2021-2023 Isiris Update

[JonMarsh]

Also, after all the supply chain issues the last two years, and 90 ships in a holding pattern outside LA harbor as of Feb 16 this year, I treat everything like my old Costco model- if you see it, and you want it, and will use it, buy it now.

Much of my current collection of guitars were bought that way, and they'd be impossible to source today.

[JonMarsh]

Progress on the tweeter board squeezed in between other work...






Hopefully I'll be able to squeeze in a little time for starting the woofer board layout also, today. A lot of non-audio stuff to do, though.

[JonMarsh]

Well, it's been a mighty busy and productive Saturday.

And I did squeeze in some time on the woofer crossover layout, and also got another shipment of parts. Isn't that how Saturdays are supposed to go? I also had some time while the wife was out to crank up some tunes on the one working speaker with the first crossover design. Saturdays don't get much better than this!

[JonMarsh]

Made some progress today- stayed focused on the woofer board, in my spare time, for the most part.






It went together fairly nicely.

A reminder, a budget oriented build for this could use the Erse SuperQ laminated steel inductors of the required inductance; I had 3 of the big Solen AWG10 inductors on hand from my earlier order, and since this might wind up being the build I use, I chose to use one here for the test.

Now I have to finish wiring up the other two boards, then will come time to test. Hopefully that will be this week...

[Scottg]

..A lot of non-audio stuff to do, though.

Me to! ops: Seems like I start moving forward only to go backward: :conveyer:

[JonMarsh]

Me to! ops: Seems like I start moving forward only to go backward: :conveyer:

That is SO TOALLY Familiar! and the real estate feels like one step forward, two steps back at time (14 issues with the new house from the 3rd party inspection, some fairly significant)

And then someone messages me about doing a design inspired by the Boressen 05, and I come back with a few suggestions, and then immediately launch off on my own tanget for a possible design using the Epique E180HE, the PTT6.5W08, and the BlieSMa T34A, which eats up a bunch of time, but I learn more things about VituixCAD and learn some things that may get retrofitted to the Isiris crossover work. And I actually may have come up with something interesting that could work well, and with available data, concocted a VituixCAD POC (Proof Of Concept) crossover that looks plausible enough to continue work- and I have most of the drivers on hand to do this, plus some unassembled Denovo cabinets for the woofers. All I'm lacking is time and the high range woofer drivers (PTT6.5W08). And yes, this is technically NOT a three way, but a 2.5 way...

Given that lead up, I suppose I should post the POC SPL curve, at least.






And maybe, for good measure, the impedance curve, because six woofers isn't a trivial task...





Code Name for this project (should you accept the assignment, Mr. Phelps) is "Saint-Saens LR3"

Yup, my weird crossover stuff.

[technodanvan]

Any particular reason you chose the Epique over an Anarchy? I only ask because I've modeled both fairly extensively for my own project (also with four woofers for the low end) and concluded that while the Epique can dig ever so slightly lower in the same box with the right tuning it's probably not worth double the cost. I'd love to revisit if there is a reason to do so.

They are very pretty though, and I have a pair of each of the Epiques to do something with, someday.

[JonMarsh]

The Anarchy's have a lot going for them with the XBL2 motor, but their design has one significant weakness if it is your intent to push out high SPL at low frequencies-- this is the linearity of the suspension versus cone travel, and the linearity symmetry (moving in versus moving out) as well as the magnetic symmetry and typical VC center offset. If you are looking for a part that has good linearity at about +/- 7mm, then you won't be disappointed. If you're expecting true subwoofer suspension performance out to the quoted Xmas, that's another matter.

The suspension stiffness also shows up in the typical FS, about 46 Hz.

OTOH, the E180HE has a typical Fs of 33 Hz, and models very nicely ported or with the right PR. In 1 cu ft, good extension and output to 30Hz is feasible with an RSS265PR.
The SPL day for the E180HE measurements for this POC analysis was taken in a 0.7 cu ft box closed. I was mainly concerned about modeling the 75 Hz and up behavior for crossover development, and I had th box on hand from testing the PuriFi midrange.

I have both 708's and 558's and 554's, and like the smaller Anarchy's for things like center channel designs with high linearity at moderate to high volume output with a limited bandwidth- and a very clean cone behavior, with an upper breakup mode that doesn't even show up in back EMF.

For reference, here is an output response/capability calculation in Unibox for 4 E180HE in a 100L enclosure wired in series parallel, (8 ohm nominal load) with 4 RSS265PR (would be on the back of the enclosure) at a peak excursion not exceeding 11mm out of the rated 14.7mm of one way Xmax.




I do think these parts have some interesting potential, and when you compare pricing to the European brands, it seems very reasonable to me.

[augerpro]

Jon, do you know of any 3rd party measurements with both the Anarchy and Epique? Someone with a Klippel preferably?

[JonMarsh]

Jon, do you know of any 3rd party measurements with both the Anarchy and Epique? Someone with a Klippel preferably?

