Official Premature Arvo Pärt Picante thread

[JonP]

Tasty... (can't think of appropriate Chinese expletive to insert here) are those flat, domes or concave? Can't quite tell from the pic. Ceramic? (or resonable facsimile?)

PS, couldn't find a reasonably priced 10kw CO2 CNC laser machine on Craigslist for ya, Jon, guess you could make do with a couple bags of kitty litter ballast instead...

[JonMarsh]

Pure alumina ceramic, and somewhat shallow domes. I want to see how they behave up higher than Jeff Bagby's 20 kHz test limit (like I hear so well up there; but then, high Q resonances often fold back into the baseband- though Jeff's distortion plots don't show any problems. )

[JonMarsh]

MLK, and made a bit of progress today...

Had a variety of things to do today, couldn't spend the whole day on this, but made some significant progress.



Did the back side relief routing today; the DeWalt 611 is such a nice size; bigger than the normal trim router, variable speed, but still able to get into somewhat tight spots.

3/4" by 1-1/2" by 11-3/4" maple strips were used above and below the wave-guide to reinforce the front panel (not shown).



The LF Module side panels were ripped to final width with 45 degree bevel; you've got to be careful handling these after the rip, as with the phenolic, the long edge is knife sharp- cut the cut fingers to prove it when I underestimated how sharp it was.




Putting the woofer module sections together is not a pretty process- the U shape woofer section previously attached to the front baffle is coated with epoxy, as is the bevel edge of the side panel; then they're mated together, with the bevel edge dropping into the bevel slot formed by the side rails. (hadn't cleaned up the epoxy yet when I took this pic).

Then to really secure things, besides the clamps, I use 1-1/4" fine thread Grabber screws spaced at 2-1/2" intervals- the sort of flat washer head is built in.

Next to assemble will be the top baffle of the woofer module and the other side panel.

The tweeter clamp boards are ready for finish cutting, and we're actually getting pretty close to being ready to mount drivers.

[JimS]

we're actually getting pretty close to being ready to mount drivers.

arty:

[JonMarsh]

arty:

More progress yesterday evening- cut new LF module lower baffles, cut and routed the main tweeter mount plates and the side braces, and did the first side panel mount to the second cabinet.

The tweeter plates were a gating issue for major aspects of the design; I wanted the woofer to woofer distance as short as possible, but the tweeter plate needed to be able to accommodate the 3-5/8" magnet cut out and still have enough meat on the shoulders to attach the 3/4" by 1-1/2" maple spars for stiffening. That dictated a minimum tweeter plate height (relative to the main panel) of 5-1/4".

Other aspects of the posted side drawing are out of date, and will be updated; this is (for me) rapid prototyping, and in some cases I'm working out things including updates as I go along, rather than a full blown paper design before starting. We'll see if that really saves time, or if I just wind up painting myself into a corner! :roll: So far it looks like I haven't screwed the pooch yet... :W

One thing that's clear, to make this thing go together well, with it's Chinese puzzle sort of construction, means cutting to about 1/32" accuracy, and being very sure that pieces are true over long measures- good thing I used iron angle pieces when clamping up and gluing the front pau ferro rails, for example. I would not call this a nice "production ready" design, certainly not easy compared to a basic box.

Took pictures but no time to get them ready for online; was updating Macbook Pro with a new SSD. That injects what was already a pretty fast machine (quad core + ATI graphics) with an additional snappiness that's quite pleasing.

I expect if there's any interest from others for this design, I'll put together a detailed PDF build guide and drawings.

[Dean100]

Jon,

It sure is nice to see you back to building some speakers, even if it isn't the way you normally do things. Looking good! :T

[JonMarsh]

Jon,

It sure is nice to see you back to building some speakers, even if it isn't the way you normally do things. Looking good! :T

Hey Dean, I hear "normal" is greatly over rated! :lol:

Just trying out something a little different in technique in order to get to the measurement and analysis stage a bit quicker.

Also intend to do something less normal about the acoustic targets; a lot of influences, there, including Quad ESL's, Magneplanar, Peak Consult, and the Version 3.2 and 3.2 Orion. IF it works out as I expect it might, that may spread out to the Ardent and the nascent Nascent. I've already been experimenting with it on the Modula Xtreme and like what I hear.

More later...

