Beyond the Ariel - Page 386 - diyAudio
Go Back   Home > Forums > Loudspeakers > Multi-Way

Multi-Way Conventional loudspeakers with crossovers

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 1st June 2008, 04:25 PM   #3851
mige0 is offline mige0  Austria
diyAudio Member
 
mige0's Avatar
 
Join Date: Jan 2005
Location: Austria, at a beautiful place right in the heart of the Alps.
Default Re: Re: Re: Re: Re: Sensitivity and Efficiency

Quote:
Originally posted by nickmckinney

IMHO - I don't know if a 10Khz test is proper against other equivalent drivers because of the vastly different impedance from inductance. How about dumping some straight DC also and maybe forcing the cone not to move? That way the coil can sit centered stationary and we can see how/if the heat is removed, that is close to the movement expected from a woofer at 10Khz (read near none)
The benefits of doing AC rather than DC is that you can use a simple soundcard with its very good resolution by simply replacing the mic by the shunt – symmetric inputs assumed.

We would like to inject a well defined power for a well defined time– so you will have to calculate the output voltage by measuring the speakers impedance - at ambient temperature - at 10kHz / 100Hz first.

No cabinet, no baffle, speaker positioned vertical in a room.
Doors and windows closed


Quote:
Originally posted by nickmckinney
So what driver(s) do you propose they go up against?
Any – its not just to do a fun competition (burning down VC's would serve that better) – its for having data available to compare.
I would hold against a two decade old 12" Dynaudio 30W54 having an outer magnet and 54mm VC and a Dynaudio 17W75 having an inner magnet and a 75mm VC.
Both speakers are well documented and represent the top engineering at their time.

Lets see how much progress was made in the meantime.



Quote:
Originally posted by Sheldon

Two thousand times, is "at least twice", no? .

Certainly.



Greetings Michael
  Reply With Quote
Old 1st June 2008, 09:07 PM   #3852
mige0 is offline mige0  Austria
diyAudio Member
 
mige0's Avatar
 
Join Date: Jan 2005
Location: Austria, at a beautiful place right in the heart of the Alps.
Quote:
Originally posted by Sheldon

The sound levels you are talking about here seem unrealistic (certainly unhealthy) for a home audio application with high efficiency prosound sized drivers. I listen pretty loud at times....
Yes, I am aware of this.

BUT
at one hand you have only 115dB SPL for very short 10sec - a 95dB sensitive speaker assumed – which is well within what Lynn wants to achieve as peaks
at the other hand – its a method of MEASUREMENT I propose – intended to serve for as many as possible speakers (home audio tweeters excluded) - and its way easier to implement 100W/1W than to propose 33,7W/0,13423562W for same strange reasons.


In addition 100W should be - for ANY real world speaker well above what is "allowed" to reach thermal equilibrium after 10sec and 1W should be low enough to allow cooling down after the 1kWs heat injection and also allow for comfortable and easy measuring at the same time.



Greetings
Michael
  Reply With Quote
Old 2nd June 2008, 12:54 AM   #3853
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Default Some Photos


Here's what the front of two XT1464's look like. A curvy shiny black horn is not the easiest thing to photograph - so for the photographers amongst us, the camera was a Pentax K10D, exposure set about 1 and 1/3 stops above the indicated meter reading, light from the window on the right with on-camera flash assist (mixed lighting), and a little bit of curve-bending in Photoshop to bring out the shadows.

The CD shown for scale is one of my favorites, the Mercury Living Presence recording of "Pictures at an Exhibition", with Antal Dorati conducting the Minneapolis Symphony Orchestra. The catalog number is Philips #434 346-2.

This recording is nearly unplayable on most high-end systems, but if the system is really on-song, the sheer beauty and energy of the performance is overwhelming. (The folks at the Audio Kinesis room at the last RMAF show will probably remember when I put this one on.)

On most high-end systems, the tremendous amount of HF energy will slew the electronics and overload the tweeters - it'll sound like a sawmill starting up in the morning thanks to all the distortion - which is NOT on the recording. If the system has very fast electronics and low-distortion tweeters with a well-designed crossover, you''ll be treated to a 10th-row in-the-hall experience - with an edge-of-the-seat performance by the Minneapolis Symphony.
Attached Images
File Type: jpg xt1464.jpg (80.0 KB, 1291 views)
  Reply With Quote
Old 2nd June 2008, 01:19 AM   #3854
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Default 12NDA520 Photo


Here's the back side of the 18Sound 12NDA520. The black object on the lower right is part of a low-diffraction mike stand system I use, discussed at more length in the "Geddes et al Measuring Loudspeakers" thread.
Attached Images
File Type: jpg 12nda520_&_mikestand.jpg (74.7 KB, 1224 views)
  Reply With Quote
Old 2nd June 2008, 01:57 AM   #3855
Salas is offline Salas  Greece
diyAudio Chief Moderator
 
Salas's Avatar
 
Join Date: Oct 2002
Location: Athens-Greece
The 18 sound products have been maturing a lot lately...I hope that they will not give you much of a hard time given a good lucky throat match with your Comp Dr.