AudioXpress has measurements of the Anarchy 708 with Kippel, which is what I base my comments above on.

I'm not aware of a detailed review of the E180HE-44 yet, BUT, AudioXpress has a review up of the E150HE-44, the baby version with essentially the same motor.

https://audioxpress.com/article/test-bench-the-e150he-44-5-5-full-range-subwoofer-from-the-dayton-audio-epique-line

The developers of the technology are held in rather high esteem by the reviewer, and I'd say so for good reason.

I'd say the performance is at a very high level out to +/- 10mm...








I think it's reasonable to say this is a fairly remarkable level of performance for this size driver. If the 7" version just replicates it, I'll be quite happy.


Again, based on this, I believe my judgement is correct in using this as a 0.5 driver or a small sub, with extended upper range for a three way.


The final reviewer comment, which seems to be completely justified:

Looking at all the data shows this driver to be in a category all on its own. It produces good low-frequency performance for a 5” driver, and indeed is capable of being utilized in two- and three-way design. Given the overall design and build quality, this is a well-crafted product specifically intended for the compact two-channel, home theater, or studio monitor market.

One other comment about the E150HE-44- the mounting frame diameter is 152 mm, making it about 5.98"- like the Anarchy 554/558 series and some other "favorite" parts of mine, it should really be called a 6" woofer, not a 5.5"- it's sort of cheating and being a bully to get compared in the 5.5" class.

[JonMarsh]

Updated LR3 crossover version

Well, how about I get back to some Isiris stuff? After ginning up the ideas for the Saint-Saens LR3 Design Study (should start a thread soon) I went back and updated the Isiris LR3 crossover version, too, working to simplify and take parts out. Dayamm, you'd think I was in marketing or something... or a bean counter!


Schematic





SPL plot






Reverse Null SPL

[JonMarsh]

And some additional data...


Group Delay and Phase





Filter Gain





And calculated Impedance...







No time for adding comments right now, but main updates are in Midrange crossover and some small tweaks in tweeter crossover.

[JonMarsh]

I did some component availability based tweaks and a few other tiny adjustments...

And ordered more parts, of course!

[JonMarsh]

Great Scott! This is turning into a very busy week! :hb

The parts for the LR3 midrange update are on the way, and should arrive tomorrow. I have started the crossover update on the original 1st version board, and plan to finish it and check it out before moving the speaker gear.


Today is our new home orientation from the builder this afternoon- all the ducks are in a row on the closing balance and we're ready to hit the banks for transfers to the title company. Hooorah!


:banana: :banana: :banana: :banana: :banana: :banana:

It should be time to celebrate fairly soon- closing March 7!

arty: arty:


But then all the moving work from storage and the current rental will commence, and I'm sure at some point it's going to seem like an endless treadmill of work... :conveyer:

That's going to include doing an epoxy coating/finish for the garage floor before moving anything in there, as well as texturing and painting the drywall and ceiling.

But I'll be able to unpack my gear that's literally been in storage for years and get the music network back together, and the big kid components...


:surround:


I guess this is what playing the long game looks like... with the payoff finally near!

[Carl V]

Congrats....
"what a long strange trip it's been" to paraphrase the Grateful Dead.

Fingers crossed, you have in the past Quoted "Many a slip twix the cup & the Lip."

Full speed ahead & damn the torpedos

[JonMarsh]

Yesterday went fine, had a long dinner with our Russian born realtor afterwards; the acoustics in the new place are much better, based on the “slap test”. Doesn’t ring like a church bell, the way this floor plan does with the all wood laminate flooring some have spec’s (like the rental we’re in, or new neighbors down the street in our rental neighborhood from Tacoma that have the same floor plan, but also a more deluxe implementation like ours in Middleton.) ETC curve is no good for music or HT monitoring! but in the new place, it seems pretty reasonable, and that before me installing the RFZ panels I have.

:T

[JonMarsh]

With everything else going on, I did manage to find some time on Sunday to modify the first midrange board prototype, updating it to the most recent schematic configuration for the LR3 design, as posted on 2-27. Obviously, this is not the layout I would use for the "production" boards, but it gives me a chance to recycle a number of parts (mostly caps) and get something together quickly that I can test... soon. Maybe even today, though we expect to hear from the Title company today about picking up keys.




I've ordered parts for updating the tweeter boards (not much there, one inductor, and some resistors) and if things look good and measure right for the midrange board update build, then I'll modify the tweeter board to match.

[JonMarsh]

I've never worked for General Electric, but progress is still one of my most important products! (I have visited them representing my employer, they being a customer)

So, let me say, I'm a happy camper on several levels today, including getting the keys to the new house this afternoon!

But! Back to the Isiris MkIII update with LR3 crossover...


First I checked the impedance by itself...





Looks like what I expected... that's always good! Means I probably didn't miswire something!


Next, lets look at the midrange output quasi near field (18") and see if that's in line with expectations also...





