dlneubec

I’ve posted in Jed’s Tombstone thread how amazed I was by the Bagby/Salk design I heard at the inDIYana event last Saturday. I’ve been doing some concept design for a "value" version of this speaker and wanted to start a thread on it. I’m hoping for sensitivity in the 92db+ range, if possible. I believe Jeff said his design was about 93db.

Here is what Jeff is using in the Salk design, according to what he told us Saturday:
12” Lambda TD12H ($259ea.), crossed at around 450hz.
6.5” PHL 1070 (?) midrange ($159ea.), crossed at 2.7khz
Morel MDT33 tweeter ($149ea.)
http://aespeakers.com/drivers.php?driver_id=6
http://www.madisound.com/catalog/pr...products_id=570

Jeff indicated that the PHL 1120 should work just as well, but he happened to already have a pair of the 1070’s (I think it was “1070” he mentioned, but it was definitely not the 1120)
http://estore.websitepros.com/1736754/Detail.bok?no=48

For those who have not seen it, here is the Salk/Bagby design.


Goals:
My goals are to do a similar high sensitivity design, with open baffle mid (or perhaps open back mid as a second choice). The Salk speaker is quite large at 15” wide x 46.5” high and I would like to make a smaller front mass to the design as well (or my wife won’t let it stay in the main listening/HT room). I’m thinking a tapered design so that the mass of the speaker is lower in the box and less intrusive in the room (see mock up below). The concept design I have below has a 76L net woofer enclosure and the size is 16” wide at the bottom, with 5º tapered sides up to a width of 8.5” at the top. Of course, the baffle width on the open baffle mid will have to be sized as needed to work at the intended crossover frequency and I have not even begun to consider that. The height of the speaker is 43” and the depth is about 18” at the bottom. I envision an 8º sloped front baffle on the woofer enclosure, with a vertical open baffle on the mid/tweeter that is separated from the woofer baffle. This arrangement allows the drivers to be approximately time-aligned. I have also attached some preliminary drawings of the speaker(see the attached pdf). Depending on the tweeter chosen, I could possibly do a shallow waveguide mount for it, like I did in my Duo design and as proposed by Zaph for using the D26NC55.

Drivers:
The drivers of the Salk/Bagby design run in excess of $1100/pair, as far as I can tell. I would like to reduce that substantially, without losing too much in either sensitivity or sound quality. Here are the possible drivers I’ve come up with to consider.

Woofers:
Lambda TD12H 12”: $259
Ciare 12.00NdW1 12”: $175
http://www.assistanceaudio.com/08_CIARE.html#neo

Midrange:
PHL 1120 6.5”: $159
B&C 6MDN44: $120
B&C 8NDL51: $136
http://www.usspeaker.com/homepage.htm

Tweeters:
Seas 29TFF/W (H131, 92.5db: $44
Peerless HDS 810921, 93db: $78
Peerless DX25TG05-04, 93.5db: $28
http://www.madisound.com/catalog/pr...roducts_id=1591
http://www.madisound.com/catalog/pr...roducts_id=1744
http://www.partsexpress.com/pe/show...tnumber=264-578

It’s unclear on the availability of several of these drivers. I have attempted to make email contact on the Lambda and have not heard anything back. The Ciare is listed as available at Assistance Audio, but I have not confirmed that. The Peerless DX25 and HDS are listed as not in stock at Madisound, though the DX25 appears to be avaialable at PE under the old Vifa name. Below is a graph from some quick modeling in Unibox with the woofers and mids at 200w input. The Ciare is at xmax at this point and the Lambda still has a lot of room to move, however, we are at over 114db spl at this point and for HT you really need a subwoofer to handle below 40hz with either of these drivers. I’m leaning towards the Ciare since it is appears to be comparable in performance, yet would save $168. The PHL mid might be hard to beat. The B&C 8NDL51 seems to be well thought of, but I’m not sure if it can be crossed in the 2.5-2.7khz range needed to work with a dome tweeter at high spl. The 6MDN44 might work, but I have not read of anyone who has used or measured this driver. I’ve not used any of the tweeters listed, but was looking for models with relatively low distortion, but with high sensitivity and these are what I found so far that were less expensive than the Morel. If the drivers selected were the Ciare woofer, PHL mid, and Seas tweeter, the driver cost will have been reduced to about $754/pair, which is a very good savings, if SQ is not compromised to significantly.

