Beyond the Ariel - Page 17 - 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 11th April 2007, 12:26 AM   #161
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Default Back Again

Quote:
Originally posted by Variac
I agree, I think our non-belligerent attitude around here attracts people like Lynn, so I would appreciate everyone keeping that in mind. Also let's not get off into general comments about perception and other topics that typically don't lead anywhere....

Lynn, when you want to start making your decisions, please feel free to start another thread that is limited to comments about your design only, as it is developed. We can keep this thread more for "Brainstorming"

Variac
Hi all, the reason I've been quiet for a couple of days is twofold:

A) John Atwood passed through Colorado on his way to the Montreal show (he loves to drive long distances). During the one-day visit, we had a lot of discussions ranging from an early four-tube phase-locked-loop circuit in the horizontal sync circuit of a 1951 RCA television, the impact of the 10 ~ 20% of the US population who are Dominionist/End Times fundamentalists will have on the future of the USA vs more-secular Asia and Europe, why Japanese Buddhist monks gave up celibacy and married (unlike Buddhist monks elsewhere in Asia), the way power distribution systems work in Vietnam, California, Nevada, and Colorado, the effect of the US combined household, government, and Asian-held Treasury Note debt on the long-term value of gold vs international currencies ... and a whole bunch of other things.

Since both of us have worked for big high-tech companies and have spent long stretches of time in Asia and Europe (and have many friends there), we had a lot to talk about besides audio. The area we were most concerned about was the de-industrialization of the USA over the last forty years - both of us had trouble imagining an economy where people do nothing more than investment-scheme paper-shuffling, making music-videos and video-games, working for Google, or flipping hamburgers, while real (physical) commodities are built elsewhere. It's an old principle of economics a village can't survive by taking in each other's washing - somebody has to grow the crops and make the stuff of everyday life.

B) It's still kind of awkward getting around the house - it takes me about ten minutes of hard work to get upstairs, so it's something I only do once or twice a day. The big Dual G5 Mac with twin 20" Samsung monitors is my preferred platform for Web surfing and chatting on the forums, but it's upstairs, and I don't get to it every day.

I'm surprised and impressed the way this thread has continued on its own momentum, instead of fizzling out and disappearing below the diyAudio event horizon.

Here are my own thoughts, which have been stimulated by the Johnincr's intriguing front-and-back symmetric horn I saw a few pages ago. Recalling the metaphor of the 1,2,3, and 4 variations of a surface, ranging from a flat baffle to a horn, and finishing with a pipe, I realized that's not the whole story.

The diaphragm is acoustically transparent, so the rear-wave expansion also affects the front. In fact, there are not one, but three sets of expansion, or surfaces, that are important:

1) The front expansion

2) The rear expansion

3) Any intervening surface between front and rear - specifically, the delay term between the edge of the front and rear surface. This is close to zero for a thin dipole, but is nonzero and significant for any other combination of front and rear surfaces.

I like to visualize a pair of expanding hemispheres coming out of the front and rear of the driver. These are reverse polarity, of course, and cancellation effects occur when they meet. The acoustic pressure of the expanding shockwave diminishes at an inverse-square rate if the expansion surface is flat, and at rates somewhat less than that if there's a horn. The pressure of the wave does not diminish at all inside a pipe, unless it is filled with absorbing material.

All of the surfaces - flat though horn through pipe - are subject to standing waves forming on their surface. The standing waves are created by surface-edge reflections; the standing waves are at their most intense if the surface-edges are radially symmetric with respect to the driver (circular). The standing waves can be reduced by:

i) Curving the edge of the surface, which is most effective at wavelengths smaller than the diameter of the curve.

ii) Lining area close to the edge with acoustically absorbent material, such as felt. Using felt in triangles similar to the Mamboni pattern or my 3-size variation is probably the best way to do this.

iii) Perforating the surface close to the edge with an array of holes, such that the holes are smaller than the shortest wavelength (1/4" or smaller), having the greatest density of holes closest to the edge, and the lowest density of holes farther away from the edge.

iv) Other surface treatments, such as roughening, patterning, EnABL, treatment, etc.

All of the standing-wave problems originate from the sharp boundary at the edge. At first blush, a doughnut or toroidal shape around the driver might be the best in terms of avoiding standing waves - if Johnincr's symmetric dual-horn had a flat surface connecting the front and rear horns, that would effectively be a near-toroid.

