Keystone Sub Using 18, 15, & 12 Inch Speakers

Dang Art. Careful with the equipment man. Of course its just not a complete day of testing without frying a sound card or mechanically breaking an $800 driver. Both things I have done.

TH distortion is more variable than what you see with traditional direct radiator systems. Some areas are very very clean and others within the fr will have much higher thd than the equivalent direct radiator. The varying loading and pressures on the cone, the xmax variance and more importantly the acoustic boosting at certain freqs causes the thd profile to be highly range dependant unlike a traditional direct box that has a predictable but comparatively smooth thd profile. TH's have much greater swings in thd level through out the operational range.
Josh,
At least when a driver is destroyed while testing one finds the limits of it's potential. Computers riding on the wind just blows.
I do have a bit of an appetite for destruction, do you have pictures of the broken driver?
Did it make funny noises before coming apart?
Are your TH enclosures built yet?

Art
 
Josh,
At least when a driver is destroyed while testing one finds the limits of it's potential. Computers riding on the wind just blows.
I do have a bit of an appetite for destruction, do you have pictures of the broken driver?
Did it make funny noises before coming apart?
Are your TH enclosures built yet?

Art

TH enclosures are still in process. Not sure on an ETA. I will be sure to drop you a note when I get them in.

Picture of the failed driver, which had the triple joint break and ripped the spiders but the coil still never bottomed, a graph of the electrical signal FR and a brief synopsis of what happened are in paragraph 3 & 4 here...
Data-Bass

I had it reconed. User error on my part. The dangers of EQ. The driver has the most displacement available in a single mass produced unit that I am aware of. I have 5 and they have been incredibly rugged over the years but that much power below tune in a ported enclosure is too much for even one of those. 5" p2p excursion is scary looking even from 50+ ft away.
 
Keystone Mouth compared to full width Exit

Hi Art,
It would be nice to see the difference in the keystone vs a normal box width terminated TH mouth.
My laptop is back, memory undamaged, screen replaced.

Below are screen shots of the Keystone exit (TH18 S glued), which is 6 inch wide at the top, 12.5 at the bottom, and 25 tall, 306.25 square inches, compared to the normal box width termination of 300 square inches, a 25 x 12inch exit. The Keystone shaped exit smooths out the response, and adds level over most of the pass band.

The next screen shows 200, 300,400 and 500 square inch exits compared to the TH18 S glued (Keystone exit). Although a bit harder to see with all the traces compared, the Keystone shape improves level and smoothness over any standard termination size.

Many (dozens) more shapes and sizes were tested, the 306 square inch Keystone exit size (which turned out to be within a few square inches of what a "normal" exit should be for this TH) had the most extended, smooth response of any.

Art Welter
 

Attachments

  • Keystone, 300 sq..png
    Keystone, 300 sq..png
    51.8 KB · Views: 3,109
  • Keystone&full width.png
    Keystone&full width.png
    70.1 KB · Views: 3,081
  • Keystone Both sides.jpg
    Keystone Both sides.jpg
    57.2 KB · Views: 3,654
Last edited:
It looks like you've used the keystone to extend the mouth opening to the driver tap?
Do you think that a TH where the driver is not recessed into the horn would have a benefit from a keystone mouth?
Yes, the thinner portion of the Keystone mouth does extend nearly to the center of the driver.
Since heat rises, having the speaker's rear cooling vents near the upper portion of the Keystone exit also gives thermal advantages.

I think many upright fold TH which have the final exit making a roughly 90 degrees turn would benefit from the Keystone mouth, whether or not the driver is located at the very end (mouth) of the horn.

The Keystone shaped exit would probably offer no advantage in the type of tapped horns where the final segment is pointed forward with the driver located at the mouth.

Art Welter
 
Last edited:
Art, did you try a similar key opening (same mouth area) in different position in relation to your final one?

(Danley seems to use a similar approach and I'm curious about those results)
 

Attachments

  • Keystone Both sides2.jpg
    Keystone Both sides2.jpg
    33.6 KB · Views: 3,048
Last edited:
Maybe not Dan, maybe it's nothing more then a design solution in the TH28. I was just wondering about the results of the "Art-key-opening" in relation to its position. If it still works the way it suppose to and if it behaves perhaps a tiny little better in the lowest octave.

An externally hosted image should be here but it was not working when we last tested it.
 
Last edited:
Art, did you try a similar key opening (same mouth area) in different position in relation to your final one?

(Danley seems to use a similar approach and I'm curious about those results)
Danley's unusual horn mouth shapes are whatever is "left over" after the folds are done.
The TH 28 exit shape is determined by the speaker baffle position, the same cover can be put on the “front” or the “side” of the cabinet to give location options.

The Keystone shape, with the horn's fold orientation, allows a compromise between a “too large” and “too small” opening.
Too large an opening provides more upper response with less low end, too small makes for a lower LF corner but less upper response.

I did not try asymmetrical openings.
The “sideways” keystone in your picture would work about the same as a 300 square inch opening at the bottom of the cabinet, as shown in the chart in post #23.

The sketches below show the various shapes tried before (and after) arriving at the Keystone shape.
The list of tests is actually only a fraction of the tests performed, many “in between” area tests were done and not saved.

