I built Cowan's 30Hz Tapped horn

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi StainlessSteve,

A nice example for this type of TH is Volvotreter's Eminence Horn:

Tapped Horns

This one also shows how to shorten L12 (throat to driver center), and a way to lengthen L34 (driver center to mouth).

There are additional nice examples of TH designs and simulations on Volvotreter's website.

Regards,
 
Are you filtering it to play only below 80 Hz? If not, then those out-of-band peaks above 100 Hz could make it sound pretty awful.

If you are filtering it, then try to measure the impedance curve and compare it to the Hornresp sim. If it's off, then there's something wrong with the build. A WT3 or DATS can do the measurement in seconds. If you don't have that, you could use REW(free) and a DIY jig (Google it) to measure the impedance curve.

I've found that horns are really REALLY sensitive to leaks.
I've been able to completely transform the sound of my horns by liberally sealing them.

Dats is essential; I just ordered my third one last week. (I've managed to break two of them)
 
I've found that horns are really REALLY sensitive to leaks.

Same here, particularly if they're around the driver.

One mistake I've made on several of my builds is to make the cutout for the driver the same size as the published requirement, even though the driver was going to be mounted in reverse (for most THs). That's a great way to introduce small leaks around the flange :-(.

Now I make the cutouts at least 4mm smaller in diameter when designing or building.
 
I've found that horns are really REALLY sensitive to leaks.
I've been able to completely transform the sound of my horns by liberally sealing them.

Dats is essential; I just ordered my third one last week. (I've managed to break two of them)

Same here, particularly if they're around the driver.

One mistake I've made on several of my builds is to make the cutout for the driver the same size as the published requirement, even though the driver was going to be mounted in reverse (for most THs). That's a great way to introduce small leaks around the flange :-(.

Now I make the cutouts at least 4mm smaller in diameter when designing or building.


Wodwork, measure twice, cut once! As for leaks, I just use some foam tape. 1mm Thick Phone Display Dustproof Black Sponge Double Adhesive Foam Tape Gasket | eBay
 
Wow, 4mm? Not sure how close mine is, but I initially cut it smaller where I had to sand it until it fit, so it was pretty damn close.
I think you guys just hit the nail on the head with leaks. Right above the driver, I drilled a small hole just large enough to run 12 AWG wire through, then filled it with caulk. That hole is very likely the cause of my leak. The box sounds very quiet, and any audio is highly undefined, a leak seems to make the most sense. I never expanded on any description of it's sound, because I assumed the issue was phase and it was a closed case, but I since tried it wired in reverse with no difference.
I'm almost certain it can't be one around the flange, because I recently pulled the driver out to check behind it and by the terminals, (mind you I screwed it in hand-tight) but the driver was in there so tight I had to yank on it AFTER removing all the screws! . The foam on the flange is pretty high, so I doubt it was that, but I do have foam tape just in case.
I sure hope it's not a leak around the terminals, if so I can always put caulk around those.
 
Last edited:
I felt uncomfortable using it to seal a hole. It's a soft, non-expanding adhesive that doesn't stick. I should have just used adhesive like I have to seal it with now

Caulk is one of those misnomer's of the car audio world. Whoever came up with the idea of using caulk for enclosures, needs a serious slap :) But the idea seems to be welcomed by the SPL clan in particular.
 
I finished this cabinet, I got a Faital Pro 12RS550 and it sounds amazing with tons of output. Here are 1 meter ground plane measurements I took with my professionally calibrated EMM-6. Blue line is 1 cycle group delay which is lower than the threshold of audibility.
Group delay
H2sd5MC.jpg

FR and Phase. I don't have an SPL meter so the SPL is uncalibrated, but this was a 8.7v 1M long sweep.
kjjzypS.png

The driver is rated to 500 AES watts and it will take about 1200 to reach maximum output:
V8pLBED.png

The cabinet is currently coated in about 3/16" of truck bed liner, has a routed recess and t-nuts around the mouth for the grill, and all the edges have a roundover on them. There's a handle on the top and bottom. It is 37 liters smaller than the TH112 (it's 10.6 cubic feet or 301 liters, TH112 is 338.5) and about 10dB louder at 31hz.
TH112 Frequency response and phase:
w7M8t9H.png
 
FR and Phase. I don't have an SPL meter so the SPL is uncalibrated, but this was a 8.7v 1M long sweep.

The driver is rated to 500 AES watts and it will take about 1200 to reach maximum output:

It is 37 liters smaller than the TH112 (it's 10.6 cubic feet or 301 liters, TH112 is 338.5) and about 10dB louder at 31hz.
StainlessSteve,

Your sub's response looks amazingly smooth for being "unsmoothed"!

Your sim would indicate around 94 dB sensitivity at 30 Hz, with no SPL meter and no calibration, why do you figure around 104 dB sensitivity for your sub at 31 Hz?

I thought you might like to see the more "smoothed" version of the TH-112, downloaded 3/9/05, the measurement date for both the one you posted and the smoother version is 3/4/05.

Art
 

Attachments

  • TH-112.png
    TH-112.png
    277 KB · Views: 527
Because I've seen how Danley measures their maximum SPL. They quote from the peaks in the FR, you can see it on Data-bass's measurement of the DTS-10 where the only place reaching 133dB is above 63hz, they don't quote from the 0dB point. My driver might have an xmax of 9.25mm, but it has an xmech of 20 and a larger SD.
 
Last edited:
Great to see measurements of your ambitious sub. Thanks. But I'm not clear about how you measured (out of doors?) and the smoothing. Distortion would be nice to post too.

Results seem good although low freq extension could be achieved simpler ways - even if not max output.

And striking that your box looks very much like my 17-foot folded pipe and about the same volume.

17 foot pipe sub 12-230 Hz ±5dB

Could folks please comment on the pro's and con's of these two physically comparable subs? Or how StainlessSteve or I could add data to compare them better?

B.
 
The measurement was a 1 meter groundplane, 1M sweep length, 7 volts at 30hz and 73f with ~38 percent humidity one thing I would never use is smoothing. The nearest object was behind it and 88 feet away, and the next nearest object was to the side of it and 376 feet away. The was a Crown XLS 1500, the soundcard was a TASCAM US-1641, the mic is an EMM-6 independently calibrated by Cross-Spectrum Labs. I wish I had an SPL meter and a better mic, then I would post distortion measurements and have SPL data.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.