The tradeoffs with a horn is the increased energy at the edge of the horn compared to a flat surface - although as mentioned earlier, a dipole has double the energy at the energy at the surface-edge since the rear-wave is in opposite polarity, and doubles the pressure compared to unipolar radiator (a closed box with 100% internal absorption).

Some impulse measurements of the same driver in a dipole vs Johnincr's symmetric (front & rear) horn would be most interesting. Although the dipole might have more energy right at that first edge reflection - thanks to the hard boundary - there's an acoustic delay between the front and rear horn (energy diffracted around the edge of the rear horn has an additional delay before it appears around the edge of the front horn), so we might see several lower-energy reflections in the symmetric-horn system. Hard to say in the absence of A vs B impulse measurements with the same driver.

Sonically, they'll sound different as well, since the null region has a different shape, and almost certainly a different frequency spectra. The biggest differences, I guess, would be in the region between 180 and 120 degrees off-axis, with interesting distortions in time and frequency response for both types of dipole. Much to be learned here, far afield of commercial loudspeakers.
  Reply With Quote
Old 11th April 2007, 01:26 AM   #162
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Quote:
Originally posted by ctrevith
Thinking of an OB with front baffle 24-28" wide at the base tapering asymmetrically to 12" at top, 48-52 tall (ala apogee). Edges of the front baffle would be rounded 1-2" rounding. Side supports behind the front baffle would start at 12-15" at base and taper to 2-3" at top. Active Deqx for crossover (80-160 Hz?) and room equalization with solid state amp for the Eminence and 300B SET for Fertin 20EX. My room is 20' W x 32' D x 8' H. Listening seat is 11' from speakers if they are 6' from rear wall.

Questions:

1. How's the Eminence KW Pro 18A for this application? Seems to be good from TS standpoint (FS=33, 95db, Qts=.6 Xmax=10mm peak, hi power handling). Seems to be very rough though beyond 200 Hz due to cone break up or heatsink? Are there others that work better? (move as much air, cost less, go lower, transition smoother, etc.) Can this driver be equalized with the deqx to a reasonable in-room bass response (shooting for 30 Hz or less?) assuming that the driver and amp have enough headroom?

2. My thinking for the baffle shape is that the asymmetrical shape would decrease undesirable baffle interaction. Is it better to have the angled side of the trapezoid to the outside or inside edge?

3. There's been a lot of discussion regarding baffle diffraction. Does this also apply to the rear of the baffle as well or is rear baffle diffraction unimportant from a time window perspective?

4. What are the best techniques for minimizing baffle vibration? My worry is that the 18 Eminence will rattle the daylights out of the Fertin.

5. How important is it to secure the drivers from the rear to supplement the basket mounts?

6. I always liked the idea of using the same type of amp for bi-amping. Would I get satisfactory results from the Eminence with the right 300B SE amp? My feeling is that it would quickly run out of gas due to the equalization necessary for extending the low end.

7. I would like to run the Fertin down as low as possible without endangering it or creating a hole in the response that has to be artificially equalized - using more of my top end amps headroom. Any ideas on where to start are welcome. I assume that I would just run a response curve on the Fertin by itself and put the cross over to one octave higher than where it starts to roll off naturally.

8. Anyone have comments on experience using the Deqx in this type of application?

Thanks
1. Yes, you will need a steep-slope (18dB/oct or more) active crossover, active equalization, and a good-quality solid-state power amp for the 18-inch driver.

2. All you need are mirror-imaged pairs. I'd put the narrow (vertical) edge closest to the center, although I'd try it both ways just to be sure.

3. Yup. Diffraction happens everywhere, although I worry more about the integrity of the front wave, since the rear waves are destined for early reflections, not the direct sound.

4. No question about it, the big drivers will REALLY vibrate things. This is where a symmetric W-baffle for the monster bass drivers is a good idea, since they can be mounted in pairs that oppose each other mechanically, thus cancelling almost all of the reaction-force vibration.

5. Good idea. If you want to be clever, you can have aluminum or brass L-beam support not just the magnet, but the side-mounting screws for the full-range driver as well. For that matter, the front baffle can be mechanically isolated from the driver, and only have a compliant gasket where the edge of the driver basket meets the cutout of the baffle.

