Arvo Pärt MkIII "Heavy Metal" Testing...

[Asterduc]

VERY odd looking impedance curve- could be a signe of box wall vibration.

Distortion would be very interesting to see...

BTW, in the DOS version of CLIO4, you can print to a BMP file on disk; this is the easiest way to get graphs out of the system. It was only with the most recent 6.5 update of CLIO WIN that I've been able to stop using the DOS version.

~Jon

Sorry Jon, I am quite limited in documentation. The system gives me curves on the screen and that is what is most important to me. Maybe got to upgrade ones, but hey, as long as it works to me ...

You are right about the wall vibration. The driver is currently installed in a composite test enclosure that I made couple years ago to do some testing on a system, called the Aquarius. This 8 mm thick composite cabinet was way to flexible.

Clio: I'll see if I can do an upgrade to at least 4.0
I thought to remember that it is possible to do a "print screen" in W95, but I am strugling to get it done.

Ed.

[Asterduc]

Jon, I think the latest impedance curve is in free air while the FR is in the box. Strange looking beasty.

That is right Dennis,

Attached are the 2 impedance curves.

VFMA26IM is in box
VFMA26I2 is in free air

[jdybnis]

Here are the last two datasets imported into Soundeasy for display.

[JonMarsh]

Still rather curious....

Well, in comparison to what I'm going to be spending on Praxis and another mic in September sometime, getting one or two of these to test is pretty cheap. Could be interesting...

I wonder, could it be a large spider with standing wave in the spider? Surround resonances are usually up higher. There's something causing some additional mechanical resonance...

Now, if you connect two of these 4 ohm drivers in series, you'd have the same nominal sensitivity, 83.3 dB/2.83 volts, but twide the power handling and half the current for the same voltage. Connect two sets of series pairs in parallel (four total), and you'd have a 4 ohm load with about 89 dB/2.83 volts sensitivity. Hook up another set of two in series in parallel, and you'd have around 2.5-3 ohms nominal- doable with many solid state amps, and maybe another 2-3 dB sensitivity. Hmmmm. Impedance rise isn't too bad to 500 Hz. The thing that's most attractive about these is the Qts and probably low weight with only a 3 kg magnet. I'd like a longer Xmax than 11-12 mm. But it may not be a deal breaker. OTOH, the price isn't all that attractive at $118- about what the Dayton RS 10" HiFi driver will come in at.

http://www.madisound.com/images/product/MA26WR09-04.jpg

[Dennis H]

I'd like a longer Xmax than 11-12 mm.

Yeah, but Asterduc said the distortion goes up to 10% below 100Hz at low signal levels so they probably can't even use very much of the Xmax they have. Strange sub driver. Clean from 100-1K and funny impedance stuff with high distortion down low. Go figure. Maybe the car guys like that bass distortion.....


THUMP bump bump Bitch, THUMP bump bump Ho'.... :driving:

[JonMarsh]

Yeah, but Asterduc said the distortion goes up to 10% below 100Hz at low signal levels so they probably can't even use very much of the Xmax they have. Strange sub driver. Clean from 100-1K and funny impedance stuff with high distortion down low. Go figure. Maybe the car guys like that bass distortion.....


THUMP bump bump Bitch, THUMP bump bump Ho'.... :driving:


:rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl:

[Aaron M.]

Hey Jon,
You've said something more than once about a dipole line source...Saint Something...I forget.
Anyway, I have two dozen Focal 5V4411 midbass thingies on the shelf that could be the midrange core for such a system. Woofers for dipole and line source use have been explained pretty well so far, but tweeters suitable for line source use seem hard to come by.

It doesn't look like domes are really usable, on paper anyway, planars I'm told sound like garbage and ribbons look to be up to the task...for a price.

If a dome is an OK thing to use, I was looking at the Vifa d26nc-55, it's supposed to be really great, but I'd need a ton of 'em to make a line...Any other possibilities?

I know you're thinking of using the Fountek 5" ribbon and that looks great but the cost is a bugger. Does anyone have distortion measurements on a ribbon like that?

Now I realize you've got to pay to play at this level, but nobody wants to blow more money than they have to.

All I want is a tweet with wicked low distortion and a tiny faceplate that can X-over at 2600Hz....for dirt cheap. "I've got a crisp five dollar bill right here...." :P

[jdybnis]

I'm not Jon, but I think your analysis of the tweeter situation sums it up 100%. It's a pickle. Even if you find a great $20 dome with a 2" flange you're going to spend well over $1k on tweeters; you're still compromising on that top octave; and you're in the price range of ribbons anyways.

I think the answer is going to be DIY ribbons. There a few guys on diyaudio.com that have built them. Here's one with a good writeup (http://home.comcast.net/~hendentures/index_files/Page332.htm). It actually looks doable to me. No harder than starting out building boxes.

[Aaron M.]

:blink: Holy crap!
That's really cool! Now if I can just make the time to do that...

[Doug Lockwood]

"All I want is a tweet with wicked low distortion and a tiny faceplate that can X-over at 2600Hz....for dirt cheap. "I've got a crisp five dollar bill right here...."

What about The Dayton 3/4" neo? I understand they were designed to work well in a line source application.

http://www.partsexpress.com/pe/pshowdetl.cfm?&PartNumber=275-030&DID=7
or
http://www.partsexpress.com/pe/pshowdetl.cfm?&PartNumber=275-035&DID=7

They showed very well in http://www.zaphaudio.com/tweetermishmash/

I have a pair of the Dayton 3/4" in an MTM OB, XO at 4K and they are incredible.

"planars I'm told sound like garbage” I would research your assumptions on planers. Every one I have heard sounded very good. As with anything, they have their limitations, and if you exceed them, the results can be disappointing.

Doug

[Aaron M.]

The Daytons spec out very nice indeed, but I positively NEED a tweet that can be crossed over at 2.6K or lower...you see, I already have the mids.

I apollogize for assuming things about planars, but I have nothing to go on accept other peoples word...has anyone posted measurements of inexpensive planars and ribbons on the web? I'd love to see that.

[mante]

Hi Thomas,

Will you give us an update on the latest version of your Arvo PartIII?

Craig

[ThomasW]

:oops: :oops: :oops: :oops: :oops: :oops: :oops: :oops: :oops: :oops: :oops: :oops: :oops:

I ended up having access to a handyman and have spent the summer doing home fix-up and remodeling. The uncut baffles are sitting in the basement awaiting construction of a table saw sled to square them up.

So as you can tell not much progress. The motivation is low since the prototypes sound so good ..... 8)

[AJINFLA]

Taaablesaawww :drool: . Ah well, maybe one day I'll have a garage so I can get one. What's a basement? :rollhead:

Cheers,

AJ(inFLA)

[Dennis H]

What's a basement?

In Florida, that would be a lap pool. ;)

[cjd]

In Florida, that would be a lap pool. ;)

Ahh, this brings back some memories.

Back in the early 80's I visited one of the foremost Harpsichord shops in the world: at the time they had the world record for largest (may still - not sure). Even have a pic of me and my sister next to it and some famous harpsichord player. Beautiful dragon on the underside of the lid.

Anyhow...

The guys that built 'em were kinda anti-cultural in some ways - they despised the stuck-up snobs that often were their clients.

When they were shopping for their new home/facility, they found a place that they just HAD to buy. Smack in the middle of some VERY upscale homes, and it had this beeeautiful large indoor pool. The pool room was *perfect* for assembly of their instruments. So they bought the house and boarded over the pool. :)

Fear not: it did not go to waste. The actual pool was converted to a studio where violins were made. :)

C

[Hank]

Aaron, Jon's got a B-G RD50 that he's going to use for midrange and lower treble and top it off with a supertweet.
jdybnis, DIY ribbons sounds intriguing. I'll read your links this evening.

[mante]

Boy, you guys sure got off track.

I've waited 14 months to start the Arvos, and Thomas teases me about how good the first version sounds. Jon says to wait until the final version is done, Thomas says no hurry, because the "old model" already sounds so good. I'm dizzy and jealous :banghead: Must make Arvo soooon.

Now I know what Hank must be feeling like......

Craig

[ThomasW]

Must make Arvo soooon. What exactly are you waiting for?

All I intend to do is shorten the design Jon posted, so there's less space between the MTM group and the upper 12" driver. This puts the tweeter close to ear level. Other than that everything will be built per the posted data.

[JonMarsh]

well, there are all those other options to consider....

Then, too, some additional cosmetic/design work. I think Craig doesn't want to build a couple of versions (or more) of these like we have.... ;)

Well, gotta get back to writing up Technical Ladder nominations for a couple of my colleagues in San Jose....

~Jon

[JoshK]

As I soon set out to make my Arvo Pärt JKIs¹, I have been spending a lot of time considering aesthetics and how I want to want them to look. I'd be interested in what you guys are considering doing with the cosmetics of yours.

¹used to denote departure from standard Arvos. I.e. planned use of waveguide, SEAS W22s & DEQX fully active.

[mante]

Jon is right; I am waiting for one of the versions to be completed. I am not too concerned with the cosmetics at this point; I just don't want to be changing out drivers. Actually, it's easier for me to build additional speaker "boxes", now upcoming changes in the crossovers and driver replacements are what have kept me on the sidelines.

Craig

[ThomasW]

Craig,

You'll find that Jon's more advanced designs, are like life in general a 'work-in-progress'.

Why?

Because mfgr's keep introducing new drivers that have better performance. And Jon's involved in the relentless pursuit of perfection... :D

Avro woofers: PE Titanic MKII woofers> SonicCraft SCC300 woofers> TC2+2 both VC's> TC2+ single VC> next possibly the new RS woofers????

Arvo Tweeters: XT25> SS9800> RS28>???

Arvo midwoofers: M8a> Vifa Nomex (now currently unavailable)

As a result over the years I've learn to grab whatever completed design is available and run with it. Other incarnations of the design will be 'slightly' better, but certainly not orders of magnitude better..... :wink:

[Jonasz]

Still asking: Is the Dayton RS28 better than the 9800???

[JonMarsh]

Still asking: Is the Dayton RS28 better than the 9800???

That all depends....

The basic sound quality from 2 kHz to ~13 kHz is VERY similar, as one might expect, considering that some of the individual components are so similar in design and probably sourced from the same vendor. The 98000 has a skosh more air in the upper registers. A "good" RS28a will handle oodles more power in the 1 kHz area, and to my ears is more dynamic if you need to run a 1200 Hz crossover. Impedance curves are similar enough, as well as basic response, that you can drop in one for the other.

Then, there's that price difference. I've bought about three pairs of 98000. Let's see, for that much money I could get a dozen RS28a.

Or, instead of a pair of SS98000, I could get a pair of Rs28a and upgrade the crossover caps to AudioCAP Theta film and foils... that would be my recommended direction.

YMMV, all the other usual disclaimers to avoid taking responsibility for my opinion influencing you. ;)
:B

~Jon

[DearS]

Craig,

You'll find that Jon's more advanced designs, are like life in general a 'work-in-progress'.

Why?

Because mfgr's keep introducing new drivers that have better performance. And Jon's involved in the relentless pursuit of perfection... :D

Avro woofers: PE Titanic MKII woofers> SonicCraft SCC300 woofers> TC2+2 both VC's> TC2+ single VC> next possibly the new RS woofers????

Arvo Tweeters: XT25> SS9800> RS28>???

Arvo midwoofers: M8a> Vifa Nomex (now currently unavailable)

As a result over the years I've learn to grab whatever completed design is available and run with it. Other incarnations of the design will be 'slightly' better, but certainly not orders of magnitude better..... :wink:


are your saying the xt25 is greater thant the SS9800? M8a is greater than the Vifa? etc etc?

[AJINFLA]

are your saying the xt25 is greater thant the SS9800? M8a is greater than the Vifa? etc etc?

LOL :alol:

On the remote chance you're not joking, its the progression of drivers from start to :?:

Cheers,

AJ

[jdybnis]

He means he made a design with the XT25 first, then switched to the SS9800, then switched to the RS28. Likewise with the midwoofers: M8a then Vifa Nomex. All are good, or they wouldn't have been chosen in the first place. Since the design started, old drivers have been discontinued and new drivers come out that are better performers and/or offer better price performance.

Edit: I shouldn't have bothered posting :P Thomas is on patrol 24/7 :T

[ThomasW]

are your saying the xt25 is greater thant the SS9800? M8a is greater than the Vifa? etc etc? Nope.........

Instead of typing, first the XT25, next the SS9800, next the whatever...

I used the > as an arrow indicating the evolution through the drive units

[DearS]

Ah....Oh righty then.

[mante]

"1) Arvo Part- FINAL driver evaluations should be completed in three weeks or so- up on the board is shootout between RS28a and Seas H1212 for tweeter duties, Peerless Nomex Exclusive 8830884 vs Seas W22 vs the defending underdog for budget priced systems, the M8a (could throw in the RS225 just to confuse things, but it won't cross high enough)(measure it in detail, anyway?); Woofer shoot out between TC2+ and RS315HF."

Sounds fantastic Jon. So today I went out and purchased your BT3100 table saw for the Arvo project. Am awaiting your "final" evaluations on drivers and am trying to find Ranger 3D board or Medite 3D MDF. Good luck finding more time and thank you for shared knowledge.;x(

Craig

[JonMarsh]

Thanks for the kind words.

I think you're going to really like that Ryobi. Have you been to BT3 Central yet?

BT3 Forum (http://www.bt3central.com/)


Look at the front page, where they have a review of the Jointech Saw train system for the Rybobi BT3X series saws. Yeah, it's a little wacko sounding to have an $800 saw table connected to a $300 table saw, but for serious hobbyist work, it really is that good. I plan to get one when I'm in slightly bigger digs (planning on early to mid next year).

http://www.jointech.com/Graphics/images/sawtrain_ryobi_bt3000_p2.JPG


http://www.jointech.com/Graphics/images/sawtrain_ryobi_bt3100_jst-4232_01M.JPG

[mante]

Yep, I've been to BT3 Central. That's where I found out about the $50 rebate. Kind of like buying Dayton speakers performance wise; 10% off coupon, $50 rebate, plus my credit card co. gave me a 100 dollar HD gift certificate . Well, for less than the price of a drill, I have a very nice table saw. I think you would have been proud. ;)

Thomas, please hang on to those drivers for me.

Thanks a lot, and I can't wait until November.

Craig

[JoshK]

I ordered a Benchdog Table router insert for my Ridgid TS. http://ec1.images-amazon.com/images/P/B00005RHP5.01._SCLZZZZZZZ_.jpg

I can't even begin to tell you how much money this forum has encouraged me to spend. :roflmao:

[mante]

I can't even begin to tell you how much money this forum has encouraged me to spend.

Josh, just wait until you finish your version of the Arvo Part. Then you will want to build the Saint Saens :B

Craig

[Evil Twin]

Josh, just wait until you finish your version of the Arvo Part. Then you will want to build the Saint Saens :B

Craig

Yes, this forum is staffed and populated by Evil people who's only goal is to lure you over to the Dark side of profligate spending on tools, test equipment, wood, MDF, and loudspeaker drivers...

all to fill what becomes an insatiable craving for the next ultimate weapon in galactic domination of home audio... DIY home audio, that is.

Do not forget that being a Jedi, whether light or dark, is about DIY- starts with designing and building your own first light saber- selecting the right core crystal, then the energy modifying tuning crystals, which give it unique color and attributes.

In this galaxy, that ritual seems to have been corrupted to building your own first speaker system... well, the path to the Dark side is many and varied.

[goskers]

When the Vifa Nomex was listed as a potential replacement was this meant to say the peerless hds exclusives?

I have not heard of a vifa with nomex cone.

Thanks

[JoshK]

I think what you mean is the Peerless Exclusives. Nomex are Peerless too and so are the HDS but the Exclusives are the new HDS topline drivers.

[RobP]

I ordered a Benchdog Table router insert for my Ridgid TS. http://ec1.images-amazon.com/images/P/B00005RHP5.01._SCLZZZZZZZ_.jpg

I can't even begin to tell you how much money this forum has encouraged me to spend. :roflmao:


If you dont mind me asking, where did you get that? I really need an insert like that for my Jet tablesaw. Also what do they cost??

