or Connect
AVS › AVS Forum › Audio › DIY Speakers and Subs › Data-Bass version 2 is live
New Posts  All Forums:Forum Nav:

Data-Bass version 2 is live - Page 5

post #121 of 244
Special request! biggrin.gif

Do you think it would be possible to retest the XXX18 in the original sealed enclosure. I thought it would be cool since you say you will revisit the 5400.

If so, the special request....haha, aren't I awful? ....is if you could drag out one of your new dual-opposed XXX18 system for a test. As a system, I think it could have the overall best SPL/extension even compared to the LLT.

Just a thought. smile.gif
post #122 of 244
Thread Starter 
Scott it's on the to do list. I probably won't do the dual driver sub cause we can extrapolate off of the single. I expect the response to be a lot more peaked in the small cab. I will be testing it in the 4.5 ft cab for 21's and RE's oversized 18's. the rest of the 18's get a 4.2 cube cab. Close enough for govt work. I only have 3 of the original 11 tests left to redo with cea2010 and more powa.
post #123 of 244
Hell yeah!

POWAAAHHH!!!

14347.jpg
post #124 of 244
Thread Starter 
And continuing with the blitkrieg of new test data....Today I put up the re-test of the LMS Ultra in a sealed enclosure just in case anyone forgot just how good these drivers are and why they ended up being so popular. Bon appetite.
post #125 of 244
Yay! smile.gif

Damn this driver is awesome. Now if only the population could figure out how to use the damn thing without breaking it. rolleyes.gifwink.giftongue.gif

I think people see, "Oh!? 4,000w power handling eh? PPOOOOWWWAAAHHHH". Like Jeremy Clarkson up above. tongue.gif
Edited by Scott Simonian - 10/19/12 at 2:31pm
post #126 of 244
"Damn this driver is awesome."

if you have a lot more space, w15gti's are going for $300 delivered on ebay. same excursion limited bottom end spl (maybe a hair more) while providing 3-4db greater sensitivity across the whole range. power spread out across 3 drivers instead of one. good inductance control, but not as much motor per driver. good distortion numbers. about the worst looking dustcap that i've ever seen though (but it can be painted black). dynamic braking makes them more durable/harder to destroy? three wired in series parallel net 4 ohms. what do you think scott? fair fight?
post #127 of 244
Quote:
Originally Posted by LTD02 View Post

"Damn this driver is awesome."
if you have a lot more space, w15gti's are going for $300 delivered on ebay. same excursion limited bottom end spl (maybe a hair more) while providing 3-4db greater sensitivity across the whole range. power spread out across 3 drivers instead of one. good inductance control, but not as much motor per driver. good distortion numbers. about the worst looking dustcap that i've ever seen though (but it can be painted black). dynamic braking makes them more durable/harder to destroy? three wired in series parallel net 4 ohms. what do you think scott? fair fight?

My models show that the 3x w15gti system would have 2dB more displacement capable output down low throughout. So... fair fight? Absolutely. Best choice for everyone. Probably not.

Btw, nice find on those for $300. They are usually at minimum $400 a piece. As of right now, I only saw six going for $300 and all others at least $100 more. A good deal for a limited time. It's too bad they aren't cheaper usually.

Speaking of... what a great driver to have Josh test out. tongue.gif
post #128 of 244
"Btw, nice find on those for $300."

the guy with them has great feedback and he seems to have them in stock from time to time. kind of hit and miss.
post #129 of 244
That's exactly the kind of system I have. Three W15GTI's wired in series parallel for a 4ohm load to my Funky Waves 2,4000W RMS plate amp. That's 800W per driver in one dual opposed 4.8cf sealed box and one 2.4cf sealed box. I realize the box size is smaller than the recomended size for these drivers but I also have them fully stuffed. I've had them for a while now and they're are as good as advertised and I've used a lot of 15" drivers. Drivers such as CCS SDX, LMS Ultra 5400 and AE AV15X, in singles though. I first used the dual opposed sub and then a couple of months ago added the extra driver(sub). I know, I need to start a review of them like I did the other subs I used.
post #130 of 244
Perhaps I can send one to Ricci later on to test and see how it performs.
post #131 of 244
You didn't want to go for the triple opposed? If you use a triangular driver arrangement you can still get an inert and balanced "box". smile.gif
post #132 of 244
Like the Paradigm Sub 2? I thought about it but ended up with the two subs with different placement for better in room response.
post #133 of 244
"I've had them for a while now and they're are as good as advertised and I've used a lot of 15" drivers."

cool. any downside other than not having six of them? :-)
post #134 of 244
Yes, they're only rated to 20mm so you might need several of them like I did.
post #135 of 244
I have used a few 15s , but my fav is w15gti.
post #136 of 244
Love the JBL's motor design.