Ok, I’ve presented my thoughts and ideas and I’d like to hear from others what they think of the concept, the drivers or others they might suggest, etc. What do you all think?

Also, does anyone have any name ideas? I thought immediately of the “PowerTowers”, but I’m ambivalent at this point.


[IMG]

This was edited to show the final concept version below:

Attached Files

PT_v1.pdf (51.9 KB, 239 views)

__________________
Dan N.

---k---

Dan,

Don't you ever take a break to sit back and enjoy the speakers before moving on to the next pair?

I don't want to drag your thread down in theory, but what do you think it is with high efficiently speakers that causes them to sound so good? Is it that they don't have to be pushed very hard, and therefore don't get into the range where distortion is an issue?

I also wonder about the definition of highly efficient. My Khans are ~91db. Stereophile would say that this is above average. People on AVS would say it is average. You're goal is 92db+. You're goal is about the same as my speakers, which you've heard. So, I'm thinking it isn't the efficiency that had your mouth gapping, but the drivers or something else.

I think this is a very cool project. There has been a lot of talk about mating pro drivers to hifi drivers. I think the concept has lots of merit and benefits. It would be great to understand what makes the pro stuff either really good or really bad.

__________________
-Ryan
The Sietecerocero Project! ND140/BC25SC06 MTM & TM
My CJD Khanspires - A Dayton RS28/RS150/RS225 WMTMW
My CJD Khancenter - A Dayton RS28/RS150/RS180 WTMW Center
My CJD In-Khan-Neatos - A Dayton RS180/RS150/RS28 In/On Wall MTW

Saurav

I've had a cheap version of a similar concept for a few years now:



Fountek JP3.0, Audax PR170M0, Peerless CSX 10" (forget the exact number). I can't find the photos from the side. The mid and tweeter are on an open baffle, passive LR4 XO @ 4.5K. The woofer is in a sealed enclosure, active LR4 @ 500Hz. Tube amp driving the top (95dB efficient), SS amp on the bottom (around 88dB, I think).

This was mostly put together with drivers I had (and the XO built with parts I had), so it's kinda ragged. But it definitely proved to me that I liked the open baffle sound. This summer I'm thinking of trying a dipole woofer as well, since I've heard so many good things about that. And use a different mid that'll allow an easier blend with the tweeter.



One theory is that they don't get into the excursion regions where power compression becomes an issue. I don't know... I just know what I enjoy, and at multiple audio shows and friends' systems, I've become convinced that small low-efficiency drivers don't give me the sound I like. Something like a big Wilson clone would have the snap and dynamics I want, but that's not happening in my living room (or budget) any time soon.

dlneubec

Hi Ryan,

I hear you and I'm not planning to start building for awhile. My wife asks the same question. These were the first speakers I've heard since I've been into DIY or that I've heard, for that matter, that really stunned me with how effortless and "live" they sounded. In my book, these were simply on a whole other level of performance.

I assume it is a combination of really good pro drivers, that have a LOT of headroom for a great dynamic range, with a great crossover design and the soundstage and other benefits of an open baffle mid design. Surely the electronics are pushed less to get the sound, which is also a positive. I suspect there are also some benefits to a simpler approach using fewer drivers to accomplish the sensitivity, i.e. basic 3way with 3 drivers versus the more typical Hi-Fi WMTMW approach. There would seem to be considerably less comb filtering an phase match issues with the 3 driver approach. I'm sure there are better speakers out there (perhaps not for the cost, however), but I haven't heard them, so I go with what inspires me to be creative.

