New MTM's in the Worx...

[JonMarsh]

Rockler has some nice tools and some nice wood- glad you had a chance to get into the Atlanta area store. You should see the Pau Ferro (aka Bolivian Rosewood) planks that I picked up for the Saint-Saens and Arvo Parts. Not cheap, but not that unreasonable, and definitely yummy.

Cherry sound pretty nice; are you thinking solid side panels, or what do you have in mind? Keep in mind Cherry can be photo sensitive, and will darken with light exposure depending on how you finish it.

This is typical of the grain in what I bought...




Well, the updated CLIO test PC seems to be fully functional now with the newest version of CLIO; it's the oldest working PC in my house! 5 years old, and couting... it's just too difficult or impossible to find "modern" PC chips with ISA slots. I have another backup, with a 550 MHz PIII that could be upgraded to a 1.5 GHz tualitin processor... may have to do that someday. Sigh...

~Jon

[Brian Bunge]

Jon,

I'd like to do something a la Meadowlark Audio with a solid wood outer baffle and top. With the size cabinet I'm talking about it would be extremely expensive to get a big enough sheet of veneer to do my standard "wrap" which consists of rounding the vertical front and back edges with one single seam centered on the back. So I thought of just veneering the back and sides with square corners and then gluing on the solid wood on the front and top and then rounding their edges. There's a local guy down the block from my dad's shop that can glue up the boards for me and has a huge sander that'll effectively plane the boards down smooth in just a few minutes.

Also, I'm a huge fan of cherry so I'm aware of it's photo sensitive issues. As a matter of fact, that's one of the things I like so much about it. Speaking of cherry, here are some figured cherry speakers that I just finished assembling for a local guy:

[JonMarsh]

They look very, very nice Brian! How does the solid Cherry you found at Rockler compare with the veneer you've been using?

[Brian Bunge]

The solid cherry isn't nearly as figured as that veneer. I've done a lot of stuff with standard cherry veneer and it looks very much like the grain pattern of the solid cherry I saw at Rockler. If I use the solid cherry I'll probably still use figured cherry veneer for the sides and back. The grain pattern is just too awesome to pass up.

[Hank]

:T Good work as usual, Brian! You are the cherry cabinet man for sure.
Jon, in my limited experience, I've found veneer to be more highly figured than solid wood. If you think about it, veneer is sliced from the primo logs, and the slicing exposes many, many layers, which increases the chance that lots of it will be highly figured, as opposed to the chance that a few boards cut from a log will expose the most/best figure. Statistics haunts us everywhere.

[JonMarsh]

: Statistics haunts us everywhere.

Ain't that the truth. I got some nice planks, but I cherry picked the best 30% out of what they had on hand. I feel sorry for the next guy that goes into that store in Pleasant Hill looking for some nice Pau Ferro. He's going to be disappointed...

[cjd]

I feel sorry for the next guy that goes into that store in Pleasant Hill looking for some nice Pau Ferro. He's going to be disappointed...

That's what the guy before you said!

And, if you tucked those RS180's in right next to the 27TDFC I bet it would sound 80% decent as a CC. If not better. I'm quite surprised at the off-axis response I am hearing out of my RS150/27TDFC MTMs with the tops of the drivers aligned and packed right next to each-other. Sadly, I can't get my PC to measure properly so I can't see how bad my ears are.

C

[JonMarsh]

You may be right.

There's several things complicating the spacing on my "vertical" MTM's; the RS180's don't have a lot of rear clearance, so I've got the panels back beveled; and I also have tweeter panel reinforcement boards made from nominal 1"X5", which means they're 4-1/2" wide; two stacked; first one has a pass hole for the tweeter as the front panel, second is solid. This makes the font panel much stiffer, and keeps the tweeter front plate from having any woofer back pressure.

Just one of my pet idiosyncrasies...

The spacing shown above works well with this contruction, and with a 1250 Hz crossover, even it may work OK as a center- but the version with the SS6000 should be a little better.

If you ever get that PC to meaure properly, we'd be curious to see how your's turned out.

~Jon

[Brian Bunge]

Hank,

Thanks for the compliment. I'm really pleased how the finish came out, and the speakers don't sound bad either! I've finished the sub as well. It uses dual Dayton Titanic 15's in a 28"W x 24"H x 22"D sealed enclosure which is built out of 1" and 1.125" MDF walls. Final weight is 155 lbs.!

I've still got to finish assembling the 3-way center and the 4 surrounds...

[Kramer]

John

A couple of questions from a new member ....

Hows the crossover complexity on these MTMs (mains)? How about the (XO) component cost?

How do you think these will eventually stack up to your M8a designs?

Thanks

[JonMarsh]

John

A couple of questions from a new member ....

