The twist with this goblet is that the rings are intertwined – they pass through each other as well as around the stem.

The goblet is made from cherry, the ring and box from bubinga.

Photos

All Three

Goblet

Ring and Box

The story of the fraternal twins.

For a small apartment in California, I thought it would be best to build something made for low to moderate listening volumes, but large enough to still have good bass extension.

The drivers – tweeter and mid-woofer – are both made by Seas, a Norwegian company that makes some of the finest drivers in the world. The woofer cone is cast magnesium with a  copper phase plug. I know… awesome.

The crossover was designed by John Krutke.

The speakers are covered in real walnut veneer and finished with a hand rubbed application of oil-varnish.

Photo time…

Construction – don’t judge yet, these are just construction photos.

All cuts made

Clamps, wood, glue…

Nearly a box

Damage and unseen fury. There was a dropping issue.

Repair in progress. I rabbeted out the damage and replaced with good wood. Then I flush trimmed to size. You can’t even tell!

Five sides are better than four

Jump forward a few days and they’re lined with sound absorbing material, the crossovers are mounted, and the wires have been run.

Clamping on the top part of the front baffle.

Bottom part of front baffle attached and front vertical edges rounded. Maybe several hours of sanding too.

Glue drying so that veneer can be ironed on. Back piece was already attached.

Veneer attached to one

And the other

Bases made, attached, and first coat applied

Seven or so coats later, with sanding in between.

Binding Posts

Drivers attached.

Post Construction

The Fraternal Twins, 95% done and brought inside for a listen.

That’s it. I hope you like.

As far as the finish, I treated the maple exactly as detailed on the Wood Whisperer site:

http://thewoodwhisperer.com/episode-32-pop-goes-the-maple/

The makore was finished with the General Finishes varnish only.

I might have sanded a little more to get the shellac and die completely off the surface because I really wanted there to be a striking contrast between the maple and makore; though, some die/shellac was still in the end grain.  I also finished the veneered curly maple baffles for the subs this way.

For the makore, I sanded to 320 grit and applied the same clear General Finishes oil varnish in the video.  I built it up by first sanding with 320 between coats and then I used steel wool between coats.  I wanted a satin finish, so after about 3-4 coats of finish on the maple and makore, I glued up everything, used steel wool to dull it down,  wiped on one final layer of finish, and then wiped it off with a clean rag (t-shirt).

I’m not going to apply wax because I might touch up a few areas, requiring a re-application of the finish. As a note, the sub cabinets and electronics cabinet were originally part of a 70′s hi-fi set that I refinished.  I pulled out all of the old speakers and built in two 10″ Dayton Reference HF subs in sealed enclosures per side.

I’m powering the ZD5s with an Adcom GFA-5400 and am powering the subs with a Carver a-500x, hooked up parallel at 2 ohms per side.  The Carver doesn’t get above warm even at loud volumes, at this load, whereas the Adcom seems to always be very warm.

Finally, thanks to my girlfriend for her help during stressful glue-ups and my friend Josh for his advice and use of some of his tools.

Anyway, here are the results:

Crossover Box:

1. Acquire curly maple.

2. Plane with router on jig to avoid tear-out.

3. Rout out recesses in maple to limit expansion/contraction issues.

4. Square up corners in recess with chisel.

5. Cut out baltic birch to fit recess.

6. Notice that, due to custom depth of recess, baltic birch also has to be planed to the correct thickness and, since I don’t have a planer, the router method must be used again.

7. Glue plywood into maple.

8. Cut out another piece of baltic birch to be inset into the cabinet.  Oh, and this is a custom depth too because I just kind of cut the depth to what looked right.  So, out comes the router jig again…

9. Align and attach the baffle to the cabinet with piece of birch fitted. Clamp and drill 1/4″ holes where driver holes will be.  Insert dowels to hold alignment.  

10. Glue pieces together, using dowels to align.

11. Cut out driver recesses and holes.

12. Round over, using 7/8″ round-over bit, on all sides.

13. Since I didn’t want to use the router to achieve the final fit with the cabinet (scary), sand until arm is jello.

Pictures of the results:

Curly maple routed out with baltic birch ready to be inset

Clamps…not enough

Results, after step 12

After step 13.  I had to make a deep fence with the round-over bit mostly covered by it because I didn’t want the entire radius of the bit to be used for the round-over.  This way, the curve on the baffle carries into the cabinet’s curve.

Test fit of the drivers, before drilling the mounting holes.  The rear of the baffles will get a 45 degree chamfers on the woofer holes after the mounting holes are drilled.