I look at speaker design as a challenging, creative outlet. Something to keep my mind alive and have a passion about. After working for a living for over 30 years, a creative outlet helps keep you sane!

My HOSS design has almost the same drivers as your Khans and though the direct radtiated energy is down at the listening position and not 91db as typically measured, the total sound radiated into the room would be quite similar to yours. I believe both of these designs were in a top five at Iowa2007 that had less than .5 pts difference between those top five. However, I can tell you, without a question in my mind, that the Salk/Bagby design struck me as stunningly better and not by just a little, IMHO. Maybe they just it all my hot buttons!

__________________
Dan N.

dlneubec

I have a couple of those Audax pro mids that I picked up used for $25ea. that I plan on doing some experimenting with as I get into this concept, but maybe try to cross them closer to where Bagby did.

__________________
Dan N.

Saurav

That would be a much better XO point for that driver. Like I said, I was working with what I had, and I had to choose between stretching the mid or stretching the tweeter, since the Fountek wouldn't be too happy down at 2.5kHz.



I've often wondered about that. Everyone runs small-signal tests on their drivers, and we look at frequency response, CSD plots, impedance, all of which helps us pick drivers and design crossovers. But what does that tell us about the "dynamic linearity" (for want of a better term) of the driver? Basically, when a transient comes along that's 20dB above the base signal level, how well does the output track the input? Does anyone test that kind of stuff? Can that behavior be inferred from the low level tests?

Anyway... good luck with your project. I'm investigating drivers for something similar, but at a lower price point, so I don't have much comment on your choices other than *drool*

dlneubec

Saurav,

I have found a couple other options that are a little better bargain. For example, the Beyma 10BRO60 is around 91db sensitive and can get up to 110db before reaching xmax, tuned to an F3 of 34hz in a 70l box. I believe it costs $99.

Put two of the Dayton ST255-8 10" Series 2 Woofers (about 98.5db for 2) in a 75L box and you can get up to 116db before exceeding xmax and they go for about $62 each, though the F3 is in the low 40s.

__________________
Dan N.

dlneubec

I forgot to ask if anyone has any experience with the Ciare drivers, especially the 12.00 NdW1? I know that Jon Marsh was considering the 18" version, I believe, in one of his uber-all-open-baffle designs.

__________________
Dan N.

Saurav

Thanks, I'll take a look at those woofers. I want to go dipole on the woofers this time around, and I'll have an SS amp on them, so I'm more concerned about Xmax and Qts than sensitivity. That Beyma looks like it might be better than the Eminence 10"/12" drivers I was running through spreadsheets last night.

JoshK

From what I've read the Audax was made by PHL and is a close variant of the 1120. Its half the price and if you have a pair its worth trying out. Beyma has some ~93db dome tweeters that are worth examining too.

fjhuerta

Dan, did I ever mention your designs are some of the most original and thoughtful out there? Every time you post something, it's hard not to be amazed!

__________________
Javier Huerta

augerpro

Cool stuff Dan. Very close to what I'm going to build this summer. One problem I'm having is deciding on the woofer. I have a pair of AE IB15, and a B&C 15NW76 in the mail. I'm also considering an AE TD15H/X as a just slightly lower sensitivity but higher xmax alternative to the 15NW76. The hard part is deciding between those two or the IB15, which has all the right numbers-except sensitivity. Like you I'm doing this to check out the whole pro audio hype. So listening to the Salk speaker do you think the "magic" extends to the bass? In other words, could I cheat and get the cheaper IB15 and still be happy? Or do I just have to accept the more expensive TD15H/X or 15NW76 and use some kind of LT circuit (these will be cardioid alignment)?

I think the Ciare is great woofer. Looking at the impedance it looks like they are paying attention to the motor. My only concern is xmax, but like you said it amy be just enough, and crossed to a sub at even 40hz is probably plenty. Within the price constraints I don't think you can go wrong with this woofer.