[JonMarsh]

Can be adjusted...

uh oh . . . just re-read the specs after seeing your progress and realized I glossed over wing width on first read, not only are these an inch taller than my part deux, they're two inches wider 8O

better sound good, cause the "I'm making smaller speakers to be sensitive to your needs" line just disappeared . . .

I will likely consider a reverse-color scheme again with veneer on the panel and TBL on the wings for the "slimming" effect.

The way they're evolving you probably could take 3" off the height at least- from the bottom! What's happened is that the upper woofer module plate has stayed where it is, but the side panels are a bit taller, and I'm probably going to prototype the passive crossover there- and likely leave it there. That means the bottom part could be cut off by 3-4", with a main bottom plate similar to the upper one about 7-3/4" underneath the lower woofer mount plate.

It will be clearer to you when I post an updated drawing. It's literally evolving as I build this one...

[JimS]

I expect if there's any interest from others for this design, I'll put together a detailed PDF build guide and drawings.

No rush on this account, I'm just excited for you that they're coming along so quickly.

I'll likely put some serious consideration into building these since I've already got the LIO and am fully bought into dipole sound, but it won't be until summer at the earliest given work schedule. The part deux are also sounding quite nice at the moment so I'm not in a rush :T

[JimS]

The way they're evolving you probably could take 3" off the height at least- from the bottom! What's happened is that the upper woofer module plate has stayed where it is, but the side panels are a bit taller, and I'm probably going to prototype the passive crossover there- and likely leave it there. That means the bottom part could be cut off by 3-4", with a main bottom plate similar to the upper one about 7-3/4" underneath the lower woofer mount plate.

It will be clearer to you when I post an updated drawing. It's literally evolving as I build this one...

That would help. Still not quite a slim tower . . .

It would also be interesting to see if an all active design could be done with the LIO, although that would likely push dsp capacity beyond the limit.

[JonMarsh]

That would help. Still not quite a slim tower . . .

It would also be interesting to see if an all active design could be done with the LIO, although that would likely push dsp capacity beyond the limit.

That really depends on what you use for program material- if someone was only doing CD, or even 88.2/24 and 96/24 max, that would effectively double the DSP resources compared with my situation, running 24/176.4 off the Fidelia player all the time.

Many things to work through yet. I also didn't want a big pile of amplifiers- bi-amping doesn't bother me too much, but things are getting out of control when you've got stere tri-amp- especially since I'll probably want to be able to carry this around occasionally, like for this year's Northern CA DIY!

BTW, I thought a fair amount about the feasibility of slimming it down, but one of the virtues or problems of this new design is that those "decorative" hardwood rails on the outside are really an integral structural component for assembling the dipole panel and the LF Module- I'd have to narrow the LF module some more, and I'm not sure of the wisdom of that. It is all about trade-offs. An admitted issue from my side is that I'm probably putting a bit more emphasis on esthetics with this design- I decided on a panel/baffle design that worked for me esthetically, and that I THINK can be made to do what it needs to acoustically, and isn't too hard to build (the latter is an open question- it's certainly time consuming- not as straight forward as a set of Statement boxes, for example).

I have a couple of easier projects in mind with the Illuminator woofer and N26c tweeter. But they won't be dipoles. And there's still the Ardent upgrade, seemingly stalled, but things I've been studying about target response in room with the Modula Xtrreme and the Ardent are going to be applied to all the projects this year, I think- starting with the Arvo Pärt Picante.

[JonMarsh]

uh oh . . . just re-read the specs after seeing your progress and realized I glossed over wing width on first read, not only are these an inch taller than my part deux, they're two inches wider 8O

better sound good, cause the "I'm making smaller speakers to be sensitive to your needs" line just disappeared . . .

I will likely consider a reverse-color scheme again with veneer on the panel and TBL on the wings for the "slimming" effect.

One last thought to throw out there this morning, about height- my experiments with the Modula Xtreme clearly showed the taller version (with the added crossover box) had a much smoother measured mid bass to midrange, and sounded more coherent, too. Height and it's influence on where it moves the floor bounce issues seems to favor height, for now.

"Upgraded" all the screw hardware last night to Highpoint XT Roundwasher head #8 by 1-1/2" in black oxide. Forgot I had a bunch of these along with the driver bits lying around...

[JonMarsh]

Both sides now

Not that much spare time this evening, but a little more done.