Wouldn't it be nice to see them pursue a new all out effort, toward a high excursion 18, employing all the goodness found across their existing lines?


Thanks for the update Josh, solid work
post #137 of 244
Quote:
Originally Posted by Ricci View Post

And continuing with the blitkrieg of new test data....Today I put up the re-test of the LMS Ultra in a sealed enclosure just in case anyone forgot just how good these drivers are and why they ended up being so popular. Bon appetite.

I'm surprised there are virtually no comments on how much higher your results are vs Ilkka's. I know they're years apart, yadda, yadda, but they really aren't even close, more than 2 to 1, actually. THD skyrocketed with Ilk's 118dB sweep, you're at 25% with a 120dB sweep.

Is the driver somehow vastly new and improved, or is it the diff in amplification? BTW, how'd you get a sustained 130V out of that amp?

Also, the compression results are a bit odd to me. They show compression below 20 Hz even at the 95dB sweep.
post #138 of 244
Thread Starter 
The k10 amps are about 95% efficient and on a 50a 240v line. 12000w vs 3600 and more efficient too .

The only time it has clipped during the sweeps was on the dual 21 and that was well over 200v . Shoulda been 270 or so without clipping.

They did improve the clearances on the driver and made some other changes to the suspension. As far as the thd differences who knows. Different mics, gear, amp, changes to the driver, etc. Maybe he had some amp clipping? Dont discount environmental noise either down below 20hz where the output isnt a lot it is much closer to the background noise. Also his distortion results are smoothed. I stopped the sweeps with the lms because it was starting to make the tap sound of the cone on the spider landing. Plus the top end was compressing 3db.

You will see a lot of weirdness on the low end of the compression results because of the background noise interfering when the level is only 60-70db during the 90db sweep. I do averaging of about 6 measurements at 90 and 95 db usually to try and mitigate this.
post #139 of 244
Quote:
Originally Posted by Ricci View Post

The k10 amps are about 95% efficient and on a 50a 240v line. 12000w vs 3600 and more efficient too .
The only time it has clipped during the sweeps was on the dual 21 and that was well over 200v . Shoulda been 270 or so without clipping.
They did improve the clearances on the driver and made some other changes to the suspension. As far as the thd differences who knows. Different mics, gear, amp, changes to the driver, etc. Maybe he had some amp clipping? Dont discount environmental noise either down below 20hz where the output isnt a lot it is much closer to the background noise. Also his distortion results are smoothed. I stopped the sweeps with the lms because it was starting to make the tap sound of the cone on the spider landing. Plus the top end was compressing 3db.
You will see a lot of weirdness on the low end of the compression results because of the background noise interfering when the level is only 60-70db during the 90db sweep. I do averaging of about 6 measurements at 90 and 95 db usually to try and mitigate this.

Thanks.

After thinking about it, it's really a no-brainer that the power plant is the #1 suspect (and it always is with subwoofers). No comparison between the Crown and the K-10, especially below 40 Hz.

What caused me to look closer was not's comments about the asymmetry in the throw of his LMS drivers, which would certainly add THD and not the opposite. I've come to trust his observations and really didn't expect to see such a jump in performance regarding the sine sweeps/swept THD

The differences in hardware are insignificant to me. A dB or even 2 and I would chalk it up to that and environment, but the differences are too significant for that to be the answer in this case, IMO.
post #140 of 244
Quote:
Originally Posted by LTD02 View Post

"Damn this driver is awesome."
if you have a lot more space, w15gti's are going for $300 delivered on ebay. same excursion limited bottom end spl (maybe a hair more) while providing 3-4db greater sensitivity across the whole range. power spread out across 3 drivers instead of one. good inductance control, but not as much motor per driver. good distortion numbers. about the worst looking dustcap that i've ever seen though (but it can be painted black). dynamic braking makes them more durable/harder to destroy? three wired in series parallel net 4 ohms. what do you think scott? fair fight?

