Want to move from MDF to real wood, any suggestions?

Status
Not open for further replies.
Re: Re: Is it live or is it Memorex?

Bill Fitzpatrick said:


I'm not dangling any bait. And I certainly wouldn't counter with the idea that the speaker is PART
of an/the instrument because it isn't.

I do like your bringing up a synthesizer though because an enclosure, unless dead as dead can be,
can act very much like an "event triggered synthesizer." Now, if you want one of these in your system,
that's OK with me. Personally, I don't want a speaker system that burps.

What rankles me about some of the asides in this thread is the suggestion that
flaws ought to be found desirable by all.

So you're saying that even when an synthesizer doesn't have a speaker hooked up and cannot make noise, it is still an musical instrument. Hmmm...

One last nudge and then it will have to be an agreement to differ: For a synth, the amp and speaker are not sound reinforcement but integral to its operation as a musical instrument.

As for the flaws, I agree that the differences of opinion can be infinitely baffling. Some will attempt to design away the flaws - concentric drivers and ultra-thick front panels. Others will try to design around them - massive amounts of stuffing. And finally, a few will simply accept that the speaker design is a lemon and will try to make lemonade - musical enclosures. Small philosophical changes but hugely different results.

I try to be in the first group when I design. My results usually have me attempting the second and with continued failure I mostly find myself looking for sugar and a spoon.

🙂ensen.
 
"There is a another similar argument going on with regard
to flute construction, with some scientists insisting that
a flute made of "inert" material, such as concrete or stone
will sound exactly like one of metal.

Meanwhile, flautists (sp?) prize silver as the material of choice."

I build speakers to reproduce as close as possible, the original source not to produce any new sound. That is my preference. In fact, my goal has always been to produce a "dead" cabinet. The "deadest" cabinets for me have always given me the cleanest sound and in my opinion the best result.

Vic
 
That's a respectable enough opinion. I simply happen
to have heard some wonderful speakers which were not
constructed so as to be dead.

The sound is the only important thing, and the fact that
high quality can be arrived at in different ways is
interesting and makes a good subject for exploration.

To announce that "dead" enclosures are the only valid
approach is dogmatic.
 
I like the idea of purplepeople(I think), looking at the tuba w/o mouthpiece. Though, one may argue a similar analogy as a reed instrument with or without the vibratile surface. "Vibratile surface" being key here. So, a human providing the necessary force to create motion on the reed's surface causing a wave of sound to travel the length of the clarinet. Then, an amplifier providing the necessary force to create motion on the transducer's cone, casuing a wave of sound to travel the lenght of the cabinet...few changes can be made though, it is just a general note.


Love the Bose comment....heehhe. "Just another kazoo."
 
The reed is part of the instrument. An adjustable and removeable part, but a component nevertheless.

The moving air is akin to the mains power going into a synthesizer. It's the energy source but not part of the instrument. When I sing, the instrument is vocal chords, larynx, tongue and mouth as they all go into modulating the energy provided by my diaphragm. When I toot my own horn, the instrument starts at the mouthpiece. In this case, my respiratory system is just a means of transmitting that energy to the modulation system (like a power cable).

Since the choice of enclosure material changes the "timbre" of the sound (however slight), I say the speaker is part of the instrument. Here's a question: Is a MIDI wind controller an instrument if it doesn't actually start the modulation process required for music?

🙂ensen.
 
Two quick thoughts--I apologize if they've already been covered previously in the thread, but I don't have time to go back and look.
--Supposing you were to take a driver and design an "ideal" dead cabinet. Let's further postulate that there is a dip in the frequency response due to the design of the driver, or perhaps an interaction with the crossover or cabinet. A possible solution would be to build a cabinet that resonates enough at the right frequncy--and only enough--to fill in the gap, thus giving flat response where a more dead cabinet would have had a suckout.
--Live cabinets tend to resonate for a brief period, even after the driver has ceased producing the note. Think in terms of a reverb circuit applied to the signal. This would add body and fullness to the sound, particularly to dry studio recordings, thus making things a little less sterile sounding. Whether the effect appeals would depend on the type of music and the listener.
I have reservations about using planks of wood, due to the variation in density from tree to tree. In the basses I build, there is an easily noticeable difference between same-species boards when held up and struck. For that purpose, I choose the boards that ring the longest and with the purest pitch--then laminate them with other pieces to kill the resonance. The result is very stiff and very non-resonant. (Note that acoustic instrument makers test their wood the same way, but they're looking for specific notes and tend not to try to damp the resonances.) The practice works for instruments because the dimensions I need are long and thin--exactly the way trees tend to grow. For speakers, you run into the need for wider pieces of wood and things might be a bit more difficult. Consistency in a commercial product would be damned difficult to attain. Good tone woods are maple (hard rock only--avoid Western soft maple, red maple, etc.), black walnut, purpleheart (from Africa, and yes, it really is purple, you've got to see it to believe it), ebony, rosewood, cocobola (a relative of rosewood--there are numerous species of Dalbergia ), kingwood, and others. At least here in the US, anything other than maple will cost a fortune. Some of these woods, e.g. ebony, tend only to grow to small sizes anyway, so you'd have trouble getting boards of any size, although ebony plywood might be an interesting thing to pursue.

