Beyond the Ariel - Page 774 - 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 17th June 2012, 11:00 AM   #7731
diyAudio Member
 
john dozier's Avatar
 
Join Date: Dec 2010
Location: columbia sc
Quote:
Originally Posted by Lynn Olson View Post
This may sound a little arrogant - but what the heck, it is my thread - but the literature is wrong fairly often. We all take it as received wisdom, but if you follow the standard models, the results don't always sound good. Go a little further, dig deeply into the measurements, poke around a little, and you'll find discrepancies that shouldn't be there.

Diffraction is harder to model than you might expect. Damping materials act oddly, and appear to have nonlinear effects that don't fall into any standard model. There's a zero in the response of vented systems, around 1~2 Hz, caused by various box leaks, and that has a strange effect on the linearity of the bass driver as it slowly wobbles back and forth at a 1~2 Hz rate.

Capacitors are genuinely microphonic, and this can be measured by using a 2-foot tube attached to a loudspeaker and beaming sound into a cap with a 9V DC polarizing charge on it. The cap acts like a very low-quality condenser microphone with several big peaks in the 1~5 kHz range. You might think, well, crossover caps don't have a DC polarization, but it's actually worse than that; the first cap of the tweeter circuit has the full voltage of the woofer across it, and LF modulation effectively makes the cap into an AM modulator for the microphonics in the 1~5 kHz range. These are real, measurable effects, not audiophile myths.

To repeat, when things sound wrong, they are wrong. Many designers fall into denial when things sound worse than expected, but that's a signal to dig deeper, instead of denying the evidence of your senses. Sometimes subtler measurements will uncover the true nature of the trouble, but don't expect to find it right away.

The top engineers at ESS took several years to discover why delta-sigma DACs didn't sound quite as good as ladder DACs - and as the inventors of the delta-sigma DAC, they had every incentive to deny what they were hearing and just market the existing DACs all the harder. That's what Philips and Sony did with 44.1/16 CD's, after all - they were marketed as "perfect sound forever" for quite a long time - until the patents ran out, and then of course DSD/SACD became "better than perfect sound for longer than forever".

TomTom has been kind enough to share his results with supertweeters. I have no idea at all why they are not sounding integrated with the mid horn - but then, it's a type of system that I have no experience with. My horns and supertweeter are freestanding and are aligned acoustically; TomTom's system evidently has a common baffle that is shared with the horn-mouth and the supertweeter, and the supertweeter is delayed electronically, presumably with a digital FIR filter with an unknown dithering algorithm and computational bit depth. That's a lot of different variables there.
Jerry Foreman of Precision Audio suggested I use only hermetically sealed Teflon caps for exactly the reason cited. Teflon caps are notoriously microphonic because of the taut winding method used. Regards
__________________
Evil looms. Cowboy up. Kill it. Get Paid.
  Reply With Quote
Old 17th June 2012, 06:04 PM   #7732
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Quote:
Originally Posted by john dozier View Post
Jerry Foreman of Precision Audio suggested I use only hermetically sealed Teflon caps for exactly the reason cited. Teflon caps are notoriously microphonic because of the taut winding method used. Regards
You don't have to spend that kind of money for a doubtful gain. Just use an external crossover behind the speaker, something I was doing back at Audionics in 1975. Tie down the caps by wrapping them with soft foam or wool felt and wrapping plastic wire ties around the soft foam or felt (the wire ties will dent the caps otherwise, and you don't want to do that). The plastic ties then go through holes in pegboard or whatever you're building the crossover on. If you want to go further, the microphonic cap can be dunked in a tray of melted beeswax, and let the beeswax slowly harden around the cap with the leads hanging out of the tray.

Along the same lines, the wax caps from Jupiter seem to be the quietest, since they are not made from tightly wrapped plastic and beeswax is not resonant. The tight wrapping seems to be responsible for the microphonics, from what I can tell, and it applies to all wrapped-plastic caps - Teflon, polypropylene, Mylar, etc. Film-and-foil construction is heavier and is mechanically a plastic/metal composite, while the metallization layer of metallized-foil types is so thin it has provides almost no damping for the tightly wound plastic.

