This time for sure! (BJM)

**Andrew Pratt** · 10 February 2002, 18:52 Sunday

Looks like a great project Jon.

**Pete** · 10 February 2002, 19:21 Sunday

Jon- The off axis inroom response looks very good, minus a little
room null.

What will be the final panel dimensions & the port in the back is
3" or 4"?.
-Pete

**JonMarsh** · 10 February 2002, 19:37 Sunday

Thanks Andrew! Like my nemesis Bullwinkle J. Moose, eventually I get things right.

The cabinet is based on a Woodstyle 123REV; it's 25" high, 12" wide, and 14.5" deep. The port is a flared 3", exiting from the rear, now. This is sufficient for the driver Xmax and Fb of about 31 Hz, for output up to about 105 dB without any port compression. I'm working on a set of cabinets this afternoon with the revised layout.

Regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**David Meek** · 11 February 2002, 10:42 Monday

Jon,

Please forgive a "wannabe speaker builder" question or two, but with a miking distance of 1 to 2 meters aren't you taking nearfield-type measurements? If so, does that minimize the effects of reflections and bass nulls?

David - HTGuide flunky
Our "Theater"
Our DVDs on DVD Tracker

**JonMarsh** · 11 February 2002, 14:16 Monday

Hi David,

True nearfield would be in the range of 6" to 1 foot. At this range, baffle step drop off in the lower midrange doesn't show up, for example, and the level of direct driver response totally swamps room reflections.

1 meter is distant enough that you see the affects of baffle step loss (where the speaker goes from 2pi radiation to 4 pi, with approximately a 6 dB drop in output. Now, in reality, with boundary loading, you don't get a 6 dB baffle step drop in the room, and typically baffle step compensation is often only done to 4 to 5 dB, instead of a full 6 dB. The MLS gate measurement uses a very short impulse type test signal, and is gated at an arbitrary interval to define the receiving window. The problem is, the LF response in the measurment is the reciprocal of the window length- the lower you want to measure, the longer the window must be (in msec; for example, a 10 msec window gets you down to about 200 Hz). But with the longer window, you see room reflections. Especially when you're using a longer measurement distance.

Now, of course, room reflections exist in the real world- that's why we go to a fair amount of hassle in setting up the speaker and listener location. That isn't the case when I was making these measurements.

The RTA shows the combination of the early sound arrivals and the reverberent sound, because it doesn't use a time window, per se- it's a continuous averaging. You don't design for flat on an RTA, either, because in the room it's normal for the HF sound to be absorbed more. What you do want to see, particularly off axis, is a smooth gentle roll off, but still with some good level on the highs. Wide, smooth dispersion off axis, without a dip in the upper mids below the tweeter crossover seems to correlate with having good imaging and a wide "sweet spot". This is often a problem for two way speakers, as even most 6-1/2" two ways are often designed with a crossover frequency of 3 kHz or so, which is beyond the pistonic region of the cone, and also in a region where the off axis response suffers- even just 30 degrees off axis.

For example, going from a 3 kHz crossover to the 1.5 kHz crossover I use here, in principle, I would get the same level of "beaminess" at the crossover point as the average 6-1/2" if I was using a 13" woofer! Now, I think most people can intuitively tell that a 13" two way might not be an optimum configruation, even crossed over at 1.5 kHz, but apparently it's not so intuitively obvious in the case of the average 6-1/2" system crossed over at 3 kHz or higher.

When setup for the "proper" speaker location in the listening room, the null I measeured in the RTA measurement will go away- but this is a function of placement of both the speaker and the listener. Location's such as Cardas suggest distribute the room modes and minimize the affect at any one frequency, as well as cutting the level of early arrival secondary reflections.

Boundary modes and lift, and room modes and nulls can be predicted fairly accurately with various software packages, (I even wrote a MathCAD doc for the boundary loading stuff years ago), so the problem doesn't have to be solved with just trial and error in your room. But in the final result, it's what you hear that counts!

Best regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**Jack Gilvey** · 14 February 2002, 11:30 Thursday

This is an exciting project, Jon, and I really appreciate the accessible manner in which you present it. Once finalized, might you be publishing a parts list/schematic so that mortals may take a stab?
Thanks,

Jack

**JonMarsh** · 14 February 2002, 12:31 Thursday

Hi Jack,

Thanks for the kind words. Yes, when I finish with this, anyone that's interested is welcome to the details. I'm also working on a magazine article based on this design. That's gone through a few iterations already.