I searched all morning in my computer for the Chinese 15s I tried about 8 or 9 years ago. They had the 2-VC config and looked very spiffy (which is why I was trying to find the pics). Pre-dated DD for sure.

What inductance controls are in the W15GTi? I never noticed anything in their section drawings, but that doesn't mean the drawing were accurate. Just curious.
post #141 of 244
"What inductance controls are in the W15GTi? I never noticed anything in their section drawings, but that doesn't mean the drawing were accurate. Just curious."

it is actually a pretty creative approach that jbl licensed for a while, but then went ahead and either bought the company or the patent rights.

jbl calls it differential drive.

they have a white paper that explains how it works:
http://www.jblpro.com/pub/technote/JBL_TN%201-33%20rev3.pdf

essentially, there are two coils and two gaps operating with reversed polarity inside the driver, which lowers and linearizes inductance significantly.

edit: addition. i think that neodymium slug sitting where you'd normally have a pole of iron helps as well.
Edited by LTD02 - 10/20/12 at 3:53pm
post #142 of 244
Thread Starter 
Dave I'm not sure what differences you refer to?

The extra output in the burst testing up top is obv just extra amp power. I see about 2dB difference at 16hz which is the largest difference in cea2010. 1 db there could easily be calibration differences. If you compare the sweep measurement to the burst results I actually drove it harder in the deep bass than with cea2010 and I could hear a slight tapping which is when you know the ultra is touching the spider landing. He did a 118dB sweep and I did a 120 both referenced at 50hz. My box is about 115L so slightly larger. The main difference I see is the distortion result for sure where his 118db sweep shot way up where the 120 in mine shows higher distortion above 30hz probably due to extra power mostly as expected, but only about 30% in the deep bass. His is off the chart so hard to tell exactly how much. I do know that the result in mine is what was measured. The sweep for the distortion is a long duration sweep also but the response range measured is up to about 4 k which is the lowest ARTA will allow so it passes more quickly through the bass range. It is possible that he may have had some amp clipping too, but I just don't know. He was using TrueRTA for the distortion or?

The THD for the LMS is pretty much all 3rd order and part of the reason it is a lot lower than others is the absence of the other harmonics at high levels. His sure seems to show that the driver was done on the 118 sweep and I know for sure that it had nothing left in the tank below 30hz in my 120 but the THD does not dramatically jump. Other than the tapping when the cone contacts the sound was not heavily distorted like I hear with other drivers. It never really runs out of Xmax until the suspension runs out and the cone bottoms. I would expect a larger enclosure to have better distortion characteristics but it is only 15L diff. Maybe I will rerun that test again but I don't expect to see much difference. I'm not really sure how much effect the amp clipping IF it was clipping would increase the distortion.
post #143 of 244
Quote:
Originally Posted by Ricci View Post

Dave I'm not sure what differences you refer to?
The extra output in the burst testing up top is obv just extra amp power. I see about 2dB difference at 16hz which is the largest difference in cea2010. 1 db there could easily be calibration differences. If you compare the sweep measurement to the burst results I actually drove it harder in the deep bass than with cea2010 and I could hear a slight tapping which is when you know the ultra is touching the spider landing. He did a 118dB sweep and I did a 120 both referenced at 50hz. My box is about 115L so slightly larger. The main difference I see is the distortion result for sure where his 118db sweep shot way up where the 120 in mine shows higher distortion above 30hz probably due to extra power mostly as expected, but only about 30% in the deep bass. His is off the chart so hard to tell exactly how much. I do know that the result in mine is what was measured. The sweep for the distortion is a long duration sweep also but the response range measured is up to about 4 k which is the lowest ARTA will allow so it passes more quickly through the bass range. It is possible that he may have had some amp clipping too, but I just don't know. He was using TrueRTA for the distortion or?
The THD for the LMS is pretty much all 3rd order and part of the reason it is a lot lower than others is the absence of the other harmonics at high levels. His sure seems to show that the driver was done on the 118 sweep and I know for sure that it had nothing left in the tank below 30hz in my 120 but the THD does not dramatically jump. Other than the tapping when the cone contacts the sound was not heavily distorted like I hear with other drivers. It never really runs out of Xmax until the suspension runs out and the cone bottoms. I would expect a larger enclosure to have better distortion characteristics but it is only 15L diff. Maybe I will rerun that test again but I don't expect to see much difference. I'm not really sure how much effect the amp clipping IF it was clipping would increase the distortion.

