My first project is a 4 way speaker...

[cochinada]

Hello guys!

Before you say I'm crazy (which I am a little I confess) please indulge me as I make a brief introduction of myself and my goals since this is my first post here.

This is also my first original speaker project although I have already executed two Zaph projects as well a pair of subwoofers entirely conceived by me. I discovered the wonderful world of DIY some years ago and became hooked. It all began after listening the system of this friend of mine that is actually some kind of electronic/audio guru and makes or modifies unbelievable good powers and pretty much everything he wants to do. He has also a pair of vintage modified B&W 801 by the way.

So, after this experience I bought some used equipment and asked him to make his magic. Afterwards I decided to built my current stereo speakers and I learn a lot on the process. They are indeed great speakers. Then I designed my current subwoofers and finally the center speaker which is also a project by Zaph.

But I wanted more and some months ago this crazy idea come out of nowhere. What if I built my dream speakers on a budget? The challenge is enormous of course as John Krutke is well known among the DYI community by his skills and knowledge that often if not always produce wonderful projects. But I like challenges and decided to give a try.

On my next post I will introduce my project. I will definitely need the help of everyone here and from the level of knowledge and experience that some members have I'm feeling optimistic that the final outcome will be find.

Thank you all

[---k---]

You're crazy! But crazy can be fun.

You're probably typing your next post now. But I have to ask what makes you think you need a 4-way over a 3-way? Even in the commercial world, you don't see many 4-ways. I'm just not sure the complexity is worth it versus an equivalent 3-way. And if you're using a sub with a receiver crossover, you have a 4 way, don't you. or you looking for a 5-way?

[Hdale85]

Personally if I were going 4-way I'd likely do a 3way with an integrated powered subwoofer.

[cochinada]

Hi Ryan! Yes I know

Well since my current speakers are a three way and I decided to try to build the best speakers I could possibly built and also because I want them to be a full range and am not afraid of the complexity I made the decision from the start: it has to be a 4 way.

I won't be using my subs nor the receiver for stereo as this one is not at the same level of my stereo gear. That stuff is only for movies and surround.

Now about my project, my initial thoughts were to go with two 15 inches woofers for the lower bass but I dropped the idea because for design reasons I could only place the drivers at the middle height and I was afraid that this wouldn't work so good. At this time I had already pictured the basic shape of the speaker and had made some decisions that I would not forfeit. Namely, the big woofers would have to be placed on each side of the speaker so that the box vibrations would be minimized (or so I hope) and there has to be an MTM in the middle. I know this poses some issues like off-axis response for example but believe me that I've simulated countless drivers and combinations to come to the best compromise possible. Basically, my goals were to minimize distortion and increase sensibility as both normally go hand in hand I guess. I also wanted my speakers to be able to full a large room if needed and last but not least to be as small as possible as my room is not that big by a long shot. All of this spending the least possible but never jeopardizing these goals.

So, after much though I finally settled for the drivers topology. It was gonna be a WWMTMWW in the front baffle with 4 extra woofers on the sides. Their sizes are 4", 8" and 12" and for the tweeter it could be either a 1" or a 1.2".

like I said, I've considered lots of different units and manufacturers and ultimately narrowed these choices to the list bellow:

Tweeters:
C25-6-158
C30-6-358
C30-6-024
D2908/714000
D3004/602010
N26CFR-A
N26MGR-G
N26CR2-G

Midranges:
C50-8-044
C90-6-724
12W/8524G00
12MU/8731T00
TI 100
ZA14W08

Mid Woofers (upper bass):
C158-8-085
C158-6-851
C173-6-191E
C220-6-222
21W/8555-00
22W/8534G00
22W/8851T00
22W/8857T00
AL 200

Woofers (upper bass):
RSS315HF-4
SB34SWNRX-S75-6
30W/4558T00
32W/4878T00

I've simulated the MTM responses using PCD from Jeff Bagby and want to thank him for the great program and also for the answers he gave me by email. With so many combinations I came up with almost 20 possible variants but ultimately reduced the candidates to the ones in green, some because of economic reasons and others because of aesthetics and vertical off-axis response. For instance the ZA14W08 is too big for an MTM in my case where the smallest tweeter AND midranges are desirabled. That would almost exclude the Transducer Lab tweeters...

PCD as great as it is has its limitations, namely there is no straightforward manner to implement a 4-way speaker nor one with 4 equal drivers per way and because I don't have any "profissional" tool for this purpose I though to start with the MTM, save the response curves and import them as a tweeter using this time as midranges and woofer my 8 inches units. Now I have a problem but I leave that for latter...

