Beyond the Ariel - Page 143 - 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 July 2007, 03:08 AM   #1421
Banned
 
Join Date: Feb 2004
Location: Portal 2012
Quote:
Originally posted by Ray Collins
Anybody cutting wood?!

Ray

Always!!!!!!!!!
  Reply With Quote
Old 1st July 2007, 03:54 AM   #1422
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Quote:
Originally posted by Norris Wilson
Possibly this thread has strayed away from its original direction.

I have not heard any more from Lynn Olson since this diffraction discussion began.

Maybe a new thread about diffration and its effects on an OB should be started.

I am interested in the latest OB design information that Lynn has
thought of.

Anything new Lynn?
Hi all, I've been wrangling with Apple about getting them to kick out the replacement computer. Since I have a visit to the Physical Trainer on July 5th, what I don't want to happen is have the long-awaited replacement arrive when I'm at Kaiser Hospital - seeing a little note from FedEx attached to the front door would create a screech that BudP would hear in Seattle!

The diffraction thread has certainly picked up an astounding momentum. The use of sims instead of measurements speaks volumes - I'm looking forward to finally using both MLSSA and SoundEasy to answer this math-heavy guesswork with real data. Let's see, at last count, we have:

1) No diffraction on a fully symmetric dipole

2) Twice as much as a monopole

3) Lossy-mesh dissipation to spread out the time artifacts

4) Dimples as on a golf ball

5) Painted-on EnABL pattern

6) Fur! Furry cones!! Furry patented cones!!!

And that's just for starters. Sheesh.

I've been poking around the Meyer Sound Labs site. Most interesting. In addition to a good discussion that de-mystifies compression drivers, I found their MTS-4A system very interesting. It uses a cascade of 12, 15, and 18-inch drivers, all operating together at the lowest frequencies. Hmm, that seems familiar, guess John Meyer got there first.

I still have serious reservations about the idea I had a while back using a pair of 8-inch drivers combined with the new RAAL ribbon. It's one thing to tolerate beaming at the highest frequencies, it's quite another in the midband - and that's what stacked 8" drivers imply. All of my instincts (yes, ole woo-woo the wizard here) tell me to stay with a single 12" driver, and do my level best to get it to combine with horn+compression driver or the new RAAL ribbon.

The recent GedLee discussion has made me aware of just how serious horn coloration really is. I guess I'll just have to shell out the money and measure both CD/horn and RAAL ribbon drivers. None of this hokey 1/3 octave smoothing, though.

Now that I've been tipped off what to look for, I am going to very carefully examine the time vs off-axis angle, in small increments. I'm going to set MLSSA or another system on fast refresh, move the microphone slowly, and look for artifacts that move around in the time domain. That would be a direct consequence of a rough wavefront emerging from the horn mouth - and is usually concealed by conversion to freq domain and 1/3 octave smoothing. These artifacts are in the time domain, and that's the place to look for them.

I'm not looking for 60, 90, or 120 degree nominal beamwidth, but for fine-grained artifacts only a few degrees wide. I've seen this kind of stuff before, when a hard dustcap radiates pencil-beam artifacts that are very narrow, or when a cone is deep into the breakup region. I take narrow-beam artifacts very seriously - in a cone driver, it's a cause for rejecting it from consideration, no matter how steep the crossover filter is going to be.
  Reply With Quote
Old 1st July 2007, 04:35 AM   #1423
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Continuing the thread, the decision between the CD/horn and the RAAL ribbon will come down to personal-preference sonics and measurements. As mentioned earlier, I'm not looking for the usual Toole industry-standard criteria, but an absence of certain flaws - narrowband artifacts in the time and spatial domains. I don't like these things, since they are almost impossible to equalize, and I find them audible and difficult (for me) to ignore.

I am not in the steep-crossover-to-get-rid-of-it school; I tried that a long time ago, in my first speaker, and didn't much care for the results. Notch filtering is something I use with caution, for two reasons.