Hows the crossover complexity on these MTMs (mains)? How about the (XO) component cost?

How do you think these will eventually stack up to your M8a designs?

Thanks

Hello Kramer, and welcome to HT Guide!

The crossover topology is very similar to the M8 designs- the topology and component count are essentilaly identical. Here's an example:



which is the current working version for the M8ta tower. Component values are different in the MTM of course, because of different crossover frequencies, and different driver impedances. Those networks are nearly done, and by sometime around the end of next week, while I'm in Denver, I can post the "release candidate" design and a BOM.

Like the M8 two way design, it will use metalized film caps and air core coils- and with the component count, it's similar in cost to a Linkwitz-Riley 4th order. The exact crossover cost depends on components- you can cut corners a bit by using less expensive caps and coils, but I'd think carefully before doing that.

ThomasW has not been a believer in "botique" caps or crossover components, but when we modified his Arvo Parts this last November with AudioCap Theta 10 uF caps in both of the tweeter caps (paralleled with the necessary GE or Solen caps to get the balance of the value) he became a believer. My Arvo's have been that way since day one. At $40 a piece per 10uF cap, it's not something everyone will go with, but I'd do that before I'd buy high buck interconnects or speaker wires. Thomas was pretty amazed at how much denser and heavier these caps are. Available through PE. Obviously, that would raise the cost quite a bit. Mine will be that way, as are my M8ta crossovers being built.

Maybe what I'll try to do is offer suggestions for "best choice", recommended, and "at least this", regarding components. Often is just comes down to the choice of type of cap - say, Dayton film cap, Solen Film cap, or?

The MTM "mains" and center channel designs have come about because we were intrested in trying out the RS180's, and because I need some smaller HT speakers in my new dedicated HT setup, but didn't want to sacrifice distortion or clarity. Though I promised myself I'd never get into building these for people, an old local friend wants a couple of sets of these for his living room and bedroom, mains and center. Compared with what Polks cost, for example, these will be very competitively priced. And better performing, of course!


There's also a local Bay area gentleman building a Thor Klone (MTM TL) using these drivers, which will use the same network design.

The RS180's have very good distortion numbers from 80 Hz through 400 Hz, comparable to much more expensive brands (like SS). But you do have to invest a bit in the crossover- the ETC cuves are good up to about 1400 Hz, but above that, regardless of what you might think based on the 1 meter frequency response measurements, they get into some energy storage problems. So, for that reason, I'm using a similar crossover frequency as the M8, around 1250 Hz.

Compared with the M8 type designs, because the Daytons have pretty high Xmax, these will play about as loud, but with the planned cabinet size won't go as low in the base. The Thor Klone version, with it's large floor standing TL cabinet, should be much better, obviously.

The M8a two way is probably still a little less expensive to build, especially if you go with the M8n driver.

There's probably going to be an M8 Mk V version I'll post in a few weeks, using the 27TDFC and M8n for maximum bang for the buck. Now, we've done MTM versions of the M8 also, with very nice performance- they'd have a clear SPL advantage for larger rooms over the RS180 MTM. But, of course, they'd be bigger. It's all about tradeoffs.

The nice thing about the MTM's with the RS180's and 27TDFC in the PE enclosures, is that there's not too much woodworking involved to get something that is seriously good performance and seriously attractive compared with name brand products. So, it's a short cut of sorts. But since building and finishing small box speakers is so NOT my thing, this is (for me) a good way to go about things.

I'm having enough trouble doing the finishing of the M8ta and Arvo's as it is!

~Jon

[Steve Goff]

Jon,

Great work! It strikes me that one might be able to come up with two versions of the MTM with bass bin, one used close to walls and one away from walls, which differ chiefly in where you roll off the 8 or 10 inch woofers. That is, in each instance the RS180s would be crossed out with a single cap with zobel, coupled with their natural roll off in the box. For speakers out in the room, the woofers in the bass bin would have considerable overlap with the RS180s, providing the boundary compensation. For speakers nearer boundaries, you could cross the 8- or 10-inch woofers out lower down, as with a conventional crossover. Is that what you're thinking?

[JonMarsh]

Yeah, that's basically it, though I don't know if Brian want's close to the wall or not- my experience is you just don't image well "close to the wall"; you've either got to be built into the wall, as in a LEDE studio, or well out from it, with normal baffle step.

But hey, we're flexible, and that's why I'm calling these Modula systems, because you can take the bits and pieces you want and put them together they way you need them. First speaker collection I built sort of like this was in the mid 80's. I was doing line array's back then, too; it used a strip of 8 Panasonic leaf ribbons on each side, then cabinets that were like stacked MTM's, except they were mid woofers and midrange domes. MB Quart 2" titanium domes. Well, I've learned a few things since then, thank heavens...