Thanks!
Robert P

[ThomasW]

http://www.amazon.com/gp/product/B00005RHP5/104-6334842-3161547?v=glance&n=228013&v=glance

[JoshK]

Thomas is right, that is where I got mine. Its a really nice insert, not cheap but worth it IMO.

[RobP]

Thanks Guys! Its alot cheaper than the one I was looking at over at Woodcrafters.

[JoshK]

Jon/Thomas,

I know that Jon was going to do some measuring soon to determine the "final" configuration of the Arvo Pärt. Myself, I was planning to employ a waveguide on the tweeter much like AJ does and we have discussed in the past. Is this something you would consider for this project if an appropriate WG were made available?

Ed L and I have been discussing offline, prompted from a thread on diyaudio, about making nice waveguides for the H1212. I think his application is going to be different than mine, as I intend to use mine for this project. I am not sure what he intends to do. Anyway, I don't wish to speak for Ed, but I know he mentioned he would make a few WGs for people who can't spin their own. He is eventually going to spin (he has a lathe and ability to spin really accurately) mine out of claro walnut.

We are working on the profile in excel exchanges and then he is going to mock up a prototype out of MDF. He is going to send me the prototype so that I can do some measuring of that prototype. This is going to be an adventure for me because I have never done measuring before, but I've been reading as much as I can. Hopefully I can ask questions and get feedback from this great group here.

I have a DEQX which comes with a calibrated Behringer mic. I am not sure how powerful, flexible or appropriate the DEQX's software is for raw measurements like this but I also have a M-Audio USB Audiophile 24/96 with built-in mic preamp and a laptop. So I am hoping if the DEQX software can't do the measuring that some of the free software I have downloaded, but haven't toyed with much yet, can. Any sage advice before I begin?

I was hoping that maybe Ed & I's work I might be able to turn into a sliver of contribution back to this community for all the great info I have gained here and the great projects I have found.

[JonMarsh]

Josh,

That sounds like an interesting investigation to persue. You might want to get Speaker Workshop if you're looking for freeware measurement tools. OTOH, if you guys can provide a waveguide profile or send one to me to measure, I can take care of that end for you. It will be interesting to see how much gains in LF extension you can get out of the tweeter, as the only real "drawback" to the waveguide approach is making room for it, then tailoring the crossover to it. The "making room for it" will necessitate a somewhat lower crossover frequency, but then it facilitates that, too. Maybe 1 kHz? That would help stay out of the range where the ETC behavior of even the W22 isn't all that hot (above 1200 Hz). Might even be feasible to use a lower order crossover, though that would require the waveguide work to a lower frequency. May be that the best overall approach is the waveguide with a moderately steep network to "unload" the driver before the waveguide unloads.

~Jon

[oneoldude]

SP Tech says,

"The Continuum A.D. MKII™ crossover frequency is set at 750Hz."

They claim the crossover is a LR 4th order.

Their MKII is an MTM with 8" aluminum cone M's and a 1" soft dome T with a wave guide that seems to have about an 8" dia mouth. I have no clue what drivers they use.

Check here:

http://www.4sptech.com/Continuum_AD_features.php

and here:

http://www.4sptech.com/continuum_AD_specifications.php

It sure would be nice to see a difference plot of a T with and without that guide (or any other guide for that matter) to see what the guide actually does.

Seems to be a real low xover. But those speakers are reputed to play REAL LOUD with no sign of T distress. Hmmm.

[JoshK]

Jon,

My thoughts were the same as yours, in order to fit the WG on the baffle the CTC widens (or the tweeter moves more to the side of the mids) which means a lower xo point is needed, but then as you say, maybe the WG facilitates that.

Thanks for the offer to help. I know you are busy like I am, but maybe I will try my luck to measure the H1212 alone and with the steel sound WGs I have and confirm my results or have you repeat one measurements (I send my tweeters and/or WG) to see if I did it all correctly. Then assuming I am not an idiot I can be of more use to the project.

By the way, I don't mean to say the Ed will make us all WGs. Ed is not here to speak for himself, but I am assuming if we can figure it out, the profile needed then we can have them made for those interested in using them.

[JonMarsh]

Jon,

My thoughts were the same as yours, in order to fit the WG on the baffle the CTC widens (or the tweeter moves more to the side of the mids) which means a lower xo point is needed, but then as you say, maybe the WG facilitates that.

Thanks for the offer to help. I know you are busy like I am, but maybe I will try my luck to measure the H1212 alone and with the steel sound WGs I have and confirm my results or have you repeat one measurements (I send my tweeters and/or WG) to see if I did it all correctly. Then assuming I am not an idiot I can be of more use to the project.

By the way, I don't mean to say the Ed will make us all WGs. Ed is not here to speak for himself, but I am assuming if we can figure it out, the profile needed then we can have them made for those interested in using them.

Don't worry, I'm not expecting to have anyone make waveguides for me, I figure we can solve that problem once we know empirically what really works.

BTW, there have been some other posts and links discussing those speakers hear at HTGuide, including the drivers, which for the midwoofers are L series Seas; I don't remember the tweeter, but I think one of our relatively long throw low distoriton low FS drivers like the H1212 ought to work very well. Dennis H had a profile and XL sheet on the waveguide profile, too, it's somewhere around my laptop... ;) YUP, found it, that was the profile design I was going to work from... I believe it was from Geddes concept/calculation, as that's how it's titled.

~Jon

[Dennis H]

I've been having second thoughts about using waveguides on a dipole. Using one would give smoother response at 45-90 degrees off axis in front of the speaker but I'm thinking it would make the total reverberant power response worse. The 45-90 degree stuff would mostly affect the sound due to early reflections and you'd probably want to kill them anyway with room treatments so maybe they aren't as important as the total reverberant field. When we look at that, the dipole bass and lower midrange are -4.8dB compared to an omnipolar radiator. The monopole tweeter on a flat baffle is -6dB because it's only radiating to the front. Not too bad a mismatch between -4.8dB bass and -6dB treble. Now, if we reduce the treble going into the room with a waveguide, we are making the reverberant mismatch worse.

Food for thought....

[JonMarsh]

One of those things that will need to be examined in detail. I recall that there has been some discussion about the total power response of the Orion in the midrange versus treble, but it totally elludes me where that discussion was... included some issues about tracking the off axis sound. Now, let's face it, any box type speaker has problems in the power response area in principle, yet we're usually very sensitive to not having enough baffle step comp and not wanting the forward presence range and treble to over power things- depends on where those early reflections occur, and whether the room is reflecting or actually absorbing (due to wall flex?) bass and lower mids.

A complex topic, but I think optimizable - at least, it's worth investigating the tradeoffs using a waveguide for launch control and distortion reduction through increase of radiation resistance.

[Dennis H]

Now, let's face it, any box type speaker has problems in the power response area in principle, yet we're usually very sensitive to not having enough baffle step comp and not wanting the forward presence range and treble to over power things

Very true. Baffle step is needed to flatten the on-axis response which seems to be a minimum requirement for a good sounding speaker. Tailoring the off-axis response gets more complicated. ;)

Here's a plot of the Phoenix main panel. It's pretty flat up to 30 degrees off axis so that should cover about anywhere you would sit and some of the early reflections. The question then is whether the widened dispersion in the 2-5kHz region sounds better or worse.

http://www.linkwitzlab.com/images/graphics/p6.gif

[JonMarsh]

OK,

Things are starting to "creep" forward again...

I have the Arvo drawing loaded up, will be playing this this more tonight.. in Ashlar Vellum Xenon. Import from AutoCAD went smoothly, now I just ahve to see if I can learn this reasonably quickly, after getting sort of up to speed with the 2D/3D wireframe Graphite. Xenon is true 3D solids with ACIS and NURBS support and all kinds of cool stuff I still barely know anything about. And this is just a trial version, expires in a few weeks, but by then if I like it I can order the real thing from Novedge, which actually does discount somewhat products from Ashlar Vellum, VectorWorks, FormZ, Alibre Design Professional, etc.

I like the drafting assistant functionality in Ashlar Vellum products; heard about it years ago, but only just tried it last month. They ship both Windows and Mac versions on the same disk, and allow installation on desktop and laptop.

I'll report back when I have more meaningful things to say other than "Gee, the pictures are pretty with the sample files!"

http://www.ashlar.com/sections/gallery/gallery-sections/consumer/images/consumer08a.jpg

(Yeah, that's a Les Paul special- design/drawn and rendered in the program, but not by me...)

~Jon

[Aaron M.]

Ahoy there!
A long while back I posted about my interest in Jons' dipole/linesource monster, (Saint something...)

First off, is there a thread more or less dedicated to that idea? I'd like to avoid threadjacking every dang time I post. :P

And now for my question...or statment.
On zaphaudio.com you can find a fresh batch of tests on some planar/ribbon tweeters.
The BG Neo3 won hands down, and this makes me wonder if the BG Neo8 is anywhere near as usable. If so, it would make for an easy and low XO for my line of Focal 5V4411s (12 per side).
Any thoughts or test data on the Neo8?

[JonMarsh]

Hello Arron,

I recall someone else getting some fairly good disstortion data from the Ne8. I have a pair which I plan on putting through the ringer also when I get Praxis setup later this month. GR Research used to offer a line array speaker called the Alpha LS which used the Neo8 and some 6" CSS woofers.

http://www.creativesound.ca/images/product/AlphaLS_lg.jpg


They used a crossover point of 1200 Hz to the midwoofers, which is more reasonalbe IMO than designs like the Linus, which to my ears have some integration problems and issues with the output of the midwoofers being beyond the pistonic region. Creative Sound Solutions in Canada does still offer the Alpha LS kit. ($1995 USD without enclosures). I believe Hank Frankenberg built the enclosures for the original demo's, pictured above.

I haven't started an official thread for the Saint-Saens yet, becuase I wanted to have a bit more driver testing done and a selected driver configuration defined. I have BG RD50's for the midrange line source, and 16 JP2 (to be crossed at ~ 4-5 kHz), and will "probably" be using the RS265HF for the lower midrange/midbass, depending on the measurement results. Other possible configurations have involved LF drivers tuned more for just midbass/midrange down to 80 Hz, and dipole woofers below that. I'm trying to see if I can simplify that, and have also been entertaining the idea of a quasi waveguide speaker to extend the bottom end of the midrange line source in a two way. Maybe that's something to consider as a project for Neo8's.

OTOH, I don't know that I'd think Neo8's any particular bargain compared with the RD50, for example- 8 Neo8 being ~$480 plus shipping, and the somwhat wider range and smoother, but less efficienct, RD50 being ~ $550. As a full range (through top end) driver the RD is a little easier to equalize, with it's cavity peak being in the 6 kHz region, instead of a higher Q one around 12-14 kHz.

These were my original measurements of the Neo8 in a large (very large) dipole baffle on and off axis (15 and 30 degrees).

http://audioworx.virtualave.net/Neo8MLS.jpg

I tended to think of this driver as suited for a WMTMW configuration, perhaps with the Neo3 or a conventional dome tweeter physically in the middle.

-Jon

[Aaron M.]

Thanks for the input,
There is a white paper out there somewhere that suggests the tweeter line doesn't have to be as long as the rest, just so that some part of said line is still at zero deg. (vertical axis) to your ear. So I was thinking about six Neo8s per side.

So far I'm basing my ideas around the two dozen Focal 5V4411s I have. I guess I should see about thier distortion figures before I dive in. Anybody have info on those?

Also, the RD drivers do look nice, people keep telling me they must be paired with tweeters 'cause of lowsy HF extension...I can't hear anything, no matter how loud, above 13.5K...should I even care?

[JonMarsh]

I haven't seen any measured data on that model Focal. I've been a little reluctant with regards to considering a 5' for an array because of the limited Xmax and sensitivity 5" midwoofers tend to have. But maybe with the right one...

I was considering going that way with the Extremis 6.8, and also have some carbon fiber 6" that unfortunately have a bit of a surround or cone mode around 1100 Hz, but were decent sensitivity and Xmax.

You can use a shorter tweeter/midrange array, depends somewhat on the chosen crossover frequencies. The trick lies in getting the midwoofers long enough to develop mirror images with the floor and ceiling so that they behave like a long array. That and dealing with the weight. ;)

There are two facets to the BG top end response. One is the roll off in the on axis response, which occurs above 16-18 kHz. The other is the narrowing of dispersion which occurs above the front cavity frequency, which is around 6 kHz. ThomasW reported that it was a bit of a toss with the Panasonic leaf tweeters- they clearly extend the upper range response, but the imagaing specificity is less, and I suspect the off axis power response isn't as smooth, as he has to cross over about 7 kHz, IMO a little high for crossing to a tweeter array located 5-6" away. At 6 kHz the have wave distance is about 2.25". At 4 kHz, about 3.5". At 3 kHz, about 4.5". 3kHz is lower than what I'd like to cross, but may be doable with one of my CE type networks, then probably a Bullock compromise 4th order on the bottom end of the RD50 to the RS265's at 400 Hz, with some additional suppression on the 10's above 800 Hz. Maybe a notch to deal with the 1.2 kHz upper mode of the 10's.

My real goals are suppressing out band stuff, and getting the response as consistent as possible out to 40 degrees off axis either direction in the crossover points as well as the driver passband. I expect an unusual panel configuration. ;^) At 400 Hz, I could tolerate up to 32 " center to center on the BG and the midwoofers. I don't expect it to be over 24".

Other than that, I'm keeping my cars close to my chest until I see how things pan out.

[Aaron M.]

The 5V4411 has 10mm Xmax (P-P) and sensitivity around 87db.
A stack of twelve should be around 5' 4" tall, not bad.
Also, I'd only expect the Focals to take the place of the RD driver (3K-400Hz).

So to sum up my thoughts:
1) I think your midbass choice is great (Dayton RS subs)
2) Gotta find out if the Focals are worthy (distortion and all that)
3) Must continue searching for worthy tweets (this could take a while)

Tweeter choices for my setup:
Newform R45- looks perfect, but what about distortion...or any specs for that matter?
BG Neo3PDR- Assuming Zaph didn't fudge things, this one just might do it, but how tall is the ribbon? How close can they be stacked?
Any of the tall ribbons- Again, Assuming Zaph didn't make a mistake (somehow I doubt that he did), They are all unsuitable for my Focals...but that's okay, I didn't want to spend that much money anyway!

Any input on these tweeters would be very cool.
Am I missing anything here?

[JonMarsh]

The Neo3 isn't all that tall- the specs are at the PE site, along with BG's white papers.

If your Focal's are 5 mm Xmax, that's pretty good; several of their 5's are only 3.5 mm.

I don't want to make promises I can fulfull (time wise), but MAYBE I could do some tests on one of your 5V4411, if you're mainly interested in FR, swept since distortion, noni-linear multi-tone distortion, and linear distortion (ETC). It depends on your timetable. I expect to be setup again for testing by Christmas time. But I'm not going to have much time off this year, and will be traveling a lot in early Janaury for biz due to CES and other commitments. Think about it, and PM me if interested. I use CLIO 6 and Praxis for measurement software, with a low distortion ACO Pacific 7012 1/2" mic calibrated to 40 kHz and a 4012 mic preamp (200V bias supply, high output current preamp).

Maybe I should pick up a Neo3PDR just to meausre- they're fairly inexpensive. ;)

~Jon

[JoshK]

The closest distance between planars of the Neo3 according to Danny Richie is 1.1". I asked on a thread about his Epiphany design. HTH

[Aaron M.]

Josh K sez:
The closest distance between planars of the Neo3 according to Danny Richie is 1.1". I asked on a thread about his Epiphany design. HTH

If that's true, the ribbon is 2.4" long...it's not the best ratio of frame size to ribbon length I've seen (2.4/3.5), but it shure beats any dome I know of...It just might work.

BTW, could I get the Epiphany link you speak of?

[JoshK]

http://www.audiocircle.com/circles/viewtopic.php?t=23488&postdays=0&postorder=asc&start=20

Not 1.1", a bit less than 1"
"A little less than 1"" ....I read that as more I guess.

[Aaron M.]

Based on that info,
A rough estimate leads me to believe 72% of the total length of the tweeter line will be actuall radiating surface if Neo3s are used...is that good or bad? sounds OK to me.