6. BAD idea, no donut. I can't think of a worse bass amp - particularly for a monster 18-driver that generates huge back-EMFs - than a tiny little 8-watt SET amp. The bass will be VERY murky and slow, and degrade the entire spectrum with all the sludge and distortion. Controlling bad-*** woofers takes an aggressive amp with lots of damping factor, a serious high-current supply, and lots of DC-coupled power. Think of a good-quality PA amplifier.

7. Retaining the quality of the widerange portion of the system is the hard part. The low-level high-pass capacitor preceding the SET amp needs to be of superlative quality, or if the amp is RC-coupled, you can play games with the value of the RC-coupling.

Setting the high-pass crossover frequency is something you might want to do by ear, although your initial suggestion is a good starting point. The subjective tradeoff is murk vs hardness.

8. Pimm's experience was the abominable electrolytic caps in the Behringer *must* be replaced with polypropylene films, at the minimum, and the op-amps swapped out for decent-quality parts. Despite the popularity of this gizmo in Europe, my audition told me it's not suitable for the high-quality part of the signal path - but there's no problem using it for low-bass crossovers and low-frequency room EQ.
  Reply With Quote
Old 11th April 2007, 01:53 AM   #163
frugal-phile(tm)
diyAudio Moderator
 
planet10's Avatar
 
Join Date: Oct 2001
Location: Victoria, BC, NA, Sol III
Blog Entries: 5
Default Re: OB with Fertin 20EX and Eminence Kilomax Pro 18A

Quote:
Originally posted by ctrevith
4. What are the best techniques for minimizing baffle vibration? My worry is that the 18 Eminence will rattle the daylights out of the Fertin.
The best technique is to mount a pair push-push (which means they will be firing out the sides of the baffle...

dave
__________________
community sites t-linespeakers.org, frugal-horn.com, frugal-phile.com ........ commercial site planet10-HiFi
p10-hifi forum here at diyA
  Reply With Quote
Old 11th April 2007, 02:56 AM   #164
terry j is offline terry j  Australia
diyAudio Member
 
Join Date: Aug 2006
hi lynn, would love to be a fly on the wall for some of those interesting and wide ranging discussions of yours!!!

Good to see you back, however I must admit that most of it still zooms way over my head ha ha ha.

Quick question, ctrevith's question made me prick my ears up, as I'm currently considering doing something a bit similar, namely to use my 18 in PHL's as a dipole bass unit. I have heard only two dipoles previously, one was the alon4's and the other the orions. Admittedly the alons only run a sealed 12 for the bass, so in the scheme of things perhaps not a good model by which to evaluate the dipole alignment.

Got to say, the orions didn't do it for me, for the life of me I couldn't pick the amazing imaging etc that is often used to describe dipoles, but that is neither here nor there and only my opinion.

HOWEVER, the bass dipole of the orions was lluuuvvvverly! So, my thinking about using the 18's only as dipole, and like ctrevith, I run my system using the deqx and as such his post had even more interest for me.(I would leave the mid and tweeter as is).

However, again when you answered his post re using the deqx, you again mentioned the behringer, and that has left me a little confused as the deqx is a totally different beast to any behringer unit, and I'm wondering if you have gotten the two products mixed up. Indeed, did ctrevith himself mean a behringer unit or the deqx unit. Praps we had better clear all that up???

One last quick question for the gurus on my possible conversion, I can understand that if I run the PHL's as a dipole they would not go very low, and so to compensate I have every intention of adding a sub, and maybe only take the PHL's down to 50 hz or so. The problem that I may run into with the PHL nay not be at the bottom, but rather at the top of the FR, as I am unable to have the mids go any lower that 300 hz.

