car driver cone never to heavy, as I've seen cones having Mmd:s approaching 400g. What makes the car drivers so suitable for THs?
Regarding light cones and the risk of shredding, is that a likely scenario even for home listening levels for a 15"?
To increase my lack of understanding of the subject further, there are sweeping suggestions like large Xmax, large Bl etc, when is Xmax large, and when is Bl large for taking two examples?
I have also seen that too high a Bl, as with PA drivers, will also cause problems, when is Bl starting to get too high? Is it too high by its force of its own, or in relation to Mmd?
I've also seen it suggested that the drivers for THs aught to have stiffer suspensions, I guess that it is the parameter Cms you are looking at then? When is the suspension stiffer than normal?
Qms and Qes should be higher than for a "normal" horn driver. Higher than?
I have also seen a ratio Fs/Qts, what's it for?
Is it insignificant in which direction the driver is mounted? That is, if the magnet is in the throat or reversed?
These were mostly driver related questions, I guess I have other questions as well, but I save those for a rainy day...
- Driver selection:
I have found that there seems to be three schools regarding the Mmd/Bl proportions. If using a ratio of Mmd/Bl you find: 1) A ratio of approx. maximum 3 where there are preferably lighter cones, say below 100 grams. 2) Heavier cones than in the first group, no apparent upper limit in Mmd, Bl said to be high. Both of these first group of drivers are mostly PA-drivers. 3) Car drivers, with very heavy cones and lower Bl than in the first two groups. Is a car driver cone never to heavy, as I've seen cones having Mmd:s approaching 400g. What makes the car drivers so suitable for THs?
Regarding light cones and the risk of shredding, is that a likely scenario even for home listening levels for a 15"?
To increase my lack of understanding of the subject further, there are sweeping suggestions like large Xmax, large Bl etc, when is Xmax large, and when is Bl large for taking two examples?
I have also seen that too high a Bl, as with PA drivers, will also cause problems, when is Bl starting to get too high? Is it too high by its force of its own, or in relation to Mmd?
I've also seen it suggested that the drivers for THs aught to have stiffer suspensions, I guess that it is the parameter Cms you are looking at then? When is the suspension stiffer than normal?
Qms and Qes should be higher than for a "normal" horn driver. Higher than?
I have also seen a ratio Fs/Qts, what's it for?
Is it insignificant in which direction the driver is mounted? That is, if the magnet is in the throat or reversed?
These were mostly driver related questions, I guess I have other questions as well, but I save those for a rainy day...
So what frequency should one typically design TH in order to get response down to 20Hz IN ROOM (HT / music purposes). I know the answer is partly room-dependant, but is a 25 Hz horn in the ballpark, or a 30 Hz horn ... ? Any practical advice from those that have built these.
I'd think that coupled with room gain, a 20 Hz TH might be a mess, in room. At the same time, should I be concerned about over-excursion or unloading at say, 20 Hz, if I were to build a smaller (say 30Hz) TH horn.
THanks
I'd think that coupled with room gain, a 20 Hz TH might be a mess, in room. At the same time, should I be concerned about over-excursion or unloading at say, 20 Hz, if I were to build a smaller (say 30Hz) TH horn.
THanks
buggsson
I'll try to point you in the right direction
"What makes the car drivers so suitable for THs?"
Probably higher fs = stiffer suspensions for the same Mmd (at least for some of those that have been mentioned in the thread)
"Regarding light cones and the risk of shredding, is that a likely scenario even for home listening levels for a 15"?"
I doubt that there is the slightest risk of damage to my PD1550 which has a compression ratio around 3:1. Something in my room will break first, I think. Probably it is not a good idea to use a much lighter cone (<<123g) if that forces you to use a higher compression ratio too.
Note that a 15" TH will be quite big (ca. 300l for a 25Hz cutoff), but very satisfying.