If the peak is minimum-phase, OK, then the notch filter corrects the time domain as well. This is easily checked with the impulse response. More significant, the peak may be directional. This is a big deal, since the compensated off-axis response is then grossly degraded, in both frequency and time domains. And if the peak is non-minimum phase, EQ is a bad idea anyway, since the driver is trying to tell you it's breaking up, and shouldn't be used in that frequency region (or at all).

Although the big 18 or 21-inch driver sitting close to the floor is appealing from the baffle-width and simplicity viewpoint, ragged response from 200 Hz on up is a huge downside. In the prosound world, they are almost always used with active 24 dB/octave filters, and that's something I'm trying to avoid. In my first speaker, the Audionics TLM-200, the KEF B139 had a huge 6 dB peak at 1.5 kHz. Even with a lowpass crossover at 200 Hz, it took a 3rd-order filter to shove that peak out of audibility - it was still audible as a subtle but annoying midrange coloration, even though the ripple on the overall response curve was almost invisible.

I don't like peaky drivers, basically. Minor slopes and ripples are one thing, easy to correct, but peaks are another, even if they're some distance out of the intended frequency range.
  Reply With Quote
Old 1st July 2007, 05:09 AM   #1424
diyAudio Member
 
Join Date: Apr 2003
Location: Melbourne
Quote:
Originally posted by Graham Maynard
Hi Bratislav,

Re your Post#1400.
Whilst I was getting my beauty sleep you have addressed me as if 'I' stated that simulation does not work.
I have not said this ! You have though put into words the mechanism I had in mind.
Graham,

sorry if this seemed like addressed towards you. It was said in general, and somewhat in jest .
I guess that is enough BS from me, so I'll keep out.

Over,
Bratislav
  Reply With Quote
Old 1st July 2007, 05:24 AM   #1425
ScottG is offline ScottG  United States
diyAudio Member
 
Join Date: Feb 2003
Location: US
Quote:
Originally posted by Lynn Olson
Now that I've been tipped off what to look for, I am going to very carefully examine the time vs off-axis angle, in small increments. I'm going to set MLSSA or another system on fast refresh, move the microphone slowly, and look for artifacts that move around in the time domain. That would be a direct consequence of a rough wavefront emerging from the horn mouth - and is usually concealed by conversion to freq domain and 1/3 octave smoothing. These artifacts are in the time domain, and that's the place to look for them.
The best/most versatile display I've seen for "dialing down" time domain performance is this one:

http://www.dr-jordan-design.com/view3D.htm

http://www.dr-jordan-design.com/Winaudiomls.htm

(..of course its not cheap, its windows only, and you already have a real-time system.)
__________________
perspective is everything
  Reply With Quote
Old 1st July 2007, 05:29 AM   #1426
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Continuing the design thread a little further, the least-aggravating path would be a smooth HF system (to be determined first) and a quartet of identical 12" drivers, in the delta pattern shown many posts ago.

A real time-waster in any speaker design is trying to make not-quite-good-enough drivers perform just a little bit better. This can take weeks or months of fruitless measuring, crossover twiddling, listening, over and over again. Better to select benign, well-behaved drivers in the first place. No serious peaks, no regions of high distortion, no directional breakups - in other words, sounds pretty decent without any EQ at all.
  Reply With Quote
Old 1st July 2007, 05:41 AM   #1427
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Quote:
Originally posted by ScottG


The best/most versatile display I've seen for "dialing down" time domain performance is this one:

http://www.dr-jordan-design.com/view3D.htm

http://www.dr-jordan-design.com/Winaudiomls.htm

(..of course its not cheap, its windows only, and you already have a real-time system.)
Already have a dual-core Athlon PC with M-Audio 192/24 soundcard in a compact and quiet HTPC enclosure bought for just this purpose. Even though I've been a Mac user since my first Mac Plus, I'm OK with using PCs for scientific purposes - they're much stronger in that area.