I was listening and tweaking the voicing last night in the test box a little more, listening to various acoustic music, including a relatively new release from Strunz and Farah (acoustic guitar duo, spanish and Middle Eastern influences, with a variety of instrumental accompaniement). I'm pretty pleased with how things are sounding now, but I can't really finalize the design until I finish the next set of boxes. Those are in process now- may be completed by this weekend. This pair is going to my honey- unfortunately, they're obviously not ready by Valentines Day!

Then I want to build a pair or two for myself- and also see how much more I can tweak up the old Mk II's. I don't know if the MkII's or the next set will wind up in the bedroom, or doing surround duties.

Best regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**ThomasW** · 14 February 2002, 12:57 Thursday

Jack

I'm currently helping a friend (PeteG from HTF) make a pair of these. Also I'm making myself a pair but using a smaller sealed box for wall mounting

Both these designs will be fully documented. So pictures, data, construction tips, etc will be uploaded to to a website as the construction progresses.

theAudioWorx
Klone-Audio

**Jack Gilvey** · 14 February 2002, 15:30 Thursday

Great, guys, thanks.

As the M8a is listed as shielded in the PE catalog, I'd assume they're ok for A/V apps. Also, as I'd be using them as sats with an ~80Hz crossover to a sub, I think I might also be interested in a smaller, sealed version...how's response look in the sealed config.?

Thanks again,
Jack

**JonMarsh** · 14 February 2002, 18:35 Thursday

Unfortunately, Jack, they're listed *incorrectly* as a shielded driver. The 6-1/2" version is; the 8" version is *not*.

The sealed version for ThomasW is planned in a 24 L enclosure; the F3 of the box is 59 Hz, the Fb is 41 Hz, the F6 is 40 Hz. Thomas W is planning on running them on a wall, which may require "detuning" the baffle step compensation a little. If used in a "normal" way, the subjective in room response would be fairly flat to 40 Hz, though maximum anechoic SPL is only 90 DB, 96 dB allowing for room gain.

The vented alignment in the box shown is capable of 100 dB down to 30 Hz in room. This may sound like overkill for a "smallish" standalone speaker, but I rationalize it as follows:

It will work quite well for most music as a full range speaker, particularly smaller rooms, like bedrooms or small family rooms.

The chosen alignment reduces cone excursion considerably on percussive bass like kick drum and low bass guitar notes, reducing overall IM distortion on rock and Jazz, and reducing "congestion" on dynamic classical pieces. :B

It facilitates using a relatively low crossover point, while maintaining phase and amplitude integrity crossing over to a sub- for example, a 65-70 Hz crossover is handled with the requisite overlap for a smooth driver transistion, (one octave on either side of crossover), even for 2nd order crossovers (my Sumo Deliah NFB crossover). This keeps the roll off of the sub out of the audible localization range. (If you cross a sub at 100 Hz, there's still some significant output up to 150 to 180 Hz. Certainly not more than -18 to -25 dB. You wouldn't want other disotrion sources in your system only at -25 dB, would you?) :roll:

In return, you have to tolerate a bit bigger box. Well, actually twice as big. :E Sigh.... there's always a tradeoff!

Regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**Jack Gilvey** · 15 February 2002, 00:17 Friday

Thanks much, Jon...no biggie about the shielding. Sounds like I should just go for the gusto with your reflex baby as you intended her to be, port 'n all. Now I'm all psyched.

I suppose, if I got ambitious (although the chances of that kicking-in at this stage in my life are small), I could try something with that 6.5" and same tweeter in a center-channel incarnation.

Regards,
The Eggman

**JonMarsh** · 15 February 2002, 10:48 Friday

You know Jack, center channels *are* are real problem. I'm thinking about doing a design for that, too, though my ideas about the specific drivers changes, the concept doesn't.

Basically, it's a WMTW, laterally, with a mid above/below at tweeter (orientation depends on whether speaker is high or low in system; for FPTV, it would be low, and tweeter on top; for RPTV/Direct view, the other way around. Most logical combination is the same 48L enclosure size, with either a 5" HiVi midwoofer, flanked by two 6-1/2", or (since I have them) Accuton C44 dome mids. The latter would make the cabinet construction easier, since no subenclosure would be required. Tweeter will remain Focal Tc120dx2, then the Td5 version. With a mid to woofer crossover at 800 Hz or so, using a 3rd order, lateral dispersion won't be a problem; (the 3rd order has a much wider beam width at crossover than the typical 4th order L-R). For the tweeter to mid transistion, I might even be able to get away with a series quasi 6-9 dB/octave network, as both the Accuton and Focal have rather extended ranges. :B

As usual, I have a lot more ideas than hours in the day! :roll:

-Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**Brian Bunge** · 15 February 2002, 14:16 Friday

Jon,

I definitely think you should move the center channel to the top of your "to do" list. There simply aren't many quality center channels out there!