[EdL]

I want us all to understand them better.
I would welcome some input in the form of requests for prototype waveguides for evaluation. I just today tracked this thread...where have I been?...Yeh, I'd like to know as well, as this seems the place to be. As JoshK said back on Nov. 11, in the interest of knowledge gained about waveguides, give me some parameters to build to for evaluation:

theta
throat depth or max. dia.
radius of curvature for the transition from the guide to the baffle
(Hint: Geddes uses 2" on his speaker w/ x-over @ 900)

Also give consideration to the throat diameter required for clearance of the surround. See the thread by zaph here:

http://www.diyaudio.com/forums/showthread.php?s=&threadid=69597

and the deeper reading...snore...here:

http://www.diyaudio.com/forums/showthread.php?s=&threadid=41178

Hang with it, as the referenced Geddes articles are available for download. See post #10. If you are interested I'll e-mail you the worksheet.

I may be dreaming but I've got one of the MCM waveguides. I would prefer to arrive at a consensus for one profile for evaluation by several. I would make and distribute several singles in mdf to those interested. If the discussion resulting from that determines a direction in which to proceed, well OK!

Who wants waveguides?

[EdL]

I think of the MCM as a "proof of concept" on our way down the yellow brick road. I admire John's work. I wish I were as well equipped as he.
I can provide a waveguide that recesses into a baffle much like a flush mounted tweeter. The backside is flat to recieve the driver of choice, or could include a recess to recieve & center the faceplate. The waveguide itself is defined by parameters called out in my previous post. I'm only limited by the capacity of my lathes, so I can turn up to ~18" diameter over the bed, more outboard.

[JonMarsh]

Ed, once we can figure out exactly what we want, I'm sure several of us will be knocking on your door.... at least, figuratively! ;)

~Jon

[EdL]

Thanks, Jon

[EdL]

Earlier today, in other correspondence about waveguides, I used the term "back chamber". I called it wrong. Should have used "side chamber".

Let me explain. This may have no merit...allow me to try anyway.
I am concerned for the throat of a horn being so far from the diaphragm that undesireable modes result. In an effort to match diaphragm and throat size here is what I envision: provide just enough relief above the surround to not interfere with it's free movement. This requires consideration given to impedance from air movement-that of the air pumped in and out of the "surround space" by its movement. Now, is this impedance altered in any way (for better or worse) by the presence of another air mass adjacent to the surround, on the other side of it (not in the throat)? This area does not constitute a front or back chamber, but a side chamber.

As my correspondent said, it would be good to see what those using waveguides in their speakers are doing. There are several variables here to be dealt with.

If the mount of a tweeter includes a flexible seal, the air gap can be adjusted with the turn of a screw.

I think this is a question that needs to be dealt with on our way through this.

What do you think?

[JonMarsh]

I get what you're "getting at" there, Ed. In practice, I think that while working with the MCM waveguides, or any other, it will have to be specifically designed/modified to work with each candidate tweeter/driver with a minimum feasible "side chamber" volume. It might be something intersting to explore by experiment, but any substantial change to the actual diameter of the MCM waveguide interface would imply a change in the angle of the throat, which should also have some loading effects.

OTOH, the idea with the waveguide is not so high a radiation resistancea as a true horn, so the "sensitivity" of the loading design should be reduced. I'm picking up a copy of SPEAK for myself for Christamas, and I'm curious to see just how much Gedlee offers/supports in wavegude design flexibility in that package, as the versions he's implemented are closer to a conventional horn than I want to use. I'm thinking that SPTechnology has figured out a pretty good implementation (I'm going to see if they're at CES this year), and understanding what they've chosen and why is one of my to-do's.



http://www.4sptech.com/version2/images/products/studio-no-grill.jpg


OTOH, they're new version using the HDS Exclusive is a tad homely, but then they're trying to offer the most bang for the buck, which means Wilsonart laminates instead of wood veneer, and a less refined front panel look. ;)

~Jon

[JoshK]

I have seen a timepiece in person. I ran my finger over the waveguide to get an idea of the contour. The budget version pictured above makes it easier to see what is going on. I imagine this will be a contour that would work well.

[oneoldude]

Ed,

You have mail.

[Steve Goff]

Amphion's version is less homely:

http://www.amphion.fi/kuvat/uusituote/argon2_max5_cherry.jpg

[JonMarsh]

argon2 is a follow up for the most succesfull amphion model to date - argon. Many consider the classic argon to be one of the best two-way mini monitors in the market. Therefore the expectations for argon2 have been especially high. In order to continue the fine argon tradition the follower has gone through painstaking R&D process and everything has been reworked: slimmer, more rigid enclosure, bigger highly precise tweeter and a bass driver with high force factor which increases the speed and tonality of the bass. Naturally also the crossover topology is totally new and now the crossover point lies at breathtakingly low 1200 Hz! Despite the 85 dB sensitivity argon2 is an easy load for any amplifier, as the minimum impedance lies at 7 ohms.


"crossover point lies at a Breathtakingly low 1200 Hz" :rofl: :rofl: :rofl:


Oh, these guys slay me! They should write standup comedy routines.... ;)


But that is a very nice looking front panel- I wonder if it's carved from solid wood like it looks?


http://www.amphion.fi/kuvat/uusituote/argon2_max5_birch.jpg

[oneoldude]

Jon,

I somewhere saw how they build these.

What they do is take strips of wood to make up the front baffle like a butcher block. On the backside there is more wood volume where the guide and back flange are to be. Then when it is dry and stable they use a monster router to cut the guide out from the front. Neet stuff!

If you look at the pictures real close, you can see the butcher block nature of the front baffle.

[Steve Goff]

I've seen and heard the Amphion line, and the tweeter horn looks like it is made of a solid block of wood. Their bigger models also have an interesting quasi-dipole midrange, used to control directivity.
http://www.amphion.fi/kuvat/tuote2002/max/xenon_cherry%202.jpg

[JoshK]

too bad they didn't go the distance and use a decent roundover on the baffle... anyway, not a bad look.

[Steve Goff]

From the NRC testing on the Argon2 done by Soundstage, the results are very good:

http://www.soundstagemagazine.com/measurements/amphion_argon_2/frequency_on1530.gif

http://www.soundstagemagazine.com/measurements/amphion_argon_2/frequency_456075.gif

Listening window:
http://www.soundstagemagazine.com/measurements/amphion_argon_2/frequency_listeningwindow.gif

THD+N at 90dB:
http://www.soundstagemagazine.com/measurements/amphion_argon_2/thd_90db.gif

THD+N at 95dB:
http://www.soundstagemagazine.com/measurements/amphion_argon_2/thd_95db.gif

[mante]

How about Monte Kay's new design?

[JonMarsh]

Don't you mean these?

http://www.mfk-projects.com/images/Seas%20dipole%20project/seas%204-way%208.jpg


I bet the crossover is fun- let's see, looks like W22's, a Seas Excel Mid, Millenium Excel, and RS315HF. 23 op amps and over 100 resistors! Yowsa!

http://www.mfk-projects.com/images/In%20the%20works/in_the2.gif


Lotsa multi-channel amps to power it all; lots of cables and connectors.... Hmmmm.

Well, it should be interesting, but you all know what a confirmed Luddite I am.

Will there be an answer to this shot across the bows? You bet! It's already in the works, and all the drivers ordered. Rather a different approach, I'm trying to do something "smaller". Sort of. ;) Smaller than the current Arvo Part MkII. It will be called the Arvo Part MkIII. I know, how original!

[Paul H]

Looks some some nice little speakers there; 4-way, a couple of dozen opamps, dipole with excels - must be a kindred spirit. ;)

I wonder though why the tweeter is mounted so high - that panel could have been inverted to put the tweeter at a more normal height, without causing woofer to subwoofer separation issues at the crossover frequency that was likely used.

Paul

[JonMarsh]

Many ways that could have been done- Mid and tweeter between the 8's, with tweeter lower for example.

[Beau]

C'mon, you cant give out teasers like that with no details, Peerless mids, RS Woofers?
Beau

[AJINFLA]

C'mon, you cant give out teasers like that with no details, Peerless mids, RS Woofers? ..SEAS neo tweeter, small form factor like the Gradient Revolution.. :W
C'mon Jon, just go nuts like me and mount it coaxially for that really compact form factor ;) .

Cheers,

AJ

[mante]

Is this it?

"Oh, well, you won't want to build any of those, you'll want to wait for the new Isiris! (New, coming soon, better than ever, not too big, either! Well, relatively. Not cheap, though) Should only be a year or two away...."

Craig

[JonMarsh]

Well, I figured, in good marketing think, stick with a "well known brand" moniker, so it became Arvo Part MkIII, seeing as it's dipole.

The proposed lineup depends on confirmation of driver test results, but will probably be as follows (sorry, AJ, no Neo's for now).

Tweeter: H1283, with waveguide. Depends on distortion performance, HF dispersion with waveguide, etc. Alternative may be an Accuton C13-6 (underhung high efficiency motor) if I don't go with waveguide.

Crossover point ~ 2.25 - 2.5 kHz

Midrange: Accuton C90/T6. Probably the low distortion champ of the world in the range from 200 Hz to 3 kHz. Much higher efficiency than I need, but that shouldn't create any problems.

Crossover in 250-300 Hz range.

Woofers: RS265HF. Four of them. Series parallel array. Probably separate crossovers for two 8 ohm series array, BUT, also looking at some other configurations. Only two will go all the way up to 250 Hz. Other two will be cut in lower.

Passive main crossover. Based partly on a melding of the Arvo and the Natlie P design (you didn't think I was going to use that weird baby just twice, in the Modula MT and Natalie P, did you?

Have passive line level equalizer worked out (balanced and unbalanced connections). Should only have about 6dB insertion loss. Just crank the volume up a little higher. Film and foil caps, of course.

NO opamps anywhere near this system (I did say I was a Luddite, didn't I?). No ferrite inductors or electrolytic caps in the crossover, either (shame on you, Gradient). The usual wiring standards and hardware will apply. Film and foil caps in the tweeters. Northcreek inductcors (AWG12) in the woofers.

Will probalby wind up around 84-86 dB nominal sensitivity, depending on how I juggle the crossover and passive equalizer. The raw woofer array and midrange driver sensitivity is 93 dB.

Recommended amplification is something like 100 watts/channel, though obviously more wouldn't hurt. Due to impednace, will need to be 4 ohm stable.

Foot print will be an inverted Superman logo outline, like the gradient, but a bit bigger. Expected height exactly 50". No sloped top module; straight sides in one baffle; will have a wood trim belt between the woofer section and midrange/tweeter. The proto test baffle (partly constructed) is laminated 1/4" HDF, like the M8ta, but the "production" models will probably be phenolilc; already have a source lined up, plan to order in February. Wood trim will be in Pau Ferro (which I have on hand, also).

Yeah, this is a little nutso, but you see, my goal is to remind my friend Chas why box speakers are so last century. ;) Even big expensive ones. Or ESPECIALLY big expensive ones.

So much for keeping this a secret until I have one working. Obviously, they'll never let me work at Apple on a secret projecct....

~Jon

[JoshK]

So just one C90/6 per side will go down to 250hz open baffle with no problems? at really loud levels?

Slightly OT: Is there any method of producing an active line level eq/xo that uses discrete components instead of opamps?

[JonMarsh]

Well, yes, but it's a bit of a pain in the butt. Zero loop feedback buffers and classic 1-2 pole filters, cascaded. It's possible to get really good bandwidth and distortion, but then you still wind up with a complex electronic crossover with lots of cables and amps. Ask ThomasW about what a pain in the keester his four way and three way setups are.

I'm not looking for ear bleed SPLs, just something a little beyond an Orion, for example. The C90/T6 is good for stuff close to ear bleed levels in a closed box system like the Avalon Isis, with very low distortion (it's 93 dB sensitive, and published distortion even at 5 watts is still very low)- that's 99 dB. With a 250 Hz crossover, the C90/T6 should be good to ~110 dB. That's enough for me- 90 dB average with 110 peaks.

The baffle is so large relative to wavelength that it's past the SPL corner where monopole and dipole output capability are equal. It would be a lot different situation if I were trying to cross over at 120 or 150 Hz. Then it would be limited to 90 - 96 dB.

There is a method behind my madness. Only the low inductance and clean upper end of the RS265HF will make this possible- can't think of another woofer I could use in place of it. The RS315HF will come close, but I can't get the size and load impedance combination I want. OTOH, that does leave open the possibility of an RS315HF based "Uber Arvo" which would have to be larger, say, ~ 68-70". That's almost 6 feet. Getting the front panel material for that would be a royal pain... but who knows... might be fun to do a proto at least.

~Jon

[ThomasW]

The (SNRC) Speaker Naming Rules Committee just adjourned and the decision is that the new project as described is non-compliant with the design and intent of the Arvo Part concept. As a result it will be necessary to coin a new name for this design.

One suggestion from the committee was "Na-AP" (Not another-Arvo Part) in keeping with current convention... ;)

[JoshK]

Since it was inspired by the Gradient, how about "Directional Derivative"? :lol:

[AJINFLA]

One suggestion from the committee was "Na-AP" (Not another-Arvo Part) in keeping with current convention
John K might sue.. :W .
Yeah, this thing is sounding tasty, but not too related to an Arvo. If you must keep the name, maybe Arvo Part "Barry Bonds Special Edition" or BBSE, you know, improved all naturally through rigorous work,etc.

Cheers,

AJ

[cjd]

Why not two tens, two twelves? ;)

Interesting, to be sure. Very interesting. If for no other reason than it solves one of the things that's bothering me still about the direction my own dipole thoughts have gone. :)

C

[ThomasW]

Why not two tens, two twelves? It's 4-10's to get the efficiency up.

[JonMarsh]

Why not two tens, two twelves? ;)

Interesting, to be sure. Very interesting. If for no other reason than it solves one of the things that's bothering me still about the direction my own dipole thoughts have gone. :)

C

That's something I've been considering also, Chris, but don't think that I can shoe horn it into 50". MAYBE.... if I was willing to increase the baffle width 2", give up the waveguide (well, actually might be able to keep it, but would need the tweeter underneath the Midrange), and give up the wainscot trim between midrange and woofer area, THEN I could shoe horn in two twelves and two tens, and keep the total height at 50", and be able to build from 48" long material. Just barely.

In that same larger baffle area, I could just fit in 6-10's. Yeah, 6 at 18 lb each. Ouch. The 12's aren't double the sd; it's about 0.026 vs 0.042. At that point I'm looking at an 18" wide baffle instead of 16".

Something to think about...

OK, so the naming committe has spoken. Na-AP doesn't really work for me, so maybe we're back to Isiris.... this whole thing started because of the Avalon Isis, but after hearing a lot of speakers at CES, I decided I wasn't going to build another box.

So, how does the voting go... two twelves and two tens, or six tens? The latter will be difficult to brace well, the former actually works out pretty well... probably no waveguide in either case. C13-6, here we come.

[JonMarsh]

It's 4-10's to get the efficiency up.

I think he meant two tens AND two 12's.

[cjd]

Yes, I meant two of each for a total of four. My personal taste is for higher sensitivity than 86dB (though I'm not sure if that's really based on anything reasonable), so anything that helps there gets my vote - this may put 6 10's as a better option, it may not. I'd also be curious if one or the other reaches down low better, which is the other side of the equation.

On the other hand, I absolutely see the value in keeping the box within standard building material dimensions - something I did not consider when I asked the question. :)

Given the driver cost, I would say that a couple extra inches of height making you buy another sheet of plywood should be a non-issue. Plus, I'm already to the point where my big 3-ways don't seem all that big (at 50x14x24) ;)

C

[JonMarsh]

Yes, I meant two of each for a total of four. My personal taste is for higher sensitivity than 86dB (though I'm not sure if that's really based on anything reasonable), so anything that helps there gets my vote - this may put 6 10's as a better option, it may not. I'd also be curious if one or the other reaches down low better, which is the other side of the equation.