Whaddya know, pretty much on the money, though it looks like a little tweak on the mid peak notch filter would be in order. Easy to try and check later...


but what about the whole enchilada?





Ok, this saying will date me horribly: "Now we're cooking with white gas!"


Now, keep in mind the tweeter section is also scheduled to be tweaked, but relatively subtlety, a new shunt inductor, and a new set of resistors.


Well, being impatient, I put some tunes on through the test system, just using the Vox player, and files I've had on that Mac for a long time. Which I'm quite familiar with.

It's quite interesting subjectively... and it's an improvement.

Now the level of integration I noted between the tweeter and the midrange is present between the midrange and woofers. The old "seamless" bit. And this is apparent in the far field response, and also at the end of our echo chamber front hall, where it sounds better balanced than any time previously.

So, I think things ARE moving in the right direction as regards the development of the LR3 crossover update.

Don't forget, I've got a POC design and boards started for an LR2 approach, which brings the system cost down. Will it work well? Will it hit the same sonic metrics, though perhaps at not quite as high a volume? Eventually we'll see, though with all the moving and prep work to be done the next two months it may be a while- though I can probably squeeze some work in on the evenings.

Sample cuts auditioned:

All or nothing at All (Daria)
Ballad of a well known gun (SACD rip)
San Marco Moderna (Al di Meola)
East River Blue (Spyro Gyra)
Let your Loose be your Lesson (Allison Krauss)
Ectopia (Oregon)
Gaia (James Taylor)

[Finleyville]

Now the level of integration I noted between the tweeter and the midrange is present between the midrange and woofers. The old "seamless" bit. And this is apparent in the far field response, and also at the end of our echo chamber front hall, where it sounds better balanced than any time previously.

So, I think things ARE moving in the right direction as regards the development of the LR3 crossover update.

This is sounding better and better. I am a fan of a three-way system. I just love the idea of most of the vocal range being reproduced by a single driver, unlike a two-way design. However, the designer now has to worry about two speaker transitions instead of just one. I have found few that do it well, and those tend to cost a good bit of jink. But hearing you say the W-M xover is now as seamless as the M-T is music to my ears. I can only hope that this project sounds as good in your new home as this sonically substandard rental!


Oh, BTW, congrats on your new home!!

[JonMarsh]

Thanks Finleyville!

I have fairly high hopes, just because of the difference in the "ambient sound" in the home with the different flooring choices we made! Carpet obviously has advantages in reducing the ETC compared to hardwood laminate.

The inductor for the tweeter update is backorder from PE, so if I want to move more quickly on testing the planned tweeter crossover tweaks, I'm going to need to order from Madisound. Looks like they have the 0.18 mH part I need in a nice Mundorf foil inductor, so time to get going on the browser...

On the LR3 version crossovers, I should prepare drawings like I started for the LR2 version layout and contraction. BTW, those boards are still awaiting time for wiring up and testing.

I'll probably put together a configuration for the LR3 midrange with updated BOM shortly, and for the tweeter network, too.

Today I have a bit of time to get caught up on tasks like that, becuase I need to camp out at the rental waiting for FedEx to deliver the Epoxy-Coat kit and sign for it. I shouldn't complain, it's coming in 4 days earlier than the original promise.

The days lately have been going by in a blur, there's just so much to do- yesterday including finalizing the refrigerator choice and ordering and paying for it, a heavy batch of Costco shopping to get in staples for the next several weeks, a trip to Middleton with items to unload from both SUV's, and a trip to the Middleton City hall to setup the city utilities (water and trash-recycling) for the new place. The gal there was surprised because I'd actually downloaded their PDF form, filled it in using PDF markup on the Mac (it wasn't a form setup for direct data entry) and had it all signed. Most folks she says arrive at the City hall unprepared. Heck, what's not to like, we even get a 62+ discount on some of the services!


the new Den with it's wide double doors is so nice- I'm even pondering the possibility of setting up the "high end" playback system in there, though I have previously designated it for music learning and instruments. Maybe I can squeeze in both... It's a much more usable space, sans closet, than the 4th bedroom in the rental. And Liubov won't mind if I fill it with sound control baffles and what not!

:rofl:

we have a tentative agreement that she will pick out the most "attractive" of my guitars and arrange them on some walls in the great room and/or entrance hall way. It's not like I have that many- only a dozen or so, plus three or four basses. My daughter's father in law has more guns than I have guitars. Draw your own conclusions. My favorite of the basses is the Al Cisneros Rickenbacher... I fell in love with it as soon as I hear about this special configuration.




well, back to work- need to make the most of today's time!

[JonMarsh]

OK, this is just under the category of crossing T's, or maybe more likely, dotting "i", but here is the calculated phase response with the pending tweeter update. Has to be a custom "unwound" inductors; some standard E24 values are just about impossible to buy off the shelf, or not in the configuration (wire gauge and conductor type) that you want. That't why God, or whomever, invented LCR meters.