Is that where my best laid plans may come undone??? ie will I get all sort of wacky response from a dipole bass unit running that high??
Lastly, I often see multiple bass units used in dipole, and assume that is mainly for help in augmenting the bass response. However, is it also to help in cancelling vibrations??? In other words, is it indeed feasible to have a single 18 doing the dipole bass??

enjoying the learning experience here
  Reply With Quote
Old 11th April 2007, 04:07 AM   #165
diyAudio Member
 
Join Date: Apr 2004
Location: Costa Rica
Send a message via AIM to johninCR Send a message via MSN to johninCR Send a message via Yahoo to johninCR
Welcome back Lynn,

The point missing from the trade imbalance discuss was that all we do is send over pieces of paper that are easy to print, and we get back real stuff. Where we get in trouble is when they send us cheap junk, and they use the paper to buy fleets of aircraft from us. Once they figure out all we send is paper, or they just decide to keep everything they make for themselves, then we're in deep trouble.

I'll be sure to include impulse responses when I start measuring. I have 2 other B200's, so I can do flat baffle too, or do I need to use the same driver?

2 things concern me about woofers mounted sideways. 1. You lose potential rear wave delay. 2. not firing the woofer at the listening position has to have consequences with open alignments. If you make it a manifold to fire essentially into a common air cushion, doesn't that affect impulse response.

I've gone both ways, but have adopted mounting the woofer to a base separate from the baffle with better success. In both alignments below, the woofer is isolated by foam rubber from the baffle, and the woofer energy goes into it's separate base, and isolated completely from the baffle structure and/or support structure of other drivers. While these are minimalist forms with limited max SPL, the same approach can be fitted to any baffle size, shape, or driver compliment. Once you don't have to put everything in a box, many things become much easier. Oh before I forget, AJ I don't have measurements for these either.
Click the image to open in full size.
Click the image to open in full size.
__________________
Everyone has a photographic memory. It's just that most are out of film.
  Reply With Quote
Old 11th April 2007, 06:44 AM   #166
diyAudio Member
 
Join Date: Jan 2005
Location: SoCal
Default Deqx question

"However, again when you answered his post re using the deqx, you again mentioned the behringer, and that has left me a little confused as the deqx is a totally different beast to any behringer unit, and I'm wondering if you have gotten the two products mixed up. Indeed, did ctrevith himself mean a behringer unit or the deqx unit. Praps we had better clear all that up???"


Thanks for the comments Lynn. Yes, as terry j mentions, the unit I was referring to was the Deqx unit: http://www.deqx.com/PDC26P-Preamp.html, not the Behringer.
  Reply With Quote
Old 11th April 2007, 07:55 AM   #167
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Yes, I was thinking of the Behringer DQX, not the DEQX. Similar names are confusing!

Naturally, with an all-digital source, you can do any kind of digital signal bending you want, limited only by the bit depth of the computing engine, and you're free to choose high-quality DACs on the other end.

As for the bass unit, well ... if you take a purist all-flat-baffle dipole approach with no W or U-box for the bass, you're going to need a LOT of equalization or a REALLY BIG (wall-sized) dipole baffle. Folks who get cute and make a room-divider folding baffle with a piano hinge are sort of kidding themselves - small air leaks through the hinge reduce the effective size to not much larger than the hinge-to-baffle-edge distance.

I still like the idea of a small, fairly deep box with an open back and and mesh termination on the sides and top. This has the merit of three surfaces available for driver mounting, a straight air-path for the front drivers, and an air-path with only one relatively unconstrained 90-degree turn for the side drivers.

Yes, the backwave is radiating into a more constrained environment, but it emerges into a combination of resistive damping material and mesh side walls. This is quite a different animal than a six-sided box with resistive filling, or a transmission line, which is a tube with resistive filling.

The diagram a few pages back with the rear-wave proportionally split between an open-back box and a folded transmission line really got my attention. It would have never have occured to me to split the rear-wave into two quite different kinds of rear loading.

However - that does prompt a thought. Instead of having an enclosure that splits the backwave into two very different lengths, why not create a continuously lossy transmission line, say with a mesh wall going all the way down the line?

So you've got a typical 1/4 wave line tuned to 45 Hz and six feet long, folded as many times as you want, but the whole thing is filled with recycled cotton waste and has a heavily resistive felt-covered mesh replacing one of the side walls. The thing "leaks" along its entire length, but that's actually a design feature!

This is kind of the inverse of the TQWT - instead of an intentionally resonant pipe with a bit of horn-gain, it's a semi-transparent tube with losses along its entire length. The gain comes from the driver array, not horn-gain, so it's a different kind of sound than a TQWT or a transmission line.