On the other hand, I absolutely see the value in keeping the box within standard building material dimensions - something I did not consider when I asked the question. :)

Given the driver cost, I would say that a couple extra inches of height making you buy another sheet of plywood should be a non-issue. Plus, I'm already to the point where my big 3-ways don't seem all that big (at 50x14x24) ;)

C

Your "big" three ways are positively dainty compared with the JBL's at the Ayre suite. Part of my issues here is my planned baffle material- getting the phenolic sheet means a huge jump in expense and hassle to go above 48". Either the 2-10" plus 2-12" or the 6-10" would be 18" wide instead of 16". (the traditional "Arvo" panel width). Sensitivity is a tradeoff between what I do in the passive crossover versus in the inline EQ. The final choice wouldn't be made until after measurements in the baffle and much wailing and knashing of teeth, I suppose. The RAW sensitivity in the upper bass of the 4-10" version should be about 93 dB, but we have to throw away much of that for LF dipole EQ. Depending. I don't want the inline EQ to be more than 6 dB insertion loss, which means only 6 dB bass boost, shelving off around 35 Hz. That might leave only 6 dB in the passive crossover, which could result in a perceived amplifier sensitivity of 87 dB. I don't want to count my chickens before they hatch, though. I'm not intending to get super duper GD down to 30 Hz like SL does; don't really thing that's needed, but I could be wrong. (has happened before). That makes the EQ requirements a lot more moderate. And drops the cone excursion quite a bit, too.

Weight and construction wise, I'd lean more towards 2-10" plus 2-12". The 10's will run up to 250 Hz, the 12's would roll off earlier, with an extra section on them. Got some different ideas to try out with the actual filter.

Many a slip twixt the cup and the lip.

~Jon

[cjd]

Phenolic? Ahh, see - I missed that link somewhere. :)

Nevermind then. :P

You already have me pondering brand new things today regarding my own dipole ambitions, however. ;)

C

[JonMarsh]

Phenolic? Ahh, see - I missed that link somewhere. :)

Nevermind then. :P

You already have me pondering brand new things today regarding my own dipole ambitions, however. ;)

C

Interesting.

Of course, a fairly credible and much less expensive version could be done with the RS150 or RS180, and similar crossover points. I'm just a midrange fanatic, and was looking for something with more extension on top than a W22, at least a 6-1/2" diameter frame, and similarly low distortion. I thought about W18, but with their upper resonance, seems like you don't really want to take them up above 1800 Hz, and I wanted something that would mate with either an H1283 or a C13-6, which really dramatically narrows the field.

[cjd]

Well, I'm actually still considering the good old Neo8. ;) But I am now thinking that with a little TLC it may be able to work with the RSS265HF. This waveguide conversation has me pondering that with a vertical MTM (not sure what tweet, maybe ribbon, maybe just Neo3). Would help keep acoustic centers coincident as well I think. About 1/4 the cost, and better top end. Because I really want to keep dipole mode up a little beyond where the RS180 wants to go (I think). RS150 might cut it, but for the rear mass issues.

Should all this pondering be in its own thread, I wonder? The Arvo really is its own thing.

The thing is, I wonder if at that point I would be better with two RS225's, two RSS315HF's.

C

[JonMarsh]

All good questions and intersting points... depends on how many crossover points you want to deal with. Anything on top of the Neo8? The Neo8 in my limited experience (I have a few, done a variety of measurements, not distortion yet, though I hear they're pretty good) could work well above, mmmm, 600 Hz. Might be able to take the RS265HF up that high, that's about the upper limit for the RS270 before it get's wonky (ETC). Imagine a line array dipole with those... only problem with the Neo8 full range is it's a bit hot on axis in the 15 kHz area, and soft off axis- looks pretty good in the 15 degree or so area. OTOH, I don't have the PDRs, but the other version... maybe the PDRs are better.

Certainly sounds like it deserves it's own thread. Probably the Arvo MkIII or Na-AP or Gradient derivative or Isiris does, too! :B Once I have something more than speculation, boxes of drivers, and 2D sketchs on my Mac, I'll have to do something about that. (well, technically, I have some 3D driver drawings already, but you know what I mean.)

~Jon

[oneoldude]

Are you talking about 2 10's and 2 12's per side plus the M and T and getting them on one baffle made from a piece 48" by 18"?

[JonMarsh]

Ah yup! Great Scott, look at what Jamo did! ;)


And now a word from our sponsor...

Got Seas H1283s?

http://www.madisound.com/images/product/h1283.jpg

They came in this afternoon. :B

[Dennis H]

Ah those engineer types. Can't even finish building one flux capacitor before they're off designing a new improved version with more jigawatts (jigawatts are way better than gigawatts ;) ).

[Landroval]

Are you talking about 2 10's and 2 12's per side plus the M and T and getting them on one baffle made from a piece 48" by 18"?
A W-frame can fit two woofers side by side. One benefit with W is also the cancelling of forces caused by cone movements. With H-frames like in the Orion, this does not happen.

[JoshK]

Jon, could you take a look at the outer circumference of the surround on the H1283 and compare that to the H1212? Are they close to similar? I mean does the extra large surround on the H1283 make up for the smaller diaphram?

[Paul H]

Plus, I'm already to the point where my big 3-ways don't seem all that big (at 50x14x24) ;)

C


I like to think I've, subconciously at least, inspired some of the huge dipole thinking going on here - I like it :T ;)



I don't want the inline EQ to be more than 6 dB insertion loss, which means only 6 dB bass boost, shelving off around 35 Hz. That might leave only 6 dB in the passive crossover, which could result in a perceived amplifier sensitivity of 87 dB. I don't want to count my chickens before they hatch, though. I'm not intending to get super duper GD down to 30 Hz like SL does; don't really thing that's needed, but I could be wrong. (has happened before). That makes the EQ requirements a lot more moderate. And drops the cone excursion quite a bit, too.

From my personal impressions, testing, playing and modelling, if (when) I build more dipoles I'll likely be targeting no lower than a 40Hz crossover to a separate sub.

This to me is the frequency where massive excursion/displacement is required to reach marginally lower at minimal return on sound quality - your ears may vary ...

I'm betting that once you get into this design you'll want at least 6 of those 10" drivers Jon.

Paul

[AJINFLA]

As I have said before, I rarely use my Rythmik servosubs, since my dipoles extend quite cleanly into the 30's. On the vast majority of music, the dipoles provide awesome clear bass with plenty of oomph. I never sense that there is something missing or they are starting to strain, possibly nearing xmax. But if I'm going to demo or just plain crank it really,really loud, or play something like St Saens, the servosubs come on and run clean below 40hz allowing the dipoles to get rather neighbor unfriendly loud. Nothing new here. SL advises to do just that. Sealed 35-45hz or so, dipole above. Cake and eat it too.

Cheers,

AJ

p.s. Paul, you have the perfect drivers for a compact hybrid dipole/monopole bass section.

[JonMarsh]

Jon, could you take a look at the outer circumference of the surround on the H1283 and compare that to the H1212? Are they close to similar? I mean does the extra large surround on the H1283 make up for the smaller diaphram?

It LOOKS like the surround is so much larger that the outer circumference of it's attachment is the same diameter as the H1212. That's holding them side by side and putting a steel rule on them. It's interesting that the H1283 has a coarser mesh grille. No phase diffuser, either.

The theory being offered by Seas is that the large suspension adds to the radiating area (wouldn't be terribly pistonic, but... then, that's all an XT25 is, surround.) Sort of the best of both worlds. Hard to argue with their supplied plots, and Seas is generally very honest about data, warts and all. I have hopes that this little guy will play out of it's league. (at least, above 2 kHz)

[JonMarsh]

Weellllll, when I was in Lost Wages I also dropped by SVS in case I could run into Ron, but he was tied up. I gave a listen to their dual driver Ultra, which is something I'd been thinking of cloning with my pair of BFD1203's.

The idea is the same as AJ's and Pauls- for music, OK to 35-40- for the gut wrenching stuff, turn on the monopole sub. I'm signed up for the program. Probably could use my Sumo Delilah crossover, too. Might have to tweak one of the capacitor sets to lower the frequency from 63 Hz to 40 Hz. (it's FP selectable on several crossover frequencies.)

Well, Paul, many of us are of the persuasion that if some's good, more is better, and too much is just enough. At the moment I'm now leaning toward the wider baffle, two 12's, and two 10's per baffle- it will have lots of Sd, and won't be quite as insane weight wise, and what the heck, I have the drivers. ;)

[JoshK]

Cool, thanks Jon, then it probably means a WG made for the H1212 will fit the H1283...maybe, just maybe we'll get lucky and get two for one...

[JonMarsh]

Oh yes, it should fit. The plate is the same size, and the mesh hole cutout is the same size, as well as the outer diameter of the surround.

[brucemck2]

For those of us salivating over using this in a home theater setup, could you envision a "matching" center channel that would be under 40" in height ... maybe something without the two 12's?

[JonMarsh]

That wouldn't be hard to do, especially if you're using for HT with bass management and willing to curtail the low end to a more reasonable range (say, 60 Hz?). The large number of drivers being discussed above is because for me this is for my music system primarily (at least initially).

I have used the last generation Arvo's briefly on HT, and I do really like what they do- you get a lot more sense of depth in the mix, things aren't so "flat" or two dimensional, and articulation in the bass and midrange part of sound effects means they sound more realistic and convincing.

I don't know that financially I'm ready to replace my Modulas in HT duties yet ( :E ) but it's an interesting thought! For CC duties it would depend on what your layout works with- whether you need a more lateral oriented speaker because of space issues, or "short" vertical is OK. The latter would be much simpler to do, and to place- I'd consider either two 10's, two 12's, or may be a 10 and a 12, depending on the results I get after measurements in the upper range of the drivers. Could probably get the height down to 36" if need be.

[brucemck2]

Either short vertical or horizontal fine ... I've got a 40" height limitation to clear the screen.

Use mulitple subs rolling in at 60hz 12db/octave, with PEQ on the subs to tame the nastiest room peaks.

I REALLY like big dipoles (and line arrays) for HT use ... used to have large Genesis towers, and they projected a sense of depth and scale that better matched the image on the big screen than my (many) box speakers since seem to do.

Now if I had the guts and funds to build a dipole line array?

[JonMarsh]

Now if I had the guts and funds to build a dipole line array?


Ah, yes, that is the question, eh? I have most of the parts to do that, but have really wanted to wrap up the Arvo series work before becoming entangled in that project. Working name is the "Saint-Saens", which those who have been on HT Guide a while will recall. RD-50 planar midrange,8 Fountek JP2
ribbons on each side, and woofers probably the RS265HF again, though I thought about looking at the Vifa MA26W, though the Xmax is a bit lower (~11.5 mm vs. 14 mm), and the efficiency is very low - ~ 83-80 dB/watt.

[JoshK]

Line arrays are cool, but I think you should really hear a couple to see if you like them. They have a totally different presentation and usually require a further back listening position from the speakers than a point source, imo, to get a realistic "view" of the sound. It something you like or you might not like. If you have the room then I think they can be amazing in their dynamic range and low coloration.

[brucemck2]

I've heard a lot of line arrays, and like them a lot.

I've been thinking of 8 of the Accutons on each side for the mids, with Founteks (array) or an Accuton (single) tweeter for the high end.

In either case would cover lower mids and upper bass with either an array of 10" (?) woofers, or, a 10" at the top and bottom of the arrays.

All crossed over to monster dedicated subs for lowest end.

[JonMarsh]

I've heard quite a few line arrays that are, to put it charitably, not ready for prime time.

Most have issues with maintaining a true line source over most of the working frequency range, and often have crososver points and drivers that would not result in very impressive results if confined to a small two way. Some well known commercial examples, like the Pipe Dreams, don't fare any better.

An example of a "line array" that works fairly well is ThomasW's setup we've evolved over the years. Even though the upper frequency crossover is not completely optimized in terms of the driver characteristics (The RD's should be crossed around or below where the cavity resonance occurs, but the Panasonics can't be run lower than 6 kHz), the results are quite satisfying to most people.

http://audioworx.virtualave.net/BGStatSS.jpg

When I see folks trying to combine line array midrange/tweeter units with point source monople bass units (one or two 8" woofers, or a 10" woofer), I have to shake my head because someone doesn't understand what they're getting into. As you point out Josh, even BG states a MINIMUM listening distance for the BG ribbons for the correct integration of the nearfield frequency response and sound field. Combining a point source bottom end and line source midrange at best can only produce a reasonable balance at one listening distance, due to the fact that the fall off with distance varies dramatically between the two types of sources.

~Jon

[AJINFLA]

though I thought about looking at the Vifa MA26W
Jon I'm interested in that driver myself for different reasons. The question is - has it survived the Vifa axe? There was a test done on it in the april 2004 issue of Voice Coil, but I can't find much on that. Madisound still has what appears to be just the preliminary data. This slightly different version (or perhaps the final data version?) I found: http://www.lautsprechershop.de/pdf/vifa/lautsprecher/26wa550.pdf
At least it has the dimensional specs ;) .

Cheers,

AJ

p.s. still can't see the logic for a line arrayed HT. Unless you have a perf screen with lots of room behind for an identical floor to ceiling line, what do you do for matching center? :scratchhead: A mini or horz. line? :roll:

http://www.openbafflespeakers.com/images/KEFdipole.jpg

[JonMarsh]

Thanks for the link to the newer spec sheet.

You've gotta admit that is the darnedest looking nearfield curve I've ever seen for a woofer that claimed such a low VC inductance- that kind of downward slope is normally only associated with somewhat high Le. The only other driver I've seen look like that is the Extremis 6.8, which has nominally low Le, due to Faraday rings, but still has a downward slope, and also has about the same Xmax AND sensitivity. Maybe a coincidence of motor design? Note the T/S Parameters have changed somewhat also. Lower Fs, lower Qts.

Four of these together might be pretty nice. Not sure I can talk myself into popping for more than a couple to measure for now. Wish they had a little more Xmax for that low sensitivty.

I hear what you say about availability; given what's going on at D.S.T, that would be my biggest concern for doing a design around them. That and rear vent noise (? an unkown factor).

[bratislav]

John,

If I may ; I think it would be a real shame to throw away wonderful efficiency of low bas/mid of your new baby just in order to perfrom shelving in the crossover. That Accuton C-90 (ouch $$$$!!!!) is simply amazing driver.
I know you hate opamps, but there are other ways to do it (by manipulating feedback network in the power amp for example). In any case biamping the speakers is not THAT bad, and gentle low pass filter is simple and can be done completely passive with handful of high quality components.
Jean Hiraga designed all passive electronic crossovers (using capacitors/inductors and resistors) almost quarter of century ago. Having two more channels of quality amplification is a small price to pay in overal scheme of things.

Bratislav

[JonMarsh]

Been there, done that, have the movie rights. ;) Over ten years ago, the last time, and a couple of times in the 70's and 80's.

I understand where you're coming from; I've also done electronic crossovers and equalizers with all discrete buffers, and passive interface circuitry (70's the first time).

Since EVERYONE else from Linkwitz on down is going active these days, why shouldn't I march to a different drummer if it suits me? (I don't want for channels of amplification, between multiple Aragon amplifiers and an Ayre V5).

The other reason is that I have some amplifier projects (complete rebuild of Aragon designs) which I'll be working on soon again, and as with the original Arvo Part, the idea was to have a single fairly full range speaker that I could listen and compare full range amplifier circuits, including power supply behvavior. Yes, speakers have much greater non-linearities in many areas (at least with simple distortion tests), but amplifiers have their own signatures and issues which are quite audible.

And BTW, I don't use a feedback network in my own amplifiers. ;) Not voltage mode, not current mode.

But thanks for your input, and welcome to HT Guide!

~Jon

[bratislav]

And BTW, I don't use a feedback network in my own amplifiers. ;) Not voltage mode, not current mode.

But thanks for your input, and welcome to HT Guide!


Thanks for the welcome ! I think HT is a bit of a misnomer, and many people will miss the DIY goldmine that resides here, simply becasue "HT" is not normally associated with ultimate in sound reproduction.