Brian

**Jack Gilvey** · 15 February 2002, 17:46 Friday

That sounds fascinating, Jon. I've long thought that the WMTW arrangement made the most sense based on my understanding of the dispersion involved, if one needs to stay with a horizontally-oriented center as opposed to one identical to the other fronts.
Maybe you can arrange for more hours in the day? I'd love to see that design also.

**PMazz** · 15 February 2002, 18:12 Friday

Another interested bystander.

Wish there was a way to help some of these projects/ideas move along. Jees, why don't you just quit your day job!

Seems to me, too, that the DIY/kit offerings are woefully missing good matching center channel plans.

Pete

**JonMarsh** · 15 February 2002, 18:55 Friday

Well, my last boss said everything would be just fine if he could just clone me! Now if I had an identical twin, like one of the quality engineers at work, that might be one solution... but weird, when you think about it!

Well, here's what's cooking on the burner's or on the prep table:

M8 two way speaker- this thread.
Modified MP-DACs- two more to build for friends. Other HTG threads.
QuatreX- an NFB balanced input amplifier design I've been playing with on and off in my SPICE simulator for a few years, which going to be built up shortly for testing, using an old Quatre Gain Cell amplifier chassis from the 70's I acquired in December.
AragonX- an implementation of the above, specific to a PCB layout to work in 8008BB's, ST's, X3's, and Palladium. Hopefully 8002, also.
Center M6+2- Center channel speaker, exact configuration to be finalized- but shooting for roughly 48L cabinet as described above. I built an 8" WMTMW with Scan Speak and MB components in the early 90's; the crossover concepts may be similar, but I'm leaing towards trying a parallel W-M crossover with a series HF. Still simulating that one, and scratching head.
Dipoles- BG75 plus?!?? ThomasW and I have been playing at this on and off. Still haven't decided on final drivers; currently with "budget" Eton 8" on midbass.
"Aerial Stryke 12" project- SW-12 Klone using Blueprint BPD1203.
AS-15- have the Stryke HE-15 driver, haven't had the time to build a box. Gee, there's a familiar pattern emerging.

So, completing the MKIII M8 is priority number one.

Completing the MP-DAC's for my buds is priority number two.

Getting the QuatreX running (last parts I *think* I needed shipped this week) and in testing/evaulation is next priority.

Everything else up for grabs.

Regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**Jack Gilvey** · 15 February 2002, 22:22 Friday

Wow, quite a full plate there, Jon. Looking forward to the following the honing of this design, and the center will get here...when it gets here.

My own plate is full of this BPD 1803 at the moment, and I still can't pull the trigger on a design :roll: ...what would you do with it?

**ThomasW** · 16 February 2002, 01:16 Saturday

Jack

after finishing Tube-Zilla and ported 1503 tube, I'd say use the 1803 in a ported tube tuned to 18Hz. The Blueprint drivers are really quite amazing. I think the ported design actually sounds best. I was going to klone the Krell MRS but got bogged down in cabinet design. That's when I decided to punt and do Tube-Zilla. The really nice thing about tube subs is that they can be whipped up fast and cheap. That's real benefit when one's stuck trying to figure out a design.

Guys

Got confirmation from PE on shipping a pair of the Hi-Vi M8a's and from Madisound regarding shipment of the cabs for both Pete's and my designs. We're only going to photo document and post my smaller sealed cab, since Jon's doing the Audio-X-Press article for the larger ported design.

Eventually both designs will end up on a website. But since neither is a klone they'll be on our new 'AudioWorx' website when the projects are complete. Most people aren't aware of the AudioWorx. It's a real company Jon and I founded in 1984 for the design and construction of original loudspeaker projects. Now that we've basically run out of fun things to klone, we're going to launch a new batch of projects that are new/original designs. We're going to target both loudspeakers and a few full blown hardware projects starting with the Hawk Audio DACs and Jon's X-mod amplifier upgrades. Some kits may evolve from these. If they don't, plans and parts lists will be complete enough for anyone to copy

theAudioWorx
Klone-Audio

**Jack Gilvey** · 16 February 2002, 09:49 Saturday

If they don't, plans and parts lists will be complete enough for anyone to copy

Excellent, thanks, looking forward to it.