You can play with the TL vs resistive-loss characteristics by having a clear air-path go through the center of the line, or filling the the entire pipe with filling material, as you wish. That controls the ratio of the energy coming out the distant end vs the leaks over the entire length - easily measured, by the way, by measuring the spectrum and magnitude of the nearfield output at the far end of the pipe. Alert readers will notice the similarity to automotive muffler design, using lossy baffles and filling material.

  Reply With Quote
Old 11th April 2007, 08:10 AM   #168
diyAudio Member
 
Graham Maynard's Avatar
 
Join Date: Dec 2003
Location: UK
Hi Lynn,

I agree with you about mesh/felt sides.

I tried similar and reported good findings to Fullrangedriver Forum, also the much improved reproduction from mounting a full-ranger against a two-three turns pipe of heavy carpet, itself 'soft' plugged with carpet.

When there are solid sides behing a driver there is no getting away from rear pressure and reflections causing cone re-excitation apparent from the front.

For 'sides' I used steel mesh (of the type used for open mezanine floors) covered with heavy carpet, and the front of driver output was very much improved. Only thing was I did not have enough materials or the capabilities to make an entire cabinet. I also wondered how the construction could be made long term acceptable in a home environment.

If the side of a TL is to have controlled leaky damping does the far end still need to be open ?

Cheers .......... Graham
  Reply With Quote
Old 11th April 2007, 08:17 AM   #169
diyAudio Member
 
Lynn Olson's Avatar
 
Join Date: Jul 2004
Location: Northern Colorado
Quote:
Originally posted by Graham Maynard
Hi Lynn,

I agree with you about mesh/felt sides.

I tried similar and reported good findings to Fullrangedriver Forum.

When there are solid sides behing a driver there is no getting away from rear pressure and reflections causing cone re-excitation apparent from the front.

I used steel mesh of the type used for open mezanine floors covered with heavy carpet, and the front of driver output was very much improved. Only thing was I did not have enough or the capabilities to make an entire cabinet. I also wondered how the construction could be made long term acceptable in a home environment.

If the side of a TL is to have controlled leaky damping does the far end need to be open ?

Cheers .......... Graham
Yes. I see no performance advantages to a closed-end TL, just more obnoxious resonance. The concept here is an enclosure that is transitional between a traditional TL and an open-baffle, as insane as that sounds. The transition is controlled by the resistance of the mesh and the percentage of the open aperture of the transmission line from front to back.

You can go even further by having a Mesh TL that is 100% TL closest to the driver (rigid walls and open space along the centerline of the pipe), and gradually becoming 100% mesh (mesh walls and all filling in the middle) by the time you get to the end. This would give the desired progressive-leak characteristic - for once, a truly non-resonant line.
  Reply With Quote
Old 11th April 2007, 08:59 AM   #170
owen is offline owen  United Kingdom
diyAudio Member
 
Join Date: Mar 2004
Location: Shilton
How on earth did I miss this thread..?

My thoughts have been along similar lines for a good while now, but the efficiency area is where we're differing.

My concept uses a smallish fullrange driver, and the OB baffle rolloff to provide the low pass filtering, and a simple 1st order or so crossover at say 10K to a reasonable tweeter. The target efficiency for this part of the system is 90 or so db/watt. Bass is augmented by a 12" prosound woofer also on an OB, and a simple inductor high pass. Target efficency of the bass stage is 96 or so db.

For a 300Hz OB (1 foot or so wide), the XO of the bass stage will be in the range of 150 - 225Hz (give or more likely take a little), taking real world bass response down to around 100Hz, with low compression, and low equalisation. Add on the fact that the bass driver could be supplemented below this by a pair of drivers, using another low pass roll off at something like 50Hz (again run OB), you'd take the 1/4 space efficiency gain of the lowest driver, and should go flat to 50 Hz, and 12 db roll off after that.

Thats as far as I've thunk on this, but If I build this, it'll be active.

Oh, as an aside, the center of the fullranger would be around 38" from the floor in something like this - ideal for some-one like me (tall)

Just the results of my addled brain.

I also like the look of something like the eminence alpha 6 as a high efficiency OB wide-ranger - as it would have the prosound headroom, and efficiency, the downside is then the bass section would need to be more efficient


Owen
  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 07:02 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