Gjallerhorn

[Josh Ricci]

If anyone wants a build print for this cab in pdf form either to build or for curiosity sake alone send me your email in a pm. The print is intended for a wood shop to use with a cnc, but all of the information is there to make any modifications necessary.

 

[weltersys]

Josh,

Interesting to compare your results of the Gjallerhorn’s output and distortion to the DSL DTS-10.

At 20 Hz, where a 5 dB increase in SPL “sounds” twice as loud, the Gjallerhorn can play 8 dB louder than the DTS-10 with less distortion. What is the volume (cubic feet or liters) of the DTS-10 compared to the Gjallerhorn?

The DTS-10 uses push pull loading, yet for the most part, 2nd harmonic distortion (which PP is purported to reduce ) is generally higher than 3rd, and also higher than the single driver Gjallerhorn.

My high power tests on Eminence Lab 12 in ported and sealed cabinets run PP compared to normal had mixed results, PP reducing 2nd harmonic at some frequencies, while increasing it and 3rd harmonic distortion at others.
http://www.diyaudio.com/forums/subwoofers/191833-push-pull-vs-normal-distortion-compared.html
Any educated guess on how much, if any, the PP loading reduces distortion in the DTS-10?

Art Welter

 

[Josh Ricci]

Well...To be honest I don't have a gut feel on how the push pull mounting of the drivers is affecting the distortion in the DTS10 cab. I haven't looked at it enough. I may do a test of this if I don't sell my last DTS10 first. The distortion in the DTS-10 cab is rather high in some places. It is predominantly H2 which is good. The much bigger lesson I have taken from things so far is that any large acoustic gain in a design will also boost the distortion harmonics that fall into that range. In TH's when you have the first large spike followed by the next 2 large boosts and rather jagged response the harmonics that fall into that range are going to get amplified mightily. Look at what happens at 27hz with regards to distortion as the second harmonic gets acted on by the massive 54hz spike in response. The DTS10 is a fairly rough response shape. I would expect the other DSL cabs which usually exhibit a much smoother response to be better in this regard.

The Gjallerhorn is 24x45x45 or 28.125 cubic ft external.
The DTS10 is 16x44x59.5 or 24.24 cubic ft external.

The Gjallerhorn is a 17hz cab where the DTS10 is about a 14hz one. The driver in the Gjallerhorn is a very low distortion unit which is of much higher quality and heavier duty than the stock DTS-10 drivers. Also it has a substantially smoother response than the DTS-10, though it is still jagged.

 

[weltersys]

Well...To be honest I don't have a gut feel on how the push pull mounting of the drivers is affecting the distortion in the DTS10 cab. I haven't looked at it enough. I may do a test of this if I don't sell my last DTS10 first. The distortion in the DTS-10 cab is rather high in some places. It is predominantly H2 which is good. The much bigger lesson I have taken from things so far is that any large acoustic gain in a design will also boost the distortion harmonics that fall into that range. In TH's when you have the first large spike followed by the next 2 large boosts and rather jagged response the harmonics that fall into that range are going to get amplified mightily. Look at what happens at 27hz with regards to distortion as the second harmonic gets acted on by the massive 54hz spike in response. The DTS10 is a fairly rough response shape. I would expect the other DSL cabs which usually exhibit a much smoother response to be better in this regard.

The Gjallerhorn is 24x45x45 or 28.125 cubic ft external.
The DTS10 is 16x44x59.5 or 24.24 cubic ft external.

The Gjallerhorn is a 17hz cab where the DTS10 is about a 14hz one. The driver in the Gjallerhorn is a very low distortion unit which is of much higher quality and heavier duty than the stock DTS-10 drivers. Also it has a substantially smoother response than the DTS-10, though it is still jagged.

The much higher clean output of the Gjallerhorn over the DTS-10, even with the 3 Hz higher corner, and slightly larger size, is impressive.

