DVC in Hornresp
Dear church of fellow bassheads,
How would one go about entering T/S parameters for a DVC woofer in hornresp? Simply use the parameters of one coil and multiply voltage by the appropriate factor (dependent on series/par config) to get effective power? or do the parameters need to be calculated for both coils and entered as two configured coils looking like one effective load?
very happy new year from switzerland... may ye all grow happy and fat bellies.
Mike
Dear church of fellow bassheads,
How would one go about entering T/S parameters for a DVC woofer in hornresp? Simply use the parameters of one coil and multiply voltage by the appropriate factor (dependent on series/par config) to get effective power? or do the parameters need to be calculated for both coils and entered as two configured coils looking like one effective load?
very happy new year from switzerland... may ye all grow happy and fat bellies.
Mike
My third TH.
Im building a small TH with a 5,25 inch speaker to my friend.
He saw this video and was hooked. http://www.youtube.com/watch?v=dKQTo0w5sNE
Im building a small TH with a 5,25 inch speaker to my friend.
He saw this video and was hooked. http://www.youtube.com/watch?v=dKQTo0w5sNE
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
This is what hornresp has to say about it.
The box will be 75 x 32 x 17,4 cm or 29,5 x 12,6 x 6,85 inch.
I dont think hornresp gives the right efficiancy but i do think it will be awsome considering the small size of driver and box. At least i hope so.
Johannes.
Since almost every tapped horn suffers from very strong frequency response ripple with their associated 180 degree narrow phase jumps (not only above 80Hz but across the whole passband) I'm wondering how are you going to cross over the tapped horns with a main system?
The experience that I have in this area is with scoop type rear loaded horns, which behave quite similar to tapped horns, and crossing them over to a midrange with proper on-axis summing is *extremely* difficult. Optimum summing is only achieved with the help of a DSP xover and 4 or more sharp parametric EQs to flatten phase. Otherwise, you may find the system summing at 60Hz, cancelling at 80Hz and summing again at 120Hz, which in practice sounds bad to terrible depending on room acoustics
(it sounds particularly bad in big reverberant spaces because if something does not sum on-axis, there are a lot of chances that it will sum off-axis).
The experience that I have in this area is with scoop type rear loaded horns, which behave quite similar to tapped horns, and crossing them over to a midrange with proper on-axis summing is *extremely* difficult. Optimum summing is only achieved with the help of a DSP xover and 4 or more sharp parametric EQs to flatten phase. Otherwise, you may find the system summing at 60Hz, cancelling at 80Hz and summing again at 120Hz, which in practice sounds bad to terrible depending on room acoustics
(it sounds particularly bad in big reverberant spaces because if something does not sum on-axis, there are a lot of chances that it will sum off-axis).No, not for War of the worlds at ref levels, but i will have to see what happens. I will probably blow a speaker or two, but its part of the fun.
This whole project has a 10 dollar budget. Its just some crap laying around that i play with. It will probably end up as a part of my übercheap PC-speaker system.
No! Not in their passband.
See what Tom Danley has to say about it in his White paper. http://www.danleysoundlabs.com/pdf/danley_tapped.pdf
Specially the part directly following "Example of a Tapped Horn" in the white papper.
Johannes.
According to Hornresp, every tapped horn has a narrow phase jump over 100 degrees in the *middle* of the passband, and two or three more jumps in the crossover region. Scoop type back loaded horns (and transmission lines) exhibit similar behaviour.
Is hornresp wrong? I don't think so because I have been experiencing this same behaviour in real systems. For example, optimum overlapping of scoop bass bins with front loaded subwoofers is really challenging.
BTW: Now I think that all this new tapped horn fashion is actually nothing but the good old 1950s JBL D55 "Scoop" bass bin reinvented. It was the most efficient/compact stuff back then (and still nowadays) but this comes at a price. Fortunately now we have DSP crossovers
Is hornresp wrong? I don't think so because I have been experiencing this same behaviour in real systems. For example, optimum overlapping of scoop bass bins with front loaded subwoofers is really challenging.
BTW: Now I think that all this new tapped horn fashion is actually nothing but the good old 1950s JBL D55 "Scoop" bass bin reinvented. It was the most efficient/compact stuff back then (and still nowadays) but this comes at a price. Fortunately now we have DSP crossovers
Hi Eva “in a secret location”
While “almost every tapped horn (show modeled here) suffers from very strong frequency response ripple with their associated 180 degree narrow phase jumps….)