The tweeter plates are ready for notching and drilling holes for tweeter and mounting to panel.



All the panels but one side were finished, and ready for "both sides now".





Next step is cutting and finishing the rosewood trim strips for the top and bottom edges of the side panels, then priming and painting the MDF, and final finish sanding of the pau ferro.

Starting to look like we might have some cabinets to mount drivers in shortly!~

[bbggg]

I'm sorry if my question is a stupid one Jon, but it's not clear to me, will you be using the MCM waveguide or the Jantzen one? I thought the MCM was not sold anymore. I'll be following this thread with much interest as I also have an LIO-8 which ls waiting for some speakers to weave its magic. :T

[JonMarsh]

I'm sorry if my question is a stupid one Jon, but it's not clear to me, will you be using the MCM waveguide or the Jantzen one? I thought the MCM was not sold anymore. I'll be following this thread with much interest as I also have an LIO-8 which ls waiting for some speakers to weave its magic. :T

Not stupid at all. It's the Jantzen waveguide- similar profile and size, very similar, but machined from solid BMC. Very nice build quality. I bought three pair in 2010, figuring I'd find something interesting to do with them, after the Modula projects. Reinforcing the MCM waveguides was something of a PITA!

[bbggg]

Did you get them directly from Jantzen, and do you remember the price perchance?

[JonMarsh]

Direct from Jantzen...

Did you get them directly from Jantzen, and do you remember the price perchance?

I got them direct, no distributors I could find carried them, or the TQWT kit, for which they were developed. They were about $50 each, with an order handling fee of $10, and EMS shipping from the warehouse in Poland of $62.00. Of course, that was two years ago, prices may have gone up. Obviously, this is more expensive than an MCM waveguide, but the quality different in construction is huge. Well worth the price if you're doing DIY for performance, not for saving bucks. And if I charge myself by the hour for futzing with the MCM waveguides to get them ready, I come out WAY ahead, in my opinion.

~jon

[JonMarsh]

Starting to look like speaker cabinets of a sort...

Some pictures taken on the measurement "turntable". They still need a lot of detail work, including sanding and painting, and wood trim on the plywood side edges and the bottom- cut from 1/8" Bolivian Rosewood from my local Woodcraft.

This is the way we'll usually be looking at them. Front baffles will be satin black.


45 degrees, not too ugly. The higher side panels should hide crossover parts when seated, and we can always put a cover of some sort on.


Direct from the side


45 degrees from the back- some clean up to do here. Will probably add something to the underneath of the bottom panel for connectors. Still thinking about that.

And yes, the woofer mounting holes are pre-drilled! Wouldn't be easy to get in to do that at this point... :roll:

[cjd]

I see your wall is just about as finished as your speaker!

Looking good! It's a pity to lose that nice LBL behind... something. But it'll look nice the way you plan to finish it too.

[bbggg]

Thanks, Jon, this is shaping up to be an excellent job. Did you have to buy the MH DSP+ license or are you making do with the standard LIO software?

[JonMarsh]

Thanks, Jon, this is shaping up to be an excellent job. Did you have to buy the MH DSP+ license or are you making do with the standard LIO software?

I bought the MH DSP+ license- there's a lot more you can do with it, and the incremental price wasn't that big a deal in the overall scheme of things.

I see your wall is just about as finished as your speaker!

Looking good! It's a pity to lose that nice LBL behind... something. But it'll look nice the way you plan to finish it too.

Yeah, Chris, I pondered doing the front baffles differently so that there was LBL to the front, but that would have meant routing out LBL, instead of HDF, which I think would have made them a bit less stiff. In the end, I decided to go with the MDF to the front, and route that out for the rebates, and do round overs on the back side.

Unfortunately, the rapid prototyping hasn't been quite as rapid as I'd hoped!

And the unfinished wall was a repaint in the living room started last February, stalled out due to crazy work stuff- two walls done, but didn't even get back to putting the hardware on the wall. Well, there's all spring hopefully to get caught up!

[dar47]

And the unfinished wall was a repaint in the living room started last February, stalled out due to crazy work stuff- two walls done, but didn't even get back to putting the hardware on the wall. Well, there's all spring hopefully to get caught up![/QUOTE]

Ah to be a single guy, If I leave one screw missing she notices and asked for a completion date. :rofl:

This is looking great Jon. My Sons friend is taking my whole theater setup off my hands this spring so I need a theater/ 2channel set for my main room and I'm eying the Ardents just not sure which version so when your done these I hope something else doesn't come up. This almost sounds like a completion date. :rofl:

[Bent]

Ah to be a single guy, If I leave one screw missing she notices and asked for a completion date.