Parts re-ordered this morning, because the previous order was stalled due to the inductor going out of stock between when I ordered it and when they went to fulfill it, apparently- I always check for stock status.

[ergo]

The project and plots start to look 'scary good'. All this is successfully creating an itch in me to build another Jon Marsh project (if only the work life would allow more hobby time).
Keep up the good work Jon!

[JonMarsh]

Thanks for the kind words, Ergo!

It's been a bit of a trick to fit in working and making progress on this with all the other stuff going on, but it's quite worthwhile for my sanity and for something to look forwards to!

No one ever warns you how much work retirement will be... especially if you move, have to relocate again, and then need to move into your brand new fixer upper (literally- brand new house, first owner, but a couple of weeks worth of projects I'm doing before starting to move in).

Now, an interesting thing about this project and why I'm putting this much work into it, is the potential for folks to do similar but modified configurations with little additional crossover work, most likely just adjusting the mid and tweeter level- I consider the PuriFi midrange and the BlieSMa tweeter to be the core of it. Of course, I'm cheating a bit by recycling my own 9 year old cabinets. But as sturdy and rugged as the bamboo is, it only made sense!

And I have been updating the top module and main drawings...

Now, several possible form factor updates for the bass fall readily to mind, considering just the range of RSS sub woofer drivers-

• Dual RS265HF-8 in a somewhat shorter cabinet, sealed.
• Dual RS265HF-8 with dual RSS315PR passive radiators. This one won't be small, but it can be tuned for a lot of bass extension
• A single RSS315HF-4 woofer with dual RSS315PR
• 2 PuriFi PTT6.5X08-NAA-08, wired in parallel, suggest considering two RSS265PR for initial analysis. This is the new aluminum cone PuriFi woofer.
• A single RSS390HF-4 with 2 RSS390PR, for those with Mad Max inclinations that want to go crazy with Cocoa Puffs!

But I'll be happy to finish exploring the options on this build and get some tunes going in the new place...

Then, ET has been making ominous noises about getting back to the Kurosawa project....

[Finleyville]

Now, several possible form factor updates for the bass fall readily to mind, considering just the range of RSS sub woofer drivers-

• Dual RS265HF-8 in a somewhat shorter cabinet, sealed.
• Dual RS265HF-8 with dual RSS315PR passive radiators. This one won't be small, but it can be tuned for a lot of bass extension
• A single RSS315HF-4 woofer with dual RSS315PR
• 2 PuriFi PTT6.5X08-NAA-08, wired in parallel, suggest considering two RSS265PR for initial analysis. This is the new aluminum cone PuriFi woofer.
• A single RSS390HF-4 with 2 RSS390PR, for those with Mad Max inclinations that want to go crazy with Cocoa Puffs!

But I'll be happy to finish exploring the options on this build and get some tunes going in the new place...

Alright Jon, you need to just stop this nonsense! Lol. How many woofer configurations do you really need?!?

Will you be building all of these configurations yourself to test or is this list just an exapmle of somthing another builder could incorporate into this design?

[technodanvan]

• Dual RS265HF-8 in a somewhat shorter cabinet, sealed.
• Dual RS265HF-8 with dual RSS315PR passive radiators. This one won't be small, but it can be tuned for a lot of bass extension
• A single RSS315HF-4 woofer with dual RSS315PR
• 2 PuriFi PTT6.5X08-NAA-08, wired in parallel, suggest considering two RSS265PR for initial analysis. This is the new aluminum cone PuriFi woofer.
• A single RSS390HF-4 with 2 RSS390PR, for those with Mad Max inclinations that want to go crazy with Cocoa Puffs!

I'm sorry, I'm looking for a dual RSS390HF / 4 RSS390PR version....yes, at 2 ohms.

[JonMarsh]

Alright Jon, you need to just stop this nonsense! Lol. How many woofer configurations do you really need?!?

Will you be building all of these configurations yourself to test or is this list just an exapmle of somthing another builder could incorporate into this design?

Just an example of easy modifications that a builder capable of adjusting L pad values, and can do a basic SPL measurement, might engage in.

You know, spread the skill base!

[JonMarsh]

I'm sorry, I'm looking for a dual RSS390HF / 4 RSS390PR version....yes, at 2 ohms.

Then you, too, might want a pair of Cambridge Edge W so you can bi-wire each channel!

[tktran]

Well Jon,

your design has definitely inspired me.

I'm doing a dual 12" Peerless XXLS 830845 (13mm x-max Klippel verified) in 100L sealed. +/- Room corrected via Dirac +/- DSP.

Midrange: B&W Kevlar FST, custom neo; or Purifi 4-6" midrange, off the shelf, and tweeter TBC.

[JonMarsh]

Well Jon,

your design has definitely inspired me.

I'm doing a dual 12" Peerless XXLS 830845 (13mm x-max Klippel verified) in 100L sealed. +/- Room corrected via Dirac +/- DSP.

Midrange: B&W Kevlar FST, custom neo; or Purifi 4-6" midrange, off the shelf, and tweeter TBC.