[cochinada]

Personally if I were going 4-way I'd likely do a 3way with an integrated powered subwoofer.

Hi there!

Right on the target! I forgot to tell that the subwoofers (12") are going to be active. That's for sure but perhaps I will not stop here. This is related to the problem I'm handling right now.

To conclude my quest for units I decided and have already ordered the following per speaker:
1 x N26CR2-G
2 x 12W/8524G00
4 x 22W/8534G00
4 x SB34SWNRX-S75-6

My primary constrains were of economical reasons of course and that's why I opted for example for the 22W/8534G00 instead of 22W/8551T00 or 12W/8524G00 instead of 12MU/8731T00. I also got a huge discount for the SB34SWNRX-S75-6. Nevertheless at least the 4" and the tweeter measure incredibly well in terms of low distortion and frequency response and the big Discovery is no slouch either. I'm talking about independent measurements/reviews like the ones on Hobby HiFi magazine and Zaph site or AudioGurman. I was a little afraid of the 22W/8534G00 but since I'm going to use 4 of them this has to be an extra factor as distortion must be looked at in perspective.

[5th element]

Interesting driver choices and plan for the loudspeaker. I think lots of us have similar systems in our homes, that would be a three way + sub type loudspeaker and the concept works very well, I have a similarly themed set up myself and am very happy with it.

Interesting driver choices, but as you say budget was important in the decision making, I am sure you will be happy with the end results. The stand out driver, as you could say, is probably going to be the transducer labs tweeter and I would most certainly want to exploit it's strengths given the type of system you are wanting to design. Because you've gone MTM and due to the tweeters large face plate this automatically places constraints on the mid to tweeter xover point with lower being much better than anything else. Luckily the TL tweeter has very low distortion and can operate extremely well with low crossover points. You are really going to want to cross over as low as the tweeter can manage whilst keeping clean at high SPLs.

[Carl V]

last year I listened to a reasonably well sorted out 4-way with a Dueland XO topoology.
some of His Commercial speakers in the Past have been Pass labs, and Big Wilsons.
he has done some other large driver three ways.

Good luck & i look forward to seeing your progress.

[cochinada]

Interesting driver choices and plan for the loudspeaker. I think lots of us have similar systems in our homes, that would be a three way + sub type loudspeaker and the concept works very well, I have a similarly themed set up myself and am very happy with it.

Interesting driver choices, but as you say budget was important in the decision making, I am sure you will be happy with the end results. The stand out driver, as you could say, is probably going to be the transducer labs tweeter and I would most certainly want to exploit it's strengths given the type of system you are wanting to design. Because you've gone MTM and due to the tweeters large face plate this automatically places constraints on the mid to tweeter xover point with lower being much better than anything else. Luckily the TL tweeter has very low distortion and can operate extremely well with low crossover points. You are really going to want to cross over as low as the tweeter can manage whilst keeping clean at high SPLs.

You are probably right. These tweeters aren't as expensive as some others that measure quite worse. From what I've been reading and looking I have also high expectations about its performance.
On my simulations I've been targeting a xo point around 1700Hz because of the reasons you mentioned. I'm afraid to go lower than that. Nevertheless I was able to get this curve with just the MTM:

Image not available

I kept the distances between the drivers as close as possible so each midrange will be at 126mm from the center.

For a moment I thought about using the rectangular version which is a tad shorter but I didn't like how it looked. So, it didn't pass the aesthetics department you could say.
The D3004/602010 for instance, would allow the drivers to come much closer hence providing a better vertical off-axis response but it is not as linear nor has the same low distortion at the same time having less sensitivity, although it is still a great tweeter!

[cochinada]

last year I listened to a reasonably well sorted out 4-way with a Dueland XO topoology.
some of His Commercial speakers in the Past have been Pass labs, and Big Wilsons.
he has done some other large driver three ways.

Good luck & i look forward to seeing your progress.

Thank you

The crossover if there was only an MTM is this:

Image not available

I'm sharing the 'frd' and 'zma' files for the MTM that I traced with Spltrace.

[cochinada]

Although it is not the final drawing it is very close as any further changes will depend mainly on the internal volumes needed according to the simulations, so here is a picture of my future speaker.
As you can see I'm planning on using baltic birch this time instead of MDF as I don't want to paint it. I'm just a little concerned about the effect of those slices at the corners mixed with the panels. I didn't want something too contrasting.

Image not available

You will also notice that the MTM is not centered as I can achieve a better diffraction this way (more about that latter on).

