TH-18 Flat to 35hz! (Xoc1's design) - Page 69 - diyAudio
Go Back   Home > Forums > Loudspeakers > Subwoofers

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 8th August 2012, 09:26 AM   #681
NEO Dan is offline NEO Dan  United States
diyAudio Member
 
NEO Dan's Avatar
 
Join Date: Jun 2008
Location: N.E. Ohio
Quote:
Originally Posted by jbell View Post
I applaud the use of 28.3v as a reference as it removes alot of ambiguity..... however even in the latest published spec that Tom just referenced. http://www.danleysoundlabs.com/danle...spec-sheet.pdf

The Text says 28.3V at 10M -- TEF graph says 100watts at 10M...

It's that pesky confusing watts and volts in the same spec page that kinda throws folks.
Click the image to open in full size.
Hi jbell,
It indicates the 4Ω driver so that's incorrect, the last version is below. What I don't agree with so much is the TEF measurement setup, the low end of the FR in the measurements looks not like what I would expect for an upright single box 1/2 space measurement.
[sarcasm mode]Maybe the outside is also a typo, could be out beside the block building [sarcasm mode/]
Click the image to open in full size.
I always understood it to be 200w nominal, but the thing I wondered is could this be a 200w/40v sweep on the 8Ω driver, to me that would make sense as the 18SW115-8 has the better LF. DLS needs a tech in house to keep the marketing folks in check.
__________________
Regards,
Dan
  Reply With Quote
Old 8th August 2012, 10:11 AM   #682
NEO Dan is offline NEO Dan  United States
diyAudio Member
 
NEO Dan's Avatar
 
Join Date: Jun 2008
Location: N.E. Ohio
Quote:
Originally Posted by Tom Danley View Post
I challenge anyone to come up with a Hornresp design that does 105dB @ 1W, or 108dB @ 2W for the range down to 40Hz. Short of making a 4m long, 2m wide fullsize horn, I don't think it's possible. So obviously I must be missing something, because I hardly believe that DSL would be lying.
Hi Tom,
I don't think anyone wants to move around a single 2m wide box so how about we break it down into some lower cost modules...
... how about 6' wide:
Click the image to open in full size.
__________________
Regards,
Dan
  Reply With Quote
Old 8th August 2012, 10:13 AM   #683
jwmbro is offline jwmbro  United States
diyAudio Member
 
jwmbro's Avatar
 
Join Date: Nov 2007
Location: München, Bavaria, Germany
Quote:
Originally Posted by Tom Danley View Post
Hi
Actually the specs are Voltage based, not an imaginary Watt;

http://www.danleysoundlabs.com/danle...IFICATIONS.pdf

A response curve and impedance curve is also given so that one can see the sensitivity at a frequency of interest, its not just one number.

http://www.danleysoundlabs.com/danle...spec-sheet.pdf
Best,
Tom Danley

Hi Tom,

good to see you still reading here, in the years I've been on this forum, I've seen you intersperse quite a few gems of knowledge amongst your posts, and I appreciate that - thank you.

I am aware of the voltage base of your specs, and applaud that as well, though I can understand you would want to clarify that for anybody reading.

Also the decision to use 28.3V at 10m instead of 2.83V at 1m seems a reasonable idea, as it definitely puts you in the far field of the sub, and well away from any near field effects.

The reason why I myself mentioned the ambiguous wattage spec, is because you measure a "nominal 4Ω" load, or in other words a minimal impedance of 3Ω (as I can't find an impedance sweep, I'm guessing around 2.5Ω DC Re) using a 28.3V signal, which corresponds to a lot more than 100W throughout most of the bandwidth of the speaker.

This is not a problem as such, seeing as you state both impedance and voltage clearly, so one can adapt to such.

For fair comparison IMO, one should measure an 8Ω design at 40V/100m or 4V/1m, the latter of which would correspond to the hornresp input. But even accounting for this power difference due to impedance, I still can't manage to simulate these 108dB/2.83V/1m in any way. Not with any existing driver I've tried, not with changing the specs to make some kind of ideal imaginary driver.

I am kind of disappointed that so much info about the DSL designs have leaked, it seems both the driver and the fold pattern are known, I liked it better when they were just a mystery box that could do 108dB@2.83v/4Ω/1m, and there was still the notion, that by changing some parameter around, you could still tickle out an extra 3dB somewhere.

Now however, the main question for me seems to be how to explain the discrepancy between hornresp simulations and measured SPL curve.

If I can find a large enough field, I think I'll do some polar plots of my tapped horn, that might offer some insight. Might be a bit tricky in the city though, so if anybody wants to beat me to it, feel free.
__________________
Regards, James
  Reply With Quote
Old 8th August 2012, 02:52 PM   #684
diyAudio Member
 
Join Date: Apr 2005
Location: Sitting behind the 'puter screen, in Illinois, USA, planet earth
Hi mwmkravchenko
We have a new website which is still an agonizingly slow work in progress, new measurements for all the products are not done yet and for example the one for the TH-118 is not done yet, there is no impedance curve shown yet.
The TH-115 might be a better one to look at as it has both the impedance and magnitude shown and is essentially the same enclosure.

