Simple Tapped Horn Tutorial using Hornresp - AVS Forum
Forum Jump: 
 3Likes
Reply
 
Thread Tools
post #1 of 114 Old 01-02-2010, 01:30 PM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
Tapped Horn Design

We're up to v34 now, and there is a new download link.

Get it here:

http://www.hornresp.net/

After reading maxmercy's excellent post on front loaded horn design, I thought I'd share what I know with respect to tapped horns.
As I've said all along, I am an amateur, I do this as a hobby. I am not selling anything, just sharing what I have figured out.
I apologize in advance for my lack of sketchup skills. I hope that I can get the idea across.

What is a tapped Horn? I don't want to go into that here, as others far smarter than I have explained that already.

http://www.google.com/search?rlz=1C1...&q=tapped+horn

Google is your friend, I recommend links 1 and 3 first. Danley's white paper is excellent too, but may go over your head at first.

Do they work?
Absolutely.

Why bother with a tapped horn?

Cabinet size. While a tapped horn may not equal a front-loaded horn in output, it can produce considerable output from a cabinet that is ~1/4 the size of a 1/4 wavelength horn.

Are they hard to build?

No, not at all. In the simplest form, it is a box with a driver near one end of a sloped baffle inside, 7 boards.



Some Caveats:
Not all drivers will work in a tapped horn
Improper designs will kill a driver quickly
Overexcursion explodes out of the pass-band (use high-pass filters)

As with any horn, boundary loading is your friend, but unlike a front-loaded horn, you don't need it to flatten the response.

Number one rule - there is no free lunch.

There is no magic here, just physics and math.

Here is an annotated diagram so you know what I am talking about. S = area, L = length. Hornresp works in metric units, like the rest of the world...



One of the primary concerns in designing a tapped horn is if a given driver will work with a reasonable compression ratio, which is the area of the driver (Sd) divided by the area of the horn at the driver entry (S2).

Small drivers can work with a higher ratio, as can drivers with a more rigid cone. Personal experience suggests that 4:1 or less is fine for small, high-excursion 6 to 8 inch diameter drivers.

Larger drivers require lower ratios, for example jbell's PA tapped horns use Eminence 3015lf 15" drivers, and have a compression ratio of 1.26:1. These are an extreme example, SPL @ 40 Hz was the primary concern.



Yes, that is a 15. These are HUGE. More details here:

http://www.diyaudio.com/forums/subwo...ped-horns.html

So - how do we design one?

Select a driver. Lots will work, I'm choosing one that I have, and one that is cheap enough for anyone to experiment with this should you choose to play along at home. For this example, I'm using the MCM 55-2421 High-excursion 8" woofer. While perhaps it is not an ideal driver, it is cheap and I already have some (that makes it even better in my book).

http://www.mcmelectronics.com/product/55-2421

These drivers are often on sale, but you have to enter the correct code to get the sale price. I paid ~$25 each shipped for the last ones I bought.

Here is a link to how to enter your own driver parameters.

This cabinet will use two drivers wired in series. I've measured several of these drivers in the past, the thiele-small parameters I have are entered into hornresp.

Select a target low frequency response cutoff. Realistically, a low corner of 66% of the drivers Fs in free-air can be reached without losing too much efficiency or sending excursion through the roof. Higher frequencies = smaller boxes and higher efficiencies. It is all about the compromises.

In our case, since I have realistic expectations for these 8 inch drivers, I'll choose 30 Hz.

Next, I determine the minimum throat dimension.

Sd * 2 (number of drivers) / 4 (max safe compression ratio)

In this case, S2 should be approximately 105, so S1 can be smaller than 105. I'll start with 100.

Then I determine the mouth area. I typically use total Sd * 2 as a starting point, some drivers want more or less.

In this case, S4 should be around 836, I'll use 800.

Now, we have to determine the lengths of the various parts of the tapped horn. We know that the drivers have to fit inside this thing, so L12 and L34 have to be at least 21 cm (one driver diameter).

Experience has taught me that an additional few centimeters is a good thing, it makes driver installation and construction much easier.
I'd suggest 32 cm as a good starting point.

What about the L23 value? Lets use 250 cm for now.

OK, we have what we need to get started. I've attached a hornresp import file to this post so we all start at the same place. 
<a class="attachment loginreq" href="/attachments/13266" title="">example.txt 0.3798828125k . file
Mfusick and diaz like this.

Last edited by lilmike; 06-25-2014 at 08:41 PM. Reason: Updated link
lilmike is offline  
Sponsored Links
Advertisement
 
post #2 of 114 Old 01-02-2010, 02:02 PM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
First - import the starting point file I have attached to the first post. Download the text file, place it in your hornresp import directory (usually c:\\hornresp\\import), then import it into hornresp.

This will provide the thiele small parameters of a single MCM 55-2421 8" woofer and the default horn record.

Here is the annotated diagram again (no sketchup yet - sorry...) of the relationships between the areas and the lengths.



To make a tapped horn model, you need three horn segments and no rear chamber. Remember the numbers above?