"To increase my lack of understanding of the subject further, there are sweeping suggestions like large Xmax, large Bl etc, when is Xmax large, and when is Bl large for taking two examples?"
and
"I have also seen that too high a Bl, as with PA drivers, will also cause problems, when is Bl starting to get too high? Is it too high by its force of its own, or in relation to Mmd?"
I don't think sweeping suggestions are all that useful. Xmax sets the maximum level for low distortion, and taste in music and level probably varies more than xmax on drivers suitable for TH use. The motor strength provides both sensitivity and damping, which can be traded off to some extent against the compression ratio in the TH. I think that is why there can be good results with low (10Tm) and high (20+Tm) BL designs with not very different Sd and Mmd.
Take a close look at the response shape and see if there is a droop in the middle, to me this suggests a non-optimum match of motor strength and horn load (but probably only in a theoretical sense, as the builders are happy with their designs). I felt the PA /high Bl route provided a design that was quite stable against small parameter errors. So even if the datasheet was not quite right the response should be not too different from the prediction. I think this is relatively important for DIY.
"I've also seen it suggested that the drivers for THs aught to have stiffer suspensions, I guess that it is the parameter Cms you are looking at then? When is the suspension stiffer than normal?"
To avoid an undamped lower end it seems necessary to have fs a bit higher than the lower cutoff (I suppose the typical factor is about 1.5 times the lower cutoff). You may as well express that by fs and Mmd as Cms (the stiffness of the suspension determines the resonant frequency for given Mmd).
"Qms and Qes should be higher than for a "normal" horn driver. Higher than?"
Really? Qms should be high as mechanical loss means loss of efficiency. Qes should be correct to match the horn. It is effectively the same thing as talking about Bl above (high Bl implies low Qes, other parameters fixed).
"I have also seen a ratio Fs/Qts, what's it for?"
It is a guide to the frequency above which the response begins to slope down in certain circumstances. I'm not sure that it is all that useful in TH where the HF response has all the pipe resonances anyway. I don't remember when it was used in the thread though.
"Is it insignificant in which direction the driver is mounted? That is, if the magnet is in the throat or reversed?"
So long as it fits - you don't want a vented magnet to be right up against the back wall in case you make a whistle.
Most of your points can be answered for yourself with a few hours playing with Hornresp (easier) or AkAbak, starting from some of the designs that have been posted. Have fun, and definitely build a TH at the end of the day.
Ken
I'll try to point you in the right direction
"What makes the car drivers so suitable for THs?"
Probably higher fs = stiffer suspensions for the same Mmd (at least for some of those that have been mentioned in the thread)
"Regarding light cones and the risk of shredding, is that a likely scenario even for home listening levels for a 15"?"
I doubt that there is the slightest risk of damage to my PD1550 which has a compression ratio around 3:1. Something in my room will break first, I think. Probably it is not a good idea to use a much lighter cone (<<123g) if that forces you to use a higher compression ratio too.
Note that a 15" TH will be quite big (ca. 300l for a 25Hz cutoff), but very satisfying.
"To increase my lack of understanding of the subject further, there are sweeping suggestions like large Xmax, large Bl etc, when is Xmax large, and when is Bl large for taking two examples?"
and
"I have also seen that too high a Bl, as with PA drivers, will also cause problems, when is Bl starting to get too high? Is it too high by its force of its own, or in relation to Mmd?"
I don't think sweeping suggestions are all that useful. Xmax sets the maximum level for low distortion, and taste in music and level probably varies more than xmax on drivers suitable for TH use. The motor strength provides both sensitivity and damping, which can be traded off to some extent against the compression ratio in the TH. I think that is why there can be good results with low (10Tm) and high (20+Tm) BL designs with not very different Sd and Mmd.
Take a close look at the response shape and see if there is a droop in the middle, to me this suggests a non-optimum match of motor strength and horn load (but probably only in a theoretical sense, as the builders are happy with their designs). I felt the PA /high Bl route provided a design that was quite stable against small parameter errors. So even if the datasheet was not quite right the response should be not too different from the prediction. I think this is relatively important for DIY.
