Building A Parsons Table Without Nails

My First Stupid Domino Project: An Exercise in Tedium and Toolporn

[festoolparsonstable]

With a halfway decent polishing workflow established, it was time to finish it all off. 

I level sanded the top and sides with a combination of Abranet and Abralon stuck to the interface pads of my RO 90 and RO 150. This process was virtually effortless once I gave up on the Granat paper, and I went through maybe $6 worth of abrasives in total before it was time to polish. Luckily, I had had the foresight to round over the corners on the outer edges a bit, so I didn't sand through to bare wood in more than two tiny little spots. I just tapped those spots with a Sharpie and called it a day.

The Festool sponges continued to perform very well, but realistically, I can't recommend anyone seek them out. The only reason I bought them was because I was able to walk into a store and get them in person and I didn't feel like waiting another week for anything else to show up in the mail when I wasn't entirely sure what I'd actually need.

I've also noticed that most Festool dealers know fuck-all about furniture finishing, so if you don't walk in there with a part number, they're not really going to be able to give you any recommendations beyond "all that polishing stuff is over there." I had to talk to car detailers to figure it all out.

While all the mistakes jump out at me immediately, nobody else seems to notice them, and overall, the result is just about what I wanted. Under normal room lighting, it's dark and monolithic looking, but when the sun catches it, you can make out the grain and inlay work underneath. 

After a thin coating of clear paste wax (Renaissance worked out a whole lot better than the Staples stuff in the picture), it's ready to be scratched all to shit by daily use and I have to go back to actually cleaning the living room again.

Many thanks to:

Peter at The Tool Nut for feeding my toy addictions so promptly and efficiently

The Rag Company for selling amazing microfiber towels that actually work better than a tack cloth

K-119 for actively stocking so many hard-to-find parts and consumables

Woodcraft of San Carlos for stocking all the other consumables no one else does

Fuji Spray for bringing back the XPC gun (after killing it off like total fools)

Magnate for making great bits that perform every... bit... as well as the others that cost 3x as much. 

And Festool for making so many portable, dustless gizmos that I can use right on top of my living room floor without making a horrible mess. 

[festoolparsonstable]

A $10 undersized Festool sponge on a regular RO 90 pad vs. a $5 Hex Logic sponge on the exact same pad. Not only do you have to pay more per sponge to cover less area, but you need to buy and store yet another smaller "polishing" pad to avoid digging into potentially soft things with the spinning rubber edge and/or exposed velcro.

The difference becomes even more pronounced when you step up to the bigger pads for the RO 150. For the cost of just one Festool sponge, you can get three Hex Logics that fit right on the standard pad, plus a bottle of citrus cleaner that dissolves dried polish and wax pretty darn well. 

The Festool sponges are honestly great, but there's nothing magical about them whatsoever. Plan ahead and save.

[festoolparsonstable]

After a couple of weeks, it was time to make it all shine. I sanded all the nibs and gnats out of the inner surfaces and quickly realized that keeping the corners of this thing as sharp as I had was a mistake. There was just no avoiding sanding clear through to the primer in several locations. I wasn't about to take it all back down and do it over again, so I just touched up the faint white hairlines with some extra-thick, partially-dried-out paint and lightly rolled over those veritable knife edges with some slow-curing CA glue riding on the end of a Q-tip. Since all the touch ups are right on the corners, no one can tell anything ever went wrong.

After finding that this coating does not respond well to MicroMesh polishing at all (I just burned through all the layers on my test board without ever getting anywhere near an actual shine) I was told to use Menzerna compounds instead. Of course, nobody carries Menzerna and I didn't want to pay $20 to ship it here. AutoZone had Meguiar's Mirror Glaze line on the shelf, and a whole lot of auto detailing nerds had side-by-side videos of it outshining Menzerna, so I went with their most recommended 105 and 205 combo. 

K-119 had the (ridiculously expensive) ultra-fine polishing sponges for my RO 90, but when I tried using them with the Meguiar's 105 Ultra-Cut, they just weren't actually cutting much of anything at all. I went back to get the medium-cut orange sponges instead, but they didn't stock them, so I had to drive back out to Woodcraft for those. The difference was like night and day. 

I kept the RO 90 around 2-3 on the speed dial in coarse/rotary mode and everything it touched with the medium-cut sponges shined up in seconds. I went back over it all with the 205 on the ultra-fine sponges I had originally picked and got the nice, glassy surfaces I was after with very little effort. 

Long story short: if you buy only one pack of Festool branded sponges, make it the orange ones. The finer "waffle" numbers are really only good for final polishing, which you can honestly probably skip if you use the Meguiar's 105 right. Its abrasive crystals break down, getting finer and finer as you burn through it–what you wind up with is usually about as shiny as you're likely to want a piece of furniture to be.

[festoolparsonstable]

Now that I know the water helps this particular coating level, I re-spray about another 20 coats on the outer surfaces and set it in the house to cure for a couple of weeks before I touch it again. 