[5th element]

With an xover around 1700Hz you are certainly on the right track and with its healthy amount of xmax and underhung motor geometry, you should be more than okay. Linkwitz crossed the SEAS Millennium tweeter around 1440Hz with a 4th order acoustic slope and he did this empirically, making sure that the tweeter would not exceed its linear limits when crossed so low, but still being capable of hitting target reference SPLs. Certainly the TL tweeter is just as capable (SPLwise) as the SEAS, so I don't think you will have any trouble with it.

The layout of the bass drivers, having some mounted near the floor and some up high will also help to smooth out floor bounce issues, a 300Hz or so xover will probably be a good idea between the MTM section and the WWs.

[cochinada]

With an xover around 1700Hz you are certainly on the right track and with its healthy amount of xmax and underhung motor geometry, you should be more than okay. Linkwitz crossed the SEAS Millennium tweeter around 1440Hz with a 4th order acoustic slope and he did this empirically, making sure that the tweeter would not exceed its linear limits when crossed so low, but still being capable of hitting target reference SPLs. Certainly the TL tweeter is just as capable (SPLwise) as the SEAS, so I don't think you will have any trouble with it.

The layout of the bass drivers, having some mounted near the floor and some up high will also help to smooth out floor bounce issues, a 300Hz or so xover will probably be a good idea between the MTM section and the WWs.

I'm very relived to hear that! About the crossover between the MTM section and the WWs I was also thinking about 300 or 350 Hz but using PCD and the text book values AND regardless of the order of the filter I chose I got scary component values. At some point I saw inductors larger than 12mH and capacitors over 1000uF and no matter what I did this seems really a mission impossible as at the same time the impedance of the system was reaching 2 ohms or so.

After much fiddling around and basically changing every component I was able to obtain a much nicer looking curve and here it is:

Image not available

The problem now is that I have no idea where the crossover points are anymore. From a visual inspection I would estimate the LP (3rd order) at 580Hz and the HP (2nd order) at 760Hz but these are wild guesses...

These are the xo values:

Image not available

Image not available

continues...

[cochinada]

Next I simulated the WWMTMWW in VituixCAD where I could import all the drivers at the same time without any constrains and after replicating the same crossovers the curves were more or less identical although not quite.
So I decided to play around and this time changing literally every component including the ones on the tweeter xo. This is what I came up with:

Image not available

and a close zoom on the frequency curve...

Image not available

and the respective crossovers...

Image not available

Now I don't even know where the crossover points are at the tweeter . Please help!

By the way, you will notice that I'm not very concerned in using real component values at this time and because of that some inductor resistances are still 0.

[cochinada]

Well I have just noticed a problem with VituixCAD. Sometimes the driver connecting mode changes from parallel to series so I've reported the problem and it was fixed minutes later! Kimmo who is the developer of VituixCAD provides a great support and from my experience reacts almost instantly. Aside from that his program allows several things like multiway with several drivers which for my project is priceless.

New curves bellow.
Seems OK...

Image not available

BIG PROBLEM -> total Impedance (in black) is dangerously low at 2 ohm!!!

Image not available

Where are the crossover points? I have no clue and worry in particular about the MTM...

Image not available

Image not available

So basically right now I don't know what to do to fix the impedance and don't know where the crossover points are...

[---k---]

Wow. That is going to be one hell of a box to try and move around. I hope you have a forklift in your living room. You should really consider making the boxes modular. Also, I've been looking at several translam threads recently. And while they all look cool, there were several that had cracking issues. I'm now a little scared of translam.

It looks like you've put more thought and effort into the crossover than most and are trying to do it right. I might have to take my "You're Crazy" back.

I know the subs look cool in the mains. But have you thought about how the room positioning will affect them. Below 100 hz, room modes are the single biggest factor in frequency response. One of the main benefits of subs is that you can place them where they have the least destructive influence (notice the wording). You're loosing this benefit. Something to think about.

A few years back, someone took CJD's Canaiolo project and added subwoofers above using an active crossover. IIRC, after the initial amazement of the size of the project wore off, there were some issues. I need to go find that thread, because I've referenced it a couple times now and still can't remember the ending.

[cochinada]

Wow. That is going to be one hell of a box to try and move around. I hope you have a forklift in your living room. You should really consider making the boxes modular. Also, I've been looking at several translam threads recently. And while they all look cool, there were several that had cracking issues. I'm now a little scared of translam.

Yes I pondered that so I decided that each speaker is made of 3 boxes where basically both bass boxes are equal with one upside down with just small details setting them apart. They all fit like a puzzle without need of any screws or attachment of any sort. I already figured that out. I may be crazy but I'm also very picky. By the way, you're right. By my calculations, each speaker will weight around 155Kg and measure 1.71m x 0.46m x 0.56m so they are not that big. Excuse my ignorance but what is a translam?