Ok, ported/18Hz it is for the 1803. 14ft^3 EBS look good to you? I'd like a 10" port, but it's quite long at 39"...whadya' think?

**ThomasW** · 16 February 2002, 12:42 Saturday

Yup a little long but it might be fine. One idea for experimenting would be to make a design like the 'Flexi-tube' sub where the endcaps are held in place with long threaded rods. This would make exchanging endcaps a snap. So try out 8" and 10" dia ports. It might end up that for most 'sane' SPLSs that 8" would be fine. For an 'OH MY GOD' sub yes EBS, and 2 or 3 8" ports.

I'm running off my backup HD so I don't have the software loaded to do any modeling right now.

theAudioWorx
Klone-Audio

**Jack Gilvey** · 16 February 2002, 13:53 Saturday

Introducing the OMG alignment, a previously undescribed variation of the EBS recognizable through its prodigious porting.

**Mark K** · 16 February 2002, 19:24 Saturday

first post.
hello all. i've been a lurker for a while.

this looks to be quite fascinating.
the m8a looks like a very interesting driver.

if i may ask,
did you think at all about using the c23-6 with it?

and an off the wall thought-what do you all think of the m8a or similar in a sealed 2 way that is essentially a virtual IB. one of the options in my room is building the drivers into the wall-the backside space is ~300 liters. (couldn't resist the pun.)

**JonMarsh** · 16 February 2002, 21:24 Saturday

Hello Mark, and welcome to HTG!

I used the C23-6 in the last version. The published curves look very nice. The measured performance was not as smooth; I wonder if they didn't measure it with the protective grilles off. It also didn't go as low and smooth as the published curves suggested. That may be due to baffle size; Accuton probably measures on an IEC baffle. In the real world, I wouldn't expect to see deviations from a baffle measurement in the 1-2 kHz region, but that was the case..

Another matter was the water fall plot I measured for the C23-6- which may also be influenced by "resonances" with the grill (there's a very definite one at 15 kHz.) This behavior is not unlike what you see with a lot of the metal dome tweeters that use diffusors or other resonantor/lens assemblies in front of the dome, ostensibly to improve the off axis response, and extend the on axis measured behavior. All in all, I don't like how that affects the sound. For example, I've had a lot of success using the original MB Quart dual magnet tweeter, but always with their diffusor removed. Eventually, they even brought out a version like that because of the requests from Avalon.

So, this brings us back to the TC120tdx2. It measures well, has a minimal diffusor which is easily removed, and it sounds good (with the right electronics- you don't want to feed it a diet of mid-fi solid state, for example, not if you've voiced it relatively flat, anyway- same applies to the Accuton).

I suspect that if the Accuton had a motor and rear chamber a little more like the Focal, and ditched the metal grille, it would be killer. But it doesn't.

So, this set is proceeding with the Focal- which I was fairly happy with in my X1 klones and a couple of other speakers. The performance I measure is quite good, and the price is fairly reasonable. So far, I've got my fingers crossed, and like what I've heard to date. (solid state gear, all NLFB) The proof will be in the new enclosures, which should be finished this weekend.

Thanks for your interest- we'll keep everyone posted.

And yes, I like the M8a driver- relatively low Q at the top end breakup, nice Xmax, good T/S parameters, reasonable price. It uses a linear spider design, and claims a relatively symmertic motor drive, though I haven't tried high power sine pulses on it to evaluate that yet. I think it would even make a good MTM or 2.5 way. I have also been considering a three way with it, using the Accuton C44 mids (I have a set I'm evaluating).

Regards,

Jon

Earth First!
_______________________________
We'll screw up the other planets later....

**JonMarsh** · 17 February 2002, 20:49 Sunday

A quick update-

besides completing the documentation for the enclosure changes and most of the construction of two new enclosures, I've tweaked up the baffle step compensation and rechecked the frequency response with a "wide band" MLS (just let it rip and to heck with the room contributions! :B )

This winds up being a bit ripply because I didn't use smoothing and the room stuff over a long integration window gets factored in, but still, we're making progress here.

Now, how have ThomasW and Pete done with their projects this weekend? :?:

Regards,

Jon

For reference: meaurements made with B&K 4133 microphone, HP Mic preamp, and CLIO for Windows.

Earth First!
_______________________________
We'll screw up the other planets later....

This time for sure! (BJM)

This time for sure! (BJM)

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment