Unisaw Torsion Box Fence Table #4: Endgrain is hard to sand.

  • Advertise with us
Blog entry by Ger21 posted 03-21-2010 05:18 PM 3326 reads 0 times favorited 3 comments Add to Favorites Watch
« Part 3: Finishing the top Part 4 of Unisaw Torsion Box Fence Table series Part 5: 7 coats of polyurethane and counting..... »

After thoroughly sealing the bottom and sides of the torsion box with a quart of polyurethane, it was time to surface the top flat. When I placed my 4 ft level on the bottom, it appeared to be flat to within < 1/64”. But for some reason, it appeared the top top had a slightly larger crown along it’s length, maybe 1/32” or so. So I lowered the bit to the surface, and started cutting just enough to not miss any spots. At first I was getting what appeared to be a very rough and torn surface, so I greatly increased the RPM, and it looked a lot smoother, but was a lot louder. I normally route at 15,000 or less, but cranked it up to about 22,000.

My CNC can only cut about 45” long, so I had to turn it around to cut the other side. But before turning it around, I recut the plate hole and rabbitt, making the rabbitt perfectly parallel to the surface. I’m going to make some custom 3/4” thick phenolic plates. I’ve got a large chunk of 10mm phenolic plate, and I’ll laminate it into two layers with epoxy, and CNC a rabbitt to fit. If I have any problems with the phenolic, then I’ll get some custom aluminum plates machined. But the phenolic should be fine, and I have enough to make about 6 plates. I keep my 7518 mounted to a plate, and a 690 as well. I also have a spare 690 base that I’ll mount. Different plates will have different hole sizes. For most of the time, I’ll use a plate with a small hole, maybe 1” or so. This should work great for 90% of the router table work I do.

I plan to use this leveling system that I found here:

I’ll make a jig with the CNC to accurately locate and drill the holes.

Once I turned it around to finish the surfacing, I noticed that there must have been a slight twist in the box. Where the surfacing passes met, it was flush on the back, but a little high on the front. I made another pass, removing .002”, and the front came out perfect. The slight discrepancy is at the end of the table where the Incra fence will mount, so even if there was more variation, it wouldn’t have really mattered.

As you can see in the pic, there were a lot of tool marks from the router removing 1/2” per pass. Unfortunately, the router is not quite perpendicular to the table, and leaves a slight ridge. Fortunately, the ridge is only about .002” high. You can barely feel it However, endgrain birch saturated with epoxy is extremely hard to sand.

As for the rough surface I thought I was getting. It turns out that as soon as I start sanding, the (perceived) roughness disappeared almost immediately. But the ridges were another story. After a quick sanding, they can no longer be felt, but you can still see them. I spent about an hour to get it to this point. I’m tempted to get out the belt sander, but am too afraid of a mistake. Belt sanders can do tremendous damage in seconds. But I do have a really good one, a 4×24 Bosch with variable speed. I might try it at a slow speed to see if it works. I also tried a scraper, which helps, but that doesn’t remove much material either. I don’t want to remove much material, but man, that .002 is HARD to sand.

Today’s goal is to get the top sanded, and rabbitt the edges and insert a bubinga border. Maybe 3/8”x3/8”. The front edge needs a bevel to match the Unisaw.

I also may need to order a piece of stainless steel angle to attach the table to the Unisaw table. Right now, there’s an 1/8” steel plate, with some small tabs bolted to it. I’d like something a bit more secure, as the current table is only supported in 3 small places at the saw. Right now, mu plan for mounting is to use countersunk stainless screws through the top, and through the steel angle on the 3 sides, and use locknuts under the angle. This is a much better method than the standard 3/4” lag screws from underneath.

One last thing. I’m really unhappy witht he spots from the CA glue bleeding?? through. Hopefully, the polyurethane finish will minimize it. Other than that, I’m pretty happy. As of right now, I have a near perfectly flat router table, that won’t sag.

And one more last thing. Before I route through the surface to expose the T-Nuts, I’ll be completely finishing the table with a few coats of polyurethane. I don’t want the polyurethane to get into the threads.

Here’s a short video showing the surfacing operation.

-- Gerry,

3 comments so far

View dlmckirdy's profile


199 posts in 4628 days

#1 posted 03-22-2010 03:23 AM


Be careful using stainless steel threaded fasteners – They SIEZE very easily, and, once siezed, are permanent. I would be more prone to using nust over tapping the stainless angle (I’ve even had taps sieze). What happens, is that the stainless absorbs and then dissipates heat so fast that it first expands to allow the threaded fastener, then the heat dissipates into the surrounding material, causing a rapid contraction onto the threads. Usually, the last 1/8 to 1/4 turn galls and the softer, hot material shrinks to conform with the displaced material on the threads, and you may as well have welded them together. Make sure to use anti-sieze/lubricant and tighten with hand tools only! Don’t get discouraged with stainless, as the advantages usually far outweigh the disadvantages – just be aware.

GOOD LUCK! Tour table looks great. Wish I had some advice on the glue “bleedthrough”.

-- Doug, Bakersfield, CA - I measured twice, cut it twice, and it is still too short!

View Ger21's profile


1100 posts in 4626 days

#2 posted 03-22-2010 03:53 AM

Thanks, but I’m not going to tap the stainless, I’ll use nylon locking nuts underneath.

-- Gerry,

View SPalm's profile


5338 posts in 5377 days

#3 posted 03-22-2010 02:28 PM

Hey Gerry. Looks good. I am too am surprised that the CA showed through so much. Hmmm.

The surfacing operation looks great. I guess I a curious that you used a straight cut rather than a overlapping zig-zag pattern to reduce the ridges. I guess in the end it really doesn’t matter.

Have fun with the sanding :>)

-- -- I'm no rocket surgeon

Have your say...

You must be signed in to post the comments.

DISCLAIMER: Any posts on LJ are posted by individuals acting in their own right and do not necessarily reflect the views of LJ. LJ will not be held liable for the actions of any user.

Latest Projects | Latest Blog Entries | Latest Forum Topics