"I've also seen it suggested that the drivers for THs aught to have stiffer suspensions, I guess that it is the parameter Cms you are looking at then? When is the suspension stiffer than normal?"
To avoid an undamped lower end it seems necessary to have fs a bit higher than the lower cutoff (I suppose the typical factor is about 1.5 times the lower cutoff). You may as well express that by fs and Mmd as Cms (the stiffness of the suspension determines the resonant frequency for given Mmd).
"Qms and Qes should be higher than for a "normal" horn driver. Higher than?"
Really? Qms should be high as mechanical loss means loss of efficiency. Qes should be correct to match the horn. It is effectively the same thing as talking about Bl above (high Bl implies low Qes, other parameters fixed).
"I have also seen a ratio Fs/Qts, what's it for?"
It is a guide to the frequency above which the response begins to slope down in certain circumstances. I'm not sure that it is all that useful in TH where the HF response has all the pipe resonances anyway. I don't remember when it was used in the thread though.
"Is it insignificant in which direction the driver is mounted? That is, if the magnet is in the throat or reversed?"
So long as it fits - you don't want a vented magnet to be right up against the back wall in case you make a whistle.
Most of your points can be answered for yourself with a few hours playing with Hornresp (easier) or AkAbak, starting from some of the designs that have been posted. Have fun, and definitely build a TH at the end of the day.
Ken
nope i tuned mine to round about 20-25hz doesn't sound boomy in room, and even at 20hz at a very loud level the cone doesn't seem to move to much
however if u make one of these use ply, i used 18mm mdf and i can feel the sides vibrating slightly
still its face down on the carpet atm, will add the rest of the bracing and the other side and see how it is
incase i need to access the driver im going to seal the top part with long strips of blue tack, that will be pushed and will mold any slight crack
i know it sounds kinda crude/stupid, but it will work (from experience)
then will paint as normal, but in a emergency i will be able to change the driver,
so so far so good, its being crossed at like 150hz at 12db/octave atm, far from ideal, but with them equilised down a tad, and my 12" sealed cabs running, u cannot hear any distortion worth caring about
buggsson, I think your approach isnt the best for getting a th sub. Trying to find subtle relationships between different TS parameters would be a hard task. My advice is: Take hornresp and simulate some horns. Use drivers with a somewhat higher fs than the intended low frequency corner. Stiff suspension is also a good idea. Large xmax only for extreme spl.
zobsky, that totally depends on your room. My answer is: Take the longest wall of your room, put the length in meters in the formula: 344 / ( 2 * longest wall ) and you get the frequency of your lowest room mode. Your room gain will not start exactly at this frequency, but maybe a half octave below the mode. Only hope to get some room gain if you have thick concrete walls and not much windows or doors.
zobsky, that totally depends on your room. My answer is: Take the longest wall of your room, put the length in meters in the formula: 344 / ( 2 * longest wall ) and you get the frequency of your lowest room mode. Your room gain will not start exactly at this frequency, but maybe a half octave below the mode. Only hope to get some room gain if you have thick concrete walls and not much windows or doors.
I wonder if a TH can be made to play into the higher frequencies, where the first few resonant peaks are, or if it sounds terrible there. Look at the picture i added. There are 4 resonances above the normal passband. If one tuned resonators to flatten them, would this extend the usable frequency range to about 20-180hz?
Attachments
Hallo Mavo,
your simulation is with a active sub modul,
only useful for~ 50 Hz 18 dB, listenable ~ 90 Hz.
i remember my post, to use the horn over 100 Hz,
you can manage this with bass horn mouth distance,
a partial delete between 100-250 Hz, by taking two smaller TH.
SAXOPHON example
your simulation is with a active sub modul,
only useful for~ 50 Hz 18 dB, listenable ~ 90 Hz.
i remember my post, to use the horn over 100 Hz,
you can manage this with bass horn mouth distance,
a partial delete between 100-250 Hz, by taking two smaller TH.
SAXOPHON example