Grey
 
What if you laminated a bunch of boards together with pipe clamps and then planed them?

🙂ensen.
 

Attachments

  • lam.jpg
    lam.jpg
    33.5 KB · Views: 438
Well then, I guess you would have some sort of laminate, but again not solid boards like the original posts suggests. I would imagine that this approach could work, but as a woodworker and not a speaker designer I can tell you it is wasteful of a precious resource. If a substrate material like mdf is not desired, I think the next logical choice is some sort of plywood. Baltic Birch plywood would be a good choice because it is very handsome looking and absent of the voids characteristic of normal plywoods. I think we are overlooking a few other considerations that make solid hardwood a less than optimal choice. Hardwood is not cheap and certainly not abundant in the widths needed. It is also sometimes frustrating to mill without the proper tools (jointer, planer, biscuiter etc.). Time is a factor because of the time required for aclimation of the wood to it's new environment before milling. I just wanted to share some of the practical drawbacks of using solid hardwood because so many of the emotional drawbacks have surfaced already.
 
The alignment purplepeople suggests in post #89 is essentially plywood. Each layer's grain goes 90 degrees from the previous layer. Not a bad idea, but if you assume that you're using 3/4" wood, it's gonna get really thick really quick. For what it's worth, that's how I build the heads on my basses--plane down the neck laminates, then build them back up with hardwood veneer layers that I laminate in situ. Very dead. Very strong. Bear in mind that if you keep that up for enough layers, you're back to a 'dead' speaker, except that it cost you a lot more than MDF would have cost. Not to mention the hassle factor.
Eric,
Interesting idea. Oak flooring is available in longer lengths. Might save time and reduce the number of butt joints. As a fringe benefit, the fact that it's mass produced will make it cheaper than most other hardwoods for folks who don't have access to surface planers, etc. Me? I buy raw sawmill planks and DIY my own boards to whatever my needs are at the time, but most people can't do that. Just another expensive hobby, and it takes up a whole lot more room than electronics.

Grey
 
Grey,

How would you tune the cabinet to resonate at the dip if the material is not homogenous enough to be predictable? With MDF, the uniform density would make it easy to tune by varying the thickness in places and even machining it into specific shapes, but with a more resonant material like a hardwood... how?

🙂ensen.

PS: Your theory could be why those little aluminum cones work for some people? The cones resonate at the dips in the frequency response of the speaker. By placing them at different locations on the cabinet, different wave nodes are doing the excitation. My example is placing a thin layer of sand on some sheet metal, then striking the metal to ring. The sand is re-arranged to match the resonant waveforms of the metal sheet.
 
Hardwood and MDF

Grey

I cannot see that putting hardwood over a substrate of MDF would get you very much. Maybe if the cabinet was loosely braced but then you would have a combination of resonant modes. Maybe, be the mid range and tweeter enclosure could be hardwood with very only a small amount of damping.

Using a hardwood for the mids and tweeter on the front baffle would tend to blur the image, however that might increase the size or the sound stage. Of course, all this is speculative. I assume either Nelson has built or heard something that makes him feel this is an alternative way to go.

Of course the only way to do the is with a Planner, Jointer table saw and a drum sander. Using small would again effect the cabinet resonances vrs, using wide boards. And as you've noted the wood, would have to be very dense.

Wood that or of very high density include:
Bubinga, Ebony, Jarrah, Lignum Vitae, Mesquite, Rose Wood, Purple Hart, Satinwood, Wenge, Tulipwood, Zircote, Coco Bola, Zebrawood and Padauk.

Woods that or medium to medium-high density are:
White and Red Oak, Hickory, Pecan, Maple, Cherry, Iroko, Teak, Birch.

In addition, some woods very in density from boards to boards these are: Mahogany, Luan,

In addition, note the type of the wood cut may make a difference in the resonance of the wood and the cabinet.
 
Hello!

[about what Eric said]

> Adrian,
> I think you missed this bit - "Like I said much earlier, unless you
> build a cabinet with panels less than 3" thick you will have to
> live with some resonances.
> We all know that this is expensive, impractical, seldom
> commercially done etc, so that means in practical terms most of
> us have to live with panel resonances of some degree.

I think you missed my reply. Any panel is going to have resonances. But we can minimize them by using the right materials and sitffening the enclosure with bracing so that they are shifted to an area where they cannot be as easily excited and are not producing audible anomalies. My actual reply to you was:

True. There's going to be resonances for any material used. The point is MDF is less prone to panel resonance than hardwood for reasons I've already described. Furthermore, thats why people can brace each panel to stiffen it up so that panel resonances are shifted to a higher frequency where they arn't as easily excited. If a subwoofer is going to be crossed at 100Hz, then a panel resonance up at 1KHz is going to be far more difficult to excite as the crossover has removed energy from this area. Or you could have a cabinet with a midrange/midbass driver used up to maybe 1.2KHz or so, so cabinet resonances at 3KHz will be less able to be excited because there is little energy in this area because of the crossover.