[festoolparsonstable]

Now the bottom needs protecting and the seams between the finishes need to be smoothed over as much as possible, so on go several more coats of clear. This time I added the same amount of water as I did of the cross linker (roughly 3% by volume) when I mixed it, and it seemed to level out a whole lot better for me. There are definitely some imperfections, but hey, it's the bottom. 

You can see the high-tech MDF router sled I used during the flattening process in the background of the top pic. 

[festoolparsonstable]

Before that dried completely, I yanked the tape and proceeded to sand out the orange peel on the outer surfaces so the corners where the two different coatings met could get smoothed out somewhat.

This was unbelievably tedious as even Festool's famed Granat papers load up almost instantaneously with this coating's dust. Even with the foam interface pad and suction on high, I had to switch discs on every single pass with my RO 150 to avoid gouging pigtails into the still-curing coating with the buildup. This gets real expensive real fast. 

Lesson learned: don't waste money on fine grit papers for finish work. Ever. Mirka's Abranet works so much better it's not even funny.

At any rate, this is about what it's going to look like when it's done, so that's like exciting and stuff. 

[festoolparsonstable]

My aesthetic goal was to get this thing to look a bit like a modernized bit of old Japanese lacquer furniture, so I went with General Finishes' Holiday Red acrylic "milk paint" for the underbelly after priming it all with a rattle can of Zinsser white shellac. 

This stuff is really, really thick and I only have a 3-stage turbine, so I wasn't sure it was going to spray at all. GF recommended the 2mm needle/aircap combo which I had a really hard time finding because nobody's actively stocking parts for the XPC guns anymore (the T Series suuuuuucks and I will not be buying one of those anytime soon), but once I got it on there, it actually sprayed pretty well. There was a bit of graininess to it because it was clearly struggling to atomize, but this sands out really easily. 

It's no alkyd enamel, but it's easier to get and I probably won't explode when the water heater kicks on. 

[festoolparsonstable]

Woodcraft carries every General Finishes product under the sun, but I've never, ever, ever been happy with any of their topcoats. Their pre-cat urethane is pretty strong, but it fisheyes like crazy no matter what I do, and if you add the cross linker to it, you have to spray it all in a few hours or it destroys your gun. 

Target Coatings' stuff lasts for up to 10 days after the introduction of the cross linker, so I decided to give it a shot even though nobody carries it in California and shipping on a gallon of liquid is beyond insane. I added some black and honey amber Trans Tint dye to their cross-linked EM9000 and let 'er rip after a thin interface coat of shellac to make sure the epoxy didn't cause any problems. 

It went on much more easily than the GF stuff with very few fisheyes, but no matter what needle/aircap combo I tried, I wound up with terrible orange peel. I was spraying, good, wet coats, it was just refusing to level out for some reason.

The inner surfaces were going to be painted very differently, so I just let it cure this way for a few days before I taped it all up and flipped it over.

[festoolparsonstable]

The beginning of the finishing. I wanted the outer surfaces black but I still wanted to let the grain peek through under the light. I sprayed two coats of General Finishes' Ebony dye stain through my Fuji gun and was generally happy with the results.

If I had to do it all over again, I would have sucked more of the yellow out of the wood before hitting it with the stain via a good soaking in some steel wool and vinegar juice. The GF ebony is a lot more purple than black and it fights with the yellow of the maple in ways I don't especially care for.

[festoolparsonstable]

While I certainly could have filled in the little gaps in the joints with some sawdust and CA glue, I decided that I'd accentuate the negative rather than attempt to cover it up with some epoxy inlay. After plowing some 1/8" grooves all over the place with an OF 1400, I stirred a bunch of Mixol into some of SystemThree's T-88 and got a nice, rich black without grinding up any sticks of ebony into powder.

I actually had to do this twice because the first go-round with the T-88 never cured as I didn't realize their "EASY 1:1 MIXING RATIO" only applied to volumetric measurements, whereas I was measuring by weight. Thank goodness I had a heat gun to get it all out of there because that was a horrible, sticky mess. 

I've since learned that Stewart-MacDonald sells black CA glue and I really wish I would have known that before I did the epoxy, because this particular formulation doesn't really ever polish up and the teeny tiny bubbles are utterly impossible to avoid. This was a painfully tedious process that could have been shortened considerably with CA. 

[festoolparsonstable]

With that mess dry, it was time for a proper flattening. I popped a Magnate surface planing/bottom cleaning bit into an OF 1400 router and built one of those flattening sleds out of MDF to ride on two rails of aluminum Kreg track--which is, again, amazingly well machined stuff for the price. After many, many, many passes, I had a nice, flat top that only needed a tiny bit of sanding to get the subtle planing marks out of it. 

It was at this stage that I started to realize the domino mortises inside the side rails from the initial lamination process were starting to telegraph through a bit. I can't really sand any more than this without serious problems. One actually did crater open and I had to cut a plug for it which was annoying.

Note the amount of sawdust on the floor. The planing bit was so big it destroyed the dust collection shroud on the OF 1400 when I released the plunge lock with it still spinning and I had to track down another one. Under normal circumstances, this would have been a lot cleaner. 