Again, my congratulations on what appears to be the the best documented, highest clean output, single speaker VLF sub in it’s size range on the planet.

You showed interest in hauling your subs to a shootout.

Your test methods show far more useful information than any I have seen at any sub shootouts, have you considered running one of your own?

 

[Josh Ricci]

Hmmmm....You know I hadn't really thought about having my own. We could probably get away with it at the normal test spot I use. It isn't much in the way of accomodations though. If Wayne has another I am going to try and take one of these out there.


Thanks for the compliments. I am pretty good at the brute force method. I am working on the finesse side of things so I can blend the 2 together and get the proverbial Iron fist in a velvet glove.

 

[weltersys]

Hmmmm....You know I hadn't really thought about having my own. We could probably get away with it at the normal test spot I use. It isn't much in the way of accomodations though. If Wayne has another I am going to try and take one of these out there.


Thanks for the compliments. I am pretty good at the brute force method. I am working on the finesse side of things so I can blend the 2 together and get the proverbial Iron fist in a velvet glove.

Wayne has not mentioned a shoot out again since he realized I had never suggested that a front loaded sub less than 1/4 the size of a 12Pi was close to the output.

If he has looked at the output vs size of our THs compared to his FLH, I doubt if he will be very interested in setting up a shootout where he knows his cabinet would be a couple bullets short .

Art Welter

 

[Josh Ricci]

Oh I don't know...I would expect the 12pi to be an absolute beast above 35hz. It is a very large sub and it is concentrated on the 40-100hz area primarily. On a 1/1 basis I don't expect many other cabs to compare well in that range. Now if we are talking about multiples of other cabs then that changes things quite a bit.


Btw I talked with Chris and it looks like the V/I box fried due to user error on my part. I must have somehow reversed the polarity from the amplifier into the box before the high power tests. An easy mistake to make if switching the V/I box in and out of the signal chain and not paying attention closely. D'oh!

 

[weltersys]

Oh I don't know...I would expect the 12pi to be an absolute beast above 35hz. It is a very large sub and it is concentrated on the 40-100hz area primarily. On a 1/1 basis I don't expect many other cabs to compare well in that range. Now if we are talking about multiples of other cabs then that changes things quite a bit.


Btw I talked with Chris and it looks like the V/I box fried due to user error on my part. I must have somehow reversed the polarity from the amplifier into the box before the high power tests. An easy mistake to make if switching the V/I box in and out of the signal chain and not paying attention closely. D'oh!

The 12 Pi appears to average 120.3 dB at one meter from 35-100 Hz, 1600 watt measurement, (20 dB added to compensate for the 10 meter vs 1 meter) as high power as Wayne seems comfortable testing it.
You are feeding the slightly smaller Gjallerhorn more than double the power, but from what I see on your chart, looks like it averages over 123 dB from 25-100 Hz.
From a SPL standpoint in singles, I’d bet on your box .
The FLH would gain some LF in multiples though, while the TH gains in forward directivity, making multiples more of a crap shoot as far as SPL advantage.

A pair of BC18SW115-4 loaded Keystone TH with extenders are about the same (stored) size, sensitivity, and frequency response as the 12Pi, and about double the power handling.

I’d expect them to also better the 12Pi by about 3 dB given double the power, and be neck and neck at the same power.

Did you test the two Gjallerhorns together?

Art Welter

 

[Josh Ricci]

No I did not test both together. Maybe some day.

 

[weltersys]

No I did not test both together. Maybe some day.

Don’t bother, today Wayne has already explained all you need to know:

“As for the tapped horn, it's a different animal. You can always say "load it with a bigger, more powerful woofer and it will be louder". You can say this about front loaded direct radiators too. But in large venues, I think the bulk-efficiency thing becomes more and more difficult - I don't think you can get them to work together as well as full-size basshorns in large groups. They're probably more useful in smaller venues.“

I guess those front loaded subs on all those “smaller venue” Clair Brothers tours like U2 in stadiums just don’t work, and our tapped horns that give lower, louder, frequency response in smaller packages than FLH are not working either.