The point is we can stiffen each wall so that panel resonance is moved out of "harms way".

[about what x.onasis said]

> Yes, a good shop can make things easier, and "cheap"?

The material was cheap for me because I bought the wood in plank form, which is a lot cheaper than in sheets. To add to that, the material was not dressed - which means thats cheaper too. Thirdly, I actually got the timber for an extremely cheap price through a discount deal from the workshop. So that made solid timber an excellent choice, and inexpensive to build my hall table.

> Are you referring to the snap-on end panels, or the keyhole
> fastening system? Since no one has commented, was this
> difficult to understand? Anyone?

Well, actually I am referring to both 🙂 I'd still like to know what you meant.

[about what Nelson Pass said]

> To announce that "dead" enclosures are the only valid
> approach is dogmatic.

Well I guess if one wants a non-flat response curve and a loudspeaker that isn't reproducing exactly was is recorded on the original program material, then you can have as many resonances as possible and be happy. I'd rather have a true representation of what was recorded however.

[about what Grey said]

> --Supposing you were to take a driver and design an "ideal"
> dead cabinet. Let's further postulate that there is a dip in the
> frequency response due to the design of the driver, or perhaps
> an interaction with the crossover or cabinet. A possible solution
> would be to build a cabinet that resonates enough at the right
> frequncy--and only enough--to fill in the gap, thus giving flat
> response where a more dead cabinet would have had a
> suckout.

It would be just about impossible to fiddle with panel resonances to fix frequency response problems as part of the design. It would be impossible, really, as you can't really tame the Q and peak of panel resonance and where they are going to all be in frequency. There's more than just one panel resonance anomaly, there are many. So the solution here is to go back to the design itself and determine why there may be a dip in the frequency response. One possible problem at the crossover that causes a huge transient dip is by adjacent drivers connected with second order networks. But panel resonane should never be thought of as a means to correct frequency response errors.

> --Live cabinets tend to resonate for a brief period, even after
> the driver has ceased producing the note. Think in terms of a
> reverb circuit applied to the signal. This would add body and
> fullness to the sound, particularly to dry studio recordings, thus
> making things a little less sterile sounding. Whether the effect
> appeals would depend on the type of music and the listener.

I can hardly think of that as adding "fullness" to the sound. If the panel resonance did act like a reverb time, then it would tend to
blur the sound by making the frequencies of panel resonance continue to make sound between the notes. It's not like real reverberation time such as in a room with little absorbtion where its the entire signal that has the reverb, but its only the frequencies of panel resonance. It would create unwanted artifacts between notes, and certainly not "musical" sounding.

> I have reservations about using planks of wood, due to the
> variation in density from tree to tree.

I agree. It will vary from tree to tree.

For my hall table I described it was a slightly different story where all that mattered was making sure the wood was all matched at the same level of color. EG: All jarrah planks of the same "darkness".

To those claiming that their resonant enclosures are awesome over a dead one - this claim should be backed up with measurements. Unless you've measured and determined that your panel resonances are in the bandwidth where they can be excited and cause those spikes and dips in the response where they are clearly audible, then it can't be said that your panel resonances are making your speakers sound better. It may be that your less-dead cabinet was still non-resonant enough so that panel resonance was shifted to a position where it isn't causing problems. Or it may be an entirely different issue altogethor that led you to this conclusion.

I can't say just how much the panel resonance is going to affect the loudspeaker, as that depends on the materials used and how well its been braced. But any resonance is a modifier of both phase and amplitude response, so there's potentially an issue with these aberrations. Best to minimize those that aren't purpose-designed as part of the system.

Adrian
 
If you build a cabinet to account for deficiencies is a driver, I see several problems. First, it seems like a pretty hard task to do and would involve a significant amount of tweaking to get the appropriate response out of an individual cabinet. Second, you're talking about defining a resonant frequency of the speaker cabinet. This would not provide the quick transient response at the deficient band of frequencies just a resonant tune. There would likely be a small and short-lived ringing at that frequency. Third, it would be difficult to have consistency across a number of similar cabinets. Obviously, there are variations in individual drivers of a particular model as there are variations in the passive components that make up the crossover. These are easier to work with since individual components and drivers can be matched.

I am not trying to be dogmatic. I just have my opinion. I strive for neutrality. Call it OCD. My preference is to avoid coloration of music as much as I can. This doesn't mean that a livelier cabinent sounds bad or is wrong. Hell, maybe because of my own personality faults I am missing out on something that sounds better. I have experimented with lots of different configurations for fun but for my permanent stuff, I still strive for neutrality.

What I listen to on a daily basis is a small set of DIY 2-ways with Audax drivers. They make Natalie Merchant sing to me personally.:angel:
 
I'm working the design of a pair of small TL subs that I was planning to make out of MDF. The possibility of using raw woods is interesting for the sheer look of it. I would only go that route if I thought I stood a reasonable chance of making it sound as good as MDF. That said, nothing I've seen here give me that reassurance, but the potential for a beautiful piece is enough to keep me probing for a solution to the resonances, especially of hardwood.

🙂ensen.
 
Status
Not open for further replies.