Above all else, don't let the caps hang from the connecting wires, where they will be free to vibrate. This is far more important than dumb things like putting little rolling-ball dinguses under a CD player. Although cap microphonics are not as directly audible as vacuum-tube noises, it is responsible for a fair amount of the "cap coloration" we hear - blurring, midrange harshness and grit, lack of dimensionality, etc.

Anyone with a spectrum analyzer can measure cap microphonics. The exciting loudspeaker has to be at least 2 feet away, otherwise the cap will simply pick up the EMF field of the loudspeaker, which defeats the point of the measurement. So use a paper or non-metallic tube to transmit the sound to the cap-under-test.

Polarize the cap with a 9V battery, and have another cap right next to the sound-card or spectrum analyzer input to block the DC (but well away from the loudspeaker). Twist the wires between the cap and analyzer so they don't pick up ambient RFI and noise.

The test setup is the same as the speaker-and-microphone test of a loudspeaker, although you are testing the "microphone" instead of the loudspeaker.

Although the "microphone" output is low, it is not zero, and the spectrum analyzer will display the frequency response of the resulting acoustic pickup of the capacitor. When you see the peaky spectrum on an analyzer, the reason for the coloration will be obvious.

Although I think it's silly to use vibration isolators on solid-state equipment, there is a bit of microphonic pickup from the electrolytic bypass caps which are universally used in solid-state electronics. The power-supply bypass caps are polarized, of course, and will pick up acoustic vibrations and translate them to low-level but quite distorted audio. So not a good idea to put a CD player in front of a compression driver.

I suspect much of the sonics of combining different types of caps - the endless cap-tweaking game that never seems to conclusive - is nothing more than combining different microphonic spectral patterns. Rather than pointless tweaking, better to keep microphonic pickup to the minimum.

Here's my Ariel crossover - not beautiful, but functional, with all caps tied down. The tweeter crossover is on the left side, the midbass on the right. The input switches select between the Marantz home theater amplifier and the Karna high-quality amplifier. The crossover board is elevated from the rug by small foam circles cut from pipe insulation.
Attached Images
File Type: jpg Crossover.jpg (229.1 KB, 565 views)

Last edited by Lynn Olson; 17th June 2012 at 06:31 PM.
  Reply With Quote
Old 17th June 2012, 07:04 PM   #7733
diyAudio Member
 
Join Date: Sep 2010
why we're discussing outboard crossovers, it reminds me of something that sometimes bothers me. Inductor orientation relative to loudspeaker voice coils and magnets - is this just an imagined problem or is it real?
  Reply With Quote
Old 17th June 2012, 07:45 PM   #7734
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
The B fields from air-core inductors spread out several inches - my rule of thumb is 90-degree orientation away from each other and at least 3~6 inches of spacing between inductors. Crossover is external (for microphonic reasons), no ferrous screws are used on the crossover board, and if there is an enclosure for esthetic reasons, it is made of wood or plastic, not aluminum (eddy currents) or steel (nonlinear inductance and unwanted interaction with air-core inductors).

The crossover is spaced off the rug a couple of inches on foam pads to damp vibration and prevent the wires underneath the board (which are the independent star-grounds for each section of the crossover) from snagging on the rug. I keep it at least a foot away from metallic amplifier chassis, or other large bits of metal.

Yes, I have wire preferences, but they're pretty simple: industrial Litz wire with cotton-cloth covering. Litz wire can be tinned and soldered with a solder-pot. If 2 feet of wire sounds different - in any way - than 20 feet, you're using the wrong speaker wire. Likewise, if the wire has to "burn in" for days or weeks, that's the plastic dielectric slowly losing the DC polarization charge that was created by manufacturing stresses. Ditch the plastic and get rid of burn-in hassles for good. If you're worried about triboelectric effects and self-motion from induced currents, better to let the wire move free than contain it in a high-Q resonant plastic straightjacket. Think about it: which is more resonant, tightly wrapped plastic (just like a plastic cap) or dead-soft copper in a Litz-wire bundle in a loose cotton sleeve?