It looks like you've put more thought and effort into the crossover than most and are trying to do it right. I might have to take my "You're Crazy" back.

I'm a bit perfeccionist but don't be so hasty yet 8)

I know the subs look cool in the mains. But have you thought about how the room positioning will affect them. Below 100 hz, room modes are the single biggest factor in frequency response. One of the main benefits of subs is that you can place them where they have the least destructive influence (notice the wording). You're loosing this benefit. Something to think about.

A few years back, someone took CJD's Canaiolo project and added subwoofers above using an active crossover. IIRC, after the initial amazement of the size of the project wore off, there were some issues. I need to go find that thread, because I've referenced it a couple times now and still can't remember the ending.

This makes perfect sense but I decided long ago not to use my subwoofers as I'm also a bit stubborn and really like the aesthetics of those nice 12 inches on the sides.

By the way, I noticed that you also built the ZD3C from Zaph. I have to take a picture and show you mine as well. The difference is that I opted for the vented version.

[knowledgebass]

... Excuse my ignorance but what is a translam? ...

Translamination. When you laminate strips of plywood and leave the plys exposed.

I used this technique in my Ansonica Build (as was CJD's original build). I also have some advice posted here for anyone thinking of going this route.

Someone also linked to this ambitious line array project that had some unfortunate issues (it was repaired in the end).

[Hdale85]

A 3 way can be perfectly full range, especially when it comes to music. With some larger mids, and larger mid-bass's (10"?) you could have something with an f3 well below 30hz with the right cabinet setup. Just throwing that out there lol, I do like what you've drawn up but that's going to be one huge and heavy beast.....As mentioned I'd probably make them in 3 sections and make it so you can bolt them together through the inside or something.

[cochinada]

Translamination. When you laminate strips of plywood and leave the plys exposed.

I used this technique in my Ansonica Build (as was CJD's original build). I also have some advice posted here for anyone thinking of going this route.

Someone also linked to this ambitious line array project that had some unfortunate issues (it was repaired in the end).

This is new to me and valuable information that I will consider very carefully.

If you look at the picture I've posted you will see that I only use translamination at both front edges of the bass boxes due to being round and the MTM box itself which is entirely made using this technique with the exception of the baffle. Since a picture is worth a thousand words let me ilustrate both cases:

Image not available

I'm planning on joining the layers with wood dowels which is the same technique I used for my subwoofers. Here is just a glimpse to show you:

Image not available

And because the subwoofer was covered with cork panels I don't know weather cracks have formed or not. Ignorance is bliss 8).

Image not available

About the MTM the case is more complex as you can see in this picture without the baffle but I will still use dowels and not rods.

Image not available

Anyway I will speak with my carpenter and let him know my concerns when the time comes. I'm sure he will know how to prevent this either using shellac or whatever other techniques that I'm not aware off. The picture of my subwoofer was taken on his shop by the way.

EDIT: Doing many things at the same time usually doesn't work very well. My subwoofer is made from MDF and as such I guess is less prone to such hazards comparing to baltic birch for instance. After all MDF is not made of multi layers and the only places where I can think that cracks can appear is between each layer if they are not properly glued.
I thought of using 30mm Baltic Birch this time, mainly because of the finishing that is much easier this way. If I'm going to use MDF I would have to use some kind of veneer and even so I don't see it feasible specially on the top and bottom of the speaker because the edge is rounded all around.

[Jonasz]

Nice project!
I have a similar system and don't regret building them at all, once you have a truly full range system, it's kind of painfully obvious it's a lot missing below, say 300hz, in the typical 7"+1" twoway...

[knowledgebass]

I also don't believe MDF would behave the same way. At least, not to the same degree as wood and veneers. People do have issues with movement of butt joints with MDF so it does do some expansion/contraction. What kind of glue will you use? What kind of glue did you use for your previous sub project?

For the baffles are you planning to leave them floating? If you bolt them down from the front and seal with a gasket that would allow the laminated sections to move freely (I see this more commonly with people using solid wood baffles). If you glue them in place you may be subject to differential expansion at the back of the box (and potential cracking). What are you going to do for the baffle finish? Wrap the baffles in fabric/leather or veneer? I think the look would will be cool. For the shapes you've given I assume you're going CNC? Part of the issues I had were a result of the time it took between cutting, and gluing and assembling the boxes. It took something like 4 months between cutting my lamination strips until I was gluing up boxes. If you're doing it in a couple days or even weeks, you probably don't need all interim finishing steps I'm suggesting. I would recommend that you get some sort of sealer on the inside and outside ASAP to stabilize the pieces, especially if it's going to take months to do the final finish. So it seems your towers will be 5 separate pieces that will then stack/lock together like Lego? For the front pieces (WW+MTM+WW) you'll have some potential vertical shrinkage but I think your subs might have more shrinkage/expansion because you'll have 4 solid plywood caps on your laminations (top and bottom per box) vs 6 cap laminations in you WWMTMWW sequence. Those caps won't expand or contract nearly as much as your 'ring' or box wall laminations which could result in a slightly different height between your subwoofer stack and the WWMTMWW stack. I'm thinking, maybe 2mm difference, but your overall change would be more than 2 cm for that many layers if left unfinished (I had ~4mm change on 8 layers of 18mm ply). Consider that for how you will affix the front and back boxes to each other and how they might be allowed to move independently so that dowels don't wrack becoming inseparable, or worse, crack laminations.

I'm sure we'll get there eventually when you start the build, but would love to see an exploded view of your cabs! The cabinet engineering is the part I understand the best

All of this also depends on where you live. My construction ranged from +30°C to -30°C in my garage, to high summer humidity (I sometimes get condensation on my granite-top table saw) and very dry humidity in my house during the winter (<30%). If you live somewhere arid (desert), or where humidity does not change drastically or seasonally (equator), this whole discussion could be much less of an issue for you. The ~4mm movement I spoke of occurred between late summer when I glued up my panels and cut the pieces for my boxes, and winter when I actually glued up the boxes.

[cochinada]

Nice project!
I have a similar system and don't regret building them at all, once you have a truly full range system, it's kind of painfully obvious it's a lot missing below, say 300hz, in the typical 7"+1" twoway...

Thanks! Nice to hear that as well.

I also don't believe MDF would behave the same way. At least, not to the same degree as wood and veneers. People do have issues with movement of butt joints with MDF so it does do some expansion/contraction. What kind of glue will you use? What kind of glue did you use for your previous sub project?

For the baffles are you planning to leave them floating? If you bolt them down from the front and seal with a gasket that would allow the laminated sections to move freely (I see this more commonly with people using solid wood baffles). If you glue them in place you may be subject to differential expansion at the back of the box (and potential cracking). What are you going to do for the baffle finish? Wrap the baffles in fabric/leather or veneer? I think the look would will be cool. For the shapes you've given I assume you're going CNC? Part of the issues I had were a result of the time it took between cutting, and gluing and assembling the boxes. It took something like 4 months between cutting my lamination strips until I was gluing up boxes. If you're doing it in a couple days or even weeks, you probably don't need all interim finishing steps I'm suggesting. I would recommend that you get some sort of sealer on the inside and outside ASAP to stabilize the pieces, especially if it's going to take months to do the final finish. So it seems your towers will be 5 separate pieces that will then stack/lock together like Lego? For the front pieces (WW+MTM+WW) you'll have some potential vertical shrinkage but I think your subs might have more shrinkage/expansion because you'll have 4 solid plywood caps on your laminations (top and bottom per box) vs 6 cap laminations in you WWMTMWW sequence. Those caps won't expand or contract nearly as much as your 'ring' or box wall laminations which could result in a slightly different height between your subwoofer stack and the WWMTMWW stack. I'm thinking, maybe 2mm difference, but your overall change would be more than 2 cm for that many layers if left unfinished (I had ~4mm change on 8 layers of 18mm ply). Consider that for how you will affix the front and back boxes to each other and how they might be allowed to move independently so that dowels don't wrack becoming inseparable, or worse, crack laminations.

I'm sure we'll get there eventually when you start the build, but would love to see an exploded view of your cabs! The cabinet engineering is the part I understand the best

All of this also depends on where you live. My construction ranged from +30°C to -30°C in my garage, to high summer humidity (I sometimes get condensation on my granite-top table saw) and very dry humidity in my house during the winter (<30%). If you live somewhere arid (desert), or where humidity does not change drastically or seasonally (equator), this whole discussion could be much less of an issue for you. The ~4mm movement I spoke of occurred between late summer when I glued up my panels and cut the pieces for my boxes, and winter when I actually glued up the boxes.

Here is an inside view of the beast:

Image not available

The drawing is also my favorite part and I have drawn several speakers, specially from Troels Gravesen site just for the pleasure of doing it it.

I'm planning to glue the baffles because it gives a more professional look but making them detachable is not yet totally out of the question but almost. I wasn't considering a different finish for them anyway. I could always paint them in mate black I guess but like I said earlier I will try to avoid painting as much as possible. First because I can't do it my self as I don't have the space for even trying and then because I don't know any trusty painter. The one that painted my Zaph SB12.3 made a lousy job. :evil: At this time I have to options: using baltic birch and a simple varnish or MDF and oak veneer or something equally nice but I have the problem with the tops as I mentioned. On the other side I would avoid the potential crack problem...

My carpenter has a CNC machine and he's already used to receive my drawings in dxf format which makes everything simpler.

Each tower is made of 3 pieces and yes, they will fit like lego. I liked that analogy . Perhaps this picture will give you a better idea:

Image not available

Bear in mind that this is not the final drawing. Depending on the crossovers I will have to re-calculate the inside volume for the drivers and make some adjusting. Some details are not yet resolved and the wholes in the braces are not final as well. I will have in any case always to separated compartments, one for the subwoofers and the other for the 8 inches as I hope it is clear on the pictures.

I leave in Portugal, famous for the nice weather and temperatures here rarely go bellow 5ºC or over 40ºC

[cochinada]

This is the top (and the bottom) that I can't figure how it can be veneered because of its shape:

Image not available

Any suggestions?

[Steve Manning]

cochinada nice looking project, can't wait to see how it turns out. I did a translam project a few years ago and I would agree with knowledgebass with his suggestions if you are going to use baltic birch. I had splitting on my project like others have, so here is my two cents worth from what I learned, they may duplicate what knowledgebass has already said .......

* If you can I would use bamboo plywood, it's a lot more stable than baltic birch.
* Seal inside as well as the outside of the wood to get even moisture absorption.
* I would not use dowels to hold the layers together, it will constrain the plywood and not allow it to move.
* Do not glue the baffle in place, same reason as the dowels, float it with a gasket. Magico did that with their original Mini.

Steve

[deewan]

This is the top (and the bottom) that I can't figure how it can be veneered because of its shape:

Image not available

Any suggestions?

What about using solid wood? Or going even crazier and using granite or forming a mold and pouring dyed concrete. It's easy for me to suggest these things since I don't have to pull off the labor.

Should be an amazing build. Good luck!!!

[cochinada]

Solid wood crossed my mind. I just wonder if wood and MDF glue well together and also if is easy to find a large piece like that but I will investigate the possibility for sure.
Thanks a lot man and keep suggesting by all means.

[Face]

Wood and MDF glued together doesn't always fare well due to different expansion rates between the two. Solid wood in general can be very difficult to work with, which is why most use some sort of ply or MDF.

[Hdale85]

How deep is the MTM box? Seems shallow, in which case rear reflections and such may cause some bad things? Granted those are monsters and it's hard to tell.

[cochinada]

Wood and MDF glued together doesn't always fare well due to different expansion rates between the two. Solid wood in general can be very difficult to work with, which is why most use some sort of ply or MDF.

I thought there might be a problem of the kind...

How deep is the MTM box? Seems shallow, in which case rear reflections and such may cause some bad things? Granted those are monsters and it's hard to tell.

Yes it is a little shallow but with some ingenious I still have a small margin of perhaps 1cm to 5cm to make it a little deeper without having to change much the bass boxes or increase the depth of the whole speaker which I don't want to do no matter what. As it is right now you have about 3cm between the driver end and the back which is not much I know but you have to look this in perspective as the driver is only a 4" and so is really small. The inside is probably going to maintain that basic shape that I empirically hope will work OK, but most certainly it will have to expand both sideways (or deeper) in order to reach the optimum volume for a Qtc of 0.707. At the moment the inside volume discounting the drivers is 4.8l and I probably have to go for 8l or 9l.

Image not available

[cochinada]

It's always a good thing to talk with more experienced people so after consulting my carpenter and telling him my concerns about veneering the top or using real wood, Baltic birch, cracking (I showed him these two pictures bellow) and so on, this is what he told me:

Image not available

Image not available

"In narrow places like on the first picture you have the same potential hazards with MDF but in the drawings you sent me it doesn't seem you have such situations.

On the second picture it seems that the crack is between two layers. If so, it can be due to bad gluing, bad glue, veneering without proper inside treatment or simply bad quality ply.

Veneering the corners and combining the top with real wood is certainly possible but I don't have the means to do it at 100%.

Ply (Baltic birch) with varnish leaves always the contrast between layers and faces. It's a matter of Googling "ply furniture" and see if you like it.

About the painting I believe I can push the painter not for a perfect job but for perhaps 99%."

And so I Googled and I liked it
99% is not good enough I'm afraid and the cost of painting would probably be too much comparing to varnishing.
So, notwithstanding other opinions I'm very inclined to go for Baltic Birch.

Aside form the fact that black in black is very hard to see what do you think between these two options?

Image not available

[deewan]

I like the contrast of the non-black cabinet. I don't remembering seeing you mention what type of room these will be used in. Is it a dedicated 2-channel listening room? Is it a home theater where you may want black speakers to 'disappear' when the lights go out?

Another idea for the top plate where veneering may be difficult, if you are open to a solid color or some sort of pattern, look into vinyl car wraps. With just a little practice and some heat you can wrap curves pieces with vinyl wraps and make it look like one solid piece. There are better instructional videos to watch, but this is short enough to give you an idea. I've done it with some crazy curved baffles using carbon and chrome wrap.

https://www.youtube.com/watch?v=dAMoCoDs1YU

[---k---]

Very beautiful renderings. I'm scared to ask how many hours you've spent on those. Really really impressive.

Problem with MDF and the butt joint shadows, I always was suspicious that the MDF soaks up a lot of moisture from the glue causing it to expand some, then as it dries it shrinks. Paint to soon and you get the joint lines. If I were doing these, I might glue up all the horizontal layers and let it dry for a week before glueing it to the large top and back that might restrain it.

I too am concerned about the depth of your mid boxes. With the complexity and beauty of what you're building, it just doesn't seem right to compromise in any area.

I'd love to know what your carpenter is charging you to build these. I would guess it is a few thousand dollars. Probably will be worth it when you're done, but if it were me I couldn't' afford to have mistakes with it. Some cheap test boxes to try out the different volumes, shapes, and construction techniques may be well worth your time and effort.

PE Tech Talk has a whole thread going on using solid wood with MDF. One guy posted showing his cracked and several people say they've done it without problems and it goes from there. Might be some useful information. I just know that the craftsman who make fine furniture spend a lot of effort to account for moisture/temperature movement of the wood. They even go through the trouble of veneering the inside to make sure the panel is balanced.

Hopefully your carpenter knows all this and has a plan. We just don't want to see you have a problem. People see to have problems randomly with mdf, and cracking and stuff, so I don't think there are clear cut rules.

[Steve Manning]

I did what your suggesting with gluing real wood to mdf and have had no problems with it at all. Of course it might be species dependent or that I just got lucky I've attached a few pics.

[5th element]

It's also dependent on how thick the wood itself is. I have done with Steve just posted, but with much thinner wood. I used mahogany. The wood did crack about a very thin section, but this isn't visible as it's concealed beneath the drivers frames.

[Steve Manning]

My vote is for the BB version ....... if you do go the painted mdf version, Ryan is right about the edges telegraphing, my last project was painted and I can see some edges. Since then I have been doing a lot of digging and the best solution I found is to seal the wood with a polyester resin gel coat first, it's used in the marine industry to seal against water. You then paint over that, Wilson Audio goes this route and have some videos you can watch. I also ran across this stuff http://www.rustoleum.com/product-cat...surface-sealer which is supposed to work very well. I have not had a chance to try either yet.

[Face]

I did what your suggesting with gluing real wood to mdf and have had no problems with it at all. Of course it might be species dependent or that I just got lucky

How long has this held up?

[Steve Manning]

How long has this held up?

Going on 8 years at this point with no cracking.

[deewan]

My vote is for the BB version ....... if you do go the painted mdf version, Ryan is right about the edges telegraphing, my last project was painted and I can see some edges. Since then I have been doing a lot of digging and the best solution I found is to seal the wood with a polyester resin gel coat first, it's used in the marine industry to seal against water. You then paint over that, Wilson Audio goes this route and have some videos you can watch. I also ran across this stuff http://www.rustoleum.com/product-cat...surface-sealer which is supposed to work very well. I have not had a chance to try either yet.

All good advice. I've used Bondo, sanded smooth, then primer and paint. Going on 4 years in a basement where the humidity changes from below 20% in the winter to 50% in the summer, and I've had no issues. But I think I may like your method more!

[cochinada]

I like the contrast of the non-black cabinet. I don't remembering seeing you mention what type of room these will be used in. Is it a dedicated 2-channel listening room? Is it a home theater where you may want black speakers to 'disappear' when the lights go out?

Another idea for the top plate where veneering may be difficult, if you are open to a solid color or some sort of pattern, look into vinyl car wraps. With just a little practice and some heat you can wrap curves pieces with vinyl wraps and make it look like one solid piece. There are better instructional videos to watch, but this is short enough to give you an idea. I've done it with some crazy curved baffles using carbon and chrome wrap.
https://www.youtube.com/watch?v=dAMoCoDs1YU

I don't have a dedicated room unfortunately. It's a normal living room and not even big.

This is a very interesting idea but looks easier than probably is. I actually have already seen a guy wrapping a Mclaren super car on youtube.

[Paul W]

Since your first project is a lot of work and expense, be confident in your acoustic measurement skills. If not, consider lots of practice on other speakers while the cabinets are being made.

[cochinada]

Very beautiful renderings. I'm scared to ask how many hours you've spent on those. Really really impressive.

Problem with MDF and the butt joint shadows, I always was suspicious that the MDF soaks up a lot of moisture from the glue causing it to expand some, then as it dries it shrinks. Paint to soon and you get the joint lines. If I were doing these, I might glue up all the horizontal layers and let it dry for a week before glueing it to the large top and back that might restrain it.

I too am concerned about the depth of your mid boxes. With the complexity and beauty of what you're building, it just doesn't seem right to compromise in any area.

I'd love to know what your carpenter is charging you to build these. I would guess it is a few thousand dollars. Probably will be worth it when you're done, but if it were me I couldn't' afford to have mistakes with it. Some cheap test boxes to try out the different volumes, shapes, and construction techniques may be well worth your time and effort.

PE Tech Talk has a whole thread going on using solid wood with MDF. One guy posted showing his cracked and several people say they've done it without problems and it goes from there. Might be some useful information. I just know that the craftsman who make fine furniture spend a lot of effort to account for moisture/temperature movement of the wood. They even go through the trouble of veneering the inside to make sure the panel is balanced.

Hopefully your carpenter knows all this and has a plan. We just don't want to see you have a problem. People see to have problems randomly with mdf, and cracking and stuff, so I don't think there are clear cut rules.

Thanks man! It's better not to ask but what takes more time are actually the drivers. Some are really challenging and can take up to 5 hours or even more. At the beginning when I was less experienced some took days but I like to draw and I think it's important to have a model as accurate as possible not only because you can see much better how the driver fits but also because you can have a much better volume estimation. This one for instance took roughly 5 hours:

Image not available

Concerning the MDF, my carpenter advised me right at the beginning not to use any kind of butt joints with it but either dowels or biscuits instead. He probably knew why...

OK, I will see what can be done to increase the dept of the MTM and yes I'm expecting that he will ask a fair amount. Thrust me that I will carefully study every detail before ordering the boxes. I have already some experience and know what kind of problems to expect. I'm also very picky and like to examine everything to the smallest detail so I don't expect any major problems. I will only move forward of course when I have simulated everything in the most rigorous way I can, knowing that before doing the final crossovers I will have to measure the drivers in the baffles but that is still far away. What I must determine first and foremost is the actual volumes so I can define the final drawings.

[cochinada]

I did what your suggesting with gluing real wood to mdf and have had no problems with it at all. Of course it might be species dependent or that I just got lucky I've attached a few pics.

That's a very nice box!

It's also dependent on how thick the wood itself is. I have done with Steve just posted, but with much thinner wood. I used mahogany. The wood did crack about a very thin section, but this isn't visible as it's concealed beneath the drivers frames.

It will probably be 30mm or 1"1/4 as you say

[deewan]

This is a very interesting idea but looks easier than probably is. I actually have already seen a guy wrapping a Mclaren super car on youtube.

If you are only needing to cover the top plate, a wrap would be pretty simple. If you are thinking of doing the entire cabinet, then yes, it would be very complex and time consuming.

[cochinada]

It's also dependent on how thick the wood itself is. I have done with Steve just posted, but with much thinner wood. I used mahogany. The wood did crack about a very thin section, but this isn't visible as it's concealed beneath the drivers frames.

The outer panels are going to be made of 30mm (1"1/4 like you say) or perhaps a little less and the inner ones 12mm.

My vote is for the BB version ....... if you do go the painted mdf version, Ryan is right about the edges telegraphing, my last project was painted and I can see some edges. Since then I have been doing a lot of digging and the best solution I found is to seal the wood with a polyester resin gel coat first, it's used in the marine industry to seal against water. You then paint over that, Wilson Audio goes this route and have some videos you can watch. I also ran across this stuff http://www.rustoleum.com/product-cat...surface-sealer which is supposed to work very well. I have not had a chance to try either yet.

Until now I believe all votes are for the BB which is also my inclining. With my current speakers I had exactly the same problem and that is why I wasn't happy at all with the painter.

[cochinada]

Since your first project is a lot of work and expense, be confident in your acoustic measurement skills. If not, consider lots of practice on other speakers while the cabinets are being made.

Good point! I have no skills with acoustic measurements. Until know I only measured my speakers at some listening positions using an USB microphone that came with my Anthem. I don't even have a proper soundcard nor any specific SW except Holmimpulse which I didn't even explore very well.