Double click S1, then enter S1 = 100, S2 = 100, L12 = 32, press Calculate, then press Save.

Notice that the next segment is has exp rather than con, so double click on exp to change to con.

Double click on S2, then enter S2 = 100, S3 = 800, L23 = 250, press Calculate, then press Save.

Double click on S3, select conical, then enter S3 = 800, S4 = 800, L34 = 32, press Calculate, then press Save.

OK, so we have our three horn segments and our driver parameters. Things should look like this:



We want to use two drivers though.

Select the Tools menu item, then select driver arrangement. We want to use two drivers in series, and we want them in a tapped horn.



A tapped horn can not have a rear chamber, enter zeroes in Vrc and Lrc. We are not using a throat chamber either, so enter zeroes in Vtc and Atc as well.

OK, we're done. Press calculate.



Not bad, but the response is a little jagged, and the cutoff is too high.

Go back to your Input Parameters (Window - Input Parameters).
Now - the cool part. Select the tools menu, then select the tapped horn wizard. (Loudspeaker Wizard in V. 26)



We want to see the frequency response, so click on the arrow next to "Schematic" and select "Response". We also want S2 to vary, so select "Horn - S2 Variable" from the second menu box.

Move the sliders and watch the magic.

(No - McBean will not be implementing wizards or sliders in any of the other horn calculators any time soon - please don't ask him) - Scratch that!

As of V. 26, Hornesp has implemented sliders for many of the other horn calculations under the Loudspeaker Wizard, so please thank Mr. McBean for this amazing piece of software!

We want a sub that will play an honest 30 Hz. Drag the L23 slider to the right.
I stopped at 350 cm.



Hmm, that response is a little too lumpy. What can we do with that? Similar to a front-loaded horn, ripples in the response are affected by the area at the throat of the horn. Let's reduce S1 and see what we get.



Setting S1 smaller certainly reduces the ripples, but it also moves the corner frequency up a bit. I set it at 53.

Press Save, that takes you back to your input parameters.

What about compression ratio? Hover your mouse pointer over S2, then look in the bottom bar. 5.27:1 is too high. Time to compromise.

Get back into the Tapped Horn wizard and set S2 a little bit bigger. Note - you will have to set the view and the S2 - Variable options each time. I increased S1 to 80. Sure there is a little bit more ripple, but it is not that bad.

Press calculate. The simulation predicts an SPL response that is pretty respectable. 105 dB? Wow - oh - wait - Ang is still set to 0.5 X pi (corner-loaded).



Let's look at 2pi and compare the responses (press F4). While the peaks and valleys get a little more pronounced, the low-end response does not droop.



Looking a little more closely - we're predicting 94 dB +/- 3 dB from 30 Hz to 100 Hz. Sure, the response gets ugly above that, use the crossover to eliminate it.

Seems to be in line with our goals that we set earlier.

How about excursion? Displacement looks fine at 1 watt. What happens when we feed it some power? These drivers are rated at 120 W RMS, and have an Xmax of 8 mm one way.

Double click Eg, enter 240 W and 8 ohms. Over 115 dB, 2pi, from 2 8 inch drivers and 240 watts (or over 125 dB when placed in a corner).





Excursion is OK but only ABOVE 25 HZ!!! Subsonic filters are not optional. I can't emphasize this enough - Do not drive these enclosures out of their pass-band unless you like replacing drivers!

Oh - yeah - how big is this thing? Click Schematic Diagram - the net internal volume is 156 Liters.



So - can 2 $25 dollar 8-inch drivers and a 6 cubic foot box really get you an honest 30 Hz at 115 dB?

No. But it will get really close. Hornresp models a truly non-resonant enclosure. In reality, there will be some losses due to panel vibrations. Actual response will typically be a dB or two lower, but it will also be flatter than predicted as a result of the panel resonances.

Remember jbell's big 3015 horn? Hornresp predicted 123 dB, he measured 122 with the prototype (unbraced). My quad Tang Band W6-1139 tapped horn was predicted to hit 95 dB @ 2.83 V, I measured 92 dB under sub-optimal measurement conditions. I also added some poly batting that significantly smoothed the response. Most importantly - the shape of the curve was as hornresp predicted, but smoother. The magnitude was certainly within the error bounds of my measurement equipment.

Do they really work? Yeah, they really do. The only failures I have experienced so far were implementation errors on my part.

If you measure the driver you model, model the driver you measured, and build the model you simulated, your results should measure very close to what you modeled.

If not - do not blame the model.

I have learned to trust Hornresp, my Woofer Tester 2, and my SPL meter more than my cabinetry skills or my ears. No magic here - It is all math and physics, and those rules are pretty solid.