One nice thing about Litz wire is there's no strand-to-strand current flow, thanks to the enamel coating of each strand. So even if the wire moves around a bit, it doesn't matter, since there's no strand-to-strand current flow and associated distortion from copper-oxide corrosion on the surface of each strand.

Last edited by Lynn Olson; 17th June 2012 at 08:10 PM.
  Reply With Quote
Old 17th June 2012, 08:34 PM   #7735
pos is offline pos  Europe
diyAudio Member
 
pos's Avatar
 
Join Date: Feb 2008
Location: Paris
Quote:
Originally Posted by tomtom View Post
Since discussion that i started moves on. I attach picture of my system. I must admit that amp are lousy AB /temporary/. Its build in concrete wall. STW on test was just temporary mounted with lots off cotton wool around to minimize diffraction. System isn't perfect - there is some room for tuning. But WTW pastern is completely without problem - as i say it has to crossed steep enough.

Lynn, i suppose in your case with lazy ribbon - the radiated power /and this is perceived at UHF/ of lazy ribbon is roughly equal of radiated power of Azura at XO, although polar are different shape. Thats why you have less pain with integration.
Just for your info i listened mostly for classical and im also concert-goer. Many thanks for tip with listening to broadband PN long time ago. There are things - very hard measureable but very easy to spot on PN.


And lastly, im not against STW actually i manage proper integration. /But at the end i was capable to get same thing without STW/ It was just not so easy that i hope so and i get many results that was very hard to choose from.
Nice looking system
  Reply With Quote
Old 17th June 2012, 08:40 PM   #7736
diyAudio Member
 
Join Date: Apr 2005
Location: Victoria, BC, Canada
Quote:
Originally Posted by pos View Post
Nice looking system
a little difficult to adjust toe-in, though.
  Reply With Quote
Old 17th June 2012, 09:53 PM   #7737
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Quote:
Originally Posted by RockLeeEV View Post
why we're discussing outboard crossovers, it reminds me of something that sometimes bothers me. Inductor orientation relative to loudspeaker voice coils and magnets - is this just an imagined problem or is it real?
It's fairly simple to validate your orientation, too. I just excite coil 1 with a low AC signal from my function generator, and coil 2 is connected to the scope input. As you move coil 1 (or 2, depending on your preference) and change orientation and distance, you can clearly see which condition provides for the lowest pickup.
  Reply With Quote
Old 17th June 2012, 11:53 PM   #7738
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Good advice. Measuring is always better than guessing.
  Reply With Quote
Old 18th June 2012, 12:15 AM   #7739
BudP is offline BudP  United States
diyAudio Member
 
BudP's Avatar
 
Join Date: Feb 2007
Location: upper left crust, united snakes
The final plus on the side of type 2 Litz wire, which is what you have on hand, is what is called "proximity effect". Basically, to induce a 60 Hz power grid signal, you have to drape the cables across a power transformer. Power cords are ignored, nylon carpets are ignored, RFI is ignored until gigahertz, it even works perfectly in low signal level interconnects as long as three meters with no shielding.

Just don't introduce it to your cat, they cannot help but chew on it....

Bud
__________________
"You and I and every other thing are a dependent arising, empty of any inherent reality" Tsong Ko Pa
  Reply With Quote
Old 18th June 2012, 02:03 AM   #7740
Pano is offline Pano  United States
diyAudio Moderator
 
Pano's Avatar
 
Join Date: Oct 2004
Location: Milliways
Blog Entries: 4
Quote:
Originally Posted by zigzagflux View Post
I just excite coil 1 with a low AC signal from my function generator, and coil 2 is connected to the scope input.
Neat idea! Thanks for that.
  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 04:47 AM.


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