Next up, I’ll be doing the same thing, minus driver holes, for the rear of the cabinets.  I’m hoping to do these in less than 32 hours.

I did end up messing up on one side when the veneer folded under as I was spreading the iron out to the edge.  I had to rip off what I had done, which involved a heat gun, scraping, and sanding.

In retrospect, I was lucky to have used Titebond I, instead of II or III, because the original’s perviousness to water allowed me to remove the glued areas that were damaged.

After all was finally veneered, trimmed and sanded, I ended up with this (the veneer is mottled makore):

Now that they are veneered, I can cut the baffles exactly to size, and finally get going with the front and back pieces.  The fronts will be three piece laminates, with the maple routed out to accept a baltic birch or mdf substrate, and this will attach to the pieces I’ve cut to fit in the square openings. The total thickness will be about 1.25″ on front and back. 

According to a representative, there are no reported failures when applied correctly using glue that is within its usable shelf life. Though, two issues that one might run into are 1) failure to heat the glue to the correct temperature and 2) exceeding the reactivation window.  Most instructions on the net just say “let dry completely” but say nothing about any activation window. For Titebond II and III, this activation window is only 12 hours.  Finally, one more disclaimer from the technical representatives:

[There is] the possibility that the adhesive holding the paper to the back of the veneer may be damaged by the heat. If it is merely softened, reversibly like our glues, then it ought to reform that bond as the assembly cools, with no problems. If, however, it is damaged or degraded, the customer could end up with our glue anchoring the backing to the core, but the veneer itself loose from the backing. With that in mind, I might pass along our lack of knowledge of the paper to veneer bond, and suggest that the customer test the veneer involved before attempting a large project.

Instructions for the iron-on veneering method:

• iron
• veneer
• Veneer tape
• PVA glue (Titebond White Glue), Aliphatic resin glue (Titebond Original Wood Glue), or PVAC glue (Titebond II Premium Wood Glue, Titebond III Ultimate Wood Glue)
• Paint roller or paint brush (used to apply the glue)
• Spray bottle containing water (optional)
• Shellac (3 lb-cut) (optional)
• Paper towel (optional)

In particular, the lack of a way to flatten the curly maple was starting to be a problem.  I gave a friend’s standard planer a shot and there was too much tear-out.   Drum sanders for planing start at $575 or so and go up quickly.  Not wanting to spend that much, I racked my little brain for ways to flatten the wood without one. 

The table saw is one option, with a tall fence, but that option is a bit frightening.  Every time I’ve tried this, with other wood, I’ve had issues aligning the cuts after flipping the piece (assumes you can’t make the cut in one pass).

Luckily, a router can do anything.  After months of looking on craigslist for a suitable drum sander that wouldn’t tear out chunks of the finicky curly maple, I came across an ingenious solution, well documented on the Internet.  I’m surprised I didn’t think of this sooner.  Behold, the router planing jig:

The tall supports are to make absolutely sure that there is no perceptible flex.  To verify that the horizontal rails for the router were flat, after the glue dried, I passed first the bottoms through the table saw, with the top of each rail against the fence, then I clamped the bottoms together, in opposing directions.  I then fed the top part of the rails, which the router rests against, through the table saw.  As you can see from the way the plies were exposed, I did not have a flat surface to begin with.

There is a small lip on the bottom inside parts of the rails that the router support slides on. 

Now for some math. 

Drum sander          $575.00

- Jig                    25.00  (15 dollars of this was the threaded rod)

-Planing Router Bit   ~30.00 (Magnate 2703 Bottom Clearing Router Bit)

                                          _______

Money Saved               $520.00

OK, so my math assumes you have a router, but they are pretty common and cheap.  

Edit: Although, I highly recommended the Magnate router bit, I have gone back to using a large straight bit because after 2 uses, the Magnate bit was dulled enough to produce tear-out and too much heat, regardless of speed, burning one of the peices.  The first two times I used the Magnate bit, it left a really nice surface and I had no tear-out issues.  with the straight bit, you’ll have to sand a bit more afterwards.

I’ll start planing the 1/2″ pieces tomorrow.  I would prefer to use them because I like the figure, but I’m worried that they are going to end up too thin, so I have some 1″ pieces on standby that I can plane down to 1/2″. 

Next up, I have some mottled makore veneer on order, which will cover the top, bottom and sides of the cabinets.

The lamination is complete for the curved sides.  As this was the most time intensive phase, I felt like I should celebrate a victory, but have enough experience to know that my enclosure design brought about some awkward situations in the workshop.