Righteous! That should be some good clean fun!

Unfortunately, a lot of the Peerless woofers have become supply chain poster Childs over here, you'd be somewhat hard pressed to buy any of the XXLS-P830845 from the distributors in the 'States right now.


I'm still working on assembling the first test boards for the LR2 versions- it will be interesting to see how they pan out, but that may wait a while as we're in the midst of early efforts for moving things into the new place and doing the garage update work this week.

[tktran]

My best woofer: 32”, 60lbs. Comes in black, white and brown.

[JonMarsh]

I can relate!

I used to have a special "woofer" too, but had to find a home for her after Beverly died... I placed Masha with a nice family with two young children, and spent most of a weekend training the parents about how to care for and handle Siberian Huskies.

[tktran]

When I was living in Vancouver I did come across huskies. They are magnificent creatures.

This is our first dog, our mutt is a rescue dog; and he is adorable.

And now, back to our regularly scheduled programming…

[JonMarsh]

When I was living in Vancouver I did come across huskies. They are magnificent creatures.

This is our first dog, our mutt is a rescue dog; and he is adorable.

And now, back to our regularly scheduled programming…

Masha was a rescue, too. Beverly usually searched for dogs 8-9 years old or older, that others wouldn't take. All of our Husky's were rescuers. Masha was very morose and quiet in the Oroville rescue facility... so Beverly thought she was older.




But when we got her back to Danville and had a checkup with our normal Vet, he said she was only 2-3 years old at most, based on her teeth- and I'd say probably the lower end of that, because she was about 60 lb when we got her, and she grew some more, no fat, and a couple of years later was a monster for a female Siberian, at 75 lb.


OK, as Bullwinkle would say, "This time for sure! (back to our normal scheduled programming)

[Steve Manning]

While we're at it ..... my 90 lb woofer, she's guarding some lbl!

[1Michael]

Who let the dogs out?
Who, who, who, who, who? :P

[JonMarsh]

While we're at it ..... my 90 lb woofer, she's guarding some lbl!


[ATTACH=CONFIG]31996[/ATTACH]

Now that’s worth having a guard dog for!

[JonMarsh]

Ok, "This time for sure!"


Back on topic... got the tweeter board tweaked up with the TA9-6 modifications (that's Test Article 9, version 6) (slow work takes time...)

Now, I have to preface this with the comment that Liubov swiped the last of my packing blankets, because I leave them out on the floor in front of the cabinet to kill early reflections from the hardwood floor... so these measurements are not directly comparable- but I've got some relatively raw plots (1/48th smoothing and 100msec) and some long and short window versions with different levels of smoothing, so as to indicate how the trends are looking more clearly.

And yeah, I did listen... actually, first, with the good electronics, with some FourPlay, and after 3 minutes of that almost didn't bother with the measurements, because I liked what I heard... In fact, I think it sounds better even at the end of the echo chamber, and I have no way to explain or justify that.


But that's not the HT Guide way... or the SMJ way, though this isn't an SMJ project.


The board modifications are only the shunt inductor and the pad resistors.

Not the prettiest board I've ever done, but not bad after two re-works.




OK, first up is the raw warts and all, lots of early reflections at higher levels clearly visible in the impulse response.




Now, with your imagination, and some experience, you can see the general trendlines, but let's make it a bit easier.

Here's 1/12 smoothing with a 20msec window, but that doesn't really get rid of the earliest reflections, and filters the LF a lot.





I really don't care for 1/3 octave smoothing, but a lot of folks use it quite a bit - here's an example, with the time domain window extended back out to 100mec.




Both of the smoothing examples make clear that we seem to have a quasi-near field presence range dip. OTOH, from previous testing, I know that we have HUGE off axis dispersion in that region, and overall the listening tonal balance was fairly pleasing to the ear.

So, at this point, no changes or updates until we get ensconced in the new digs.

However, time permitting, assembly of the LR2 crossover solution will continue- I'm working on the midrange section right now.


================================================== ========================

Several hours later, listening to several albums, Spyro Gyra "In Modern Times" was a particular treat... that LR2 crossover is about 3/4 assembled!

[Finleyville]

Both of the smoothing examples make clear that we seem to have a quasi-near field presence range dip. OTOH, from previous testing, I know that we have HUGE off axis dispersion in that region, and overall the listening tonal balance was fairly pleasing to the ear.

So is that 1k-4k dip something to be worked on or will you be keeping that response because it "just sounds right?" Also, for the layman (me!) what does that off-axis dispersion mean if listening primarily on-axis?

[JonMarsh]

Good questions!

Yes, you may be listening primarily on axis... but unless you have your listening space setup like a LEDE studio (Live End Dead End) the reverberant power response including the balance of off axis to on axis has a might big influence on how things are perceived.

I helped design and build a couple of small recording studios in Boulder back in the 70s, and this was something I investigated a fair amount back then, especially the trade offs in speaker baffle configuration and driver range versus dispersion.