The TEF machine measurements do at times look different than sequence based measurements as it has superior noise immunity and the jagged ups and downs (measurement grass) which goes away when you average a number of measurements, isn’t present.

The custom had been to use “Watts” as the frame of reference but that was based on a nominal impedance (either 4 or 8 ohms) but in commercial sound Voltage is now the preferred term as Voltage is very easy to measure while power is not.

Pull up the data sheet for a TH-115 which has a clearer curve and has the impedance curve. When a curve says 100W, it was made with 28Vrms. This cabinet has a 4 Ohm driver but notice the impedance curve is above 5 Ohms over most of its range and only dips to 5 around 80 Hz.

For an 8 Ohm driver, the minimum Z might be as low as 5.6 - 6 Ohms but 5 ohms is too low to call it nominal 8 ohms and too high to call it 4 ohms so it’s another reason we used voltage as the reference. We measure at 10 meters and 10X the normal voltage because the enclosures take up a fair amount of space in the near field so 10 meter measurements allow a much more accurate prediction for greater distances used in commercial sound.

The measurement resolution for the TH-115 is 3.5Hz, the curve smoothing is 18% which is about 1/5 -1 /6 octave (hearing resolution mid band is one Bark wide or about 1/6 octave. If one were to take the measurement with a typical sequence based system, there would usually be a fair amount “noise features” which are not actually present and go away with a number of averages.
The woofer measurements are taken in a large flat area near the shop, the microphone (an earthworks) and HP voltmeter are used along with the TEF to take the measurements.

Jwmbro, yes there has been a great interest in “what’s inside”, a few have arrived at a close approximation what’s inside too. The number of tapped horns in the DIY area (and a few companies now) is funny considering that also a number of people have suggested it is “the same as” the Transflex (a design I had never seen or heard of) but it is clearly different.

Perhaps they have not actually read that patent as the Transflex patent calls out that the efficiency is the same as the direct radiating drive and the driver can be mounted on the cabinet exterior.
With the Tapped horn the sensitivity can be 10dB or more greater than the direct radiating case as it is a horn and the throat and mouth dimensions are part of the design. A detail like that may not matter to some but does to the patent office.

Lastly, the strength of a computer model is only as great as it’s ability to predict what you measure from the real thing.
Even Akabak does not exactly predict what you measure from a Tapped horn or bass horns unless you add some extra elements and the hf behavior is rarely right. All of our products were built going back and forth between computer models and the wooden reality, adjusting the computer model to predict what is measured, a proper measurement trumps the computer model.

If interested in what we are up to and have facebook and headphones on the computer, go to the facebook site, go to “posts by others” hit “see all” and scroll down to the video Mike posted from Michigan state University stadium. This was recorded at 700 feet from the speakers.

Danley Sound Labs, Inc. | Facebook

Best,
Tom
__________________
Bring back mst3k and futurama
  Reply With Quote
Old 9th August 2012, 10:05 AM   #685
diyAudio Member
 
Join Date: Jun 2007
Quote:
Originally Posted by jwmbro View Post
I still can't manage to simulate these 108dB/2.83V/1m in any way. Not with any existing driver I've tried, not with changing the specs to make some kind of ideal imaginary driver.
Hi jwmbro,

If it is any consolation to you, I can’t get near the 108 dB half-space sensitivity figure either - it seems that Tom still has some secrets up his sleeve :-).

It is interesting that the Klipschorn has a claimed sensitivity of 105 dB @ 1W/1M - I'm not sure how closely this equates to the TH-118 2.83V/1M result, but presumably the Klipsch bass corner-horn design would be more efficient acoustically than a tapped horn.

It would be nice to be able to see the actual specifications of the driver used in the TH-118 :-).

Kind regards,

David
__________________
www.hornresp.net
  Reply With Quote
Old 9th August 2012, 11:15 AM   #686
Mark Kravchenko
diyAudio Member
 
mwmkravchenko's Avatar
 
Join Date: Jan 2002
Location: Perth Canada
I did a considerable amount of R&D for a company modeling all of the Klipsch products. They do not meet that sensitivity figure for much of their passband. And the frequency response andimpedance models were extremely close to the prototypes.

Same can be said of tapped horns. I have modeled dozens, and built enough to greatly respect the accuracy of hornresp. It is not perfect as Tom said, not really reliable on the top end. But otherwise for low frequency reproduction it is very close to the response and the efficiency of a real product.

Looking at available information I would say that the driver inside the TH118 is probably from B&C. The highest BL version would make the most sense.
__________________
Mark
www.kravchenko-audio.com
  Reply With Quote
Old 9th August 2012, 11:22 AM   #687
djk is offline djk
diyAudio Member
 
djk's Avatar
 
Join Date: Feb 2001
Location: USA
"It is interesting that the Klipschorn has a claimed sensitivity of 105 dB @ 1W/1M"

In 0.5Pi, Tom is speaking of 2Pi.