In any case, the amplifier HAS to have negative feedback implemented somewhere, otherwise we'd be discussing oscillators. It is relatively simple to add few components to alter its frequency response to suit. Most amps will only have two resitors and a capacitor there (to limit DC gain).
But as you say, there's a lot more ways to skin the cat (an awful saying indeed - I have two cats that tolerate me in their home :B )

Bratislav

[Dennis H]

the amplifier HAS to have negative feedback implemented somewhere

Uh oh........ ;)

[bratislav]

Did I say something bad here :confused:

If there's a way to make an amplifier without the negative feedback I'd like to be educated. Even so touted 'feedback free' SET designs are not without feedback, no matter what their sales department tries to (s/t)ell you
(plate to grid voltage forms a natural internal negative feedback).
Not having an overal feedback loop doesn't mean an amp is free from feedback. It resides somewhere, only some are more easy to spot (and modify :B )

[JonMarsh]

Well, your exact words were "feedback network", which would be commonly construed to be a feedback network connected from the output of the amplifier back to some point near the input state, as is common for voltage mode opamps and current mode amplifiers (the opamp like devices using feedback into a transconductance input; often used for specialized very wideband applications).

Solid State Commercial amplifiers NOT having loop feedback networks include Ayre Acoustics, Theta Digital, Dzarteel, and BAT (Balanced Audio Technology). The first amplifier of that type which I know of was designed by a gentleman by the name of Andy Rappaport, but due to limtations of semiconductors and the topologies he was using, had to be biased at some pretty hideously high levels in the output stage to get acceptable open look distortion, and they tended to be thermal handgrenades as a result. The AMP-1 was the model/name, I believe.

The Ayre designs, my own, and the to the best of my knowledge the Theta Digital and BAT only use single stage degenrative feedback at each stage. There is no interstage feedback connection of any type. Some of the Theta models use sever loops for DC stability. Ayre doesn't, I don't.

~Jon

[bratislav]

The Ayre designs, my own, and the to the best of my knowledge the Theta Digital and BAT only use single stage degenrative feedback at each stage. There is no interstage feedback connection of any type. Some of the Theta models use sever loops for DC stability. Ayre doesn't, I don't.


Fair enough. Note that feedback loop transfer function modification was one of the suggestion offered (simplest and perhaps most obvious to yours truly).
Most modern (and some very high touted designs too, including Pass designs, Dan D - a-ka Mr Krell - which includes some, if not most of the Aragons, Levinsons I've seen, all of the VTL, most of AR/CJ), will have easily accessible feedback network for budding experimentators.
Having passive filters inbetween gain stages or pre - amp is also simple, alhough one is then excluded from changing anything in the chain (change in input & output impedances will alter crossover points and/or slopes where it counts).

In any case, I don't want to be seen as preaching to anyone - I had only best intentions :oops:

I'm in absolute awe with work done on Arvo Part (how does one get the 'a with dots' character?). In fact it was inspirational enough to investigate deeper into this fascinating composer - double thanks for that (as well as wonderful alternative(s) to SL's Orion!).

Bratislav

[ThomasW]

I'm in absolute awe with work done on Arvo Part (how does one get the 'a with dots' character?). If you're using a Windows OS go into the Accessories>System Tools> Character Map. Then click on the character you want and hit 'copy'. Next type the word you want and when you get to the letter use the 'paste' command and you get Arvo Pärt

[JonMarsh]

Fair enough. Note that feedback loop transfer function modification was one of the suggestion offered (simplest and perhaps most obvious to yours truly).
Most modern (and some very high touted designs too, including Pass designs, Dan D - a-ka Mr Krell - which includes some, if not most of the Aragons, Levinsons I've seen, all of the VTL, most of AR/CJ), will have easily accessible feedback network for budding experimentators.
Having passive filters inbetween gain stages or pre - amp is also simple, alhough one is then excluded from changing anything in the chain (change in input & output impedances will alter crossover points and/or slopes where it counts).

In any case, I don't want to be seen as preaching to anyone - I had only best intentions :oops:

I'm in absolute awe with work done on Arvo Part (how does one get the 'a with dots' character?). In fact it was inspirational enough to investigate deeper into this fascinating composer - double thanks for that (as well as wonderful alternative(s) to SL's Orion!).

Bratislav


Glad you enjoyed getting into the music, too- he's not particularly well known or touted, but some of the recordings of his works are truly sublime, IMO.

Now, this is coming from someone who likes a very wide range of music, including classic jazz, bluegrass, old progressive rock, acid jazz, early 20th century classical, blues, zydeco, and many others. I either have no taste or very wide taste... ;)

I'm really hoping to have the recent work on Arvo's and pending items finished and documented in the next couple of months. There's a possible mini Arvo, depending on some test results on 830883 and H1283, the latter also with waveguides, and I still have to look at the Peerless 830884's in some detail. Then there's the W22's I'm sitting on now... well, not literally. ;) How all of these work together and how desirable remain to be seen.

Will probably even get a web site up finally (have the tools now, just been lacking time). I'm also pretty excited by the new system; I haven't heard the C90/T6 yet, but ThomasW did at RMAF in the new Avalon Isis. The measured data look very good, and I'm also going to test some C13-6; I think Accuton's new underhung motor structure is a big more forward for them, compared with their older drivers I've worked with.

I just don't want to do box speakers anymore except for cost/performance type designs, like the Natalie P.

~Jon

[bratislav]

I get it. Thanks. I was kind of hoping there's another option (kind of 'add character' like add icon from the list), but cut/copy & paste works too.
Let me try

Б. Ћурчић :T

Magic !

[JoshK]

my way is to hold down the 'alt' key and type '132'.

[bratislav]

Now, this is coming from someone who likes a very wide range of music, including classic jazz, bluegrass, old progressive rock, acid jazz, early 20th century classical, blues, zydeco, and many others. I either have no taste or very wide taste... ;)


Well, I have simple taste - I like good music ;)
That happens to include J S Bach and Tom Waits, W A Mozart and Frank Zappa, Eric Satie and King Crimson, Miles Davis and Clash, Egberto Gismonti and John Butler.
None's perfect :^x

Cheers !

[bratislav]

Been there, done that, have the movie rights. ;)

I've just done something I should have done first - read the archives :oops:

Tons of amazing stuff, and no surprise, passive crossovers mentioned several times, a-ka
http://htguide.com/forum/showthread.php4?t=3311&page=6&pp=35

Apologies for the brainf@#t

My head is still spinning from the amount of information and variety of the projects seen in here ! ;x(

[capslock]

my way is to hold down the 'alt' key and type '132'.

Or use "set keyb gr+", then the ä is two keys to the right of the L :P

[Dotay]

I've just done something I should have done first - read the archives :oops:


A bit off topic, but is there a specific archive section or are you just referring to reading old posts?

[JonMarsh]

I think he means old posts in the forum- they can be searched all the way back to the forum founding, around 2000.

[Dotay]

I think he means old posts in the forum- they can be searched all the way back to the forum founding, around 2000.
Thanks Jon, that's what I figured he was talking about but I thought I'd be sure I wasn't missing the "Holy Grail" of info tucked away in a some obscure section of this forum that I had overlooked for the past year and a half. ;)

[JonMarsh]

When you start tabulating the amount of words written and topics discussesd, it almost gets a little scary- it was only a few months ago I noticed I was in the low 3000 post range- how did it get up to 4,170 already?

"Hello, my name is Jon, and I'm an HG Guide Junkie".

[mante]

Swap out the C-90 for the RD-50 and you have the Debussy. Such wonderful choices. :T

Craig

[TacoD]

"Hello, my name is Jon, and I'm an HG Guide Junkie".

Do you want to talk about it?

[JonMarsh]

Or, as Dr. Phil would say,

"How's that workin' for ya?" :B

BTW, Mante, it's pretty scary that the C90 and the BG RD50 cost about the same.... it's just easier to add the tweeter to the C90....

~Jon

[borderdad]

Hi, I was hoping someone else would ask this saving me from looking like a dumb-ass... :oops: The Oaudio T2's are on sale and I would like to start on the Arvo, but I wanna make sure I'm building the right ones. I have the info (Arvo Data) but when it came out on the thread, the next entry was to not use it. I also have the xover plans put out later. Is either of these OK? Also, the single voice coil sub T2 is on sale and since I think the latest used only 1 voice coil is this a better buy? Lastly, the RS tweet is on sale. This is still in the final plans, right? Sorry for so many questions. Thanks. Kevin.

[JonMarsh]

Kevin,

Your questions are NOT D-A in any way at all. But the Arvo has been a complex story, and is still in some evolution, besides the set that ThomasW owns. His have a passive crossover for the tweeter to mid, and active else where.

Mine have been using the dual Voice Coil TC2+, and with one crossover I used both voice coils (each wired in series, both drivers in parallel), but the most recent experiment has been with single voice coils from the dual VC drivers, as this lowers the inductance and changes the response profile in a way I think is advantageous for a dipole system. However, that also results in the need to wire the two woofers in series, getting an 8 ohm load, which is lower sensitivity on the woofer. I'd partly been persuing that development path because then the crossover might be fairly similar between a TC2+ in single VC mode and RS315HF.

I have never measured a single voice coil model of the TC2+. Some recent data another friend has measured (a well known guy) has me scratching my head, again the issue seems to be consistency in driver types. I've got to check back with him and see what version he was measuring. They didn't come in as well as the ones I took to his place.

So, I'm not sure what to say about the single voice coil version of the TC2+. I've never had one, and obviously I've never measured one. Note the difference in Qts, Qes. The 12" SVC seems a driver designed for ported enclosures, or small sealed boxes. It doesn't have the T/S parameters of the DVC version, and it won't work the same. So, though I like Todd and the TC2+ is a good driver, that version wouldn't be recommended for the Arvo. Not anymore than I'd recommend the RS315HO for it.

The RS28a on sale is a good deal- no change expected there. Only real alternatives are SS 98000 or Seas H1212, or Millenium Excel- the latter two requiring a new crossover. Not done, not planned.

I'm still going to evaluate the 830884 Peerless for mids, and possibly W22. Will likely have crossover versions for all.

~Jon

[mante]

Hi Jon,

I was at Madisound last week, and Larry said they were out of stock of the 830884 Peerless woofer. Do we know if Peerless is going to continue producing this driver? Also, will the Arvo Part Isiris be completed before the Arvo Part II?

Cj

p.s. glad to see your alter ego has returned ;) (I think)

[borderdad]

I appreciate the response. I can live with a work in progress and wait a bit more. Just wanted to jump on some sale items if possible. Thanks again and I'll stay tuned. Kevin. :)

[oneoldude]

Kevin,

Not anymore than I'd recommend the RS315HO for it.

~Jon



But would you reccomend the RS315HF for the Arvo or other dipole use?

[JonMarsh]

But would you reccomend the RS315HF for the Arvo or other dipole use?


Yes, because the T/S parameters are good, and the VC inductance is much lower.

It's a little tricky when these companies bring out different versions of what they're calling mostly the same part number, becuase their behavior is quite different. Look at the RS265HO vs RS265HF, big difference in T/S parameters (Qts is what we're most concenred with, which derives from difference in Qes), and even in Xmax. For the RS315HF vs HO, it's not quite as big, and Xmax is the same. But for a sealed sub or dipole application, you'd want the HF, especially running up higher in frequency. The HO is low inductance compared with most other 12" subs, but not compared with the HF.

On the subject of 8's, that's a pretty good question regarding the Peerless- have you checked with Josh or Adam if they're on order, and what the status is? I got my 884 from Solen, and my 883 from Madisound. I've got another couple of weeks to wait (I'm told) for my C90s, and W22 are out of stock (I've got two I'll be measuring also). It would be ironic if the final "standard" Arvo Pärt Mk II use W22 just becuase they're the best thing consistently available, and price be damned.

[oneoldude]

"Hello, my name is Jon, and I'm an HG Guide Junkie".

Please learn to live with it. :T

We would all miss you greatly if you were cured. :B

[JimS]

Hi all,

First - thanks Jon & Thomas for a very informative site & special thanks to Thomas for his patience & turning me on to some good bass ideas. BFD has made a huge difference and I'll be working an IB into my basement plans in a year or two.

Same questions as above - I'm very interested in building the Arvo's and would like to take advantage of current sale prices. Sounds like the RS28s are a buy at this point. Mids I'll hold off on until further testing is complete. RS315HF appears to be on sale 'til mid-Feb. Is it worth taking a chance at the sale price or waiting until the final version is released? Any other tidbits on mini-Arvo you mentioned?

Thomas - one thought on short-term solution while the design is finalized & I build would be adding the 4 12"s as dipole bass to my monitors & going to box sub at 40-50hz. Would buying CX2310 (with BFD for shelving if needed) be sufficient? Is RS meter enough to dial this in? What would you recommend if sensitivity adjustment needed?

Thanks again for a great resource!!

Jim

[JonMarsh]

The Arvo MkI was a "mini" design in comparison to the current one, using two Dayton "10" woofers that were really 11's, and two 8" HiVi M8a, and the Vifa XT25 (that was years ago, now, wasn't it? And it was "announced" in an April Fools thread, too. ;)

I've been asked about the possibility of a smaller "Arvo Part, and were I to go that way (under consideration), it would probably have two RS265HF, but might go with 2 RS315HF in a smaller baffle. The rest is up in the air.

One possibility is dual 830883 with RS28a. Crossovers at 150-175 Hz and 1.4 kHz.

Another possibility is a single 830883 or XG-18 with H1283 in waveguide, crossovers at 225 Hz and 2 kHz. Probably more inclined to use the RS265HF with that concept, as they should go higher before issues, and this could be the real "mini-me" of the Arvo series. This wouldn't be a very high sensitivity speaker, of course. Would wind up like a compact floor standers with full BSC which wind up around 82 dB REAL WORLD efficiency, not the BS numbers the manufacturers often publish. Which would mean desired amplification in the 75-250 watt range. Have a couple of ideas about how to get that to work better, and improve (lower the impedance) of the amplifier load. Not ready to talk about that, yet. More in the way of back of the napkin things right now.

~Jon

[JoshK]

I think the mini-me could also make for a superb HT speaker where the lowest duties would be done by the LFE channel. With the mini-me, if the cost wasn't as high as some of the uber Arvo's we have been dreaming up, it might be feasible.

If I ever get around to building the add on room to my house that will be my theater (small theater) then I might consider something like this for rears or sides.

[Evil Twin]

Craig,

I will have a longish response to your PM to JM later this weekend- for now we are emptying out the Imperial labs from the accumulated rubbish of the last three years (Princess Elaine's past luggage and retinue) so that the resumption of normal research activities will be possible. There IS a master plan, but too much discussion on the boards may enlighten the Rebellion.

Moff Tarkin's current dipole plans are suffering from an abundance of ideas, and a paucity of established test data. Without Incom and Sein-Seinar labs to call on, and a new Sr. Director to deal with in the Emperor's chain of command, engineering progress has been substantially reduced. This will change.

[JonMarsh]

I think the mini-me could also make for a superb HT speaker where the lowest duties would be done by the LFE channel. With the mini-me, if the cost wasn't as high as some of the uber Arvo's we have been dreaming up, it might be feasible.

If I ever get around to building the add on room to my house that will be my theater (small theater) then I might consider something like this for rears or sides.

That would be even easier to do, and might allow raising the sensitivity somewhat, as in that case the crossover parameters might adjusted for optimum output down to 60-80 Hz, and not caring too much below that.

Yet to complete are FR, Z, distortion and CSD plots on the potential drivers in standardized test conditions. I have a little bit different take on how to do that than Mark and Monte and some others; we'll see soon how valid that is.

Personally, I'm of the same viewpoint as Widescreen Review, in that all corners should be equal speakers or nearly so with identical voicing vs dispersion, and that a phantom center is better than a mediocre center. A shorty mini-me woulds probably be a good center channel for FPTV.

~Jon

[JimS]

Yeah the H1283/830883 you mentioned a week ago was the one I was referring to . . . not the april fools :) (somehow can't get the quote thread function to work :oops: )

If its still in the napkin stage, I'll likely look towards one of the others that's a bit closer.

[JonMarsh]

For Jim and anyone else who has joined since mid 2002, a little bit of history at today's humor link....


Arvo Parrot Introduction (http://www.htguide.com/forum/showthread.php4?t=5264)


Don't say you weren't warned.... there's a reason I have a day job.


~Jon

[Dennis H]

Bump

[JoshK]

good idea making this a sticky :T



now playing: marilyn manson - lest we forget...