A lot of conventional box speakers in that day and age were literally designed using mainly on axis measurements in something close to an anechoic chamber. The problem with that is what happens off axis for different drivers and designs in a real room. A speaker with fairly flat on axis response but irregularities off axis, especially a roll of in the presence or high frequency range, will be perceived quite differently than one which is fairly consistent with both. That, for example, is why JA at Stereophile tests for that.

Now, another point to bring up- BBC dip. This is the reputed dip in the presence range of commercial BBC style speakers like the LS5/9.



This is in someways not dissimilar to the voicing filter discussed in other posts here, employed with good results for various speakers including the Wavecor Ardent, but it restores the top octave level on axis- possibly because many drivers have dispersion and power response issues above 5 kHz or so.

A lot of the effort put into some designs I worked on in the 70's was trying to optimize the on and off axis behavior throughout the midrange to top end, and using faceted baffles to accomplish that. I gave a set of those to Charlie Hansen, who worked for me while he was going to school at CU in the late 70's, and he returned a "favor" by introducing me at times as the grandfather of Avalon.

Note that I mentioned I don't plan to make any final crossover decisions until I have these in the new house to listen to and work with = one that has carpeting, and will have RFZ wall absorbers and some diffusers.

In this "reverb chamber" we are renting, with a hardwood laminate floor and hard walls and ceilings, things are rather lively, and the measured response as I show above seems to be the key to having far field sound on the money and the overall acoustic being tolerable. And sitting right where I measure it (rather "near field" for a speaker like that) things resemble one channel of a high end headphone, as I mentioned, so I think that this is a configuration which I'll keep on tap and may recommend consideration depending on your listening setup.

The baffle setup (from the Avalon Isis) and drivers chosen for the Isiris give it remarkably lucid sound even 90 degrees off axis. The T34 gets soft there, of course, but the rest holds up very well.

This illustration conveys the basics of LEDE, EXCEPT that the main monitors should be build in to a diagonal baffle at the intersection of the front wall and ceiling- little box speakers on the console are just to reference that with more mundane speakers, the mix is still artistically and commercially viable.




For a music listening room, I would recommend a Cardas style speaker placement based on golden means, and at least the use of wall absorbers like the RFZ panels placed so that from the listening position, if you had a mirror at the panel, you'd see the speaker on the same side. This is to reduce early reflections that cause comb filtering in the response- such as you see with my first measurement below, even though near field sound should be dominating due to the short distance from the speaker system.

The more that you can make the room act like a LEDE setup, you may find the imaging improves quite a bit, because you can hear the acoustic of the recording, not your room. I had a setup in a house in Boulder that the Sound Gallery receptionist and I were renting that was fairly magical once I got it setup right- imagine having the equivalent of two living rooms, with a very side opening between them and sliding doors to close it off if you want. Well, pull the doors open, put the speakers just in front and inside of the opening for the doors, and put all your heavily absorbent couches and chairs in the rear room... it was like a Cardas setup, without even having worked the details out, and with a bit of isolation of the rear wave. Symphonic works and chamber music recorded properly (not multi-mic'd) were just amazing.

[Finleyville]

Yes, you may be listening primarily on axis... but unless you have your listening space setup like a LEDE studio (Live End Dead End) the reverberant power response including the balance of off axis to on axis has a might big influence on how things are perceived.

I love talking room acoustics!

Alright then, so because designers know that any speaker design will probably not be placed in an ideal listening environment (LEDE), they have to plan and account for not only a level response on-axis but also off-axis to account for first and second-order boundary reflections. Honestly, this seems counter-intuitive to me since logically, I would think designing a speaker with linear but a greatly reduced off-axis response would help its sound by reducing the energy levels one would hear from those reflected surfaces.

Sorry off-topic, but your Sketchup rendering of a listening space seems like it is ideal to have absorption panels to cover the first reflective points in front of the listener and then mostly diffraction panels behind the listener to scatter the remaining back wave and secondary reflection points. Is this correct?

[Gerald Jansen]

Dear Finleyville,

I like your subscript "BE ALERT! The world needs more lerts."

Kind regards, G

[JonMarsh]

Yeah, it's not my sketch, but it mostly gets the basic idea right. And yes, Live End (at the rear) and Dead End (at the front).

Let's keep in mind this is a studio concept that has been around a LONG time, and it's not necessary the last word in how to do a home setup, but some of the basic ideas are good to keep in mind.

Now, as to that limited off axis dispersion, one type of speaker can do that well over a wide frequency range if designed properly- a dipole. Planar magnetic or electrostatic, or multi-way dynamic, like the Orion and others. They have their own challenges, but will minimize the front room interactions.

With a normal/typcial box speaker, you've got a problem in that the bass through low midrange will be pretty omnidirectional, but the upper end won't - so if you restrict that off axis a lot (some designs try that) then you do have an odd tonal balance in the reverberant sound field-

An interesting example of a box speaker that does it's best to have it's cake and eat it too, in that regard, is the Dutch & Dutch 8C. It embodies a number of principles I like, but also somethings I don't like (built in DAC's, external clock synchronization issues, and limited capability from the rear facing LF drivers).







It gets controlled front dispersion by using a cardioid loaded low midrange/midwoofer and waveguide tweeter- crossed over at about 1250 Hz- the rear woofers are intended to be put close enough to the wall to operate in pressure zone below the room Schroeder frequency (typically between 100Hz and 200Hz, depending on the room) and are cut in at 100Hz.

Here is a link to a pretty comprehensive and fairly positive review. Note that the plotted output levels are scaled, and are actually a lower level than what is plotted. It's explained.

https://www.audiosciencereview.com/forum/index.php?threads/dutch-dutch-8c-review.21016/

Stereophile's review is here:

https://www.stereophile.com/content/dutch-dutch-8c-active-loudspeaker-system

Kalman's comment about setting it up is revealing:

After this education process, and having cleared the other speakers out of the way, I repositioned the 8c's and Sound Anchors closer to the front wall, positioned as D&D recommends. I switched to the AES3 input (and, ultimately, to Roon via LAN); there's no justification for using an analog connection between my digital server and these speakers—they'd redigitize the signal anyway.

So, if you are happy with their DACs, this might be an interesting solution- at moderate or lower SPL. Hoffman's Iron law always wins, and with a cabinet this small going for this much extension, and with a single driver Diople, SPL level and distortion suffer a bit.


Earlier reviewers have had problems with things like the networked input, and with synchronizing the clocks between the two speakers with networked audio.


Bottom line for me, I think it's an interesting concept, but I wouldn't use all the DSP, I'd use a different driver complement (such as CSS SDX10's on the LF; got them in storage) and I'd use a different driver the dipole and a true waveguide and high performance compression driver (all tucked away in storage, though some bits are now at the new house stacked up in the dining room area). And it would be quite a bit bigger, of course... for $12,500 a pair, I'd take a different road. Would it work better? Who knows? Some day I will, hopefully. It's the journey, as much as the destination. And I'd still be able to use my TotalDAC or Terminator and Brainstorm DCD8 and Rubidium reference oscillator...

I'm lucky to have the opportunity to explore options some day... as long as my "long game" pans out.

Many a slip twixt the cup and the lip, you know!

[Scottg]

I'd started with LEDE (more than 20 years ago) - and found it to be completely wrong for a small-room acoustic. ops: Well, at least for 2-channel listening. Multi-channel is often better with the "front" (3rd or so) of the listening room with more absorption at higher freq.s..

LEDE is really there for sub-optimal mixing "control rooms": the kind where you don't have much space and you tend to create a quasi-in-wall speaker monitor system (basically shoving your mixing console up against that monitor wall). The effort of course is to improve lateral tracking, but ironically most of those setups still had the console itself as a large very near-field reflector (even with typically poor absorption over the front of the desk) screwing-up lateral tracking. Even then the LE portion (of LEDE) is a bastardization of Concert Hall Acoustics - as if you will generate an improved spatial characteristic (ambiance) with such short reflections in a small-room context (..which doesn't happen). At best LE in a small mixing room makes the room sound less disturbingly "dead".

By contrast, in a 2-channel listening system the best result is often the one where you can pull your loudspeakers as far from the "front" wall as is practical (..the wall behind the speakers). Space the speakers about 7-9 feet apart and make sure the side-walls generate the same sort of reflection near the near speaker (ex. L-wall L-speaker), and the listener only a bit longer than the 7-9 foot distance from either loudspeaker (..almost a perfect/equilateral/equilangular triangle). Only a few feet forward of the loudspeaker (to the listener) should absorption start. It's almost the polar opposite of a "LEDE" room.

Of course don't take my word for it - always try-out it out and "see" (if possible), starting with just the loudspeakers/listener placement (and if preferred, potentially adding absorption as mentioned). If however you are "stuck" with shoving 2-channel listening speakers against the "front" wall - then LEDE might be the best option. ops: Better-still would likely be quality in-wall or controlled directive (as low as practical) like the Dutch & Dutch.



Note: a LOT of the analysis that Toole and Olive have done was with a comparatively "dead" *front wall with the loudspeakers fairly near that wall. It completely skews much of their resulting design conclusion (the most obvious ex. being dipole loudspeakers and their resulting low preference).

*..and it probably got worse with the "turn-table" for demoing multiple speakers in a single testing sample.

Among many other test situations, I rather like Hawksford's - his idea was to simply move the loudspeakers as far from any boundary/reflection-point as possible in a small-room context and ended-up with a test setup (minus the absorption near the listener) very similar to what I've described (..though I *think* it was flipped with respect to long-wall vs short-wall orientation - just to get the left speaker further from the left wall and the right speaker further from the right wall).