So - how the heck do I fold one up and build one? That's another post. (Coming soon - I promise.)
maxcooper and diaz like this.
lilmike is offline  
post #3 of 114 Old 01-02-2010, 02:14 PM
AVS Special Member
 
brandonnash's Avatar
 
Join Date: Jul 2003
Location: murfreesboro, tn
Posts: 3,942
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 24
Eagerly awaiting your next post on the building of one. Very good write up. Between the folded horns and the tapped horns I'm beginning to want to build some. I'm beginning to believe that a 30-40 hz horn on each channel of my system would be pretty sick with the DTS-10 filling in the low end.
brandonnash is offline  
post #4 of 114 Old 01-02-2010, 02:17 PM
AVS Special Member
 
soho54's Avatar
 
Join Date: Sep 2005
Posts: 3,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
Just cause I already had it ready to go.
diaz likes this.
soho54 is offline  
post #5 of 114 Old 01-02-2010, 02:33 PM
AVS Special Member
 
Oklahoma Wolf's Avatar
 
Join Date: Nov 2006
Location: Eastend, SK, Canada
Posts: 1,745
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 26
Nice tutorial - I'm still fiddling with my design. Got the size down while still keeping all that nice performance I was after. Maybe now I can drop the corner down a bit.

Found an alternate driver that will work... will be looking for more of them yet before I post my thread.

Edit - found another one - Tang Band's 8x12" W8Q-1071F. Not sure it fits (need two of them), but by golly does it ever bring that horn to life. 128dB or so at 20Hz, 200W in. Compression ratio may be too high. Too bad it costs so much. Designed a 15Hz tapped horn with it, and it still brought 120+ dB to the table. Big horn, but wow was it ever impressive. Somebody needs to build a TH with that bad boy.
Oklahoma Wolf is offline  
post #6 of 114 Old 01-02-2010, 03:08 PM
Senior Member
 
Antripodean's Avatar
 
Join Date: Apr 2009
Location: The Antripodes (OZ)
Posts: 328
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by lilmike View Post

Tapped Horn Design

Some Caveats:
Not all drivers will work in a tapped horn
One of the primary concerns in designing a tapped horn is if a given driver will work with a reasonable compression ratio, which is the area of the driver (Sd) divided by the area of the horn at the driver entry (S2).

Great work lilmike!

I was wondering if there is some key equation or critical (range) T/S parameters to determine what driver is best for a tapped horn?

explore the music
Antripodean is offline  
post #7 of 114 Old 01-02-2010, 03:39 PM
AVS Special Member
 
dbldare's Avatar
 
Join Date: Jan 2001
Location: Broomfield, CO USA
Posts: 2,518
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 15
lilmike,

Thanks for the great write-up! Can't wait for more!


dbl

Let's all go to the lobby
....Let's all go to the lobby
........Let's all go to the lobby
............To get ourselves a treat!
dbldare is offline  
post #8 of 114 Old 01-02-2010, 05:19 PM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
Quote:
Originally Posted by Antripodean View Post

Great work lilmike!

I was wondering if there is some key equation or critical (range) T/S parameters to determine what driver is best for a tapped horn?

If there is, I have not found it yet.
lilmike is offline  
post #9 of 114 Old 01-02-2010, 07:42 PM
AVS Special Member
 
soho54's Avatar
 
Join Date: Sep 2005
Posts: 3,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
Quotes from Tom Danley from the interweb:
-
"I can maybe give you guys another clue or two which will help understanding how it works. As one approaches the low corner on a conventional bass horn, the horn acts more and more like a slug of air, it has mass which appears to increase as the frequency falls. In a normal horn, that reactance is adjusted by changing the “t” and ideally exactly off set by the compliance of the driver in a sealed box. With the Tapped horn, one has no rear volume so to get the lowest low corner, one needs the driver’s Fs to be somewhat above the low cutoff. For example, it wouldn’t be unusual to have an Fs of 45 or 50Hz with a low corner of 30Hz.
The other thing which you will find is that for a fixed low corner and driver, increasing the length of the portion past the front driver Tap moves the “wiggles” higher in frequency AND makes the driver appear to be a heavier, stronger driver."

-
"The whole point of the Tap is to provide additional acoustic loading in the region where a normal “too small” bass horn has low corner consisting of two peaks and a deep dip.
In the region where one had a deep saddle, one now has both sides of the radiation adding in phase. At the low corner (quarter wave resonance), only the “end” position feels any significant loading so it is sort of like having T&S parameters (SD?) which change with F, so far as the horn is concerned."
-
"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 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."
-
"A few thoughts, one can get at least 9+ dB of acoustic gain (so far) over a direct radiator version using the same drive depending on the Tapped horn and driver.
That 9dB corresponds to an added acoustic load which cuts excursion to about 1 / 3 as well. "
-
"So far as excursion, this is a factor here too, the peaks are usually not as high and the dips not as deep as what is generally predicted, if your going to focus on getting the max out, you must also adjust your models to fit measured reality as closely as possible."
-
"The quarter wave resonator is similar but it is the > half wave horn that is “efficient”.