I just don't let them run around and play freely on the Internet, where they can get in bad neighborhoods - although I use Ad-Aware SE and Avast! just to be sure. By keeping most of the software clutter off WinXP SP2 and its Registry file, it runs pretty well, with no slowdown issues to contend with.

By using different machines for different purposes, maintenance issues are simplified. I've been measuring on PCs and transferring the pretty graphics over to Macs since my days at Tektronix in the late Eighties, where we did the same thing with Unix hosts and Macs for documentation. (I was usually the network guy that configured the TCP/IP software for the tech-writing group.)

Thanks for the links, will look into the software, which looks interesting.
  Reply With Quote
Old 1st July 2007, 07:06 AM   #1428
BudP is offline BudP  United States
diyAudio Member
 
BudP's Avatar
 
Join Date: Feb 2007
Location: upper left crust, united snakes
Hi Lynn,

Indeed the fur did fly while you were fighting the good fight. In all of the driver posting I have looked at no one seems to have mentioned Gary Pimm's favorites, the Eminence Beta 8, so here is a link. Scroll down to the Acrobat file for specs.

http://www.partsexpress.com/pe/pshow...Number=290-404

If any pro sound driver is going to match Raal's ribbons this looks like a good candidate.

Bud
  Reply With Quote
Old 2nd July 2007, 02:48 AM   #1429
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Default Driver QC

Good news ... Apple has sent the replacement 2.66GHz quad-core machine, the tracking code says it's in Denver already, so it should be here Monday morning. After that I attach the Firewire cable and use Migration Assistant to selectively move parts of the G5 machine to its new Intel environment. At least the file structure is the same, and Rosetta is supposed to emulate the G5 for the older software. Much of the power of the quad Core Duos will disappear into Rosetta, at least until I shell out the bucks for the latest versions of Photoshop and various other applications.

------------

I'm going to be writing a short series on the steps I'm going to be doing as the design gets under way, probably later this summer. The most probable configurations are 1x8" + 3x12", and 4x12". I'll be measuring and auditioning the 18Sound 12NDA520 and 8NMB420 on a large flat baffle, like a door.

After measuring each at different angles, I'll be auditioning a stereo pair and comparing them to the Ariels, which have a smooth response curve that fits into a 5 dB window. Similar low-coloration candidates would be the Celestion SL600's, Quads, BBC monitors, or other speakers with smooth curves. I don't think comparing the drivers to speakers with rougher responses - such as many audiophile speakers or vintage prosound - would reveal very much, except to compare one set of colorations against another.

The full-range auditions will use pink-noise, comparing the 12NDA520 against the Ariel, the 8NMB420 against the Ariel, and the 12NDA520 running in parallel with the 8NMB420 against the Ariel. This should give a general idea where the dominant colorations of the drivers are, and how well they combine with each other. Only after assessing the general character of these drivers will I move on to musical assessment, and even then I'll be using my Denon HT receiver just to see how they sound on average-quality electronics. I don't see any point to using fancy amps when making go-nogo decisions about drivers - the basic character of drivers are pretty audible with nearly any amplifier.
  Reply With Quote
Old 2nd July 2007, 10:29 AM   #1430
Pano is offline Pano  United States
diyAudio Moderator
 
Pano's Avatar
 
Join Date: Oct 2004
Location: Milliways
Blog Entries: 4
Quote:
Originally posted by Lynn Olson
In addition to a good discussion that de-mystifies compression drivers
Nice link, thanks Lynn. Interesting that they say "Not many people make high drivers at this quality level - there's only really TAD and JBL ." Leaving out one of their suppliers- Yamaha! There are a lot of other CD out there, wonder why Meyer didn't like them?

It must be the 4" JBL that sounds so good in the Meyer CQ-1 boxes. Very nice. But not in the CQ-2. Same driver, different horn, different crossover. Sounds rough. Have not yet heard the new Meyer CDs - AFAIK.

OK, back to my fur lined speakers.
__________________
Take the Speaker Voltage Test!
  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 10:25 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