The first order of business was to trim the overhang from the top, bottom, front, and back of the enclosure.  Given that there were no square sides in easy reach of a table saw fence, this was only going to be easy in my head.

My first attempt involved simply extending some wood out from the front and back of the horizontal braces/forms, and then slide this makeshift face along an over-sized fence.

Since the forms are square with what will be the front and back, it should have resulted in straight cuts. The plan was simple, elegant, and failed miserably.  

Freedom of motion on a table saw is a bad thing.  Ideally you want your workpiece to be secured in such a way that it can only move forward.  The problem with my plan was that I had to hold the enclosure firmly against the fence while sliding it forward.  Since the sides it is sliding on are curved, I also had to prevent rotation while pushing against the fence.  I was able to get the front trimmed without too much of an issue, but the back side attempt was when the woodworking gods lashed out.

The problem is that the saw blade cannot reach all the way through the back because, while laying on the table with the front and back at 90 degrees, the saw blade does not reach high enough.  Here was my first tactic, drawn up a little hastily:

In case you can’t tell, I tried setting the enclosure and saw blade at the same angle, then sliding the enclosure through.  Wrong!

Since my hands bristled when I asked them to go nearer to the saw blade, I had to push into the fence nearer the top of the enclosure, which made it hard to keep the bottom part firmly seated against the fence.  My cut was less than straight, despite curses, concentration that would cut through steel, and firm determination. Am I the only one who’s built an enclosure this way?  All the other attempts I’ve read about have seemed to go swimmingly.  I acknowledge that I’m not a real woodworker, but not every builder of curved speakers can be a professional, can they?

Anyway, I finally got smart, finished my crosscut sled, and adapted it for this job.

First I used it to trim top and bottom closer, so that routing flush wouldn’t require so much routing, then, after routing, I had a flat face and was able to easily flip the enclosure around and cut the front and back.  I used a digital angle gauge to make sure the angle of the blade always matched that of the enclosure.  I drew marks on the sled where I had the speaker lined up previously  so that I could get the correct depth when flipping it around.  Not having to maintain control over several axes and simply focusing on pushing was luxurious.

I did get a bit of tear-out and kicked myself for not using tape or something, so I’m repairing that as I type.  Also, The front has turned out to be wider than the original design.  Wide enough that I’m in danger of overshooting the width of my maple baffles if I trim back any more.  Because of this, I’m going to have to adjust the thickness of the front baffle, probably requiring a three part lamination when accounting for the maple.

Until next time, here are some more pictures…

Top and back

Front mostly finished

Close-up of lamination after cutting

In addition to the top cauls, I also attached a narrow strip underneath, along the edge, because the cauls and attachment method limited the amount of pressure I could apply to the edges.

In the top picture, the center band clamp that runs under the jig helps hold the board in position when positioning the jig, and keeps the center from bowing.  So, the only problem is that I can only do one at a time.  I tried doing two at a time, but by the time I got everything glued evenly, the glue was starting to set up.  Also, I don’t have enough clamps.  Since it takes the glue 24 hours to reach full strength (according to the container), it’s going to take 10 days to get the speakers fully laminated.

You may ask yourself, why didn’t he just use kerfed MDF? I think it’s either because I’m stupid or because I wanted the enclosure to be very rigid.  I think that this method will produce that, though I could have glued one layer of 1/8″ plywood around the outside of the kerfed MDF to add rigidity and  smooth out the flat areas you get with kerfed wood. Another reason is that I don’t like working with MDF.  It dulls saw blades and router bits and produces nasty dust.

Unfortunately, I can’t work on the front and back pieces yet because their finished widths are going to be cut according to how wide the enclosures end up being.

Until then…

Please note, what you see below looks like it was easy.  It was not.  That result was achieved after much cursing at the people who designed the band clamps, cursing in general, and telling all of the various clamps what I thought of them. 

The band clamps suck due to the absurd, cheap attachment method used to connect the band to the spindle.  I had to modify all three mid glue-up because they slipped when attempting to ratchet them tighter.  Of course, none of this nonsense happened during the test clamp phase.  

Also, that is the second time I clamped everything.  On the first try, I had put down some parchment paper that got pinched inbetween the sides and the forms.  Having to losen everything was also a source of harsh words.

Finally, Titebond III has  a 10 minute set time.  I think Titebond IV should have a set time of 1 hour.

They’re starting to look like speakers:

Tomorrow night, side two with lessons learned about cheap band clamps.