The 6MDN44 is a good choice, although I wonder if the 6NDL38 might be a better sensitivity match to the tweeter? Depending on baffle step and xo point you could probably run this without BSC for the full 92-93dB, which would match both tweeters without padding. Tough call, both nice woofers.

I can't recommend the 810921 enough. In fact I've been thinking this tweeter is just screaming for a waveguide. I really think this would be a winner.

__________________
~Brandon

DriverVault
Soma Sonus
Old Driver Tests

exojam

In open baffle designs like this one, is placement really critical? I ask regards to the speaker needing to be a further distance from the wall in comparison to enclosed designs? Also, when one uses speakers such as these, what do you do for a center channel? Thanks.

dlneubec

Thanks Javier! I do try hard to be creative and make speakers that have some artistic value as well as auditory value. Not everyones cup of tea, however.

__________________
Dan N.

dlneubec

Hi Brandon,

There seem to be more 15" drivers that will do the job well than there are 12", but there is no way I'd get 15"ers into the living room and keep my wife happy, unless I took the stereo subs out, which I don't think is wise from an HT standpoint.

I think either of the B&C 6.5" drivers might work, but I know that the PHL will for sure, so I'm leaning that direction, unless someone else gives a review on how those drivers sound in a hi-fi setup. I still have to give it more thought. Are you still going to use the 8NDL51? I wish they were all round so they weren't such a p.i.t.a. to mount.

My approach at this time is to go safely middle ground, somewhere comfortably between the typical hi-fi and hi-spl pro audio, hence the dome tweeter and 92-93db goal. IIRC, you are going solidly pro with a compression driver, big waveguide, etc. correct?

On the Ciare, I'll be surprised if 115db from one speaker at 1 meter in a 15'x20' room is not enough, given the output I heard from the Salk/Bagby design in a much larger space. I need to find out if that woofer is still available. PHL says the max spl on the 1120 is 113db, so if that is accurate, they are going to pushed too hard before I reach xmax on the Ciare. I doubt my amp will push them hard enough to do that anyway, unless these end up closer to 4ohms.

I don't know what to tell you in regards to the bass. The TD12H sounded great and the whole package was unbelievable to me. I know they had tried unnamed 12" Dayton and JBL drivers in that design at one time and were not happy with it and liked the TD12H much better. I guess you could always biamp the IB15, if you have an extra amp to run the woofers.

__________________
Dan N.

dlneubec

As I understand it, they are sensitive to room placement, and especially need some room behind them so first reflections don't reach the listener too soon. However, the nulls they create to the side probably means they can be a little closer to the side walls than typical box speakers.

Since the center channel is mostly dialog based, I would assume it should still be the traditional closed box for best results.

__________________
Dan N.

dlneubec

Brandon,

I forgot to ask, have you looked at using 2-10's or 2-12's as an alternative to the single 15's?

__________________
Dan N.

augerpro

Bah! Plenty of PHL designs (and opinions) floating around. But none on the small B&C or 18 Sound drivers. You're the pioneering type Dan, check it out!



Yeah I hear ya. I considered the 10NW64 instead but the frequency response just doesn't do what I need.



I'll be using the 8NDL51 mid and DE250 in either the DDS waveguide or a 10" one from Geddes, in an OB. Woofer will be a damped U frame, as I said not sure what I'll use here.




Hmm. Maybe I will use the pro woofer and live with the increased cost and need to EQ the bass. Bummer. This design is in fact biamped because the W will be actively crossed to TM. So I could use the IB15, but it gets away from the whole idea of checking out the pro audio hype. I have a feeling the IB15 will be used for a later, cheaper version more along the lines of what you are doing.

As far using multiple smaller woofers, yes I considered it. But the cost got quickly out of hand since I was just looking at the smaller versions of the already expensive 15" drivers.

__________________
~Brandon

DriverVault
Soma Sonus
Old Driver Tests

---k---

Dan,

Do you do any distortion testing of drivers? It would be interesting to see how the pro drivers you end up with compare to some of the better hifi drivers, like the Scans Jed is using. Especially if tested at a little higher voltage than is typically used for testing.