To make a quarter wave horn efficient, requires a different set of driver parameters, one with a much stronger motor and greater mass and not so suitable above ¼ wl."
-
"The quarter wave resonance is what defines the low corner of the response for a TH.
Essentially, one finds that wherever the knee in the response is, is where the quarter wave resonance is.
As with a conventional horn, having a compliance volume at the throat combines with the throat mass and forms a low pass corner. To a degree, this volume also shifts the lowest resonant frequency downward slightly."
-
"The Tapped horn is a way to have the driver properties effectively change, at the low cutoff, only the end radiator face “feels” the load from the quarter wave stub.
AS the frequency climbs, the other face begins to add constructively.
Without the front tap, these horns still behave like horns, it has a series of resonances, not like a BP speaker, and there isn’t a low pass filter as is normally the case with a BP system.
Adding the front tap doesn’t add a low pass corner either but when everything is right, it fills in the saddle.

The reason the Fs is best above the low cutoff is that like a conventional horn, one finds that the air in the horn acts like a capacitance who’s value increase as the frequency falls approaching the cutoff. In a conventional horn, one can choose the right driver, use an appropriate Hyperbolic expansion to tailor the shape of that curve, combined with the proper rear volume and that results in “reactance annulling”, which extends the useful low corner downward.. Here, the compliance volume conjugates the horn’s mass by producing a broadly resonant condition.
In the Tapped horn, there is no adjustable rear volume to play with but that required spring force can be had in the drivers suspension.
"
-
"I would say that while horns or “tapered pipes” have some similar properties to a Helmholtz resonator , there are also differences. For example, a woofer driving a Helmholtz resonator “on tune” acts as an inverter, while a quarter wave stub produces a motional transformation but exhibits half the phase shift (90 degrees).
Also, where a Helmholtz resonator has one primary resonance, a “tapered pipe” has a number of them and alternate between pressure max and velocity max, in a horn that’s “large enough” radiation resistance makes them much less visible or even invisible."
-
"The Tapped horn works because the acoustic impedances all (when everything is right) all compensate each other, the dip between the first and second peak on a normal “too small horn” is gone.
Those peaks in a conventional horn reflect the changing load presenter to the cone, they correspond to peaks and dips in the electrical impedance and that governs how much power is delivered vs frequency.
Where the dip would normally be, now both side of the cone radiate fully additively within the horn and that load is present in the impedance as an increase in delivered power."

-
diaz likes this.
soho54 is offline  
post #10 of 114 Old 01-02-2010, 07:44 PM
AVS Special Member
 
soho54's Avatar
 
Join Date: Sep 2005
Posts: 3,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
Continued;
"When you make a bass horn “too small’ and by that I mean a mouth that is significantly less than a wl in circumference in full space and by making it a quarter wavelength long instead of ½ wl., you have a number of tradeoffs.

By making the horn shorter, only a quarter wave long at the low cutoff, you prevent the system from having high electroscoustic efficiency conversely you’ll find that the short bass horn does reach efficiency until it’s about an octave above low cutoff, again where it is a half wl or greater. The classical horn, needs to have the driver at the motional maxima, which begins at a half wavelength.

Also, one finds the response of such a horn typically droops off instead of being “flat” down to the low corner. When the mouth is made smaller in area, then one finds the horn’s modes become increasingly revealed by the reduced radiation resistance (which fully damps them in a larger horn).

When you make the horn mouth a convenient size (too small), then you have a response curve with a large peak defining the low corner, with a deep valley then another peak with a series of smaller and closer together peaks after that.

It is the magnitude of these peaks and dips which define how small you can make a horn.

Acoustically, what you see is where there are peaks, there is the maximum acoustic loading on the driver, the point where the motion and impedance is minimum.

Where the deep valley is, there is little load on the cone, motion and impedance are maximum.


The quarter wavelength bass horn can’t be efficient down low because the driver’s parameters that are ideal for operation where it is ½ wl long or longer, are totally different than those needed for the quarter wave mode which is a motion minimum instead of motion maximum (1/2 wl)

The Tapped horn idea began when I was thinking about the quarter wave length reflection that places a notch in the response of the side mounted drivers (as used in the synergy horns), that notch defines where the usable upper response corner is.

I wondered, “what happens if I substitute a source of the opposite phase for that reflection” Having that source already available in the drivers rear radiation, I made a computer model.

It took a long time before I had anything I felt like building, but occasionally I would see predictions of a nice sensitivity.

The Tapped horn idea is that you use a driver which can efficiently drive the quarter wavelength mode which defines the low corner. This driver is too massive for efficient horn loading normally.

At the low corner, the front driver is approaching 90 degrees out of phase from the radiation at the throat. The acoustic load is nearly entirely on the throat side of the driver at the low cutoff. As the frequency climbs into the range where one has the deep valley, now the shorter wavelength means the front and rear radiation are separated by 180 degrees of path length and now are additive in the horn.

“When everything is right” (the horn dimensions and driver parameters), then where the valley was, is now flat, filled in by the added radiation load on the cone.

Compared to a vented box, these typically have about half the group delay for a given low corner and they often have a roll off slope between a vented and sealed box.

In a conventional horn, one normally sizes the rear volume to put the Fb somewhat above the system low corner. The drivers suspension spring is in parallel with the box compliance or spring. By choosing the horns hyperbolic or shape, one can adjust the acoustic mass the horn presents to the driver as you approach the low corner. The proper relationship between the two results in “reactance annulling” which lowers the low corner to its lowest value.

In the Tapped horn, the acoustic mass in the horn also exists but with no rear volume, the drivers suspension spring does the entire job. As a result the ideal Fs for a Tapped horn is usually the better part of an octave above the low corner.

Anyway, like a normal horn, once you have “all” the parameters right, they can work very well and provide flat smooth response, like a much much larger horn..
"

Last one...
"==============

"Hi John, all
I would certainly second Marshal's horn paper, I have made dozens and dozens
of horns based on his math (for the driver / horn
relationship) over the last 12 years or so and his alignments
(relationships) work well, even into the HF range.
Bottom line I suppose is that the "important" part is that when you build a
real horn that it measures like the computer prediction
and since most horns I have built were not "full ideal size", having
Marshal's math as a starting point, followed by modeling of the
real thing has proven to be most powerful.
For example, when the horn mouth is smaller than ideal, it may turn out that
some other profile (other than the t = X as defined
by the computer) may provide better results. Similarly, as in the LAB
project, for cases when fewer than "ideal" numbers are
used, a somewhat different driver gave better results. With that project I
started with the "ideal" driver and then on the real horn
model used that driver as a starting point, looking at a set of 2, 4 and 6
boxes and diddling the parameters to get the best
results.

As for curves vs reflectors, it would depend entirely on the acoustic
dimensions.
As a rule of thumb, when an obstacle is under about 1/4 wl in size, sound
goes around it, no problem, the object is acoustically
invisible..
If an object is say 10 wl across, it begins to make a pretty good reflector.
In a folded horn, the bends have a direct bearing on the results higher up.
When the difference in the acoustic path length between the inside and
outside radius of the bend reaches 1/2 wl, there will be a
deep cancellation notch in the response.
This is why most all the folded bass horns I design have minimum angle bends
and those bends tend to be wide rather than deep
(to push the cancellation F's as high as possible).
For a real woofer horn though, this corner issue is a non issue.

Once we had a customer who distributed our servodrive horns in Italy.
He wanted to make his own boxes to save shipping and we said ok send us one
to look at / measure first.
He was very meticulous, had all the corners glassed in, all of the inside
was smooth and aerodynamic like an airplane part.
It really did look cool and obviously he spent a lot of time making it
perfect.
Our cabinet shop had made up a sample horn with no curves (using flat
reflectors) and another with nothing at all in any corners
except the final bend.
I measured each box with the same driver module, in the same location, with
the same mic and TEF machine.
I measured at rated power and 1/10 to see if there were differences.

The TEF machine is repeatable to a tiny fraction of a dB so the most
informative way to look at the plots was to overlay them.
The highest SPL's were from the prototype without corners, second was the
normal production model and last (about 1.5 dB
down) was the aerodynamic one.
Distortion was essentially identical for all units.
The results were sort of a surprise however on examination, it was clear
that even thought the BT-7 could produce a large
acoustic power (about 200 acoustic Watts each, in a group of 4), the air
motion in the throat and passageways was not a
sufficiently high speed to benefit from the aerodynamic treatment.
Also, that filling in the corners reduces the actual volume of air in the
horn making it "smaller" than the unfilled unit.
Granted, this box is never used above 120 HZ and usually under 80 Hz so hf
performance is not normally a concern.
One can improve some aspects of real hf performance in a folded horn of
large dimensions by using a reflector style corner.
Reflectors are common in microwave frequencies in antennas because wave
theory also applies.
With sound in a folded horn, one finds that no matter how it is done, there
is a region between where there is the 1/2 wl path
length cancellation and where the reflector really works (and remember the
acoustic path for a reflector and mass flow are also
different) that is undesirable (chaotic) so where you see reflectors used in
antenna's, they are limited to the range where they
ARE reflectors, just like with sound, the in between range is undesirable.

If one was designing a folded horn for use higher up in frequency, a second
thing to be aware of is the mode translation which
means that when the parallel walls are acoustically 1, 3, 5 etc. 1/2
wavelengths and large enough in area, there is a deep notch in
the response. This is because there is a standing wave between the two walls
which the desired acoustic output drives.
Driving the resonance also taps out the desired acoustic signal's power and
so at the mouth is a notch.
This is the source of the notch on the lab sub who's internal width is 21
inches, 1/2 wl @ 21 inches ~ 1/2 wl @ 323 Hz.


As also mentioned, the front volume (trapped between the radiator and
throat) can be sized to extend the hf response.
This uses the air volume compliance acting against the air mass in the
throat to make a low pass filter like the vent is on a vented
box.
The difference here is that the resonance defines the upper cutoff on the
horn and lower on a vented box.
This "filter" if it can be sized correctly, has an additional advantage.
The Unity, LAB sub and BT-7 all have the front volumes sized to produce the
hf extension BUT after the high cutoff the roll off
is 1 order steeper than before.
This means that the harmonic distortion that is inevitable with any driver
is attenuated before being radiated (as those harmonics
are above the cutoff).
In the Unity, the driver placement on the horn walls also adds an additional
low pass filter which further attenuates out of band
signals from being radiated. The horn loading combined with this effect are
in the Unity's which typically measure about 1/10 or
less the midband THD of anything I have measured.
On the oother hand, often it is not practical to use this trick, like hf
compression drivers where this air volume is smaller than can
be achieved as it is.

Some one also asked about compression ratio and what effect than has, I
would suggest they dig through the design part of the
LAB sub as I had written about that (from my perspective) there.
In short, compression ratio is a "transformer" and one is adjusting the
turns ratio.
Why that effects BW and efficiency is in the LAB stuff and too much to write
now.
Well, I have speakers to work on, got to run"
diaz likes this.
soho54 is offline  
post #11 of 114 Old 01-03-2010, 10:58 AM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
@soho54

Thanks, it is great to have all of Danley's tidbits added in one place.

Tutorial Update:

I have the folding tutorial written (not typed yet, I said written - like pencil on paper - you know - old school) and I am working on the diagrams (which are also pencil on paper now). As I have prepared it, it will work with any single-taper conical tapped horn where L12 and L34 are equal. With minimal tweaking, I can also make it work when L34>L12, I'll see if I can add that in.

L34

I definitely suck at sketchup, but I am working on figuring it out.

I also have plans for the example sub nearly done.

Got a bunch of other important things to do today, so it might be a couple of days before I get the rest posted.

Sunday = brew day! Time to make the beer.

Primary issue with the tutorial progress right now is a cranky computer - I have to get that addressed before I can really complete the folding tutorial, because I use a spreadsheet to do the math for me. This is really not a big deal, but yet another delay.

It's coming.
lilmike is offline  
post #12 of 114 Old 01-03-2010, 11:37 AM
AVS Special Member
 
soho54's Avatar
 
Join Date: Sep 2005
Posts: 3,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
EDIT: Read "total system usable bandwidth" instead of "passband" below.

I thought a little about phase shift should be added here. In a horn you want to get things down to a 90deg phase shift.

What this means is the from the top of your passband to the bottom there is no more than a 90deg difference. This can be -90 to 0, or -45 to 45. In a tapped horn it is never going to be as pretty as a regular horn, but you can get it close.

This is the phase response of a dual driver TH I'm working on, it has two 21" drivers and an 20Hz low corner, 180Hz high corner:


This is the resulting impulse response:
soho54 is offline  
post #13 of 114 Old 01-03-2010, 11:49 AM
AVS Special Member
 
soho54's Avatar
 
Join Date: Sep 2005
Posts: 3,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
Here are another set using a single 15" driver. The lowcorner is around 20Hz, upper limit ~95Hz
As you can see the phase dip is gone in the IR, but the cone damping is not as good. I still need to work on that some.


soho54 is offline  
post #14 of 114 Old 01-03-2010, 11:57 AM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
Thanks. I appreciate the additional information.

(And I learned something too!)
lilmike is offline  
post #15 of 114 Old 01-03-2010, 11:58 AM
AVS Special Member
 
Oklahoma Wolf's Avatar
 
Join Date: Nov 2006
Location: Eastend, SK, Canada
Posts: 1,745
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 26
Quote:
Originally Posted by lilmike View Post

Thanks. I appreciate the additional information.

+1 - that helped a fair bit with mine
Oklahoma Wolf is offline  
post #16 of 114 Old 01-03-2010, 11:59 AM
AVS Special Member
 
soho54's Avatar
 
Join Date: Sep 2005
Posts: 3,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
Another note:
The Auto Phase in the program sometimes picks the wrong delay settings. Take your raw horn path length and convert it from cm to feet. Then check the delay time. There should be a negative second for every foot.

If it is way off you can manually set the delay of the phase graph in the tool tab.
soho54 is offline  
post #17 of 114 Old 01-03-2010, 03:23 PM
AVS Special Member
 
maxmercy's Avatar
 
Join Date: Nov 2008
Posts: 2,336
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 22 Post(s)
Liked: 131
Mike, Soho,

Thanks a bunch for this great info. I am working on a How to Fold a horn in Sketchup tutorial....it should be ready in a few weeks, it will be a series of videos...

This thread has been linked from the OD horn design thread...

OK folks, get to designing!!

JSS
maxmercy is offline  
post #18 of 114 Old 01-03-2010, 07:35 PM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
So, the modeled response is all lumpy. Reality might be a completely different picture. Remember, hornresp models a completely non-resonant enclosure with no losses. I know I can't build one like that, lucky for me, some resonance and losses help things out immensely

This is a quad tang-band W6-1139 tapped horn, designed for 30 Hz.