Tell me about it - a creative mind must be free to wander, reining it in stifles the creativity.

[JonMarsh]

And the unfinished wall was a repaint in the living room started last February, stalled out due to crazy work stuff- two walls done, but didn't even get back to putting the hardware on the wall. Well, there's all spring hopefully to get caught up!

Ah to be a single guy, If I leave one screw missing she notices and asked for a completion date. :rofl:

This is looking great Jon. My Sons friend is taking my whole theater setup off my hands this spring so I need a theater/ 2channel set for my main room and I'm eying the Ardents just not sure which version so when your done these I hope something else doesn't come up. This almost sounds like a completion date. :rofl:[/QUOTE]

Getting back to the Ardent update is Numero Uno after I get these basically running. (I plan to do an ASP unit for these also, but that can be later this year). Things I've been learning about target response and system behavior from the Modula Xtreme will be applied as well as the new Illuminator woofers! Those targets and philosophy will also be implemented in this project.

And yes, life is much easier not shared under one roof! My girlfriend is very particular about her esthetics, to the extent that she just sewed up a custom cover for the 50" Panasonic Plasma we found her at a great price, so that it matches the decor in her family room and is not readily identified as a TV; looks more like an art piece of some kind. Her speakers are little things tucked up in the four corners of the room at the wall and ceiling junctions. that does some good things (no comb filtering from wall reflections) and some strange things, as you can imagine. She has a setup she bought used, with a early Yamaha processor that has just one optical input for DD decoding. She also listens to all her music on vinyl- got her a new turntable Xmas 2010. She doesn't believe in prematurely abandoning technology; She still uses in daily life a Palm III from the mid nineties...

[Bent]

my kinda gal...

[sdl2112]

Jon, your project is looking great! I like your mix of materials...right part for the right job...that is what engineering is all about! For some added incentive here's something I've been listening to, a remaster of....

Like you I am busy at work. I am in the middle of designing the power electronics for a 750kVA UPS. The IGBTs might be from a company you are familiar with

[bbggg]

Jon, I was looking at your X-over sims. Surely you're doing an active setup, aren't you? In this case all you'll be needing will be a series cap for the tweeter, for DC and thump protection, and optionally a couple more for the mids, for extra safety. Alternatively you could use capacitor-coupled amps for the mids and tweeter, and conventional DC-coupled stuff for the woofers. All the heavy lifting, ie driver level setting, slopes, and EQ, can be done by the DSP+, I think.

[JimS]

It's [almost] alive :E . . .

Jon, I was looking at your X-over sims. Surely you're doing an active setup, aren't you? In this case all you'll be needing will be a series cap for the tweeter, for DC and thump protection, and optionally a couple more for the mids, for extra safety. Alternatively you could use capacitor-coupled amps for the mids and tweeter, and conventional DC-coupled stuff for the woofers. All the heavy lifting, ie driver level setting, slopes, and EQ, can be done by the DSP+, I think.

I'm not Jon, but he's planning to run passive T/M and active M/W and likely active dipole correction. The major issue here is available dsp space on the LIO for 4x rates (24/192 and 24/176) running that much processing, secondary issue as he noted above is extra amp channels. All active would likely be an option if your sources are CD-quality or 2x rates and if you have +dsp.

[JonMarsh]

Jon, your project is looking great! I like your mix of materials...right part for the right job...that is what engineering is all about! For some added incentive here's something I've been listening to, a remaster of....

Like you I am busy at work. I am in the middle of designing the power electronics for a 750kVA UPS. The IGBTs might be from a company you are familiar with

Now that's a REAL man's UPS! I'm just working on some LLC converter building blocks for offline power and a medium power Class D with our new 200 V OptiMOS. Got a little interleaved PFC that's languishing a bit.

I love that album! Good motivator- just added it to some CD's I'm ordering from Amazon next Wednesday! Now if only HDTracks would get it... Their "Tea for the Tillerman" (new, got the 24/192 version) is great.

[JonMarsh]

It's [almost] alive :E . . .



