Last time we checked in I had determined that the table was indeed pretty flat, so I was going to proceed. So I did.
I was going to do some glue-up for another project which meant my work surface (table saw) would be unavailable for torsion box work, so I ripped some 2″ wide MDF strips, cut 6 pieces 29″ long and glued them up three and three for end risers.
The next day I finished my other project, trimmed up the risers on the saw to 1-5/8″ wide, and drilled the holes in them.
Next I proceeded to glue the top layer onto the table. First I measured out the holes and drilled them, then I spread glue all over the table, positioned the top layer and placed some weights on it.
Unfortunately the top piece wasn’t quite square and I lined it up using one side instead of the front edge which I had based my measurements on so it turned out that my holes didn’t quite line up. I didn’t discover this until assembly time and I managed to open the holes up enough from the top, but it was annoying to discover this mistake.
After gluing up the top I chucked up a flush trim bit in my router and trimmed the edges that were overhanging. Then I put a 3/4″ diameter bit in and started cutting the slots for the T-Slot channels. I used my not quite straight cutting guide because it’s not super important that the slots be straight. I cut the slots on 3″ centers.
I had forgotten how much I hate using a hand-held router. I don’t have an adapter for the (probably useless) dust port on the router, so the dust just goes everywhere. My shop, already a disaster, turned into a huge mess. This is only about half of the dust.
Anyway, I soldiered on and got all the slots cut. They came out great. The rails sit about .030 below the table top.
Then I spent about 30 minutes cleaning up the dust.
On to bolting the machine to the base!
First I had to make some room in the shop. I had built a new table when the S3 arrived but I decided it would be more convenient to put the S3 where my S1 was currently sitting, so I had to play a bit of a re-arranging game to clear off the new table and put the S1 on it. Once I got the torsion table in place I started by test-fitting the S3 end plate.
I wanted the riser and end plate to be flush with the front of the torsion table so that I could clamp a work piece to the front if I wanted to. This meant that I needed to open up the holes in the end plate and riser a bit more to get them in the right position.
Once I had enough adjustment room I clamped some straight edges to the end plate and bolted it down tight. This end plate was to be my zero and all the other squaring would reference off of it.
Then I figured that I had better screw down the T-Slots before I mounted up the S3 and it was in the way. The T-Slot description said they came drilled on 3″ centers for bolting down but the ones I received had no holes, so I had to drill my own. Oh well, I have a drill press. After drilling a bunch of holes I started screwing them to the table. I spaced them 3/4″ from the riser so I could easily slip the hold-down bolts in.
Then I had to pilot drill all the holes again into the torsion table and put in all the screws. 80 of them I think.
Once the T-Slots were screwed down I loosely bolted the other end plate and riser to the table and mounted up the S3 to them. I left all the bolts loose but wanted to do a quick check to see if the router height seemed okay.
I slipped the router into the mount and put one of my fixtures onto the table just to see how it looked.
I actually think it looks a little high, but I’m going to run with it for now. I will mount the router about half way up the body so there is less of a lever arm into the Z plate and use pieces of MDF to lift the work pieces to a more reasonable height. If I decide that it’s too high it’s pretty simple to make new risers that are only two strips high.
The other thing I will want to do is figure out how to add a stop and probably another limit switch to the Z axis so it doesn’t get over extended and come off the rails at the bottom.
On to squaring up the machine! The first thing was to pick a front corner, make sure the Y rail was square to the table and tighten the screws.
Next I measured the height using my Snap-Check gauge (which worked great for this application) and moved over to the other side to try and make them the same.
They were, of course, not. They were about .015″ different, which might not sound like much, but all the errors in square tend to accumulate. So I spent some time moving the two ends up and down until they were within .003″ of each other. I called that good enough. I also was careful to pull the carriage forward and back to make sure the rails were parallel and the correct distance apart.
When the carriage was forward I discovered that it didn’t meet the end plate on both ends, so that meant that the X axis was not square to the Y by quite a bit. So I pulled on the far end plate to yank it into square a bit more.
Squaring up the machine is the most fiddly bit and I spent a bunch of time checking all four corners to make sure they were the same height as I tightened them down. I couldn’t get the far right corner to be the same height as the others but since I didn’t feel like taking the end plate off again to open the holes up some more I used some brass shim stock under the corner to raise it up.
Once all four corners were within .003″ of each other it was on to squaring up the X axis.
First I checked to make sure the rail wasn’t twisted. I loosened up the screws on both ends and then tightened one end square to the table.
After that it was using the Snap-Check gauge and moving from one end to the other and adjusting them until they were the same height.
I also measured from the rail to the front edge of the carriage plate to make sure they were the same distance in.
After the X rail was squared up I moved on to the Z. I took off the Z plate (easy to do with all the clearance underneath) and checked that the Z rails were square to the table. They, of course, were not, so I loosened them up and squared them.
One thing that I haven’t shown was measuring the squareness of the X and Y axis to each other. One reason is because it takes me two hands to check it and I can’t take a photo at the same time. Suffice it to say that I made the X and Y axis as square to each other as I could. I’ll try and use some clamps to arrange a photo of how I checked it.
The other reason is because tensioning the Y drive belts will also have a lot to do with how square the X and Y are to each other and I haven’t done that yet.
The next step was to put the Z plate and router mount back on and make sure it was square to the table.
After that I mounted up the router and attached my test indicator to it to see how things looked. (No photo.)
Things looked horrible. The Z axis was not perpendicular to the table in the YZ plane by quite a bit! Whisky-Tango-Foxtrot?
At that point I stopped. It was getting kind of late in the evening and I didn’t really want to get into it.
This morning I went down to take a look at the state of things. I pulled off the Z plate and measured the square of the Z rails to the table in the YZ plane. They were off a little bit, but not a ton. The Z carriage plate was closer to square. So I assume my Z travel plate is possibly bent or the router was not mounting up square to the plate.
Unfortunately while looking things over I discovered that I have the V-Wheel washer clearance issue reported here.
Which means that any more work trying to square up the Z axis would be pointless at this time. So I’m stopped until I hear back from Carbide 3D support.