Ah, shucks..

 

[jbell]

Should be easy enough for you to prove or disprove, perhaps you will do a test with the same speaker to verify how close or far apart the response is.

Based on what you have posted, my money is on a pretty close (within 2 dB) level between a ported cabinet and the tapped horn 45-125 Hz.

I hope you prove me wrong, I want to believe !

I guess those front loaded subs on all those “smaller venue” Clair Brothers tours like U2 in stadiums just don’t work, and our tapped horns that give lower, louder, frequency response in smaller packages than FLH are not working either.

Ah, shucks..

You've come a long way art.....

 

[Josh Ricci]

I am not sure that I buy into the whole "Horns experience a huge transformation when moving to multiples" thing. Isn't the flattening and what some think is a lowering of response corner just due to the efficiency ceiling being met in the peaks of the horns response causing less gain there than at the lower output regions in response? That is the way I have been leaning but I am no expert on the phenomenon.

Is there a definitive research paper or data collection for this theory? I have seen a few 1/2/4 cabinet tests but they always seem to be missing something to me.

 

[weltersys]

Originally Posted by weltersys
Should be easy enough for you to prove or disprove, perhaps you will do a test with the same speaker to verify how close or far apart the response is.

Based on what you have posted, my money is on a pretty close (within 2 dB) level between a ported cabinet and the tapped horn 45-125 Hz.

I hope you prove me wrong, I want to believe !


Quote:
Originally Posted by weltersys
I guess those front loaded subs on all those “smaller venue” Clair Brothers tours like U2 in stadiums just don’t work, and our tapped horns that give lower, louder, frequency response in smaller packages than FLH are not working either.

Ah, shucks..

You've come a long way art.....

Well, I have made BR cabinets which have more LF output per cubic foot than FLH cabinets, and TH cabinets that have more output per cubic foot than FLH cabinets, but BR still can put out more LF per cubic foot given more power and more cones.

Although I found the TH can raise efficiency by about 6 dB over a ported cabinet with the same low corner, it takes almost double the cabinet volume.
Double the ported cabinet, you gain 3 dB efficiency, double the power, you gain six dB, the two BR can put out about 3 dB more level than the TH in the same truck space.

This is why BR is still the choice for most touring companies, and of course, people are used to the sound of BR helps too.

Although I'm happy with the efficiency gain going from BR to TH, the SPL output gain using the same volume of cabinet is not what I’d hoped.

Art Welter

 

[weltersys]

I am not sure that I buy into the whole "Horns experience a huge transformation when moving to multiples" thing. Isn't the flattening and what some think is a lowering of response corner just due to the efficiency ceiling being met in the peaks of the horns response causing less gain there than at the lower output regions in response? That is the way I have been leaning but I am no expert on the phenomenon.

Is there a definitive research paper or data collection for this theory? I have seen a few 1/2/4 cabinet tests but they always seem to be missing something to me.

Most of the whole "Horns experience a huge transformation when moving to multiples" thing is naiveté, a lot of people don’t seem to understand that doubling cone area and power gives a six dB advantage whether horn loaded or not.

Going from one cabinet to four adds 12 dB of low frequency, over twice as loud.
Hence the oft heard comments “you really need four of (insert cabinet name) to work.”

With “too small” horn mouth FLH cabinets, the low corner does go down a few Hz, and the LF level does increase a few dB in multiples in addition to the expected cone area and power increase . The smaller the FLH cabinet, the more noticable the difference.

I have not seen a definitive research paper or data collection for this phenomenon, but Hornresp predicts the change fairly accurately on the bottom for FLH (but exaggerates it for TH) but AFAIK ignores what happens up at higher frequencies in multiples.

The large FLH array, still within a 1/4 wavelength from one side to another, reinforces the LF, but at the upper end, destructive interference starts to occur. At 100 Hz, a quarter wavelength is only 2.83 feet, less than a meter wide.
A LF source wider than 1/4 wavelength starts loosing upper response, while the LF increases, making the LF gain seem even greater by comparison.