In reality, most computer models over estimate the Q magnitude of horn resonances, in some cases, they are fully absent in the measured response and it is possible not to have any at all in the operating band.
Best is to build what you model, measure what you build and revise the model to fit what it actually does when measured and do this loop until done.
“The experience that I have in this area is with scoop type rear loaded horns, which behave quite similar to tapped horns,”
I would wonder, if you had any experience with Tapped horns however as they do not exhibit any of the issues you speak of.
In reality, what stops a scoop from being a tapped horn is that the direct radiating side of the driver is usually way too far from the horn passage to adequately couple to it or coherently and normally has the wrong parameters so it produces nulls and cancellations and exhibits a ragged response..
In the tapped horn, a key to making it work is having both sides of the driver in the horn passage so both sides add coherently and having both the driver and horn parameters correctly chosen..
As a result of the physical relationships only the sum can be radiated from the mouth, there is no radiation mechanism to produce beams or cancellation nodes. The “Tapped horn” idea was a result of a casual observation working on the Unity and Synergy horn which has multiple drivers which also add fully coherently in to one radiation source.
An explanation of that is here;
http://www.danleysoundlabs.com/pdf/danley_tapped.pdf
The basic idea is that since a horn can’t be fully efficient until its about a half wl long and has a sufficient mouth size and are usually a quarter wave at the low cutoff, the Tapped horn driver arrangement can accommodate the large change in acoustic load the small horn presents traversing from its quarter wave cutoff to where it can be efficient.
With the tapped horn, one has two driving points in the horn allowing a variable addition between the two sides. If all the proportions are right, one sees a large improvement in response shape and output compared to a conventional bass horn of similar size.
As an extreme example, compare the 1W 1M measured response to another horn of your choosing and see what something of similar size and response looks like when done conventionally.
http://www.danleysoundlabs.com/pdf/Danley TH Mini Spec Sheet.pdf
Keep in mind, compared to a Vented box too, the Tapped horn typically has about half or less the G.D and no Vented box port choking.
Take a look at the measured response and box volumes of big name pro sound subwoofers for more of perspective.
Alternately find a measured response for a conventional horn of a similar physical size that has the low cutoff, sensitivity, output and unusual “not tilted” response shape (not typical of bass horns) of this one.
http://www.danleysoundlabs.com/pdf/Danley TH-50 Spec Sheet r1.pdf
Best,
Tom Danley
While “almost every tapped horn (show modeled here) suffers from very strong frequency response ripple with their associated 180 degree narrow phase jumps….)
In reality, most computer models over estimate the Q magnitude of horn resonances, in some cases, they are fully absent in the measured response and it is possible not to have any at all in the operating band.
Best is to build what you model, measure what you build and revise the model to fit what it actually does when measured and do this loop until done.
“The experience that I have in this area is with scoop type rear loaded horns, which behave quite similar to tapped horns,”
I would wonder, if you had any experience with Tapped horns however as they do not exhibit any of the issues you speak of.
In reality, what stops a scoop from being a tapped horn is that the direct radiating side of the driver is usually way too far from the horn passage to adequately couple to it or coherently and normally has the wrong parameters so it produces nulls and cancellations and exhibits a ragged response..
In the tapped horn, a key to making it work is having both sides of the driver in the horn passage so both sides add coherently and having both the driver and horn parameters correctly chosen..
As a result of the physical relationships only the sum can be radiated from the mouth, there is no radiation mechanism to produce beams or cancellation nodes. The “Tapped horn” idea was a result of a casual observation working on the Unity and Synergy horn which has multiple drivers which also add fully coherently in to one radiation source.
An explanation of that is here;
http://www.danleysoundlabs.com/pdf/danley_tapped.pdf
The basic idea is that since a horn can’t be fully efficient until its about a half wl long and has a sufficient mouth size and are usually a quarter wave at the low cutoff, the Tapped horn driver arrangement can accommodate the large change in acoustic load the small horn presents traversing from its quarter wave cutoff to where it can be efficient.
With the tapped horn, one has two driving points in the horn allowing a variable addition between the two sides. If all the proportions are right, one sees a large improvement in response shape and output compared to a conventional bass horn of similar size.
