Thanks, that response looks terrible.
Looks like the Lab 12s will stay in the ported boxes .
My gut says it may be more than negligible difference.4? I am not sure if they would fit without extensive modification or at all. 3 might be workable. I posted the HR input data. You could give it a whirl yourself you know.
Yeah I beefed the bracing a lot but I don't think it will affect things much whether it is skeletonized or not. My gut tells me it is negligible.
Think in terms of very long ports, a slotted port (port using a cabinet side wall) will tune lower than a “normal” cylyndrical port of the same area and length.
Anyway, you may be right, the difference may be negligible, no way to know for sure without an A/B test.
As far as the me inputting the data, Hornresp is on a computer I have not used in several months, I’d have to figure out how to use it again (and would then want to download the new version of Hornresp), so I was just being lazy
.Art
Art,
Part of the reason that I did all of the modeling for this and the Gjallerhorn in Akabak was because of questions like how the bracing will affect the response. I have modeled the paths when split by bracing as separate segments. In the Gjallerhorn this type of detail actually caused me to run out of script. I was told that it was unecessary to go to such lengths but I wanted to be sure. The differences in the simulation with solid bracing and without are very small. Real world driver parameter differences and build differences will amount to more. That's why I am not worried about it too bad. At this point it is time to look at real as built results.
Part of the reason that I did all of the modeling for this and the Gjallerhorn in Akabak was because of questions like how the bracing will affect the response. I have modeled the paths when split by bracing as separate segments. In the Gjallerhorn this type of detail actually caused me to run out of script. I was told that it was unecessary to go to such lengths but I wanted to be sure. The differences in the simulation with solid bracing and without are very small. Real world driver parameter differences and build differences will amount to more. That's why I am not worried about it too bad. At this point it is time to look at real as built results.
Josh
Nice work!
I'm interested by the Sd/S2 compression ratio that you are using. 4.19 : 1 for a 21" driver.
Most designs posted on the forum seem to use a lower compression ratio.
As a rule of thumb I used approx 2.5 : 1 on the Xoc1 TH18.
This seems to be even more compression than the Gjallerhorn 3.88 : 1.
How does the compression ratio effect the performance of the speaker and driver?
Regards
Martin
Nice work!
I'm interested by the Sd/S2 compression ratio that you are using. 4.19 : 1 for a 21" driver.
Most designs posted on the forum seem to use a lower compression ratio.
As a rule of thumb I used approx 2.5 : 1 on the Xoc1 TH18.
This seems to be even more compression than the Gjallerhorn 3.88 : 1.
How does the compression ratio effect the performance of the speaker and driver?
Regards
Martin
In practice these are the compression ratios that seemed to provide realistic clearance for the drivers maximum theoretical excursion and a smooth response. A lower compression ratio in the throat results in larger magnitude variations in response. The cones and motors of the drivers used here are very strong. I had some reservations about whether the aluminum LMS ultra cone would hold up in the Gjallerhorn with its extreme amount of stroke but there have been no problems. If it was going to fold up or fail it would have during testing when it had 130 volts dumped into it with sine sweeps and 1/48th octave stepped sine tests covering 10-200hz. I have been using them with out issue. During the conception of this cab I asked Tom D. through email about his thoughts on whether the 21sw152 would take a 4/1 compression well. He thought that it should be fine. I am confident. I sort of march to my own tune, preffering to think about and try things myself rather than accept others doubts as fact. These are not the cones of yesteryear.
Josh,
Although I agree with most all you have said, my most early “expensive mistakes” were back in the days when aluminum (rather than titanium) HF diaphragms were used in the JBL line.
The aluminum diaphragms would develop stress cracks from fatigue over time, in my two years time with Eclipse Concert Systems, I became very adept at changing (and aligning) aluminum four inch 2440 and early 2441 diaphragms.
The aluminum diaphragms would all fail over time, time depending on excursion. There are plenty of 50+ year old aluminum diaphragms still working in low excursion home use, but none in high excursion PA use.
When the titanium diaphragms were introduced (JBL 2425, 2445) we started to use them at STS, and had no failures due to cracking from 1982 through 1992.
I had about twenty 1.75” 2420 aluminum diaphragm drivers which failed from cracking, all failed over several years, but when replaced with the titanium 2425 diaphragms, never failed (other than coil burnout) again.