I think it was Berlioz Symphony Fantastique on Mercury that I listened to, on an XT120 driven by HD1040 2 years ago, flanked by PHL 8 inch. There was a Karlson aperture PHL 15 inch sub too.
When those Mercurys get busy around 8kHz, they really do get busy! At a point it verged to shrill. But it was just a test system with powerful solid state.
  Reply With Quote
Old 2nd June 2008, 02:36 AM   #3856
Salas is offline Salas  Greece
diyAudio Chief Moderator
 
Salas's Avatar
 
Join Date: Oct 2002
Location: Athens-Greece
Default Re: Some Photos

Quote:
Originally posted by Lynn Olson
This recording is nearly unplayable on most high-end systems, but if the system is really on-song, the sheer beauty and energy of the performance is overwhelming. (The folks at the Audio Kinesis room at the last RMAF show will probably remember when I put this one on.)

On most high-end systems, the tremendous amount of HF energy will slew the electronics and overload the tweeters - it'll sound like a sawmill starting up in the morning thanks to all the distortion - which is NOT on the recording. If the system has very fast electronics and low-distortion tweeters with a well-designed crossover, you''ll be treated to a 10th-row in-the-hall experience - with an edge-of-the-seat performance by the Minneapolis Symphony.

I have another recording with really strong HF content that when it is being replayed right is captivating to recommend. This one Lynn I think that you will like to use. Great on big ribbon tweeters. There is really much information needing to come out.

It is Harmonia Mundi La Fete de L' ane
  Reply With Quote
Old 2nd June 2008, 02:36 AM   #3857
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Matt Kamna and I measured the output of a Burr-Brown PCM-63K DAC a few years ago. We were curious just how fast the state transitions really were, so we used a 1500V/uSec digital-video buffer following a 100-ohm passive I/V converter, to make sure there was no slewing in the analog stages of the DAC. The test track on the CD was 20 kHz at full modulation, so there were lots of +1 to -1 transitions to fill up the spectrum.

Matt owned (and probably still has) a Hewlett-Packard RF spectrum analyzer with 80 dB of on-screen dynamic range and flat response to 100 MHz. We connected the output of the video buffers straight to the input of the spectrum analyzer - they are designed to drive coaxial cable into 75 ohms, so no problem there.

What astonished both of us was seeing a flat comb spectra extending out to 20 MHz! After reaching 20 MHz, it gradually rolled off until it descended into the noise at 50 MHz. We did some quick back-of-the-envelope calculations and found that Red Book CD's required at least 1000V/uSec in order not to slew the analog electronics following the DAC.

Now, this test was done with a traditional ladder DAC with the resistor array directly connected to the current-mode output of the DAC. Most modern DACs, and certainly all delta-sigma hybrids, have built-in opamps with slewing rates 30 to 75X slower than the 1000V/uSec figure we calculated was the minimum required - so naturally, if you measured a modern DAC, you'd never see that 20 MHz comb spectra, since the built-in DAC would slew and limit it!

It must be kept in mind that slewing and lowpass filtering are not at all the same. One represents 100% distortion for a very brief interval, and the other is a linear process that simply lowpass filters the signal. Accurate reconstruction of the original analog input signal requires accurate, undistorted low-pass filtering in the reconstruction stage. This is why passive filtering sidesteps the slewing issue, and restricts the amount of RFI entering the analog electronics - audio electronics simply do not handle MHz signals gracefully - what you get is are broadband IM products that fill the entire spectrum and effectively raise the noise floor with program-induced garbage.

After this experience we started auditioning all of our "tough dog" CD's that we thought were hopelessly distorted (this was before the current craze for overmodulated CD's). Well guess what, they weren't distorted after all! Just bright and over-equalized, but not actually distorted. If you replaced the ultrafast buffers with generic 13V/uSec opamps, the screaming distortion returned. (Audiophile-approved CD's tended to sound the same - they didn't stress the analog electronics as hard.)