As an extreme example, compare the 1W 1M measured response to another horn of your choosing and see what something of similar size and response looks like when done conventionally.
http://www.danleysoundlabs.com/pdf/Danley TH Mini Spec Sheet.pdf
Keep in mind, compared to a Vented box too, the Tapped horn typically has about half or less the G.D and no Vented box port choking.
Take a look at the measured response and box volumes of big name pro sound subwoofers for more of perspective.
Alternately find a measured response for a conventional horn of a similar physical size that has the low cutoff, sensitivity, output and unusual “not tilted” response shape (not typical of bass horns) of this one.
http://www.danleysoundlabs.com/pdf/Danley TH-50 Spec Sheet r1.pdf
Best,
Tom Danley
Circlomanen said:
Happy New Year and all that jive!
Choose the desired driver diameter and click 'go' to get a selection of drivers to review: http://www.thielesmall.com/database.asp
GM
Hi Tom
As I said, efficiency is very good and size is compact but this comes at a price. Your datasheets show this point quite well (although the response plots are obviously smoothed).
The TH-50 shows the 100 degree phase step in the crossover region, a 4th order LR crossover at 70Hz won't be sharp enough to overcome this and get optimum summing unless two bands of parametric EQ are employed.
The SH-25 shows another interesting phenomena related to placing the drivers away from horn throat. The wide 10dB 220Hz notch (big disaster for voice) is produced by reflection and cancellation isn't it? (isn't LF output 35cm away from throat?). This is like the other 10dB jump at 1200Hz (more disaster for voice). Also, conical horns of that size progressively "destroy" anything above 1Khz due to internal reflections and built-in comb filtering. They result in several paths of different lenght to the same listening angle. This is demonstrated by the 6dB beaming at +/-10 degree at 2Khz and 4Khz in your sample polar plots. Measurement at higher frequencies with more angular resolution would reveal even worse lobing. Clean "single-path" high frequency output requires a separate specific horn.
This weekend I had to use 8 parametric EQs to make two big 150Hz horizontally-arrayed midrange conical horns with 12" drivers sum reasonably with the high frequency horn. Crossover was LR24 1400Hz and couldn't be set lower. This is an example of how I've learnt to hate conical horns and horizontal arraying for midrange and trebble. Fortunately you use smaller drivers.
As I said, efficiency is very good and size is compact but this comes at a price. Your datasheets show this point quite well (although the response plots are obviously smoothed).
The TH-50 shows the 100 degree phase step in the crossover region, a 4th order LR crossover at 70Hz won't be sharp enough to overcome this and get optimum summing unless two bands of parametric EQ are employed.
The SH-25 shows another interesting phenomena related to placing the drivers away from horn throat. The wide 10dB 220Hz notch (big disaster for voice) is produced by reflection and cancellation isn't it? (isn't LF output 35cm away from throat?). This is like the other 10dB jump at 1200Hz (more disaster for voice). Also, conical horns of that size progressively "destroy" anything above 1Khz due to internal reflections and built-in comb filtering. They result in several paths of different lenght to the same listening angle. This is demonstrated by the 6dB beaming at +/-10 degree at 2Khz and 4Khz in your sample polar plots. Measurement at higher frequencies with more angular resolution would reveal even worse lobing. Clean "single-path" high frequency output requires a separate specific horn.
This weekend I had to use 8 parametric EQs to make two big 150Hz horizontally-arrayed midrange conical horns with 12" drivers sum reasonably with the high frequency horn. Crossover was LR24 1400Hz and couldn't be set lower. This is an example of how I've learnt to hate conical horns and horizontal arraying for midrange and trebble. Fortunately you use smaller drivers.
Hi
“As I said, efficiency is very good and size is compact but this comes at a price. Your datasheets show this point quite well (although the response plots are obviously smoothed).”
Smoothing? the Mini and TH-50 don’t, they are straight TEF measurements, what is the “price” do you refer to? To me it still appears to have nice response and a lot of output for its size no?
“The TH-50 shows the 100 degree phase step in the crossover region, a 4th order LR crossover at 70Hz won't be sharp enough to overcome this”
The phase step is associate with the bump in amplitude, fix one, the other is fixed too.
I personally use the TH-50’s and have been able to cross them into SH-50 with no phase step fwiw.
“The SH-25 shows another interesting phenomena related to placing the drivers away from horn throat. The wide 10dB 220Hz notch (big disaster for voice) is produced by reflection and cancellation isn't it? (isn't LF output 35cm away from throat?).”