__________________
-Ryan
The Sietecerocero Project! ND140/BC25SC06 MTM & TM
My CJD Khanspires - A Dayton RS28/RS150/RS225 WMTMW
My CJD Khancenter - A Dayton RS28/RS150/RS180 WTMW Center
My CJD In-Khan-Neatos - A Dayton RS180/RS150/RS28 In/On Wall MTW

Dennis H

That's the thing about pro drivers. When they measure distortion, it's usually at something like 110-115 dB (100 watts). Hifi drivers are usually measured at something like 90-96 dB (2-3 watts) and would totally puke if you tried to play them as loud as the pro tests.

dlneubec

I have done a litte distortion testing and have the capability, but have not really spent the time it takes to understand how to do it right. Mark C. has done a test of the PHL 1120 and compared it to two high end hifi drivers, the Seas M15 and the ScanSpeak 8545. I'd say it did quite well in some pretty good company. Here is his summary:

"PHL 1120
For comparison's sake, graphs of the Seas M15 are also shown, and in selected graphs, the SS8545.

Comments
The PHL 1120 did reasonably well. It had strikingly low nonlinear distortion at 850 Hz and reasonable numbers at 400 Hz. Neither did well at 150 Hz. The linear distortion curves are also good. Not as good as the M15, but as good as, or maybe better than the SS8545. Where the PHL excels is sensitivity. I measured a sensitivity of ~95 dB for the PHL. Certainly, if you need the sensitivity, this is a respectable driver. If you don't need the sensitivity, there are other drivers that test better."


Brandon (Augerpro) has also done some testing of various pro mids, such as B&C, EighteenSound, etc. For example, the B&C 6NDL38 had lower distortion figures than the TB w4-1337 and as good as or better than the RS150, IIRC. See the link under his posts above.

__________________
Dan N.

Dennis H

Again, Mark tests at pretty low SPL. I'm not saying the pro drivers necessarily do better at high SPL but I think you need to test up there before you can draw a conclusion.

augerpro

Dan do you have a link to those PHL results?

__________________
~Brandon

DriverVault
Soma Sonus
Old Driver Tests

tktran

IMHO we will always be KNOCKED over by a high sensitivity design, especially if we are used to typical 85dB/W speakers with only 100W behind them. The amps will be clipping far more often than we know, or expect.

I believe that the reason that Jeff Bagby's design is a refreshing change from others in the DIY circles (apart from the careful XO work) is the big, high sensitivity woofer in a decent sized cabinet for a WMT 3 way. Not subwoofer, not a mid woofer, but a real woofer that can cover 55Hz (A-1) to 440Hz (A-4) with authority.

Each time I come home from a classical concert, I look at my speakers and think "I need speakers WAYYY bigger than this to replicate that kind of dynamic range"

At the listening position our system should be able to play cleanly everything from the quiestest pianissimos (?40dB) to the thundering crescendos (?110dB).

Of course, getting it WAF friendly is another matter.

dlneubec

Yes I do. Here you go.

PHL 1120

__________________
Dan N.

dlneubec

I think you are right and the latter is the real challenge

__________________
Dan N.

dlneubec

I think you are right Dennis. Push the HiFi drivers close to their thermal limits during high db output and you might see completely different results. The experience I had with Jeff's speakers was nearly the opposite. They sounded great at what I thought was a pretty high level and then they turned them up a bunch more, like a live performance, and there was not a hint of strain that I can recall.

__________________
Dan N.

dawaro

You might want to shoot Bagby an email about the B&C speakers. IIRC he had onced posted his impressions of one of the B&C drivers. From what I remember he said it was one of the best sounding true mids he had ever heard.

__________________
I am not Dawaro the muslim state in Ethiopia...Just DAvid WAyne ROberts

Undefinition

Boy this thread exploded in just one day! You must have hit a nerve, Dan!