Exactly, you get +3dB more output with very low THD while he shows THD running off the scale to an unknown amount (which actually has always peeved me, I wish you guys would expand the scale to show the actual THD sweep if you're gonna push the sub that hard, show the result)

lmsilkvricci_zpsc3205f3c.jpg

No, TrueRTA doesn't have a THD sweep function. I don't know what program he used. I always found that result to be curious. I really think it's the difference in amplification, but wanted to get your thoughts (and other's, if anyone had any thoughts),
post #144 of 244
Quote:
Originally Posted by LTD02 View Post

"What inductance controls are in the W15GTi? I never noticed anything in their section drawings, but that doesn't mean the drawing were accurate. Just curious."
it is actually a pretty creative approach that jbl licensed for a while, but then went ahead and either bought the company or the patent rights.
jbl calls it differential drive.
they have a white paper that explains how it works:
http://www.jblpro.com/pub/technote/JBL_TN%201-33%20rev3.pdf
essentially, there are two coils and two gaps operating with reversed polarity inside the driver, which lowers and linearizes inductance significantly.
edit: addition. i think that neodymium slug sitting where you'd normally have a pole of iron helps as well.

Yes, I've read this 'paper' before. You know me and JBL papers. They get 100% 3HD at 20 Hz with 1/3 rated power, so they just don't mention it and instead focus on some better performing result. Okey-Dokey then.

They don't show any test results for how 2 coils 'linearizes' inductance. They just say it does. Actually, they seem to be attributing most of the reduction in flux modulation to the Neo magnets, showing test results.

I always feel like I have to take a shower after reading a JBL 'paper'. tongue.gif
post #145 of 244
"Yes, I've read this 'paper' before. You know me and JBL papers. They get 100% 3HD at 20 Hz with 1/3 rated power, so they just don't mention it and instead focus on some better performing result. Okey-Dokey then."

the drivers mentioned in the paper are the short throw, high sensitivity mid-basses that are run in ported enclosures.

the w15gti is more of a subwoofer. distortion on it looks pretty good.

this one is at 35volts, which is around 400 watts.



as for whether or not the coils do anything, i suppose that it could just be the neodymium slug, but i'm not sure why they would have bought the company or the patent on the differential drive technology if that were the case.

the frequency response of the driver shows a characteristically smooth response through the bass and no top end rolloff even out past 1khz. that is an indication of good inductance control, so is the extremely low mid-bass distortion. distortion is down 50db at 100hz at 115db spl. that is only 0.3%. that is about as good as i've ever seen.

the normalized inductance (le/re) is 0.36, which is pretty low and a further indication that something is reducing inductance in this driver.

i know that you don't like jbl and that's cool.
post #146 of 244
Quote:
Originally Posted by Scott Simonian View Post

Yay! smile.gif
Damn this driver is awesome. Now if only the population could figure out how to use the damn thing without breaking it. rolleyes.gifwink.giftongue.gif
I think people see, "Oh!? 4,000w power handling eh? PPOOOOWWWAAAHHHH". Like Jeremy Clarkson up above. tongue.gif
Simon, in my case I did not realise that you could fry a driver with too much power. Every sub I have ever owned (mass produced) would always just distort baddly and nothing else. I just new that the driver could not handle any more volume. So when I decided to go DIY, I did not know any better and I think that could be the case for other people too. I mean, I did not even know about free air testing until yesterday.redface.gif

One good thing came out of me frying the driver though, and that was adding 4 RE XXX 18's. If I did not fry the driver I would still be with a lowely 4 LMS ULTRA's! tongue.gif
post #147 of 244
Quote:
Originally Posted by LTD02 View Post

"Yes, I've read this 'paper' before. You know me and JBL papers. They get 100% 3HD at 20 Hz with 1/3 rated power, so they just don't mention it and instead focus on some better performing result. Okey-Dokey then."
the drivers mentioned in the paper are the short throw, high sensitivity mid-basses that are run in ported enclosures.
the w15gti is more of a subwoofer. distortion on it looks pretty good.
this one is at 35volts, which is around 400 watts.

as for whether or not the coils do anything, i suppose that it could just be the neodymium slug, but i'm not sure why they would have bought the company or the patent on the differential drive technology if that were the case.
the frequency response of the driver shows a characteristically smooth response through the bass and no top end rolloff even out past 1khz. that is an indication of good inductance control, so is the extremely low mid-bass distortion. distortion is down 50db at 100hz at 115db spl. that is only 0.3%. that is about as good as i've ever seen.
the normalized inductance (le/re) is 0.36, which is pretty low and a further indication that something is reducing inductance in this driver.
i know that you don't like jbl and that's cool.