These cabinets are never used alone and always are next to a mate or SH-50’s.
As the frequency falls there is mutual coupling between boxes so in order to have 2 or more boxes sum to flat response the low end must be made for multiple box use hence that shape. What you see is a mistuning effect, the first woofer reflection is above crossover at about 450Hz.
They like the SH-50 have according to the users, remarkable clarity, much better than the arrays of self interfering boxes and line sources they replace.
While a single horn is conceptually simpler, there is unfortunately no way to coherently cross over between two separate horns, without large lobes and nulls being produced.
Look at the same spherical measurements for any regular multiway speakers and see (most are presented at much less resolution than ours too fwiw).
“Also, conical horns of that size progressively "destroy" anything above 1Khz due to internal reflections and built-in comb filtering.”
While this can be true it isn’t always true, it depends how the horn is driven, a conical horn can also have the least coloration and distortion of all too...
Contrast your impression of conical horns chronic flaws to a reviewer and others impressions who actually heard / used the SH-25’s sister the SH-50
http://www.proaudioreview.com/pages/s.0037/t.8805.html
http://www.aes.org/sections/chicago/feb07review.html
http://db.audioasylum.com/cgi/m.mpl?forum=prophead&n=38622&highlight=sh-50&r=&session=
Don’t know what you’re used to but if you haven’t heard these, you haven’t heard what a horn system can sound like or see one that can reproduce a square wave broad band.
http://www.diyaudio.com/forums/showthread.php?threadid=71824&perpage=25&highlight=&pagenumber=1
Best,
Tom Danley
“As I said, efficiency is very good and size is compact but this comes at a price. Your datasheets show this point quite well (although the response plots are obviously smoothed).”
Smoothing? the Mini and TH-50 don’t, they are straight TEF measurements, what is the “price” do you refer to? To me it still appears to have nice response and a lot of output for its size no?
“The TH-50 shows the 100 degree phase step in the crossover region, a 4th order LR crossover at 70Hz won't be sharp enough to overcome this”
The phase step is associate with the bump in amplitude, fix one, the other is fixed too.
I personally use the TH-50’s and have been able to cross them into SH-50 with no phase step fwiw.
“The SH-25 shows another interesting phenomena related to placing the drivers away from horn throat. The wide 10dB 220Hz notch (big disaster for voice) is produced by reflection and cancellation isn't it? (isn't LF output 35cm away from throat?).”
These cabinets are never used alone and always are next to a mate or SH-50’s.
As the frequency falls there is mutual coupling between boxes so in order to have 2 or more boxes sum to flat response the low end must be made for multiple box use hence that shape. What you see is a mistuning effect, the first woofer reflection is above crossover at about 450Hz.
They like the SH-50 have according to the users, remarkable clarity, much better than the arrays of self interfering boxes and line sources they replace.
While a single horn is conceptually simpler, there is unfortunately no way to coherently cross over between two separate horns, without large lobes and nulls being produced.
Look at the same spherical measurements for any regular multiway speakers and see (most are presented at much less resolution than ours too fwiw).
“Also, conical horns of that size progressively "destroy" anything above 1Khz due to internal reflections and built-in comb filtering.”
While this can be true it isn’t always true, it depends how the horn is driven, a conical horn can also have the least coloration and distortion of all too...
Contrast your impression of conical horns chronic flaws to a reviewer and others impressions who actually heard / used the SH-25’s sister the SH-50
http://www.proaudioreview.com/pages/s.0037/t.8805.html
http://www.aes.org/sections/chicago/feb07review.html
http://db.audioasylum.com/cgi/m.mpl?forum=prophead&n=38622&highlight=sh-50&r=&session=
Don’t know what you’re used to but if you haven’t heard these, you haven’t heard what a horn system can sound like or see one that can reproduce a square wave broad band.
http://www.diyaudio.com/forums/showthread.php?threadid=71824&perpage=25&highlight=&pagenumber=1
Best,
Tom Danley
Well, interesting stuff folks.
I am very familiar with the problems that Eva refers to, and agree that they can be a pain. However, (probably at the weekend now), I will be measuring the box I built and it will be interesting to see how it relates to the HR simulations, then once the magnitude of the out of band ripples ( ) are confirmed I can look at the filter situation. I was looking at 48dB slopes to start with anyway, so to add one or two bands of notch if required shouldn't be an issue.
