Greetings once again. I apologize for the long post. I didn’t know how to make it concise and still paint an accurate picture.
So many people here know so much more than I do I thought I might see if anyone can offer an insight into a problem I am having.
I have been building my machine for a long, long time… Mine was one of the pre-orders I know… I am slow. But I only have a few hours per weekend. I have tried to build it as carefully as I can for accuracy.
My spoilboard system is supposed to consist of 7.5” x 32.125” MDF pieces separated by T-Tracks. These will be mounted to a piece of BB .75 plywood with ¼-20 bolts into threaded inserts installed on the back side of the plywood. The plywood is in turn bolted to a torsion box. (See picture)
In order to be consistent, I used the 1F to make a spoilboard drill template. What I mean by this is that I didn’t want to attempt to use the 1F to make the holes in the plywood that would eventually hold the spoilboard. A tremendous amount of work went into that plywood in that each bolt insert holding the plywood to the torsion box had to be carefully drilled and recessed. In addition, there were inserts for the 1F itself. So, I didn’t want to take a chance that I might jack up the programming and ruin all that work. So instead, I set up a temporary spoilboard and then used the 1F to make a template.
Essentially I very precisely cut two 7.5” x 32.125” pieces of MDF. I moved the gantry all the way to the back and placed one piece against the y-rail. I then butted the other piece up to the piece against the y-rail and clamped it down. (I did this so that the template would be parallel with the y-rail)
I was having trouble with the touch probe being consistent even though I slowed it down so I decided to go with a center origin in my design. So I took a piece of drill stock and sharpened it to a point. I turned the router to its slowest speed and very carefully aligned the point to each side of the board, did a little math and found the center. (I checked the center X on each end of the Y as well.) So I set the Zero on the X and Y and used the touch probe to set the Z.
I then created my template. (See attached CAD and CAM for the template.)
SB Drill Template CO.nc (162.2 KB)
SB Drill Template CO v4.f3d (185.7 KB)
Oh how I wished I had checked that template more closely.
I then used the template and the T-tracks and very carefully drilled the Plywood. This was last summer sometime. It has taken me this long to finish and get to this point.
So I used the spoilboard template CAD created earlier to then create the CAD file for the actual spoilboard. All I did was create counter bores for the bolt clearance and also some holes for additional hold-down locations. (See attached files.)
SB Top mm.nc (404.7 KB)
Spoilboard Final v11.f3d (257.4 KB)
Note: There are features on both sides of the SB. I only included the CAM for the top side.
So I very carefully aligned a new piece of MDF up to my shiny new CNC and run the program. When I drop the spoilboard in place between the t-tracks the holes are visibly off! I measure the distance from the holes to the edge and they are off by about 1mm. Grrrrr… So I go back in and check my CAD and CAM and everything (to me) looks good. So I go back out and check where my threaded inserts are located in relation to the t-tracks and sure enough, every @#$@ one of them are shifted 1mm to one side.
My first thought was that I didn’t build the template correctly. That seems the most logical. But if that were the case, I would have had to have misaligned the actual spoilboard in the same way because it was off the same amount. (I came back in at that point and made sure that the holes in the the CAD for both the templae and the spoilboard were correct. They are.)
What is most telling, is that if I spin the spoilboard 180 degrees, the holes line up perfectly. This tells me that the holes in both my spoilboard template and the spoilboard itself are off by about 1mm.
If you knew me, you would know how much this is going to make my brain itch. It is an itch that would require much money time and trouble to scratch. I think I am just going to have to live with it and either make my spoilboards smaller or just make sure they are lined in the right direction.
But what happened??
So I go to the G-Code for the Spoilboard itself. I look at the code for the first hole and I find the maximum and minimum X distance. I then did a little math and figured out that it was correct. I subtracted the maximum from the minimum, divided it by two and added it to the minimum. This should be the distance from the center. I went back to Fusion and checked the CAD. The numbers match exactly. (The actual files are in mm rather than inches.) (G-code attached.)
So then I thought that the 1F itself may be off.
I took a clean piece of MDF and laid it across the X direction. I then Zeroed the machine and used the manual interface to move the X axis to 50mm I then very carefully plunged the sharpened drill stock barely into the MDF… just enough to make a tiny mark. At this point I moved the X axis to 800 and did the same thing. I then took the MDF and measured it with a very accurate 1 meter ruler. It measured 750.3 mm. In other words, the machine is accurate to within .3mm over a 750mm span.
So I am really baffled by this.
The worst thing is that I ran into this when I made my first test cuts a long time ago. I contacted Onefinity about this and as best as I can remember they just told me it was a hobby level machine and I shouldn’t expect too much. I think they may have also suggested I use Carbide Create rather than Fusion.
So… if anyone has any suggestions I would love to hear them. This afternoon I am going to try to do some testing and try not to cry over all of my spoil board mounting bolts being off by 1 mm. (At least they are ALL off by 1mm. 