I was especially disgusted to learn the $5000 Sony SACD-1 used - you guessed it - ancient Signetics 5534 (13V/Usec) opamps throughout all the analog circuitry. And there's no question that SACD/DSD has RF content going well beyond 20 MHz. It is painfully obvious that the people who design these products do not have RF spectrum analyzers, do not have RFI engineering experience, and don't seem to understand how undesirable slewing distortion is, and how easily digital state transitions induce it.

(Note: the output of the PCM-63 that we measured was already deglitched - we confirmed that with a 500 MHz Tektronix scope. It was the state transitions that we were seeing on the RF spectrum analyzer.)

These slewing problems are routinely ignored by the RF-unaware audio design community. They can't measure it, aren't aware of what it would look like, and have no RF design experience, so they don't care. Out of sight, out of mind, ignorance is bliss.

One of the nice things about working with Matt Kamna, Gary Pimm, and other Tek and HP alumni was that these guys understood RF and how to deal with it. Unfortunately, most other audio engineers are in a state of denial, or more frankly, ignorance when it comes to dealing with signal content above 1 MHz.

The other common problem are tweeters with lots of IM distortion thanks to poor out-of-band rejection by the crossover. The fact is that tweeters just don't perform very well at 500 Hz - but they can make plenty of IM distortion if any significant lower-mid content arrives at the speaker terminals. The rule of thumb with all drivers is to limit out-of-band energy, most importantly at the low end of the band, where it can induce a lot of IM distortion at higher frequencies.

Combine massive slewing distortion with tweeters with lots of IM distortion, and many large-scale symphonic and choral works turn into a screaming mess. Hifi exhibitors get around the problem by never playing that kind of program material at hifi shows - instead, you get "minimalist" audiophile-approved rock, jazz, and blues recordings, and they all have one thing in common: a sparse HF spectra, with the instrumentalists thoughtfully taking turns. Even MP3 compression has no problems when the spectra is so simple.
  Reply With Quote
Old 2nd June 2008, 05:17 AM   #3858
ScottG is offline ScottG  United States
diyAudio Member
 
Join Date: Feb 2003
Location: US
Default Re: Some Photos

Quote:
Originally posted by Lynn Olson

Here's what the front of two XT1464's look like.
When you play with them, get them optimized, and then decide that it isn't what you are looking for:

Then try them as a dipole (i.e. two drivers w/horns back to back and out of phase), and repeat the process. That might get you to where you want to go.
__________________
perspective is everything
  Reply With Quote
Old 2nd June 2008, 02:04 PM   #3859
Banned
 
Join Date: Feb 2004
Location: Portal 2012
Quote:
Originally posted by Lynn Olson
Matt Kamna and I measured the output of a Burr-Brown PCM-63K DAC a few years ago. We were curious just how fast the state transitions really were, so we used a 1500V/uSec digital-video buffer following a 100-ohm passive I/V converter, to make sure there was no slewing in the analog stages of the DAC. The test track on the CD was 20 kHz at full modulation, so there were lots of +1 to -1 transitions to fill up the spectrum.

Matt owned (and probably still has) a Hewlett-Packard RF spectrum analyzer with 80 dB of on-screen dynamic range and flat response to 100 MHz. We connected the output of the video buffers straight to the input of the spectrum analyzer - they are designed to drive coaxial cable into 75 ohms, so no problem there.

What astonished both of us was seeing a flat comb spectra extending out to 20 MHz! After reaching 20 MHz, it gradually rolled off until it descended into the noise at 50 MHz. We did some quick back-of-the-envelope calculations and found that Red Book CD's required at least 1000V/uSec in order not to slew the analog electronics following the DAC.
So Lynn, which A/D's have this slew rate?
  Reply With Quote
Old 2nd June 2008, 04:10 PM   #3860
mige0 is offline mige0  Austria
diyAudio Member
 
mige0's Avatar
 
Join Date: Jan 2005
Location: Austria, at a beautiful place right in the heart of the Alps.
A fulls scale DA out 1/0/1/0 is suposed to have a large spectrum.


http://www.kettering.edu/~drussell/D...r/Fourier.html


To bring this down is what the reconstruction filter is good for (if it works ).
One of the reasons the Behringer has (pretty much) room for improvement


Swiching power amps have some HF left too - with the same ignorant argument that its way beyond our hearing range - not considering the fold back into audio band.

Look at the pretty well regarded Hypex modules for example - other brands may be nothing better.


Greetings
Michael
  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



New To Site? Need Help?

All times are GMT. The time now is 12:03 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