[festoolparsonstable]

The actual glue-up was a whirlwind of pounding and cursing, so there was no stopping to take pictures until after it was done. The strap clamp that had worked so beautifully at cinching everything up in perfect unison during the dry-fits broke almost immediately once I actually added glue to al the mortises as there was just too much resistance for it to deal with. 

I did not cut extra wide mortises as many recommend for this particular joint. To ease insertion, I rounded over the edges of the dominos ever so slightly so I had just enough wiggle room to get each tenon started in its hole before moving on to the next corner. There were many, many, many false starts and do-overs, but it was well worth it for the extra strength the tighter mortise gets you.

Ultimately, two joints were left with hairline gaps I just couldn't close because of subtle variances from 90 degrees on the top's corners. If I had done the top first and waited to cut the miters until the day of the gluing, I'm pretty sure I could have compensated for this much more easily. 

I also wish I'd used a slow-curing epoxy rather than Titebond, not only for the increased open time and strength on the endgrain, but also because it seems to keep the seams from telegraphing through the finish much better.

Nevertheless, it dried strong as hell without a single metal fastener and that was all I was after.

[festoolparsonstable]

After weeks of waffling on whether or not I wanted to do a plywood torsion box, I decided to keep it solid maple throughout. I ripped a whole bunch of fairly narrow strips and alternated the grain so as to minimize any cupping, then jointed and dominoed them all together. 

I hand-planed it down just enough to get it about as flat as it really needed to be to get a decent reference surface for further dominoing knowing that the whole shebang would need more serious flattening further down the road. Some degree of warpage when clamping a thing like this together is kinda unavoidable. 

Note that this construction approach adds more difficulty to the glue-up, as literally every single part has to be brought together simultaneously on account of the locking joints at the corners. You can't glue the frame together and then nail/screw in the top after that's perfect--it all has to happen all at once or not at all. The way it's dry-fit right now, it is physically impossible to insert the legs due to the angles at which the tenons need to travel.

The last picture is all of the sawdust I had to sweep up after this process was complete. The collector sucked up everything like a dream. 

[festoolparsonstable]

Everything cut and dry fit pretty nicely. My laminated rails ended up being about 2.25" wide, so I moved the mortises back a bit and used the 8mm bit.

I wasn't too sure how I was going to do the top at this point, which I later realized was a mistake. The entire glue-up would have gone much smoother if I had cut the rails to match the top's lengths and angles rather than vice-versa, but I didn't want to buy (and waste) all that wood if I ended up being too mechanically incompetent to pull the frame off. 

[festoolparsonstable]

While I did a pretty good job of getting the laminated sticks flat and square as they were ripped, they weren't good enough for the miter cuts I knew I was going to be making. Several of them needed to have 1/8" or more shaved off of one side to be truly, truly square.

I tried doing this with a good old-fashioned Lie-Nielsen jointing plane, and while it worked beautifully, I only have a few hours each weekend to work on this thing, and hand planing was slow as hell.

I knew Festool made a plane that had a neat little jointing table and set off to K119 to pick one up. Once I actually saw it, the jointing table was so small and so rickety, I wasn't about to pay $300 for it. I lopped off a hunk of Kreg jumbo aluminum track (which is L-shaped and absolutely perfectly square), drilled it out so it could mount to the planer, then stuck a hunk of phenolic coated birch with a t-track on top. This gave me a much better infeed surface to work with, and the featherboard kept my fingers out of harm's way.

Even though there was only one screw hole to hold the whole thing onto the planer itself, this worked beautifully, save for the part where I had to jam a Superball in the planer trigger because there's no lock on it. Seriously, Festool, what's up with that? 

[festoolparsonstable]

I didn't want to hunt down a whole lot of thick, long maple stock that may or may not twist on me, so I went the lamination route with whatever 3/4" stuff I could find on the shelf at the Home Depot Pro.

I ripped it all down with a TS-55 Req loaded with a Freud blade (Festool's "Panther" blade just made a goddamned mess of everything and the stock blade couldn't get through the maple without scorching) and I used dominos to assist in the alignment. This would turn out to be one of my biggest mistakes. As the project progressed, I shaved more and more wood off the surfaces of these things, and eventually, the domino pockets started to telegraph through. 

Even though it seemed like there was more than enough room for them at the time, if I ever do this again, I'll use the 4mm dominos and far, far, far fewer of them. What I did here was total overkill. 

[festoolparsonstable]

In May, I bought a Domino joiner because it was cool as hell and on sale for hundreds of dollars less than it usually goes for. I didn't have an immediate use for it but when I saw the locking Parsons joint diagram, I remembered how I had tried and failed to pull this off several years ago and decided to see how I fared now.

I bought a $2 furring strip and knocked one out on my first try. I haphazardly glued it together without clamping and it dried strong enough to make this seem like a really good idea.

I want to emphasize the fact that I am a complete fucking amateur. I write software for a living and have absolutely no formal training in any of this business, so I make a lot of stupid mistakes. 

I also want to emphasize the fact that I live in a very small house with a very small garage that I have to get a car into, so as much as I'd love to "get a cabinet saw" and "build a spray booth," it's just not going to happen. If I can't carry it up into the attic when I'm done with it, it doesn't get purchased.