~Jon

[Steve Goff]

I concur. But I was thinking that, if you want 5 or 7 full range speakers for surround sound, some of them could be away from the walls, but the surrounds (and perhaps the center) might have to be near the walls.

[Steve Goff]

One of the first speakers I built was a passive variant of Linkwitz's original two way speaker featured in Wireless World. I had to come up with the passive fourth-order crossover myself, which was much harder in those days, especially with the available measuring systems. Instead of active delay I slanted the front baffle, and instead of active baffle step compensation I used a parallel inductor/resistor combination, in series with the KEF B110, as in the LS-35A. With pure luck I was able to make these sound really good, and a friend used them for demos several years at CES, actively crossed over to some woofers desgned by Mila Nestorovic, another friend from the early years at CES. Within a year or two Dave Wilson came out with the WATT, which looked quite similar.

[JonMarsh]

I concur. But I was thinking that, if you want 5 or 7 full range speakers for surround sound, some of them could be away from the walls, but the surrounds (and perhaps the center) might have to be near the walls.

Definitely.

Actually, I'm planning on the center being done with two crossover configs- one next to all, as needed for my FPTV, and one out from the wall, as my friend locally who wants a set needs.

The various MT's I'm looking at right now will have similar options- fortunately, it's easy to get the box data either way, then just adapt the crossover.

We're thinking to put them up in some "sticky" threads, sort of a designer/builder's library.

We'll see how it all works out.

~Jon

[JonMarsh]

One of the first speakers I built was a passive variant of Linkwitz's original two way speaker featured in Wireless World. I had to come up with the passive fourth-order crossover myself, which was much harder in those days, especially with the available measuring systems. Instead of active delay I slanted the front baffle, and instead of active baffle step compensation I used a parallel inductor/resistor combination, in series with the KEF B110, as in the LS-35A. With pure luck I was able to make these sound really good, and a friend used them for demos several years at CES, actively crossed over to some woofers desgned by Mila Nestorovic, another friend from the early years at CES. Within a year or two Dave Wilson came out with the WATT, which looked quite similar.

That sounds like a pretty interesting combo for the time. Speaker design was a lot tougher in the 70's and 80's- I can remember when I got my first copy of CACD, a DOS mode speaker design program with a basic optimizer- geez, I thought I'd died and gone to heaven!

OTOH, I've still got the same HP preamp and B&K mic I got in the late 70's... somethings don't go out of style!

[GrahamT]

A builder's library with your projects is a good idea. As well as a library of your posts explaining cauer elliptics and other commonly asked questions. It would probably save you a lot of time in the long run, and it would also probably help with the whole irascibility thing.

[Kramer]

Thanks for the reply John.

[PMazz]

PE now shows the RS due in on the 19th. I guess it's time for another GE cap buy at Maddy.

Jon, any advantage.....or disadvantage.......for me to move the Daytons closer together for a center channel? I have a 14" high baffle to work with and can offset the SS6000 so the Daytons are literally touching. Will that mess up your xover work?

Pete

[JonMarsh]

Hi Pete!

What you plan sounds OK- let's take it a little further, though. If you can give me your complete baffle dimensions, I'll run it through Baffle Diffraction Simulator, and optimize the layout on both midwoofers and tweet. Most likely, it's going to be about 1/3 from one cabinet edge to the midwoofers, and one third from the other edge for the tweeter. Won't mees up the crossover design at all to have the midwoofers closer - it's better that way, but with the PE cabinet I have some contraints, and have to also consider baffle strength, and having some room for a tweeter back panel.

~Jon

[Brian Bunge]

Jon,

Whatever I build, the front baffles will be at least 2' out from the wall. My RPTV is roughly 24" deep and there ends up being another 2-3" behind that due to cables and such. Also, my audio rack is roughly 20" or so deep and I have it flush with the front of the RPTV and it'll be within an inch or so of my left front speaker. So I'll have the front baffle an inch or 2 out from there. If full baffle step is in order for something like that then that's fine with me. Especially if it ends up taming the efficiency of the 10's enough to go with an all Dayton RS woofer design!

On that note, if the 10s' efficiency becomes an issue, do you think dual 8's would work out better in my case?

[cjd]

huh...

I was pondering what to build for my HT as I was trying to drift off to sleep last night and it looks like it may be similar to your goals here Brian. I think ~90dB efficiency with a single RS275 is about max with partial baffle step (~3-4db vs 6). Dual RS225's will manage ~90dB with full baffle step I think - I have a preliminary WWMT design sketched out using a pair of 225's, a 150, and the 27TDFC. It's fairly big. Not usable in my HT since I have to keep anything that gets into the screen height at about 6.5" width maximum.