As modeled:



As built 2.83V 2pi measurement (with a 30 Hz. highpass in place, as well as some batting in throat):

lilmike is offline  
post #19 of 114 Old 01-03-2010, 11:43 PM
AVS Special Member
 
soho54's Avatar
 
Join Date: Sep 2005
Posts: 3,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
Now it we could just get a DIY rotary sub going for <20Hz output we would be all set.
soho54 is offline  
post #20 of 114 Old 01-05-2010, 10:38 AM
Newbie
 
sixgen's Avatar
 
Join Date: Sep 2009
Posts: 1
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
thank you so much lilmike for the nice guide.
So far this is the best guide i have seen, well because its the easiest to understand.
The other guides are just too hard to understand especially for a non technical person like me.
I know this guide will help alot of people in understanding hornresp and TH a lot better.
I cant wait to see the rest of the tutorial
Can you also make a guide like this for rear/back loaded horn??
sixgen is offline  
post #21 of 114 Old 01-05-2010, 01:40 PM
AVS Special Member
 
soho54's Avatar
 
Join Date: Sep 2005
Posts: 3,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
Quote:
Originally Posted by sixgen View Post

Can you also make a guide like this for rear/back loaded horn??

Nothing to them. Use the Nd horn setting, and zero out all chambers fields out. Use the Vtc for the rear chamber, and Atc as at least equal to the driver diameter. Now when you calculate the SPL go to the Combined Response setting and input the path difference between the horn mouth to your LP and the driver to your LP.

Don't forget to count the wall bounce distance if the driver and mouth are on opposite sides.
soho54 is offline  
post #22 of 114 Old 01-06-2010, 09:15 PM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
Back to the learnin'!

Now that we have a simulation of our little sub, let's take a few minutes and make it better. It is easier to change things now than to build enclosures over and over. A few days ago, soho54 pointed out that we had not talked about phase, then provided some great information.

Based on that information, it looks like our model could be improved. How? Well, the frequency response looks pretty good, and meets our goals. The size fits our needs too. The phase response? Well, let's take a look:



Well, that is not that bad - we're within 90 degrees from 30 to ~100 Hz.

How about that impulse response?



Well, I've seen a lot worse, but there is room for improvement here. We'd really rather see a clearly defined initial sinusoidal pulse reaching 1.0 on the positive and nearly 1.0 on the negative. Well, our amplitude is OK, but the waveform is a little ugly. Additionally, the horn rings a bit, the resonances are a little higher in amplitude than we'd like to see, and they do not damp out quickly enough.

What to do?

Well, it turns out that the length and geometry of the throat are the primary influences on the impulse waveform. In our case, the throat may be too long, causing destructive interference. Let's see what we can do to fix that.

Cutting L12 in half, adding the extra 16 cm to L23, and maintaining the areas, (I typed the entries in, I did not use the wizard) nets us this:

SPL


Phase


Impulse


Well, that is definitely an improvement, and it does not look like we broke anything while fixing the rest. Let's see if we can take this a little further.

Dropping L12 to 14 and increasing L23 to 368 yields this:

(No discernible change in SPL)

Phase


Impulse


So, an 18 cm change in the location of one side of the driver relative to the throat end makes a significant difference in the impulse response, but has little effect on the SPL (none in band), excursion (none), or phase plots (limited shift).

But wait - how are we going to fit two drivers into an L12 that is only 14 cm long? Side by side. There is plenty of room, the radius of our driver is approximately 10.5 cm. L12 is the distance from the throat to the center of the driver cutout.

Now L12 and L34 are different lengths - what will that do to the folding calculations? So long as L34 is longer, it is an easy fix, simply block off the excess horn path. Sure, there could be some wasted space in the box, we could also just make the mouth that much larger.

While this enclosure is not damped as well as I'd prefer, I have seen (and built) worse, and they actually sounded (and measured) pretty good. At this point, I believe that this simulation is sawdust-worthy, so the next step will be folding it and drawing up a set of plans.

If anyone can offer any criticisms or comments thus far, please chime in. I'd rather hear them now than change things later. Remember, I am an amateur like most of the rest of us here, and I am learning too. Please let me know if I have omitted anything or overlooked something.
diaz likes this.
lilmike is offline  
post #23 of 114 Old 01-06-2010, 09:23 PM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
Quote:
Originally Posted by sixgen View Post

thank you so much lilmike for the nice guide.
So far this is the best guide i have seen, well because its the easiest to understand.
The other guides are just too hard to understand especially for a non technical person like me.
I know this guide will help alot of people in understanding hornresp and TH a lot better.
I cant wait to see the rest of the tutorial
Can you also make a guide like this for rear/back loaded horn??

Patience - let me finish this one up....

So long as no one finds any fatal flaws or room for improvement in this one, I think folding and plans come next.

Seriously - I've never made a BLH, I'm sure I could model one, but would rather stick to the stuff I know. soho54 posted the basics - they are not that tough to model. Hornresp does all the hard work.
lilmike is offline  
post #24 of 114 Old 01-06-2010, 09:27 PM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
Would have helped if I had remembered to post the model.