I'm not Jon, but he's planning to run passive T/M and active M/W and likely active dipole correction. The major issue here is available dsp space on the LIO for 4x rates (24/192 and 24/176) running that much processing, secondary issue as he noted above is extra amp channels. All active would likely be an option if your sources are CD-quality or 2x rates and if you have +dsp.

+1

Please read the very first post. I did a proof of concept of a 2.5 way MTW crossover for the mid woofers and tweeter, which includes a fair amount of impedance correction and frequency shaping. As my "normal" esetup is Quad rates on the input with 174 kHz input from Fidelia, using Izotope SRC, DSP processing power is at a premium. I would not run a straight MTM as the top Bart of a three way with the intended crossover frequency.

I do also plan a MiniDSP solution and an ASP design with PCB's; I may order a small run of bare PCB's if there's interest, considering I can buy four or six as cheaply as one.

[bbggg]

Ok, I reread it more carefully. It's just that six RLC's per speaker seemed a bit excessive but then my experience in these matters is next to nil compared to yours. I guess the concept is pretty much finalised, but still I wonder if a simpler 3-way all-active were preferable if the DSP+ can accommodate the processing necessary.

[dlneubec]

+1

Please read the very first post. I did a proof of concept of a 2.5 way MTW crossover for the mid woofers and tweeter, which includes a fair amount of impedance correction and frequency shaping. As my "normal" esetup is Quad rates on the input with 174 kHz input from Fidelia, using Izotope SRC, DSP processing power is at a premium. I would not run a straight MTM as the top Bart of a three way with the intended crossover frequency.

I do also plan a MiniDSP solution and an ASP design with PCB's; I may order a small run of bare PCB's if there's interest, considering I can buy four or six as cheaply as one.

Jon,

I've been thinking 2.5way on a WMTM I have in the planning stages. I'm assuming you're making the top M the .5way for the obvious distance issues from the low mounted woofers, which is my case as well. The .5way approach also lets you deal with baffle step for a mids crossed below BS freuqency.

I've not yet looked at what it does to the lobing of the MTM. I presume it may tend to tilt it somewhat downward, and you'd have to address that in the XO, though with the second M out of the XO region, it might presumably lead to less narowing of the frontal lobe at the MT XO point, behaving more like a traditional WMT. CTC distance between M's would perhaps be less of an issue as well. It would still maintain some of the vertical directivity benefits on the low end prior to the .5way roll off, so floor and ceiling bounce would still be reduced.

If you did a matchting center channel, a 2.5way laid on its side might offer some horizontal off axis improvements also, I would presume.

Anyway, I'm curious what was your reasoning for choosing this route?

[JonMarsh]

Lessor of Evils?

Jon,

I've been thinking 2.5way on a WMTM I have in the planning stages. I'm assuming you're making the top M the .5way for the obvious distance issues from the low mounted woofers, which is my case as well. The .5way approach also lets you deal with baffle step for a mids crossed below BS freuqency.

I've not yet looked at what it does to the lobing of the MTM. I presume it may tend to tilt it somewhat downward, and you'd have to address that in the XO, though with the second M out of the XO region, it might presumably lead to less narowing of the frontal lobe at the MT XO point, behaving more like a traditional WMT. CTC distance between M's would perhaps be less of an issue as well. It would still maintain some of the vertical directivity benefits on the low end prior to the .5way roll off, so floor and ceiling bounce would still be reduced.

If you did a matchting center channel, a 2.5way laid on its side might offer some horizontal off axis improvements also, I would presume.

Anyway, I'm curious what was your reasoning for choosing this route?

Not to be flippant, but as your thinking out loud points out, there are many tradeoffs involved here, and sometimes it comes down to the "lessor of evils".

For example, why two sevens instead of one 8 with close to the same Sd?