[JimS]

Thanks for the history (didn't see that thread before) . . . and here I had you pegged for a marketing guru doing some tech stuff on the side . . .

Actually I have great respect for your day work. I had an opportunity to tour the new 300mm fab in richmond a few weeks ago . . . wow a plumbing schematic there would make your crossovers look straightforward :E

Got tweeters at the nice price :) and eagerly await the rest . . .

[JonMarsh]

Yeah, that Fab started out as White Oak Semiconductor, a joint project between Motorla and Siemens. How much has changed since then...

Now, we're going to carve out the memory operations as a separate entity, in preparation for possibly spinning it off as a different company worldwide. That's gonna cost a few Euro's this year and for a while- for example, we've spent the last couple of years consolidating a lot of our US wide support functions in Cary, NC. But that facility is otherwise 95% inhabited by memory guys and our memory applications group. By July that is supposed to be a separate legal entity, and the people from other parts of our company (like my direct boss, one of our main marketing guys, our subgroup's national sales manager, and SE FAE will all have to go to another facility. July's not that far away.

I'm an Applications Engineering manager for power semiconductors, so sometimes I think my day job isn't all that different from my activities here, which is a little disturbing in some ways... ;)

~Jon

[JonMarsh]

Just a few points and measurements I've alluded to in other threads, sort of OT in those cases.

The Arvo Pärt MkII uses the TC2+, because of it's good nonlinear distortion performance, and a relatively stiff cone with high breakup frequency. When I was testing Ascendent 12's and the TC2+, both with dual VCs, I saw some interesting behavior with regards to whether one or two voice coils were connected. While the TC2+ is a dual voice coil design normally intended to have both 4 Ω VC's connected in series for home use, I have to say the behavior of the driver with only one VC is interesting.

Now, the Arvo baffle design has a deeper than average U, and this gives some rising response in the 150-200 Hz region- on purpose. I've made some measurements on drivers with what I think of as a basic test baffle, 18" wide dipole baffle (about 1-1/2" thick), with a mic distance of 6", so it's sort of near field, but not "tight" nearfield where you'd mostly see the local cone behavior- there's enough distance for some integration.

The following three plots are made under identical test conditions and drive levels- note that the TC2+ are from my Arvo protype, and have been played a while, whereas the RS315HF is brand new.


http://audioworx.virtualave.net/TC2Plus8ohmSeries.jpg

TC2+ with VC's in series- 8ohm total


Sorry that the absolute pressure level isn't calibrated- drive level was 2.8 VRMS. Is the gradual roll off due to VC inductance? Bump at 300 Hz may be due to re-inforcement around baffle panel.

The next graph is same driver, only 1 VC connected.

http://audioworx.virtualave.net/TC2Plus4ohmSeries.jpg

TC2+ with one VC- 4ohm total

As expected, the sensitivity increases due to lower impedance. The LF corner behavior is also quite a bit different (higher Q, I'd say), and note how much less the response above 100 Hz rolls off- probably a function of VC inductance.



http://audioworx.virtualave.net/RS315HFSPL.jpg

RS314HF 4ohm

Here's the RS315HF- not the same Q and LF corner behavior- this driver has not been run in, so possibly that will make some difference. Pretty flat from 50 Hz to 800 Hz or so- without EQ, it would sound lighter in the bottom end, but I bet the lower midrange definition will be fairly good. Will need to do some cluster tone nonlinear distortion and shaped sine burst linear testing as a follow up shortly.

[JimS]

Great to see the testing up & running :T

Slightly confused on "sort of nearfield" testing . . . is the dipole rolloff pretty well captured here or would it be greater at listening position?

Sounds like another fun year at work . . . stability is overrated anyway!!

[JonMarsh]

This is nearfield enough (6") to see primarily the driver repsonse and not much dipole characteristics except perhaps at very low frequencies. The idea with this measurement is to get the driver characteristic, not an enclosure or baffle (within reason), but also to have not the nearfield cone characteristics, but the rough radiated power. At a greater distance there will be more LF roll off, of course.

The working method I prefer is two different measurement approaches- one like this, is for understanding the innate driver characteristics, the other, with the actual enclosure/baffle of the target system, is for executing the system design.

Something I find pretty remarkable about the cone of the RS315HF and RS265HF is that it is very "dead" in the midrange if you tap it- you don't hear any bending modes as a kind of midrange upper midbase thunk sort of sound. I suspect that means it will be very clean handling the lower midrange. We'll see.

[JimS]

Got it. Thanks. Been through SL's site a few times, but dipole response & measurement still makes my head hurt a bit . . . especially on long weekends.

Realize you have a lot on your plate, so just FYI, for us retail-type customers the RS subs go off sale this Thursday :cry: (currently $104 each for the RS315HF, about $80 savings on the quad).

[JonMarsh]

Hey, that is a pretty nice price. :T

Is the sale on through Thursday, or does it end earlier? Mante thought it ended today. You see, I have a project I need 8 for... but I dont' have $900 in my piggy bank, unless I don't take advantage of a limited time offer on my 3D CAD software that will save me about $1500. Sheeeshh.

Well, it will have to wait for now- won't be ready for that project for a few months anyway.

[JimS]

It's one of their monthly specials & according to the flyer the sale price is good through the 23rd. If you're looking at 8 of the 265HF for the Isiris(?), those are even better at $96 each . . .

[Beau]

Jon,
What sensitivity are you getting on average from this setup? Do you think it would be sufficient for a project that doesnt require brain melting SPL's? Looks from the graph that rolloff starts around 50hz,is this correct?

Beau.

[JonMarsh]

What kind of project are you thinking about for the RS315HF? In a dual 12 dipole configuration, SPL output in 40-60 Hz region is about what you'd get from a good long throw (Xmax 10mm+) 10 in a box- maybe a little more. Actually, it would be about double the typical 7 mm throw 10" woofer (say, two 10's in a box) Above that, capable of much more.

The driver is about 86-87 dB/2.83 VRMS. No one will confuse it with a JBL, but it's not bad. The example tested hasn't been run in under power- and in this configuration the Q is about 0.4 open baffle; the roll off rate would be higher at a listening position, and requires some EQ. How much depends on the baffle design- with the Arvo MkII there's actually a rise from 100-200 Hz due to the U baffle, but the passive crossover rolls that, and contributes to some additional low end lift. (well, it does with the TC2+, haven't tried it with these yet).

The real reason I'm measuring the RS315s with such interest is another variation of the Arvo (dont' want to give away the secrets here, look for an "announcement" possibly on April 1) with 4 twelves per side working up to 250-300 Hz, and not as deep a U baffle; much shallower. But a bit wider; basic front panel of 20" wide, 60" tall, with very short angle backed baffles. You'll understand when you see a CAD drawing. There will be a smaller version, too, with the RS265HF, assuming they measure like the prototypes. I thought I'd get done early here today and scoot home, but a problem at Dell has kept me in th eoffice longer than intended, considering how early I get here. Was going to test those tonight before going on my business trip this week.

[Dennis H]

look for an "announcement" possibly on April 1

Oh sure, you're going to give us the real scoop on April Fools Day. Yeah right, the check's in the mail. We all believe that. ;)

[Beau]

So in a single woofer setup, one could expect similar output to a single sealed driver, say a Dayton RS270 or Seas w26 correct? Im still throwing Ideas around for woofers to mate to an M15ch001 @ around 300Hz in a flat baffle similar to the one you just referred to.
Beau

[JonMarsh]

Oh sure, you're going to give us the real scoop on April Fools Day. Yeah right, the check's in the mail. We all believe that. ;)

Well, you'll get SOME kind of scoop... maybe choclate, maybe vanilla, maybe Cherry Garcia, depending on the mood I'm in and what I accomplish the next few weeks. But we're making some progress here- Praxis is working with the new ACO Pacific mic and preamp, I've got a deal lined up for upgrading to the full tilt version of the CAD program I bought a few months ago for these new projects (much more productive and flexible 3D interface, plus some zoomy features for creating 2D tooling drawings from 3D models). A few things seem to be "falling" into place. Got lots of drivers to measure, still waiting on some.

Since Apple is probably going to have big product announcements on their 30th anniversary (April 1), I thought it would be appropriate to have an Arvo Pärt announcement, since the platform is stimulating my productivity. ;) After all, it's about time we had a follow up to the announcement linked above on April 1, 2002. This time we'll do it in Mission Possible DIY.

Who knows, maybe even the new improved, miniaturized flux capacitor with a built in Mr Fusion willi be ready? :B

[tg3]

Why don't you try with one VC shorted, and again with 100 ohm or so of resistance? As I'm sure you're aware, Adire Audio has a white paper on varing the driver Qts with this technique - they call it RDO (resistively damped operation).

http://www.adireaudio.com/Files/TechPapers/RDOOperation.pdf

[JonMarsh]

Yes, and so did Ascendent. With the Ascendent Atlas 12" DVC, it did not change the Qts, only changed the output level. Not what I expected, but reality trumps theory every time.

[JimS]

The real reason I'm measuring the RS315s with such interest is another variation of the Arvo (dont' want to give away the secrets here, look for an "announcement" possibly on April 1) with 4 twelves per side working up to 250-300 Hz, and not as deep a U baffle; much shallower. But a bit wider; basic front panel of 20" wide, 60" tall, with very short angle backed baffles. You'll understand when you see a CAD drawing. There will be a smaller version, too, with the RS265HF, assuming they measure like the prototypes.

Hi Jon,

4 per side and 20"x60" might push the budget & WAF too far . . . (especially if you're looking at the accutons up top)

Are you still planning on trying to work them into the "regular" Arvo instead of the TC2+? Guess I'm really asking if it makes sense to order 4 at the nice price tomorrow based on preliminary results or sit on my hands patiently until the testing is finalized . . .

What size might the smaller version come in at if the RS265HF work out?

[JonMarsh]

The "smaller" version is looking to be 48" tall, the main baffle ~16", with angled back wings ~4". Shorter than the existing Arvo design. The one with 4-12 should be my "reference system", will be under 60" tall, main baffle 18"-20" wide, with similar angle back wings. It'll make more sense when you see drawings. Ordered an udate to my CAD software this week to make it easier to design and document.

I'm in DFW airport as I write this, gotta go catch a plane! Which means I'm going to have to run really, REALLY fast, as those things are pretty quick, I hear....

~Jon

[JonMarsh]

See the "Testing 1, 2, 3" thread for updates on driver testing relevant to the Arvo design.

[JonMarsh]

Well, some elements of the Arvo Pärt design have been evolving, based on a desire to revisit the midwoofers, further refining the baffle interaction in the 500 Hz to 15 kHz area, and to move away from a deep U baffle and it's affects in the upper bass (gradual peaking then a big drop). The midrange to high frequency baffle behavior should be very nice (according to simulation) for both types, even better than the original.

Midwoofers are 830883; this layout is presuming RS28a tweeters, no waveguides. RS315HF or TC2+ (DVC using only 1 VC) woofers, though I think I prefer the former.

There have been two directions; one based on a shallow side wings (in hardwood), called the Type 2, and one based on the work I've been doing for the Saint-Saens baffle simulation, which is derived somewhat from Apogee general layouts. Truthfully, both could be successful, possibly in slightly different ways, (esthetically and acoustically), and I'm having a little trouble deciding on which to go with. Might have to try both.

I'll have more comments later about how hardwood/trim would be handled in each design. I'ts pretty much all worked out in my head. I have a big stack of solid Pau Ferro (sometimes called Bolivian Rosewood), so think in terms of 5" Rosewood wings on the Type 2, and a top plate bridging the wings and front baffle of the same; plus a back panel strip where input connections will be made done in Rosewood (yeah, where no one will see it). For the Type three, the edges will be tapered sections going from 1-1/2" thick to ~1" or 3/4"; think big Magneplanars. With edges may be an inch or two wider than drawn here, depending on if I can shrink in the grille panel space. Black grille cloth panels, magnetically attached, with felt inset a'la M8 bookshelf or Avalons. I'll have full CAD drawings in a couple of weeks or so- (I hope!) these are just napkin concept sketches in EasyDraw.

Type 2

http://audioworx.virtualave.net/ArvoType2Concept.jpg




Type 3

http://audioworx.virtualave.net/ArvoType3Concept.jpg

In both speakers, the tweeter is toward the center of the stereo pair; i.e., the straight baffle edge of the Type 3 is toward the inside, and the whole design is mirror imaged for both types.

[TacoD]

I think type 3 is nicer to look at, technically I don't think there will audioable differences :). Don't you get some interference from placing the tweeter almost besides the two mids?

[JonMarsh]

There will be some diffraction, but OTOH, it's better for integration of the main lobe in the crossover region. OTOH, if a waveguide is used, there will be minimum diffraction... Maybe this weekend I'll focus on cutting and machining up a couple of MCM waveguides for some basic tests. That's probably the critical path at this point.

I'll probably test a couple of upper panels only before building the next full test baffle. At this point I'm leaning toward the Typ 3, at least for a first pass.

[dlneubec]

FWIW, I like the look of the Type 3 also.

I'm farily new at this hobby, so please excuse me if my questions show my ignorance.

I'm confused by your description a bit. Will the baffle itself have a tapered thickness from 1-3/4" to ~1", or are you talking about another part that is tapered.

Am I correct in assuming these are going to be passively crossed for the MTM section, and bi-amped or actively crossed for the Subs?

Thanks, these look great!

[JonMarsh]

Hello dlneubec,

Welcome to HT Guide. No problem with your questions, I don't always write all that clearly.

The baffle will be constant thickness, but as currently conceived, the hardwood edges, which will be about 2" wide (in the width of the front panel), will be flat for about 3/4" at full thickness, then have a slight taper (~15 degrees) toward the outside edge. Similar to a Magneplanar 20.1, which if you're not at all familiar with those (and who is besides audio wonks?), probably doesn't mean much. ;)

Being the Luddite that I am, the whole thing will probably be passive crossover again(as the earlier generations), unless I run into some unexpected problems. We have done Arvo's the way you describe (active crossover between woofers and mids) in the Arvo Part installation at my friend ThomasW's in Denver. I use an inline EQ for LF equalization, in the last version, I designed both pure passive version and with buffer; the EQ circuit for the reuqired LF boost is pretty simple.

If this all seems unlikely to work, consider that the Gradient Revolution does something similar with two twelves without any LF EQ, but uses higher Qts drivers. I prefer the sound of a lower Qts driver with some LF boost in a shelving function, which is pretty simple to impelment with the necessary slopes with RC networks, and followed by an open loop unity gain buffer with so that cable length isn't an issue.

You see, I have some moderately high power monoblocks (Aragon Palladiums), and originally conceived this project as a test speaker for those and working on some of my own amplifier circuits. I like big wonking coils and caps, and though I've done the multi-channel active crossover thing several times in the past (70's, 80's, 90's), I didn't want to go that route for this system. Not yet, anyway. With the right baffle design and driver and crossover frequencies, one can manage this without a lot of complicated EQ, IMO.

[dlneubec]

Thanks for the detailed response. I'm glad to hear you are possibly going all passive. It seems like whenever I read about Open baffle/dipole speakers I see active designs and always wondered if that was necessary. Obviously not.

I've only done a couple projects to date and all have been passive, except my sub pairs (that my main MTM's sit on), which have their own plate amps and the HT reciever controls the crossovers with the MTM's via bass management (I hope I got that right!). I've not done any EQ on the subs to date, so that would be a new twist.

I'm also quite interested in the prospects of using a waveguide with the tweeter, so I'm anxiously awaiting how your design finally shakes out.

Dan

[Jonasz]

Jon: Are you also planning a W22/RS28 version of the MTM-top?