Early in the testing one shape seemed promising, the extended upper response turned out to be a clamp resonating on a piece of wood.
After discovering that problem, each test required walking a 200 foot round trip twice, once to set the clamps and pieces of plywood in place, back to the shop to turn the pink noise on, then back outside to listen for any rattles and correct them if necessary. In retrospect I could have just unplugged the cabinet and saved half of those trips, but probably would have worn out the plugs from hot patching.

Of particular interest, the shapes in “E” and “N”, although similar in cross sectional area in relation to the horn path, did not smooth out the response, or increase SPL like the keystone shape.
 

Attachments

  • Test Sheets.jpg
    Test Sheets.jpg
    64.2 KB · Views: 1,596
  • TH Tests.jpg
    TH Tests.jpg
    107.7 KB · Views: 1,228
  • Keystone Fold.jpg
    Keystone Fold.jpg
    11.2 KB · Views: 1,280
Good stuff Art.

I am surprised at how much of a difference the shape made in some of the tests. Obviously some of it is changing the driver relationship in distance to the mouth and the mouth area can also have a profound affect but still...Filling in by as much as 7db in some areas and at the knee some major differences in extension and hence the low frequency output changed by as much as 8db below 35hz.

Why do you think the keystone pattern was the best compromise? I mean how is the shape / location affecting the performance in such a positive way as compared with a regular rectangular or square exit which is more common? Did you try a rectangular mouth of the same area which also started in the same spot in the path and ended in the same spot without the expansion?
 
Good stuff Art.

I am surprised at how much of a difference the shape made in some of the tests. Obviously some of it is changing the driver relationship in distance to the mouth and the mouth area can also have a profound affect but still...Filling in by as much as 7db in some areas and at the knee some major differences in extension and hence the low frequency output changed by as much as 8db below 35hz.

Why do you think the keystone pattern was the best compromise? I mean how is the shape / location affecting the performance in such a positive way as compared with a regular rectangular or square exit which is more common? Did you try a rectangular mouth of the same area which also started in the same spot in the path and ended in the same spot without the expansion?
Changing the mouth on a TH is like changing the port on a BR cabinet.
As you know, changing the shape of a port also can affect tuning quite a bit.
The comparison BR cabinet used a "top hat" kind of port, the addition of the "top hat" made the port tune way lower than I would have thought, same as what happened with your "25 Fb should have been 30" box.

I did try a rectangular mouth of the same area which started in the same spot in the path and ended in the same spot without the expansion, test "R".

The Keystone shape seems to be the best compromise for this particular cabinet, I don't know how to explain it other than the way Goldilocks did her testing- too hot, too cold, to hard, too soft, just right.

I'm sure if I had extensively studied laminar flow patterns and such I could give a more technical explanation.

I found the sound output of the Keystone can throw up large rooster tails of sawdust in the shop.
It would be interesting to use a strobe and smoke and watch the air movement patterns around the exit.

Although LF sound behaves in a cabinet more like laminar flow than ray tracing (light waves), if one visualizes the last portion of the horn, a "keystone" correction needs to be applied for the horn exit to appear rectangular from inside the horn. A regular rectangle or square is actually "too big" at the top of the exit. The overhead photos show this effect, when the keystone shaped exit appears more square, the cabinet itself looks keystone shaped.

It is possible another TH with different dimensions (but similar fold pattern) may work better with a slot or "A" shape, but I am fairly sure none would benefit from a "V" shape, as tests "E", and "N" resulted in worse response curves than a standard opening of the same area.
Those results support my "keystone correction" analogy.

Test "F" was poor too, it seems the sound waves are disturbed with side openings rather than center openings.

Mouth shape, size and positioning all have interrelated functions, it took a lot of tests to work them out.

Art
 

Attachments

  • Test Sheets.jpg
    Test Sheets.jpg
    64.2 KB · Views: 2,319
  • Keystone Correction.jpg
    Keystone Correction.jpg
    23 KB · Views: 2,301
  • Keystone Fold.jpg
    Keystone Fold.jpg
    11.2 KB · Views: 2,291
Last edited:
Fair enough... I like the way that you planned to have a removable panel at the mouth so that you could experiment with different configurations. That is good thinking and planning ahead.
The entire front of the cabinet was left off for most of the testing, as I was also experimenting with corner ramps too.

I'm glad to be done with the Keystone, lugging out the cabinet, all those clamps and bits of wood started to get old after a few days. Or weeks.

Art Welter
 

Attachments

  • Clamps.jpg
    Clamps.jpg
    98 KB · Views: 2,358
  • Done.jpg
    Done.jpg
    60.7 KB · Views: 1,325
tapped horn / bandpass ?

Hi weltersys,

This is all very interesting. It looks like your keystone mouth converts the driver-to-mouth horn section into a tuned chamber, as you would find it in a bandpass. Did you take any measurements w/ the mouth at very end of the horn (i.e.: the bottom of the enclosure)?

Regards,
 
Hi weltersys,

This is all very interesting. It looks like your keystone mouth converts the driver-to-mouth horn section into a tuned chamber, as you would find it in a bandpass. Did you take any measurements w/ the mouth at very end of the horn (i.e.: the bottom of the enclosure)?

Regards,
Post #26 and Post#29 but maybe if enough people ask then Art might....perhaps...