Click the image to open in full size.

113.67dB/W/1M in 0.5Pi
__________________
Candidates for the Darwin Award should not read this author.
  Reply With Quote
Old 9th August 2012, 05:56 PM   #688
diyAudio Member
 
Join Date: Apr 2005
Location: Sitting behind the 'puter screen, in Illinois, USA, planet earth
If it is any consolation to you, I can’t get near the 108 dB half-space sensitivity figure either - it seems that Tom still has some secrets up his sleeve :-).

Hi David
Keep in mind, we do not use the 1w1m measure any longer, it is a Voltage based measurement, Voltage being vastly easier to measure than power and normally an impedance curve is also provided..

Generally, a small conventional bass horn with a low cutoff has a tilted frequency response while the Tapped horns usually don’t.
That is the only advantage to a Tapped horn and it goes away once one can make a normal horn large enough.
That being the case, the Tapped horn can be more efficient near the bottom corner of it’s response while the conventional horn more efficient up high. Also, in the days of the K horn, the driver selection approach was much less precise than say starting with Marshal Leach’s horn / driver selection criteria.
The reason the lab sub is still popular 10 + years on is because that driver was specified using the Leach approach and tweaked based on the predicted response of a finite enclosure.
One nice freedom is being able to ask a mfr to adjust a driver for ones needs.
A real problem with modeling is there are things which are not included when you look at a cabinet drawing and make a model.
For example, a flexing cone (they all do) and wood that is not infinitely ridged both add invisible elements to an acoustic model.
The response features in a prediction are also normally a higher Q than what one measures from a real speaker. I have not used your program (being so used to AKABAK) but that program while comprehensive, still needs some extra stuff to make a prediction very close to what you measure.
Also, I have seen many DIY tapped horn models where they were missing another key aspect of the operation "reactance annulling", any driver can work to make sound in a tapped horn and only the right driver and horn parameters produce the highest efficiency.
One might examine the impedance curves as well as response curves to see how close one is to a production cabinet fwiw.
Yes, the drivers we use for Tapped horns like the th-118 are from B&C.
Best
Tom Danley
__________________
Bring back mst3k and futurama
  Reply With Quote
Old 9th August 2012, 06:56 PM   #689
NEO Dan is offline NEO Dan  United States
diyAudio Member
 
NEO Dan's Avatar
 
Join Date: Jun 2008
Location: N.E. Ohio
Hi Tom,
Methinks most everybody speaks in "normal" "nominal" values here

I understand the marketability of voltage sensitivity into low "nominal" impedance loads, the problem comes with dealing with the folks who can't read between the lines.

Did anybody ever produce actual constant power measurements that you are aware of? I've never seen a spec or sell sheet done that way...

It was mentioned earlier that you had tried the Eighteen Sound, did they ever get that driver working to your satisfaction?
__________________
Regards,
Dan

Last edited by NEO Dan; 9th August 2012 at 07:00 PM.
  Reply With Quote
Old 9th August 2012, 07:43 PM   #690
diyAudio Member
 
Join Date: Apr 2005
Location: Sitting behind the 'puter screen, in Illinois, USA, planet earth
Quote:
Originally Posted by NEO Dan View Post
Hi Tom,
Methinks most everybody speaks in "normal" "nominal" values here

I understand the marketability of voltage sensitivity into low "nominal" impedance loads, the problem comes with dealing with the folks who can't read between the lines.

Hi
In the market we are in, many companies made the switch to Voltage based measurements without including an impedance curve which can be misleading. For others the idea of a single number and not response curve or impedance curve is more appealing

Did anybody ever produce actual constant power measurements that you are aware of? I've never seen a spec or sell sheet done that way...

Constant power measurements normally do not deliver anything like flat response, loudspeakers as we use them are Voltage referenced devices already, the idea of “power” delivered is a remnant of the tube days before low impedance voltage sources existed and matching the source / load impedance was the objective..


It was mentioned earlier that you had tried the Eighteen Sound, did they ever get that driver working to your satisfaction?
I really liked the parameters of the 18sound woofer in question but they had a problem at high powers the edge suspension / cone attachment would fail too often. They may have fixed (they were working on it) that but we did not go back to using that one. They do make some very nice drivers though.
Best,
Tom
__________________
Bring back mst3k and futurama
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
18 inch labhorn design. mulletdude Subwoofers 23 27th November 2013 09:43 PM
Ciare 18 bass horn design soundhead Subwoofers 5 22nd September 2013 09:33 AM
2 18 bass design simpleacoustic Subwoofers 101 15th May 2011 10:22 AM
I want my Avalanche 18 to go flat to 15hz thadman Subwoofers 10 11th November 2006 03:54 AM
Looking for suggestions on a 1300W Dual 18 sub design. mjcohen Subwoofers 5 21st December 2004 08:38 AM


New To Site? Need Help?

All times are GMT. The time now is 01:06 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2