The wider the array, the more the loss, or lack of gain, in the upper response.

An old test of my C horns from back in the pre Smaart era shows this, around 18 dB gain from two cabinets compared to to eight at 31.5 Hz, but only a 5 dB gain at 100 Hz.

That’s a 6 dB gain at 31.5 Hz over the expected 12 dB, but a 7 dB loss at 100 Hz over expected.

The C horn is a FLH with about a 65 inch path length, 26.5 x 30 inch mouth, 30 inch depth.
The 12C were loaded with Lab 12, the 15C loaded with EV150X.

The four cabinet tests of the “C” horns had the cabinets stacked 2 high with the flat sides of the horn together, previously found to be the optimum configuration for four.
When all 8 were tested together, they were two side by side stacks , 10 feet wide. Microphone was at 25 feet from the front center of the arrays.

 

[boydon_lepasci]

While the various types of horns really interest me, I wonder if any of them make sense for pro or semipro use. The "unwashed masses" who must be catered to because they pay the bills are unlikely to appreciate what they are hearing and will probably mistake it for weak bass versus direct radiators.

 

[weltersys]

While the various types of horns really interest me, I wonder if any of them make sense for pro or semipro use. The "unwashed masses" who must be catered to because they pay the bills are unlikely to appreciate what they are hearing and will probably mistake it for weak bass versus direct radiators.

No one hearing a tapped horn with a similar frequency response would “mistake it for weak bass versus direct radiators”, though they might find it “sounds different”.

I have been providing professional production services since the mid 1970’s.

At many levels, the “unwashed masses” who must be catered to have no opinion as long as they can hear whatever they want as loud as they want it, whether hip hop, heavy metal, or a corporate speaking engagement.

There are many in market sectors that still can choose the tools they find are best for the job.
I just provided sound for an act, using tapped horn low cabinets, the time before provided bass reflex, their engineer had no comment in either case, as he was able to get what he wanted.

Though I have always utilized my own proprietary cabinets, there is a “sound by the pound” sector of the business that requires rider conformity.

If I still was catering to that portion of the business, I’d try to provide the most popular subs, and the most popular subs continue to be bass reflex.

 

[boydon_lepasci]

Thanks for the perspective Art. I was ~familiar with your background and mostly tolling for your feedback, but I welcome more responses to widen the experience base.

 

[weltersys]

Thanks for the perspective Art. I was ~familiar with your background and mostly tolling for your feedback, but I welcome more responses to widen the experience base.

It would be interesting to chart the many ups and downs of horn woofers in the Pro market through the years.

Back in the days of small amps and 100 watt speakers, touring bass horns were pretty common, in some markets over 50% of bass came through horns.

Now, if I were to guess,in the USA I’d put the ratio at around 80/20, and TH would only be a small fraction of the 20%, as they have only been available only a relatively short time in a market resistant to change.

Art Welter

 

[18Hurts]

Now, if I were to guess,in the USA I’d put the ratio at around 80/20, and TH would only be a small fraction of the 20%, as they have only been available only a relatively short time in a market resistant to change.

Art Welter

I have noted the explosive growth of line arrays, do you think the much lighter weight of BR works better with that setup? As far as physically moving speakers around, BR has it over horns when it comes to ease of moving the lighter cabinets.

The thing I ponder is what type of sound systems do they use in IMAX theaters? Since they are a more permanent installation--do horns dominate new construction?

 

[Upstate NY low life]

That's some very nice work there. Maybe even a slight rounding of each corner, including the insides of the turns by chamfering, could help a lot with the aerodynamics. With all that folded length scrunched so small, I wonder if a gloss finish on the whole inside would help, too. I was thinking about that for the Graham Holliman, whose instructions stress the perfect rounding of the one 90 degree fold that has.