Here, it comes down to the technology available in specific drivers. The aluminum cone Illuminators have some rather different characteristics that I believe will make them easier to use or higher performance than W22 (mind you, I have a pair of W22's, on hand for several years- it's not an academic discussion or cost driven from buying the W22's)

• The Illuminators have a light weight motor assembly with a rather stiff but open basket frame- OTOH, the W22 DO have a pronounced motor/frame physical resonance in a critical part of the midrange, necessitating SL's mounting by the magnet to avoid exciting this resonance
• The W22 has a very prounced upper frequency resonance that is low enough in frequency (4.5 kHz) that filtering it and avoiding the distortion resonance amplification is something of a problem; the energy storage further more shows up as a dip in the response below resonance, and energy storage at lower frequencies in the ETC curve.
• Nearfield measurements of the W22 show out of phase issues due to cone behavior in the 1800 Hz region- note that another 8" metal cone woofer, the HiVi M8a, does NOT have this dip; it is not a function of radiating diameter, but may be related to how a cone without a dust cap structure behaves.
• The Seas W22 is rated for 10 mm Xmax, with a conventional overhung voice coil. The Illuminators have an underhung motor with a physical nominal linear travel of 6 mm but by BL product droop have a working travel of 9 mm for Xmax
• A true MTM would be undesirable, IMO, because of the change in vertical directivity in the crossover region. As you point out, by moving to a 2.5 way mid setup, we have more options, and in effect the vertical source height in the mid to tweeter crossover region is cut in half.
• The additional crossover complexity would have things getting out of hand with an all active implementation, as it would effectivly become a four way.
• A conventional MTM dispersion lobe would be tilted downward as you describe, due to the usual/normal operation in phase quadrature, as well as issues with driver offset. But here the waveguide corrects much of the tweeter time alignment, and the crossover type determines the phase behavior- in this case, the alignment modeled was for the driver array with the drivers offset like an MTM, physically, but with the separate feeds.
• Both mid woofers are operational in the low frequency area, with the same LF cutoff, (which as you notes aids controlling floor bounce and ceiling bounce in the lower frequency range) but the 0.5 has a down tilt in the response; I don't think it should matter if it's above or below, and for coherence and setting the vertical window, it may be better to have the "full range" mid woofer on the top. That I intend to experiment with and evaluate.
• Ideally, the change in vertical height of the radiating size shift over the frequency range so that we have a somewhat consistent, if not perfectly so, vertical pattern. Many a slip twixt the cup and the lip.

So, we'll see how this all turns about- it may wind up being a triumph of development over initial design. I.E., if I made a booboo, hopefully I can fix it as I go along! :B

[Hank]

She doesn't believe in prematurely abandoning technology; She still uses in daily life a Palm III from the mid nineties...

Hmmm...seems like enviably low maintenance. Has she got a sister?

Enclosure is looking good, BTW. It'll be interesting to see your DSP board implementation. Mark is still trying to convince me to go active on the next speakers I build.

[Jonasz]

Ok John, I've probably missed it, but what is the NatalieP Scandinavia? As a Scandinavian I am oc interested! :B

Those dipoles look REALLY nice btw!

[JonMarsh]

Hmmm...seems like enviably low maintenance. Has she got a sister?

Enclosure is looking good, BTW. It'll be interesting to see your DSP board implementation. Mark is still trying to convince me to go active on the next speakers I build.

Hey Hank- thanks for the kind words about the enclosure! Definitely a work in progress.

She does have a sister, but, in this case, the concept "Evil Twin" applies remarkably, or so I'm told. The stories I've heard...

Mind you, Beverly is fairly technical (IBM Systems engineer for 25 years, for example; programmer at AT&T Bell Labs back when I was just playing in Rock bands...)

While her daily driver is a Passat, her "real" car and garage queen is a bright red Porsche 356. She looks at things like my mid 90's NSX as being far too new fangled...

[JonMarsh]

Ok John, I've probably missed it, but what is the NatalieP Scandinavia? As a Scandinavian I am oc interested! :B

Those dipoles look REALLY nice btw!

Thanks!

The NatalieP Scandinavia is actually a precursor exploration/project to the Arvo Pärt Picante; imagine we take the upper driver complement and put them in a sealed box a big bigger than the original NatalieP. I've done one prototype with the 6640, but now am leaning towards the N26C for a real build with custom cabinets. Could probably do a version without the waveguide, but why?

That will be an update to the Nascent concept - again, the N26c tweeter. No Waveguide. I'm may have to rebuild the nice looking prototype front panels, but that's the price of progress. I think with a little luck I'll be able to adapt the originals. Crossover derived from the original NatalieP series/parallel concept.



This system is intended to integrate with a sub or bass bin, but should be fairly decent in a smaller room. Like my bedroom?

[Face]

Here, it comes down to the technology available in specific drivers. The aluminum cone Illuminators have some rather different characteristics that I believe will make them easier to use or higher performance than W22 (mind you, I have a pair of W22's, on hand for several years- it's not an academic discussion or cost driven from buying the W22's)

Jon, I know it's late, but did you consider the W22NY(Nextel)?

[JonMarsh]

Jon, I know it's late, but did you consider the W22NY(Nextel)?

Gosh Darn, I know there was SOMETHING I was forgetting!

Actually, I didn't really consider them - I vaguely remember looking at them when they first came out, and thought, interesting, but I think I would have jumped on them, but maybe I came across some measurement or other data I didn't like...

Oh yeah, it's that impedance wrinkle at 700-900 Hz, and the next one at 1800 Hz. That's indicative of a cone or surround resonance. Lots of people like to tell you it's just surround... I call BS. Why? Go look for impedance wrinkles in the RS180's or 225 in that region... nada. Compare the Paper Illuminator and the Aluminum Illuminator 7" mid woofers- 1.2 kHz glitch in the paper, clean impedance to 2.2 kHz for aluminum, and a barely discernible ripple there. I mean barely...

Same surround, but not the same cone.


Is this as significant as I make it out to be? Maybe not- there's a lot of discussion and disagreement between reasonable people on this sort of thing. But, since the late 90's I've been avoiding mid woofers or midranges that exhibit that kind of behavior, and the miss always seem to be more transparent and less speaker like to me. Eerily so with my current big guys (that big Accuton cone is very well behaved, as is the one in the C79).

Just my 0.02. Everyone's gotta take a stand someplace.

[EdL]

Oh yeah, it's that impedance wrinkle at 700-900 Hz, and the next one at 1800 Hz. That's indicative of a cone or surround resonance.

It seems to run in the family. From Zaph's tests upon the JA8008, which looks to be in the same frame from SEAS as the W22:



It appears necessary to dope the cones...or the surrounds...or both...

Q for Jon and/or CraigJ: Are there links to a better understanding for the correct sizing of a W-frame to a particular pair of woofers? I'm heading that way as well...

[JonMarsh]

It seems to run in the family. From Zaph's tests upon the JA8008, which looks to be in the same frame from SEAS as the W22:



It appears necessary to dope the cones...or the surrounds...or both...

Q for Jon and/or CraigJ: Are there links to a better understanding for the correct sizing of a W-frame to a particular pair of woofers? I'm heading that way as well...

No links, but here's my rationale for governing criteria:

• The approximate throat areas of the w legs combined and the center opening should be equal
• The baffle has to be wide/deep enough to accommodate the driver diameter
• Each leg should be as short as possible to raise the frequency of the peaking as high as possible, trying to get it outside of the working band and minimizing the size of the compensating notch filter
• The height has to be tall enough to accommodate the depth of the driver from mounting plane to back of the magnet assembly
• Last, don't forget to drill the mounting holes before putting it together!

[CraigJ]

EdL,

I used the dipole woofer designed for the Phoenix by Dr. Linkwitz found here; http://www.linkwitzlab.com/woofer.htm and here; http://www.linkwitzlab.com/woofer3.htm As Jon mentioned, if your woofer is a different size than the original Peerless driver, some minor changes can be made. I had to change the height of mine to accommodate my Aura woofers.

I'm not sure I'd make the woofer out of 1/2 ply primarily for its cheap looks.

Craig

[EdL]

Thanks Jon & Craig,

I have a quad of 10" woofers that will be put into service. That is sufficient guidance to get me on the way.

[JonMarsh]

Jon, your project is looking great! I like your mix of materials...right part for the right job...that is what engineering is all about! For some added incentive here's something I've been listening to, a remaster of....

Like you I am busy at work. I am in the middle of designing the power electronics for a 750kVA UPS. The IGBTs might be from a company you are familiar with

Now, this is the remaster of Romantic Warrior I've been waiting for! The last one was an improvement over the original CD, but still not there!

[JonMarsh]

Thanks Jon & Craig,

I have a quad of 10" woofers that will be put into service. That is sufficient guidance to get me on the way.

What 10's are you going to be trying out? I almost went with the set of NS10's that I had, but figured, hey, it's Christmas, time to treat myself. Also, I had another different type of build in mind for the NS10's until I rediscovered I have four NS12's.