I was just looking at your choices of drivers. I don't know how much longer Peerless/V-line/whatever will be making the DX25, but it seems like your best bet right now. Also, as far as midranges go, you might also want to check out Eminence; although B&C certainly has more prestige, Eminence certainly gets the job done in a lot of pro sound speakers.

I also got one question for you: what is the advantage of doing a "micro-horn-load" of the tweeter (using the roundover bit), as you did on the Duos?

The design looks very nice! Good luck!
Paul

John_E_Janowitz

Hi Dan,

wow, our web host picked a bad week to have issues. They were down for 2 full days so we had no email or website operational. Once they got back up, emails started coming in as they got through the backups. I spent an extra day trying to fix our forum, online store, and replace all the lost data. Then the flood gates opened up. Jim Salk told me the reaction to the speakers was great and we've had a lot of requests about the Lambda drivers this week. A couple quick things. Purchasing the Lambda woofers in pairs does drop $50 off the total, or $25 per driver. We used to discount more, but our materials costs keep going up and up. We are still currently at the same pricing as Lambda was selling at 7 yrs ago. I've tried to push off any price increases.

I'd like to link you to a page on the Lambda motor design, but that was unfortunately lost in the server crash. I don't have the files at home other than the original first page as I was testing the style sheet. Here is the link that should be working by mid morning tomorrow as I get to the shop to upload the files again. It just gives an overview of the Lambda motor and what was done to keep distortion low.

http://www.aespeakers.com/Lambda001-1.php

One of the biggest differences is how the copper shorting ring fixes the flux in the gap and how it decreases inductance. We actually have quite a few people mating TD drivers (mainly TDM's) to compression drivers, ribbons like the RAAL, etc in 2way systems up to 2KHz or so. The TD15M for example is quite flat up to 2KHz at as much as 30 degrees off axis and to over 4KHz on axis. As an extreme way to save some money if you plan to have a small sweet spot, you could away with doing a 2 way system with very good results.

John

augerpro

I tried the DX25 with the MCM waveguide:



NOT a success. The DX25 has a drop in response in the same area the wg causes a drop. The combination was not good. Now I did not take measurements off axis, so this may have smoothed out, as proper wg's do. But seeing this isn't promising.

BTW congrats on the reviews of the Aethers!

__________________
~Brandon

DriverVault
Soma Sonus
Old Driver Tests

---k---

Yeah, that is what I was implying. Once you and Jed get your drivers in hand, it would be great if you both could do some distortion testing on them, both at a normal level then at an agreed upon higher than normal.

Dan, I think what separates your work from many others is the amount of thought and research you put into it. You definitely deserve a big . People like you (and many others here) are what makes this a great community. It would have gotten borning long ago without new trails being blazed.

__________________
-Ryan
The Sietecerocero Project! ND140/BC25SC06 MTM & TM
My CJD Khanspires - A Dayton RS28/RS150/RS225 WMTMW
My CJD Khancenter - A Dayton RS28/RS150/RS180 WTMW Center
My CJD In-Khan-Neatos - A Dayton RS180/RS150/RS28 In/On Wall MTW

---k---

Could I get a little crossover education? I'm looking at the frequency response of the PHL driver and an RS180:



The RS180 is pretty flat up to 2-3khz, it makes sense to me how to design for this driver (at least I think I understand it). While the PHL has a big hump from 500hz to 1700hz, but is then fairly flat to ~3k. How do you deal with that big hump in the midrange?

__________________
-Ryan
The Sietecerocero Project! ND140/BC25SC06 MTM & TM
My CJD Khanspires - A Dayton RS28/RS150/RS225 WMTMW
My CJD Khancenter - A Dayton RS28/RS150/RS180 WTMW Center
My CJD In-Khan-Neatos - A Dayton RS180/RS150/RS28 In/On Wall MTW

dlneubec

I did correspond with Jeff a bit after last Saturday. He said he has no expereince with the B&C drivers, so it must have been something else he posted on. Maybe it was the PHL they ended up using.

__________________
Dan N.