Here it is as a hornresp export.

Oh - hornresp was updated again (bugfixes), so go update.

 

example.txt 0.3798828125k . file
lilmike is offline  
post #25 of 114 Old 01-06-2010, 09:56 PM
AVS Special Member
 
soho54's Avatar
 
Join Date: Sep 2005
Posts: 3,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
soho54 is offline  
post #26 of 114 Old 01-06-2010, 10:33 PM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
Impedance:



Scaled:



The range seems pretty high to me, especially that 12.5 Hz peak. On the last tapped horn enclosure I made and measured, I saw values that were lower than hornresp predicted in magnitude, but generally in the right spots frequency-wise. The last project did not use this driver - perhaps this driver is the issue?

It appears that I can tame the peaks within the passband (36, 79, and 109) a little bit by reducing the mouth size and playing with the lengths a bit, but the huge peak @ 12.5 Hz is interesting. It is also outside our passband, so is it a concern?

Edit -

I swapped the very similar Tang-Band W8-740C into the cabinet, as I have measured specs for these drivers too. It appears that the low-frequency impedance peak is more cabinet-related than driver-related, as the Tang Band driver also simulates with a 100 ohm peak in that region (16 Hz).

Edit 2 -

Scratch that - this looks to be purely a driver-related phenomena. The impedance magnitude is a result of the rather high Qms value measured for the MCM 55-2421 (mine actually measured lower than published though...), meaning that the driver diaphragm is not well damped mechanically and has low mechanical losses. Qms affects the Rms value calculated by hornresp. Changing Rms definitely changes the height of the impedance spikes. As to whether or not it is an issue - that is beyond my ability to determine. My (rather basic) understanding is that typical amplifiers are voltage sources, and are current limited. As such, we are usually more concerned with low impedances, where amplifiers try to deliver more current to the driver.

Clearly, I'm learning too. Perhaps another driver would be a better choice?
lilmike is offline  
post #27 of 114 Old 01-07-2010, 06:25 AM
AVS Special Member
 
Ricci's Avatar
 
Join Date: Mar 2007
Location: Louisville KY
Posts: 5,093
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 204
Those impedance peaks will never be that high in reality. I think that goes back to the fact the HR assumes unobtainium strength enclosure walls and surfaces with perfect loading and sealing too.
Ricci is offline  
post #28 of 114 Old 01-08-2010, 12:41 AM
AVS Special Member
 
soho54's Avatar
 
Join Date: Sep 2005
Posts: 3,329
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 17
Yeah, the impedance is a little funny. I wouldn't worry about anything outside of the pass band.

As you noted the peaks here will not be as high in real life, but those are some crazy high numbers. I am more interested in the shape the dips take. The same with driver displacement bumps.
soho54 is offline  
post #29 of 114 Old 01-13-2010, 06:10 AM
Exclusively Servo
 
Rythmik's Avatar
 
Join Date: Aug 2005
Location: Austin
Posts: 1,831
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Liked: 72
First, very interesting thread. The correlation between excursion bump and valley and SPL output gives us clue on how to modify the BL value or Qts to get smoother output. For instance, at excursion peak, if the SPL output is too high, increasing BL value can reduce the SPL output at that frequency. On the other hand, at excursion valley, if the SPL output is too high, reducing BL value can reduce the SPL output. Overall, it is a process of how we can pick a BL value to get the smoothest output. Of course other factors such as fs and Mmd affects the general shelving direction.
Rythmik is offline  
post #30 of 114 Old 01-13-2010, 09:40 AM - Thread Starter
AVS Special Member
 
lilmike's Avatar
 
Join Date: Dec 2009
Location: Pacific Northwest
Posts: 2,189
Mentioned: 0 Post(s)
Tagged: 1 Thread(s)
Quoted: 15 Post(s)
Liked: 85
Quote:
Originally Posted by Rythmik View Post

First, very interesting thread. The correlation between excursion bump and valley and SPL output gives us clue on how to modify the BL value or Qts to get smoother output. For instance, at excursion peak, if the SPL output is too high, increasing BL value can reduce the SPL output at that frequency. On the other hand, at excursion valley, if the SPL output is too high, reducing BL value can reduce the SPL output. Overall, it is a process of how we can pick a BL value to get the smoothest output. Of course other factors such as fs and Mmd affects the general shelving direction.

Thanks, and thanks for your insight. I'm still trying to wrap my head around driver parameters and their influences on the response in a tapped horn. While Bl is key, I think Bl by itself only tells part of the story, it needs to be looked at along with Re at the same time.

Some manufacturers present the ratio of Bl/sqrt(Re), which seems to have a good correlation with a driver's performance in a tapped horn.

I have a reasonable approximation of the tutorial tapped horn already under construction, and I have several other sets of 8" drivers in addition to the MCM 55-2421s. My current plan is to swap out the pairs of drivers and compare specifications, predictions, and measured performance later in this thread.

Any other suggestions as to how I could explore this further?
lilmike is offline  
Reply DIY Speakers and Subs

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off