In fact, JBL wrote in the 2425 spec sheet:
“This titanium diamond structure combines the ruggedness of phenolic and composite type diaphragms with the outstanding frequency response of the fragile aluminum and exotic metal diaphragms. Nontoxic titanium has no fatigue limit, It can last forever if not overdriven.”
It would be interesting to look at the LMS ultra cone under a strobe with a sine wave playing and see if the movement is purely pistonic, or if it does flex some.
Although your short term tests show the aluminum LMS ultra cone has held up fine, I hope they fare well over the long haul.
Art
Last edited:
We will see whether they will or not. If they fail a year or 2 from now from long term aluminum fatigue then I will know. One thing to remember is that these drivers were originally intended for car audio applications. Now that being what it is the one thing I do like about most car audio market drivers is that they are built to take a severe beating in competitions and from general neglect, tough environmental conditions and plain old user idiocy in some cases. We will see.
Car audio can be severe, but horn loading is more severe as far as higher and more uneven pressure. I am not all that familiar with the current range of car audio, but I do know that it would take a huge alternator, battery bank and capacitor storage to provide the kind of power that is common in pro PA use.
Do you know what the thickness of the 21sw152 aluminum cone is?
My mistake, you had mentioned "During the conception of this cab I asked Tom D. through email about his thoughts on whether the 21sw152 would take a 4/1 compression well. He thought that it should be fine. "
Do you know what the thickness of the LMS ultra cone aluminum cone is
?
I have an answer...The LMS cone is 24mil aluminum.
I have been playing with the new toys that have been put in HR. There are some severe pressures developed inside of horns at the throat. At the max 120v input the throat particle velocity reaches 20.7ms at 100hz and the throat pressure reaches 11,122pa at 29hz (174.9db) and at 110hz (>177db). Even less high pressure/power horns easily get up near 170db in the throat theoretically.
I have been playing with the new toys that have been put in HR. There are some severe pressures developed inside of horns at the throat. At the max 120v input the throat particle velocity reaches 20.7ms at 100hz and the throat pressure reaches 11,122pa at 29hz (174.9db) and at 110hz (>177db). Even less high pressure/power horns easily get up near 170db in the throat theoretically.
While 170db is no joke for our ears, 11,000pa is like 1.5 psi. This is a lot of force over the total cone area, but it doesn't seem high enough to cause point ruptures in materials such as aluminum or quality paper.
I feel like most of the dammage incurred in horns is due to unloading and overexcursion. Horns with compression chambers, and horns with higher bandwidth don't tend to suffer this fate because either the xmax is limmited due to a sealed chamber, or the dammage to the driver becomes plainly audible such as a kick bin with midbass duties.
Does this make sense?
Tade
I feel like most of the dammage incurred in horns is due to unloading and overexcursion. Horns with compression chambers, and horns with higher bandwidth don't tend to suffer this fate because either the xmax is limmited due to a sealed chamber, or the dammage to the driver becomes plainly audible such as a kick bin with midbass duties.
Does this make sense?
Tade
170 + db
@ Josh Ricci et al.
WOW, that's a heck of a lot of db loss from around 170db in the throat to approx 105 db etc @ the mouth !
Also it's a heck of a lot of initial gain too.
So "on paper" etc, we see around 65 + db of overall gain, then lose it ALL on the way to the mouth. Somehow it doesn't "sound" right ? I'd love to hear an explanation as to why this "could" be, & also graphs etc if possible
@ Josh Ricci et al.
WOW, that's a heck of a lot of db loss from around 170db in the throat to approx 105 db etc @ the mouth !
Also it's a heck of a lot of initial gain too.
So "on paper" etc, we see around 65 + db of overall gain, then lose it ALL on the way to the mouth. Somehow it doesn't "sound" right ? I'd love to hear an explanation as to why this "could" be, & also graphs etc if possible
24 mil is about .61 of a millimeter thick.
At well over 1 acoustic watt/sq cm, my guess is the 24 mil thick 21 inch diameter cone will not be purely pistonic in motion.
It will be interesting to study the stress cracks
.Art
Josh is talking about 120V, about 3600 watts into 4 ohms, not one watt producing 170 db.@ Josh Ricci et al.
WOW, that's a heck of a lot of db loss from around 170db in the throat to approx 105 db etc @ the mouth !
Also it's a heck of a lot of initial gain too.
So "on paper" etc, we see around 65 + db of overall gain, then lose it ALL on the way to the mouth. Somehow it doesn't "sound" right ? I'd love to hear an explanation as to why this "could" be, & also graphs etc if possible
IIRC, at 2 meters, the Gjallerhorn output with 120V is around 130 dB, using the inverse distance law, each time the distance is halved, six dB is gained.