So many projects, so little time (and space to work!)

C

[PMazz]

Image not available

The center channel enclosure is part of the soffitsub. The enclosure is fixed size and I'll make up a baffle to apply to it. Not sure if it's worth running thru BDS with that in mind. Only thing I may do if a flush baffle doesn't work acoustically, is make an angled front baffle assembly to point down at the seating area. We'll see.

Pete

[JonMarsh]

In this situation Pete, I'd recommend you consider the possibility of an angled baffle which extends out from the bottom- probably between 15 and 30 degrees from the vertical, depending on the optimum angle to the primary listening position- plan on using something like heavy felt or Whisper mat on the ceiling adjacent to the speaker.

The reason why is becuase the proximity to the ceiling will create a strong mirror image reflection that is time delayed in the upper frequency range- this will result in comb filtering with a veiled or coarse sound, unless you take some preventative acoustic measures.

With this overal configuration there should probably be no baffle step comp in the crossover, since you're basically launching from a plane boundary across the whole frequency range.

[JonMarsh]

Jon,

Whatever I build, the front baffles will be at least 2' out from the wall. My RPTV is roughly 24" deep and there ends up being another 2-3" behind that due to cables and such. Also, my audio rack is roughly 20" or so deep and I have it flush with the front of the RPTV and it'll be within an inch or so of my left front speaker. So I'll have the front baffle an inch or 2 out from there. If full baffle step is in order for something like that then that's fine with me. Especially if it ends up taming the efficiency of the 10's enough to go with an all Dayton RS woofer design!

On that note, if the 10s' efficiency becomes an issue, do you think dual 8's would work out better in my case?

Well, my original plan was based on requiring full baffle step, so this configuration shouldn't be an issue.

I'm going on a measuring binge this weekend, but the missing data will be RS270's in a suitable box- however, I can probalby work with the PE data initially. While I'm at ThomasW's next week, I'm going to spend a lot of time cranking through these designs and documenting stuff. I've already gotten the basic MTM networks worked out in LspCAD6, its just a matter of working with the exact measured data instead of something that's just really close...

Dual RS225's is something I thought of too, but it seemed with those one would be better off ported. If you do that, you wind up with a box as big as the sealed box for the 10's. With baffle step, the net efficiency on the 10's will drop back down to about 90 into full space - but as they're running mostly below 100 Hz, there is room gain to consider. I usually use a placement about 2 meters from the wall; this moves the rear wall boundary effect down to the 50 Hz area or so. When you say two feet, do you mean from the rear of the enclosure, or the front? The radiation boundary resistance is determined from a combination of driver and port location if ported.

Later - maybe next week? - we should start a seperate thread just for discussion about this three way system.

~Jon

[Brian Bunge]

Jon,

If you look back at my first sentence, you'll see I said "the front baffles will be at least 2' out from the wall."

Yeah, I remember dual 8's ported ended up being around the same volume as the dual 10's sealed. The 8's would obviously have better extension down low, but the 10's should have lower distortion and greater potential Vd.

Here's another thought. Maybe add one of Adire's new UCD amps to each cabinet to power the 10's. Mount it and a power supply, some RCA's, gain control, etc. to a 1/8" plate and you're ready to go. That way I assume you could adjust the gain control to properly mate the 10's with the rest of the speaker. I know, I know...I'm really blowing my currently non-existant budget with this line of thinking.

[JonMarsh]

Yeah, but I like the way you're thinking- creativity at it's best - there's usually always a way to come up with the money if you've got good ideas.

That would also be an easier way to go IF you don't have an industrial grade amplifier, becuase then you wouldn't have to worry about the paralleled loads of the drivers. OTOH, my concept originally was a Wilson Audio like approach; the low bass cab would be rolled off slow initially, and fill in the drop from baffle step on the 7's. This is how the Wilson Watt Puppy, the Wilson X1 SLAMM, and the Wilson X2, as well as the Maxx work. This approach gives some advanates, because in the frequency range where you're most likely to have some issues from floor bounce, you have overlap on the drivers and a tall source, which minimizes floor bounce effects.

We'll continue to discuss...

~Jon

[Brian Bunge]

I assume the approach that Wilson has taken saves some efficiency in the mids at the expense of some efficiency for the woofers. It appears most larger woofers from the same line seem to have higher efficiency than their smaller counterparts. If this is the case, I assume this would be a good trade-off.

I did have an industrial amp (K2) but sold it when I sold my Tumult. I was first thinking about using the lower buck PE plate amps but don't know if they'd be worth using up to frequencies that the 10's would be running. I'm sure most of the powered towers on the market use plate amps crossed around 80Hz so sound quality probably isn't an issue.