[JonMarsh]

By contrast, in a 2-channel listening system the best result is often the one where you can pull your loudspeakers as far from the "front" wall as is practical (..the wall behind the speakers). Space the speakers about 7-9 feet apart and make sure the side-walls generate the same sort of reflection near the near speaker (ex. L-wall L-speaker), and the listener only a bit longer than the 7-9 foot distance from either loudspeaker (..almost a perfect/equilateral/equilangular triangle). Only a few feet forward of the loudspeaker (to the listener) should absorption start. It's almost the polar opposite of a "LEDE" room.

And this is very similar to the Cardas setup, which I recommend all the time.

My main point is that early reflections and comb filtering are a big problem, and unfortunately most of the pictures I see of HT Guide users setups will promote that, and won't begin to realize the potential of their system. But a key issue remains in that critical early reflections in a small room are at much higher levels than for a large one, and have greater impact. This is why using the mirror trick (place asorbent baffles on the walls and ceiling where you can see the speakers from the primary listening position) can be so helpful. RFZ (reflection free zone) baffles from vendors like RealTraps can be very helpful- placed correctly. This picture shows baffles, though not optimum positioning for conventional home listening. I have a number of these in storage...



Having well designed speakers is important- but a well designed environment is just as important, IMO.

[Scottg]

..another rambling post from me (sorry): ops:

I actually haven't experienced much audible contribution of higher freq. room direct reflections other than loudness. (..impulsive high freq. sounds with a really bad room having "slap-echo" would be an obvious exception (..beyond/lower-freq. than that are floor/ceiling bounce and modes that do of course).)

I would note though that it is perceptual: most minds just *disregard higher freq. reflections other than their contribution to loudness (in a small room context), BUT there are a few people out there that have problems doing that processing - even some of whom have quite literally altered their hearing by focusing on those particular aspects. (..I remember one of the forum members on diyaudio having exactly this problem and he does professional mixing.)

*the most notable exception here is moving into an unfamiliar room for the first time - that can take a bit (usually less than a minute) of time to acclimate to.

Even within the "summing localization" region <2ms (ex. left loudspeaker near left wall at 1.5 feet and right speaker near right wall at 1.5 feet), before the precedence region, those reflections very near the average tend to just alter loudness (assuming the near-wall that's reflecting isn't altering the angle with a non-flat shape ie. diffusive vs. specular) making the loudspeaker's output just sound louder rather than generating any audible combing.

HOWEVER, those reflections off the near wall in such close proximity (and notably intensity as a result), to the loudspeaker will not only reflect at higher freq.s from the loudspeaker itself (as a secondary reflection), but also diffract (a bit lower in freq.) around the profile of the loudspeaker - altering the diffraction result of the loudspeaker (or its **"signature character"). This (from my experience) has resulted in some alteration to imaging (though the shape of the loudspeaker and its lack of directional behavior vs. freq. is key). Note: This is a good area to experiment with - spot-"padding" the wall near the loudspeaker vs. just "padding" the area between and a little closer to the loudspeaker (relative to the near wall): with that "divider" approach don't move the padding divider further forward to the listener than the loudspeaker's baffle.

Still though, in most instances I setup with loudspeakers at least 3 feet away from their near-wall and rarely have any negative substantivly-audible contribution, though again - a "padding"/high-freq. absorptive divider can sometimes be audibly advantageous.

**it can also be interesting to try padding on the loudspeaker's exterior side-wall closest to the listener.

(..for a broader HF band absorber I prefer Ultratouch Insulation: a cotton fairly loose "re-bond" recycled product, often with some sort of frame and loose-weave fabric cover.)

With respect to floor-ceiling reflections at higher freq.s - with carpeting on the floor (in-between the loudspeaker and listener) I tend to have no problem with that (..it's often also a very near reflection - but it depends on the height of the loudspeaker). The ceiling's contribution I don't seem to notice unless it's a large space with and irregularly shaped ceiling (or the height of the loudspeaker is such that the higher-freq. drivers are closer to it than the floor).

This is all an area where I largely ***agree with Toole's research (and lot's of others like Ando). A dissent (or perhaps more of a deviation) here would be from the late great Linkwitz.

***I like less of an angle from near side-wall reflections toward the listener - maybe as much as 30 degrees (with respect to Toole's graphical displays on the subject), and I don't like much in the way of contralateral side-wall reflections toward the listener (left speaker reflecting to right wall closer to the listener and right speaker reflecting toward left wall closer to the listener). This is why I suggest side-wall padding/absorption starting a few feet "in front of" the loudspeaker. This really isn't about combing but rather loudness/intensity with respect to listener perceived stereo "channel separation" perception at higher freq.s. (..a.k.a.: "cross-correlation" depending on the freq. vs. "head shading".) It's all preference, and it's always a good idea to experiment with that total reflection angle as it moves forward of the loudspeaker to the listener: for some like me - a few feet from the loudspeaker, and others - several feet from the loudspeaker (before starting absorption on the side walls) to near or exceed a 60 degree angle.