First, any data above 80 Hz is irrelevant. The input signal will be down -15dB at 100 Hz and the driver will never see signal beyond that in its lifetime. That means that virtually every driver Josh will ever test will have inaudibly low THD at 100 Hz. If 100-1000 Hz is what you want to discuss, this is the wrong venue and 'subwoofer' would be a misnomer.

Take the lowly SSD-18 for around the same $$. At 20 Hz, your graph indicates (the data is not shown) at least 30% THD with 35V giving 94dB output (normalized to 2M). The SSD matches the 94dB output with 23V and 17% THD (the data IS shown). With similar THD at 20 Hz (28%) and 42V input, the SSD gives 99dB at 20 Hz. That's 4 times where it counts (20 Hz) with far less THD (we can only imagine the THD at 20 Hz with higher voltage input into the GTI).

And, I trust Josh's data. If you trust the JBL data, good enough, but I'll lay odds and money on the table that Josh would get a different result. No, you have it completely wrong, I have always been a fan of Jimmy B Lansing's products and their ascendants. It's the modern specsmanship I detest. Pro-sounders are notoriously blatant liars at worst and misleading at best. Harmon spews the most baloney and gets the highest praise from its fawning fans. Just a pet peeve of mine. Nothing to do with their overpriced products.
post #148 of 244
Thread Starter 
Quote:
Originally Posted by bossobass View Post

Exactly, you get +3dB more output with very low THD while he shows THD running off the scale to an unknown amount (which actually has always peeved me, I wish you guys would expand the scale to show the actual THD sweep if you're gonna push the sub that hard, show the result)
lmsilkvricci_zpsc3205f3c.jpg
No, TrueRTA doesn't have a THD sweep function. I don't know what program he used. I always found that result to be curious. I really think it's the difference in amplification, but wanted to get your thoughts (and other's, if anyone had any thoughts),

Hmmm. I have no idea what program he was using for THD then. Actually the way that the THD is shaped in his 118dB sweep is odd. After looking at this stuff so much there are a lot of common themes and interesting details that start to appear. With sealed subs measured at high volume there is almost always a notch of lower distortion levels around the impedance maxima. You can see this at about 33Hz in my measurements for the Ultra. Actually these show up for most types of subs not just sealed, anyway the effect typically gets more exaggerated the harder you drive it. The profile of the THD in the 118dB sweep is more L shaped and I don't see any evidence of this notch. Also the transition frequency or corner in the L is much higher and closer to 45Hz.

My money is on the amp situation.





Let's talk about the distortion graphs. I originally was going to try a 0-100% scale but to me is obscured too much in the 0-20% range where most of the time the information will be above 25Hz. I switched to a 30% upper limit. The distortion by harmonic chart does go up to 100% but is logarithmic and harder to read really where the difference between 0.1 and 10% takes up as much room as from 10-100% on the graph. I thought about switching this to a linear chart originally.

Let me get some votes on this. Here is 0-30% scale on the THD chart versus 0-100% scale. Which is more useful? Perhaps something in the middle should be used like 50%? I can change this without too much trouble.



100% scale






I need opinions on this too Should THD by harmonic be using a log scale or linear? Anything else that should be changed?




Linear scale

post #149 of 244
Thread Starter 
I like JBL and the DD motor system does seem to work well. There are lengthy patent documents with example data available on it. I wish they would do a little more with it. They do have the 2269H driver which is a pro 18" with a 4" DD motor and 19mm xmax but it's $1K and only available as a replacement part.
post #150 of 244
Hey, Josh. I kind of think it would be cool to have both graphs for the THD. You could get a micro and macro look at the whole profile at the same time. I think that would be cool.

The THD by component works best as it is, imo.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: DIY Speakers and Subs
AVS › AVS Forum › Audio › DIY Speakers and Subs › Data-Bass version 2 is live