I am very familiar with the problems that Eva refers to, and agree that they can be a pain. However, (probably at the weekend now), I will be measuring the box I built and it will be interesting to see how it relates to the HR simulations, then once the magnitude of the out of band ripples (
) are confirmed I can look at the filter situation. I was looking at 48dB slopes to start with anyway, so to add one or two bands of notch if required shouldn't be an issue.
I mean that the downside of so high efficiency/size ratio is uneven frequency response, particularly notches and lobes.
It's true that your systems don't seem to have major notches between 1500Hz and 3000Hz. This really helps a lot with perceived clarity, while other systems do have plenty of notches in that region, and I have suffered them from time to time
Actually with two separate horns you can still get a reasonably wide sweet spot across the crossover region with the lobes happening only off-axis (and off-axis null lobes at 2Khz are a wonderful thing for clarity in reverberant environments), but this requires careful phase matching and a mid horn with smooth and coherent off-axis response (something hard to find).
It's true that your systems don't seem to have major notches between 1500Hz and 3000Hz. This really helps a lot with perceived clarity, while other systems do have plenty of notches in that region, and I have suffered them from time to time
Actually with two separate horns you can still get a reasonably wide sweet spot across the crossover region with the lobes happening only off-axis (and off-axis null lobes at 2Khz are a wonderful thing for clarity in reverberant environments), but this requires careful phase matching and a mid horn with smooth and coherent off-axis response (something hard to find).
The last picture of the inside before I glue the lid on it.
An externally hosted image should be here but it was not working when we last tested it.
Hi Eva, pinkmouse
While I have not used McBean’s program to design a Tapped horn, you normally wouldn’t find high Q resonances at low frequencies in a small horn as its higher internal losses set the actual Q’s. I wouldn’t be surprised if the measured hf response was not identical to the model.
Eva, you said “I mean that the downside of so high efficiency/size ratio is uneven frequency response, particularly notches and lobes.”
Your missing a key element of the design it would seem.
When you have multiple sources that are too small to have directivity and you put them less than ¼ wavelength apart, they add coherently, feel each others radiation pressure and radiate as one source. Do this in a horn and they act as one driver and one gets the radiation pattern the horn defines with no interference.
You’re right about smaller driver too, to follow what is needed here a 12 inch driver can only be used up to about 300-400Hz.
“Actually with two separate horns you can still get a reasonably wide sweet spot across the crossover region with the lobes happening only off-axis (and off-axis null lobes at 2Khz are a wonderful thing for clarity in reverberant environments)”
Perhaps in the home lobes and nulls might be ok but in commercial sound (where these speakers are used) the off axis lobes caused by incoherent addition are to be avoided if possible because they directly excite the reverberant field..
For voice intelligibility with all other things being equal, the greater the ratio of direct to reverberant sound the better.
For one conventional horn over another, one can make a very nice transition at crossover at listening height.
Best,
Tom Danley
While I have not used McBean’s program to design a Tapped horn, you normally wouldn’t find high Q resonances at low frequencies in a small horn as its higher internal losses set the actual Q’s. I wouldn’t be surprised if the measured hf response was not identical to the model.
Eva, you said “I mean that the downside of so high efficiency/size ratio is uneven frequency response, particularly notches and lobes.”
Your missing a key element of the design it would seem.
When you have multiple sources that are too small to have directivity and you put them less than ¼ wavelength apart, they add coherently, feel each others radiation pressure and radiate as one source. Do this in a horn and they act as one driver and one gets the radiation pattern the horn defines with no interference.
You’re right about smaller driver too, to follow what is needed here a 12 inch driver can only be used up to about 300-400Hz.
“Actually with two separate horns you can still get a reasonably wide sweet spot across the crossover region with the lobes happening only off-axis (and off-axis null lobes at 2Khz are a wonderful thing for clarity in reverberant environments)”
Perhaps in the home lobes and nulls might be ok but in commercial sound (where these speakers are used) the off axis lobes caused by incoherent addition are to be avoided if possible because they directly excite the reverberant field..
For voice intelligibility with all other things being equal, the greater the ratio of direct to reverberant sound the better.
For one conventional horn over another, one can make a very nice transition at crossover at listening height.
Best,
Tom Danley