[JonMarsh]

Right now I'm looking at a few options- the next spin will be the 830883's becuase of their relatively clean behavior up to 3 kHz. The changes in the baffle will allow more flexibility in woofer to midwoofer choices. With dual 7's there's nothing wrong with a 180 Hz crossover, but I can probably take the crossover as high as 250-300 Hz without problems. That opens up other possibilities, as hinted at previously, such as Accuton midwoofers or midranges, even just in singles. I'm hoping to get some tweeter/waveguide combo together that works well down to 2 kHz in a reasonable form factor, with good pattern control. Then I'd be shooting for something I can run up to 2 kHz. The W22 won't do that, and many other drivers I wouldn't try that either, like the RS180. The 830883 may be OK, the C88-6, or C89-7, or C90/T5 should be a walk in the park to go to 2 kHz clean. On the bottom, larger versions with more woofers will be considered. There are other "napkin sketches" in existence for some time.... ;) Realistically, it might be better to keep the main panel at a reasonable size with just two woofers, then add on push-pull W frame dipole "subs". :B That would entail an actual active crossover, but I've got a couple of those around, so no big deal. I need more room at home before that is feasible. Something else on my to-do list for this year.

Imagine a system configuration rougly like the Type 3 above, 300 Hz crossover to Accuton mid, waveguide loaded tweeter above that (thank you, AJ and SPTech) but with different woofers: 2-3 dB higher sensitivity, slightly shorter Xmax 12's, crossed at about 75 Hz to a W woofer array of four twelves on a side. Might be fun....

[Beau]

Jon, I dont want to throw a wet blanket on all of this but have you seen Troels Gravesens Dipole study?
One of the test baffles had a shape similar to the type 3 you have designed and it didnt fare too well against a straight sided baffle. OTOH it may be just particular to that design of his. Anyways, worth a read if you are interested.

Beau

[JonMarsh]

Actually, I've seen Troel's study, about a year ago. It was quite good justification, in a sense, for the Arvo designs which existed to date.

But what he did was empirical experimentation (which is not a bad thing). What I do is simulation using baffle boundary and diffraction tools, and it can predict all of the results he gets. In fact, that's why you see the asymmetric driver location in the orignal Arvo (2002) and all the one's since, and why I know pretty much in advance before building it whether there will be any screwy little dips or peaks... A symmetrical location is not good, nor generally is a symmetrical baffle, though you can work around that.

The design as executed is based on distributing the modes all very uniformly (basically at a series of golden mean ratios), so that the baffle rise, and subsequent fall in the region below 1 kHz is uniform and predictable, and so that the area above 1 kHz is as flat and smooth as possible.

Better living through computing. ;) :B

[ThomasW]

For those that don't know......
http://www.troelsgravesen.dk/OBS.htm

[Beau]

8 minutes between responses from the other side of the world, What an age we live in... Good or bad Is a matter of opinion I suppose.

Beau

[Paul H]

... What I do is simulation using baffle boundary and diffraction tools, and it can predict all of the results he gets. In fact, that's why you see the asymmetric driver location in the orignal Arvo (2002) and all the one's since, and why I know pretty much in advance before building it whether there will be any screwy little dips or peaks...


I thought the jury was still out on being able to do accurate simulations for dipoles using LSPCad, etc - your statement would suggest otherwise. Is thiis using LSPCad?

Paul

[Dennis H]

Jon can answer this better than I but I think the baffle diffraction effects above a few hundred hertz are pretty similar for a dipole and a monopole. Going back to SL's stuff, he only does the 6dB dipole EQ from about 400 Hz down and that's with a pretty narrow baffle so it will be even lower with a bigger baffle.

[JonMarsh]

Dennis is on the money- what I've found is that above the dipole frequency, the baffle boundary and diffraction response is pretty much the same. The best tool I know of is BDS from the FRD consortium. LspCAD is not as detailed or flexible for predicting driver boundary behavior. LspCAD can be used to model the basic diopole bass response situation, but you can rule of thumb estimate that fairly well, anyway. In the Arvo, we shoot for having a midrange response that's pretty solid down to about 250 Hz, and little "special" crossover work is required above that- with a good baffle design and driver positioning. One question to answer is can I get the net power response from 1 driver as clean as from two- having the second one actually makes it easier, in my opinion.

SL believes that the diffraction stuff at the higher frequencies the midwoofer works at aren't too important, because in principle the front and rear waves create opposing pressures which cancel out. That should be true if the waves are exactly equal, but just having the diaphram behind the baffle ensures that's not really the case at some frequencies. So, I figure do it as carefully as you can to begin with.

If you put the pieces together right, the crossover winds up being fairly simple, and for the midwoofer not much different from what I do in a closed box system. In an untreated (no felt or other stuff) there are some ripples, for example, in the midwoofer response, but they're at a pretty reasonable level - +/- 2-3 dB.

http://www.htguide.com/forum/attachment.php4?attachmentid=2494&stc=1

You can gauge that from the passband of the M8a between 200 Hz and 1400 Hz. Because the drivers are wired in series, the component values are different from the M8 box MTM, but the network topology is the same, and there are no traps or other elements to compensate for baffle induced problems. Because the panel is fairly large, the diffraction effects are moved to lower frequencies than you see with a small box.

~Jon

[jdybnis]

Good or bad Is a matter of opinion I suppose.

Beau

No, this is not a subjective thing. You are interpreting Troels results over broadly. The baffle that Troels tested was like the one Jon is proposing, but it is not the same. Jon says above, his computer simulations are consistent with Troels' study. They show the same results. Jon may be using the same shape as one Troels tested but the exact dimensions of Jon's design are different. That makes a difference. The problem Troels saw will not show up in Jon's design.

[Dennis H]

SL believes that the diffraction stuff at the higher frequencies the midwoofer works at aren't too important, because in principle the front and rear waves create opposing pressures which cancel out.

In fairness to SL, he acknowledges the diffraction effects but says they are only a factor (at least as calculated) on axis and everything changes as you move off axis. He aims the speakers well behind his head so that's his fix for the "problem".

[JonMarsh]

No, this is not a subjective thing. You are interpreting Troels results over broadly. The baffle that Troels tested was like the one Jon is proposing, but it is not the same. Jon says above, his computer simulations are consistent with Troels' study. They show the same results. Jon may be using the same shape as one Troels tested but the exact dimensions of Jon's design are different. That makes a difference. The problem Troels saw will not show up in Jon's design.

And the driver(s) positioning on the baffle is very important- note the asymmetry in driver positioning in Troels' tests which produced the best overall results. Nothing wrong with what he did and his basic observations, perhaps he just drew some conclusions from some simple tests that weren't fully justified.

[Beau]

Jdybnis, Sorry, I wasnt referring to Troels study when I said Good or bad is a personal thing, I was referring to the communication between 2 hemispheres instantaneousy... In any case im in no position to question the Guru's of this board, I havent even finished building my Mic preamp yet

[jdybnis]

Sorry about that on my part. I overreated a bit. 8)

[JoshK]

I'd vote for Type 2 for looks. Type 3 is far too modern for my decor, but that is what makes this DIY I guess.

[JonMarsh]

I'd vote for Type 2 for looks. Type 3 is far too modern for my decor, but that is what makes this DIY I guess.

There seems to be a consistent division along those esthetic lines, especially when I include private communications. This will be even more evident when the CAD drawings and renderings are completed. But, shouldn't we have freedom of choice? I'm not even sure yet which one I'd prefer... :B

Does that mean I'll have to build both? (groan....!) ;)

[Hank]

Hi, I just ducked in to check on any "hot" news. Jon, I like design 3. 'Get 'er done', then on to Saint-Saens nirvhana.

[JonMarsh]

Type 3 is inspired partly by what I've been doing in Saint-Saens development, and of course, wanting to have a clear product brand identity.... as if that matters here! :rofl: :rofl:

But, I agree with the goals. Still, the Arvo series will be "completed" to keep me honest when I start working in earnest on the Saint-Saens. Also have probably found the woofers I'll use on that, as a result of these latest Arvo investigations. Will keep the weight down... underhung neodymium magnet structures, 12" cone with break up mode at nearly 2 kHz. We'll see if they live up to the specs... I have my fingers crossed.

The Saint-Saens will wind up looking like an Arvo Type 3 on steriods... serious, Barry Bonds style steriods. Or it might be two part on each side- I'm concerned about the woofer panels talking to the mids and tweeters. Have some investigations to do. Think Acoustical short circuit.

[Paul H]

.. Also have probably found the woofers I'll use on that, as a result of these latest Arvo investigations. Will keep the weight down... underhung neodymium magnet structures, 12" cone with break up mode at nearly 2 kHz. We'll see if they live up to the specs... I have my fingers crossed.
..



Sounds like a prime candidate for my thus-far theoretical 2-way open baffle dipoles for HT surrounds - and they are which driver? ;)

Paul

[AJINFLA]

http://www.aurasound.com/pdf/NS12-513-4A.PDF

My guess ;). Cross it to a NSW2 @ 6-700hz and there's your surround.


cheers,

AJ

[JonMarsh]

Tha'ts the one, AJ! :T All I want out of it is 300 Hz clean, and maybe a little higher effective sensitivity than a TC2+. The plots claim to be at 1 watt into 4 ohms, not 2.83V into 4 ohms. If true, could be in the money, especially since they weigh about 1/2 of an RS315HF.

[CraigJ]

The Arvo Pärt Tall Type 3, with Type 2 to follow :D


http://www.htguide.com/forum/attachment.php4?attachmentid=5803&stc=1

[ThomasW]

The Arvo Pärt Tall Type 3, with Type 2 to follow Cool......... :T

Don't bother with the tweeter, as I get older I find this high-frequency thing is way over rated.... ;)

[JonMarsh]

Cool......... :T

Don't bother with the tweeter, as I get older I find this high-frequency thing is way over rated.... ;)


That's why we don't let ThomasW do any of the critical cable A/B testing anymore.... ;)

[CraigJ]

Tweeter, what tweeter? Actually, I can't believe how good "it" sounds without the tweeter, though imaging is not the best ;) Type 2 is done and drying in the basement. Per usual, you guys are right, MDF sucks. The final version will be made out of Baltic Birch.

Regarding the Arvo Pärt Tall Type 3, because the baffle doesn't have much support above the woofer, the upper baffle vibrates a lot. Will this be an issue? Also, IMHO, these speakers are heavy enough.

Finally, I couldn't help but wonder what these would sound with three or four (or more) Fountek tweeters :D

Jon and Thomas, thank you.

Craig

p.s. I'm so happy to have joined the "Dark side". :party:

[cjd]

I bet I could sneak my upper-mid/tweeter array into this one!

C

[Evil Twin]

Tweeter, what tweeter? Actually, I can't believe how good "it" sounds without the tweeter, though imaging is not the best ;) Type 2 is done and drying in the basement. Per usual, you guys are right, MDF sucks. The final version will be made out of Baltic Birch.

Regarding the Arvo Pärt Tall Type 3, because the baffle doesn't have much support above the woofer, the upper baffle vibrates a lot. Will this be an issue? Also, IMHO, these speakers are heavy enough.

Finally, I couldn't help but wonder what these would sound with three or four (or more) Fountec tweeters :D

Jon and Thomas, thank you.

Craig

p.s. I'm so happy to have joined the "Dark side". :party:


Young Padawan,

While you are clearly progressing in the ways of the Force, there is much yet to learn. Even though you are lacking Imperial Duralloy, do not think to get by with a single layer of 3/4" MDF.

While boron reinforced carbonite would be the most suitable alternative, you'll find that other Padawan constructors have had reasonable success with Birch Ply layered with HDF, using a 1-1/4" thick construction. Extending the 90 degree vertical side spar to the top of the panel will also substantially improve the vertical stiffness of the panel.

I trust these simple suggestions will be sufficient for you to redouble the quality of your efforts... the Emperor does not suffer half measures for Imperial sanctioned designs. We have a reputation to consider, Dark side and all.

On the otherhand, if the construction of this simulcrum is merely to investigate the esthetic outlines of this design, then not devoting your full energies and capabilities to it is understandable... for now.

Review the last Arvo construction pictures if you need guidance for technique beyond the power of the Force beginning to stir in your hands....

[CraigJ]

Dear Evil Twin,

I'm glad to see that the force is so strong that you can hear the birth of two dipole speakers through hail and twisters. It proved to be a good weekend protected with tools in one's basement in the midwest.

While boron reinforced carbonite would be the most suitable alternative, you'll find that other Padawan constructors have had reasonable success with Birch Ply layered with HDF, using a 1-1/4" thick construction. This Young Padawan used two layers of the Borg's finest 5/8" MDF glued with Contact cement. And yes, the construction of this simulcrum is merely to investigate the esthetic outlines of this design. This is why I only made one of each types.

http://www.htguide.com/forum/attachment.php4?attachmentid=5809&stc=1


http://www.htguide.com/forum/attachment.php4?attachmentid=5811&stc=1



Two layers of three quarter inch BB and a sheet of HDF are waiting for your approval of the "final" Arvo Pärt Tall Type. On second thought, that's probably an oxymoron.

Young Padawan in training

[Evil Twin]

Impressive, young Padawan. Though you are not yet a Jedai Master, clearly you are moving in the ways of the Force, and your focus and dedication shows in the fruits of your efforts.

You will find Baltic Birch ply (not just Birch veneered ply) to be a most suitable fabrication material. It is open to discussion still among the masters whether using a contact cement or an alphiatic resin glue is the best assembly approach- the former will result in a lower Q of the panel vibration, but is not as stiff and is more prone to voids or assembly issues. Even for veneering, there are those of us who adhere to the old ways before the Clone Wars, and use alphiatic resin even for veneering. If you have some wood scraps around, you may wish to experiment and draw your own conclusions.

There are still some ongoing investigations into the optimum choices for mounting the midwoofers- all will be revealed in time.

[JoshK]

I am realling digging the looks of the Type 2. I think I will use this padowan's prototype for my own inspiration.

[Jonasz]

I think the type 2 is more beautiful but I would build the type 3 anyway. The type 2 is just gonna look so BIG with that wide top.

Slightly OT, does anyone know if you can run the XLS12 up to around 150 hz with good performance?

[CraigJ]

As soon as I get the tweeter in and installed, I will listen to see if my ears can hear a difference in the designs. Certainly, Type 3 with only five pieces, is embarrassingly easy to make ;) O.K., don't try to make a metric router jig and cut holes, after sniffing Contact cement indoors :thud:

Chris, you are always welcome to come and visit. Just bring your testing equipment. :D

Craig

[cjd]

A pair of Neo8's and a nice ribbon tweeter in an MTM configuration would slip right in next to those 8's quite nicely, don't you think?

C

[Dennis H]

The type 2 is just gonna look so BIG with that wide top.

I'd be tempted to taper the wings down to nothing at the top, both for looks and to spread out the baffle effects similar to the taper of the type 3.

[JonMarsh]

That could work... the CAD concept in place has those wings in hardwood; in Pau Ferro it will be a little dark; in light Oak or Maple it might lighten up the apparent width. Both layouts simulate pretty well in BDS. The Type 2 will have a more "traditional" look as many have noted. Traditional as in 60s/early 70's. ;) The Type three will be more "modern", as in 90's. Maybe there's a place for both?

[cjd]

What happens if you introduce some curves into the equation? Probably should toss a sketch out to show what I'm thinking. :)

C

[CraigJ]

I'll dust off the jig saw. BTW, you didn't take my bait to come back to Mad town :dancenana:

[dlneubec]

I like the idea of adding some curves to the version 3 design. FWIW, here are a couple ideas. I'm assuming a hardwood piece added over Baltic Birch, etc. I hope I attached these correctly and they show up

Dan


http://www.htguide.com/forum/attachment.php4?attachmentid=5820&stc=1


http://www.htguide.com/forum/attachment.php4?attachmentid=5821&stc=1


http://www.htguide.com/forum/attachment.php4?attachmentid=5822&stc=1

[JonMarsh]

Interesting looking concepts. Have you tried simulating any of them? The problem with arc boundaries roughly equi-distant from driveers is that they concentrate baffle boundary effects at one frequency, and the frequency response of the driver on the baffle is less smooth.

Unfortunately, BDS can't handle a shape this complex- you'd probably have to build and measure and compare.

~Jon

[dlneubec]

Hi John,

No, I'm just a novice at this. :oops: I was simply trying to help out with some ideas on alternative looks, you know make some kind of limited contribution to the effort. I was looking for a way to break up the large size of the baffle visually and perhaps inspire someone else who can take it to the next step.

In my very limited knowlege, I was assuming that the closeness to the woofers would not be an issue due to their low freq. output. and long wavelengths.