This would put the one meter level around 136 dB.
At the mouth, the level may be around 142 dB (or more, the inverse distance law does not apply in close proximity to the mouth of a large horn) then there is a long horn path within the box that is also subject to the inverse distance law.
Let’s say that horn is 5 meters long.
At 2.5 meters inside, level is 148 dB, at 1.25 meters 154 dB, at .625 meters 160 dB, at .3125 meters, 166 dB, at .15625 meters, 172 dB.
No laws of physics are broken in the above example
.One consideration in high compression, high power horns is air actually becomes non linear at very high SPL, at sound levels of power densities above 1 acoustic watt/sq cm [160 dB] the compression and rarefaction of air becomes progressively non linear.
The maximum SPL that can be achieved in air is 194 dB, increases over 160 dB or so require progressively more power to overcome air non linearity.
I wonder if Hornresp takes air non linearity into account?
Art Welter
@ weltersys
Hi, thanks for posting.
Right, i see what was meant now, working backwards via the -6db method Even so i find it incredible that any Single off the shelf driver could/would Actually produce such high SPL levels, & also maintain them day in/out for hours at a time & for years !
Food for thought ? Take for eg an HF compression driver on a horn, rated at 112 db @ 1M for 1W. Using the reverse -6db method = 172db @ 0.9765625mm. At 64 Watts @ 0.9765625mm = 208db ! The compression would be Enormous, & i would expect the distortion to be off the scale too.
Something must be in error in all this, & i'm not afraid to be corrected if it's me
I'm unaware of any published data that has Actually measured such high levels at the throat etc, for Any driver/combo. If anyone knows of such, please post a link
Hi, thanks for posting.
Right, i see what was meant now, working backwards via the -6db method
Even so i find it incredible that any Single off the shelf driver could/would Actually produce such high SPL levels, & also maintain them day in/out for hours at a time & for years ! Food for thought ? Take for eg an HF compression driver on a horn, rated at 112 db @ 1M for 1W. Using the reverse -6db method = 172db @ 0.9765625mm. At 64 Watts @ 0.9765625mm = 208db ! The compression would be Enormous, & i would expect the distortion to be off the scale too.
Something must be in error in all this, & i'm not afraid to be corrected if it's me
I'm unaware of any published data that has Actually measured such high levels at the throat etc, for Any driver/combo. If anyone knows of such, please post a link
24 mil is about .61 of a millimeter thick.
At well over 1 acoustic watt/sq cm, my guess is the 24 mil thick 21 inch diameter cone will not be purely pistonic in motion.
It will be interesting to study the stress cracks
Art
The 21 is paper. The Gjallerhorn 18 is the aluminum one.
No stress cracks yet.
The inverse distance law does not continue to an infinitely small distance, with an infinitely loud SPL.@ weltersys
Hi, thanks for posting.
Right, i see what was meant now, working backwards via the -6db method
Food for thought ? Take for eg an HF compression driver on a horn, rated at 112 db @ 1M for 1W. Using the reverse -6db method = 172db @ 0.9765625mm. At 64 Watts @ 0.9765625mm = 208db ! The compression would be Enormous, & i would expect the distortion to be off the scale too.
Something must be in error in all this, & i'm not afraid to be corrected if it's me
I'm unaware of any published data that has Actually measured such high levels at the throat etc, for Any driver/combo. If anyone knows of such, please post a link
208 dB is impossible, 194 dB is the maximum level possible in air, and far exceeds any SPL possible with compression drivers.
That said, the quad mid and high driver arrangement as used in the EV MT4 is capable of high enough SPL to cause "throat distortion" at high power levels. Although the single MF and HF horn eliminated the comb filtering associated with using eight separate horns, the air non linearity was responsible for distortion which is far more objectionable in HF devices than LF, and also caused compression at high levels, reducing headroom.
It is possible to have "too much of a good thing"
.Art Welter
Josh,
How many hours on the Gjallerhorns so far
?Looks like the voice coil former to aluminum bond would be an area of stress concentration.
Is the former glued directly to the aluminum, or is there some build out material in that area ?
Attachments
I tend to agree with you.
Even fairly modest bass horns reach up into the 165-170db range in the throat at certain frequencies and some only require 30 or 40v to do so. So the pressures may be somewhat higher here, but the forces do not appear to be out of the ordinary.