Another thought I had was that a change in baffle material, and a step out, might also allow for a modular approach, with the top baffle of the MTM overlapping the WW baffle slightly, to cover the seam between the two, so you might have two pieces that look like on once stacked. They might be easier to move around if in two pieces. Folks could possibly start with building the MTM and match it to an existing bass box they already had. Just some random thoughts.

Dan

[JoshK]

Some random thoughts, may be helpful, may be trivial and understood.

The freq affected by diffraction (call this w) is inversely related to the distance to the baffle's edge from the center of the driver (call this d). As the freq's go low, to "spread out" w, you need larger and larger deviations in d. So for practicality's sake, it is easier to vary d for the MT section.

In Dan's last example, the top left corner's arc isn't parrallel to the the top woofer's circumference so d still varies over the arc, i.e. the w's aren't bunch up too much. Put another way, the d over the arc length isn't constant as it would be for concentric circles. A rectangular shape may vary the d more so that this example but I kinda think, not that much so, but this is a guess. Then again it is only for a small portion of the full 2pi radiation. Certainly it will matter more for the tweeter's freqs than for the M's.

If you describe the function of the distance from the center to the edge, d(r), from r = 0 to 2pi for a baffle, then the freq's affected w(r) are S/d(r) where S is the speed of sound constant. To spread out the w's, is the opposite problem of least squares, in that you want to maximize (not min) the w's around the center w. This is not a well defined problem and solutions would look wacky and crazy, so the practical thing to do is to examine the sum of squares deviation in w for each baffle example and determine the penalty in bunching with the aesthetic value.

[dlneubec]

I just realized that it may not be clear what I was suggesting in these examples. I envisioned the dark red baffle of the MTM section to stick out in front of the WW baffle by 3/4" and therefore overlap the WW baffle a bit. However, it could be done in the reverse, with the WW baffle out in front of the MTM baffle. In my novice mind, I just figured the former would present less significant diffraction problems than the latter.

[cjd]

My idea of curves is, of course, different. But I like the concepts, particularly the latter two. ;) I have way too much headache right now to think I'll be doing much other than sleeping when I get home today, but at some point I may toss out some of my own thoughts, just for fun.

Mix-n-match some - stay away from parallel edges for best results.

C

[JonMarsh]

Sorry to hear about the headache- I know how nasty that can be, though I haven't had a bad one for a few years.

Lots of good ideas and discussion- I like that. :T

[Rudolf]

Unfortunately, BDS can't handle a shape this complex- you'd probably have to build and measure and compare.
~Jon
EDGE can handle any odd shaped baffle (with a little bit of effort) - but only for drivers of the same type. So a simulation with EDGE would not reveal the exact effect of those radii, but could give a first impression of how big the change of the frequency response might be.

I remember some discussion on the madisound forum month´s ago, when the accuracy of EDGE was found to be equivalent to the BDS sheets.

Rudolf

[JonMarsh]

I remember seeing something once about that discussion, but I was traveling on business a lot when that came up at Madisound, and later I coudln't even remember the name of the program. Thanks for the memory "tickler", I've located it and have downloaded to take a closer look.

~Jon

[JonMarsh]

OK, since the proof is in the pudding, and since these will never get done if we don't test stuff, there's a pair of the Aurasounds 12's heading here from Madisound. These are NOT to replace the RS315HF or TC2+ in the general Arvo design (including "Tall" Type 2 & 3), but are a possible woofer in the Grande/Venti configuration for main panels. Just a possibility at this point- as Willie the Shake would say, many a slip twixt the cup and the lip.

http://www.solen.ca/pics/aura/ns12_513.jpg

[BobEllis]

EDGE can handle any odd shaped baffle (with a little bit of effort) - but only for drivers of the same type. So a simulation with EDGE would not reveal the exact effect of those radii, but could give a first impression of how big the change of the frequency response might be.

I remember some discussion on the madisound forum month´s ago, when the accuracy of EDGE was found to be equivalent to the BDS sheets.

Rudolf

It may not be worth much from a guy whose measurement setup isn't calibrated, but EDGE predicted and measured response were virtually identical the last few times I used it. I haven't used BDS.

Bob

[JonMarsh]

Thanks for the input, Bob. I've used BDS a fair amount; I can see Edge can be easy to use if you're willing to model each drier on the baffle seperately. I'm going to give it a whirl.

[Hank]

Regarding the Arvo Pärt Tall Type 3, because the baffle doesn't have much support above the woofer, the upper baffle vibrates a lot.

That, and Jon's noting that he may do a two-piece on the Saint-Saens because of "acoustic shorting", and my concern over weight, all seem to point towards two-piece, but darn, I sure like the look of a seamless baffle (Rosewood, of course). The Aurra NS12 makes sense here. Dontcha just love neodymium motors? No basket compression. :T

Despite my being totally consumed at work, I'm making a tad of MDF dust. A local guy had me cut the MDF for a sub recently and now I'm cutting the parts for a dipole that a designer just did for him (no, it's not Danny). 45" tall baffle (double 3/4" MDF) x 12" wide. MTM dipole upper with 2 Accuton C82/T8's and a B.G. Neo3 PDR. Bottom bass section will house an ExcelW22 woof. The BG is not offset - it's a standard MTM layout. Haven't seen the crossover, and may not. My guess is that if it's successful, the designer may make it one of his kit (or finished) offerings. I'm anxious to hear my first dipole in a home setting.

[JonMarsh]

Glad to hear you're doing something outside the office! It gets harder and harder, doesn't it?

I just got in a pair of NOS Accuton C88-6 I'm going to be testing; consider it a "warm-up" to the C90/T5 (if they ever get here), and who knows, it may wind up doing midrange duties in one of the upscale versions. Even in Dipole mode only going down to 250 Hz or so, it may be useful. Or it may take two (it's a 5" also, but a little more expensive than the C82/T8). I'll be curious to see what you think of the one you're working on... smells like Rick Craig, to me. Or someone emulating him.

Keep us posted if you can.

~Jon

[Hank]

Glad to hear you're doing something outside the office! It gets harder and harder, doesn't it? Yeah, it's going contrary to my opinion, which is that it should get easier as you get older.
:laughat:

smells like Rick Craig, to me.
You're a perceptive rogue. ;)

More expensive than the C82? 8O The C82 is the first $200 midrange I've held in my hands. The ceramic diaphram is so thin, it's translucent.

I'll eventually post a pic or two (if the customer lets me).

[JonMarsh]

Yeah, it's going contrary to my opinion, which is that it should get easier as you get older.
:laughat:


You're a perceptive rogue. ;)

More expensive than the C82? 8O The C82 is the first $200 midrange I've held in my hands. The ceramic diaphram is so thin, it's translucent.

I'll eventually post a pic or two (if the customer lets me).


I hope he does. ;)

~Jon

[JonMarsh]

Email from Adam at Madisound as well as their fullfillment shipping service- the C90/T5's are in shipping and I should have them Friday or Monday.






:T :banana:http://www.madisound.com/images/product/C90-T6.jpg


Update: UPS "confirms" Monday delivery... I feel a "medical" day coming on. :B

[JonMarsh]

View from above- note double roll surround, damping pads, and disched in grille- much stronger and less resonant than for other's I've seen from Accuton.

http://audioworx.virtualave.net/T90-T5_Top.jpg


These weigh nearly 10 lb a piece; whatever they're doing with the neodymium, it's NOT about weight reduction or a small magnet assembly. ;) Rather massive pole piece and rather large rear vent, nicely radiussed. May also take some tricks to avoid rear masking. Certainly a back bevel, maybe a mounting plate.

http://audioworx.virtualave.net/T90-T6_Back.jpg


Not a very good picture, just trying to show the size (55 mm diameter) of that titanium VC.

http://audioworx.virtualave.net/T90-T6_Side.jpg

Have business travel next weekend, so I'm not sure when I'll get to testing.

~Jon

[CraigJ]

We've waited 90 days for the Accutons, and that's it? ;) I'll just have to go back and listen to my Arvo Pärt Tall Type 2 and Type 3's:D . Thomas, thank you again for encouraging me to "what are you waiting for , just make them". The Arvos are really an enjoyable speaker.

Thanks Jon,

Craig

p.s. I'm going to join Hank in waiting for the Saint-Saens, my next diy speaker in "the other room".

[JonMarsh]

We've waited 90 days for the Accutons, and that's it? ;) I'll just have to go back and listen to my Arvo Pärt Tall Type 2 and Type 3's:D . Thomas, thank you again for encouraging me to "what are you waiting for , just make them". The Arvos are really an enjoyable speaker.

Thanks Jon,

Craig

p.s. I'm going to join Hank in waiting for the Saint-Saens, my next diy speaker in "the other room".

Yeah, just some expensive white, black and silver drivers that don't look much different from everything else we play with around here. :B


Well, that's just the start. Next, measurements. In parallel with that, baffle design development. Then construction. Then more measurements. Heck, might even have to design and build a pasive crossover- I don't have any of those new fangled toys like that DCX2496 you're got! :lol:

And I still have to work on the waveguide issues- but the C13-6 evaluations and mods are coming along- see "Testing 1, 2, 3...."

As Dr. Tihanyi used to say before his passing, "Slow work takes time..."


~Jon

[JonMarsh]

OK, don't know what happened the last time I tried to post this... :huh: Somehow it went into the ether...

But speaking of baffle design, "The Edge" gets two thumbs up from this reviewer- easy to use, easy to setup and tweak, quite quick even in emulation in Virtual PC on my Mac (see screen shot), and even explicitly models dipole based roll off in bass as well as box diffraction/boundary roll off. Did I mention it's free? What's not to like! ;)

http://audioworx.virtualave.net/EdgeVPC.jpg

It does have electrical calculators to model a basic EQ circuit for BSC or diople compensation, but the real value I see here is being able to quickly model baffle designs and driver positioning to come up with the smoothest response without/before any additional EQ; I think in many cases that stuff can be taken care of in the cossover. It doesn't model multiple individual driver responses, but I can work around that. It will model summed driver responses, such as the dual midwoofers in an Arvo Part Tall.

~Jon

[JoshK]

That's it? :huh: What do you have a job or something? :huh:

:rofl:

[JonMarsh]

well, it's obvious I don't have a life.... ;)

Did evalauate a variety of panel layouts and sizes in "The Edge" (sounds more like an amusement park ride to me!)

[Gerald Jansen]

OK, don't know what happened the last time I tried to post this... :huh: Somehow it went into the ether...

But speaking of baffle design, "The Edge" gets two thumbs up from this reviewer- easy to use, easy to setup and tweak, quite quick even in emulation in Virtual PC on my Mac (see screen shot), and even explicitly models dipole based roll off in bass as well as box diffraction/boundary roll off. Did I mention it's free? What's not to like! ;)

~Jon
Hi Jon, I'm wondering about the secondary source at x=0 (and I ask you because I did not yet find time to try the Edge). Shouldn't they (@ x=0) be turned off for floorstander simulation (actually placed at negative infinty)? Is that possible in the Edge?
Regards, Gerald

[JonMarsh]

Well, that IS a drawback to this and BDS, is that neither seem to handle interaction with the floor. Particularly that has an impact for the woofers (say, dual 12's near the floor); there's going to be something like a mirror image reflection, depending on the material.

OTOH, for monitors on a stand, the approach BDS and EDGE use will work fairly well, though not pefectly- that's why boundary calculations for LF re-inforcement can be quite useful.

Probably the only way to do a more accurate simulation I can think of with available tools is CARA, at least, it should work for a box source.

In practice, though, using BDS or The Edge will get you most of the way there for the really tricky styff (midrange and tweeter interaction with the panel), and the LF stuff is somewhat modeled OK; in the end you have to measure the response in room to confirm what you're getting and develop the crossover.

The thing is, you can identify issues and get MUCH closer using these tools to an optimized solution than just by "guessing"; this is particularly critical, it seems, when running a single midrange source; the dual driver Arvo setup is a lot easier to optimize for an overall smooth (easy to deal with in crossover) midrange passband.

I've got three separate sims for the woofer (dual), mid (single) and tweeter with The Edge, and just used Grab to capture the screens after saving the data (one missing feature of Edge is printing or plot saving, but it's easy to work around). Experimented with a number of other shapes, and variations on size and form factor; the affect on the midrange is the most critical.

[Evil Twin]

View from above- note double roll surround, damping pads, and disched in grille- much stronger and less resonant than for other's I've seen from Accuton.

http://audioworx.virtualave.net/T90-T5_Top.jpg


These weigh nearly 10 lb a piece; whatever they're doing with the neodymium, it's NOT about weight reduction or a small magnet assembly. ;) Rather massive pole piece and rather large rear vent, nicely radiussed. May also take some tricks to avoid rear masking. Certainly a back bevel, maybe a mounting plate.

http://audioworx.virtualave.net/T90-T6_Back.jpg


Not a very good picture, just trying to show the size (55 mm diameter) of that titanium VC.

http://audioworx.virtualave.net/T90-T6_Side.jpg

Have business travel next weekend, so I'm not sure when I'll get to testing.

~Jon


I would be more favorably impressed if the diaphrams were black....


This, this may be a promising development.

I SUGGEST you accelerate your testing schedule, lest you find the Emperor displeased and forced to make a visit to discuss your lack of effort with you personally...

...this is an event I would not encourage were I you.

[TacoD]

Wow great looking driver, is the whole magnet assembly from neodymium? Or are the pole pieces very thick/ massive?

[JonMarsh]

I suspect that every thing we can see here is just pole plates; this is usually the case externally with neodymium designs, and part of how they are inherently self sheilding. I wonder how big the actual gap is; my suspicion is that these have an effective Xmax of only a couple of mm, unlike the NS12-513A. The sensitivity is mighty high, about the same as the C24-6 underhung tweeter, and the C13-6. The C13-6 is a real feather weight, and uses a smaller form factor than other Accuton tweeters, including the C12.

The "Edge" data last night does have me pretty excited about moving forward and building a test panel for this system. The response profiles are just what I'm looking for with the passive crossover approach I have in mind, though I'll probably (for the benefit of the forum crowd) do an LspCAD "active" design, which will make the design considertations more obvious. I think I may be able to simulate my line level equalizer in LspCAD, too.

I also have to get the waveguide testing done to see if that will pan out. I'm thinking a thin sheet of neoprene as a "gasket" between the waveguide and tweeter face plate; also considering re-inforcing the frame of the waveguide to reduce any residual resonance.

One thing for sure, if these don't turn out the way I'd like or hope, I won't be able to point at the drivers and say it was their shortcomings that held me back! :roll:

[cjd]

I won't be able to point at the drivers and say it was their shortcomings that held me back! :roll:

Subcontract! Subcontract! You do not want the Emperor's displeasure aimed your way... ;)

C

[dlneubec]

Hi Jon,

I wonder if there would be a good way to reduce any residual resonance on the waveguide by treating or applying some damping material to the back of it, like rope caulk or something.

Of course, that particular material would be a rather unattractive solution from an aesthetics standpoint.

[JonMarsh]

I'm thinking in terms of some additional build up with fiberglass and resin. It's really not that bad to begin with, but I just want to "take it to the next level", so to speak.

[CraigJ]

Jon,

I've seen the Dark Side, I'll be the Sub! ;x(

The Arvo's continue to get better. I added four inches of height to the Type 3 (48 to 52 inches); now Type 3 matches Type 2, and imaging has improved :D Could be my ears 8O

Craig

[JonMarsh]

Last night I was simulating the panel layouts for the Type 3 with the C90/T5 positioning- surprisingly sensitive to panel height to get the best midrange. I'll have to take a look and see if LspCAD can generate DCX coefficients (I think it does), and maybe we can put together a more optimized crossover specific to your panels. The dual driver configuration is easier manage the response.

[Dennis H]

I'm thinking a thin sheet of neoprene as a "gasket" between the waveguide and tweeter face plate
The Force is strong with felt. No need to join the two pieces perfectly if felt absorbs all the nasty reflections/refractions at the throat. You could even use a hole in the felt smaller than the dome to control off-axis dispersion (make it similar to the on-axis), like the Neo3 with the internal felt. Of course you'd need some Darkside XO magic to get the whole thing flat.