CNC Tool Offset Test

The tool offset - that is the specified bit diameter when creating toolpath for a CNC can be critical.
A project like my Tahoma Decagon Frame the correct tool offset is very important. If the individual segments of the decagon shape are each off just a little - multiply that by ten and the accumulated error will come back to bite you!
When the bit diameter is not called out correctly an inside cut will be different than an outside cut . This can be a difficult concept to envision. In the video I have featured a drawing that hopefully will make this clear.

[@rbyt2010]

Thanks Darrell. Surprising to see a new bit so far out of wack!

[@DarrellPeartFurnitureMaker]

It was an Amana Spetkra bit no less

[@rbyt2010]

@@DarrellPeartFurnitureMaker jeez, I like those

[@DarrellPeartFurnitureMaker]

@@rbyt2010 me too! I guess it’s possible one got by

[@fadi9575]

High end bits, e.g. Amana Tools or BitsandBits, are extremely accurate to what say. The diameter is not the issue. The issue is the tight tolerances that the CNC cuts to in wood, which you have to account for. What you did is this: 0.375-0.365=0.010 inch difference in diameter, that is 0.005 in radius offset added to both sides to allow for the parts to mate. In Vectric software you have the option of adding “offsets”. What you did was to trick the software to add that offset. When I inlay in the CNC or do finger joints, I have to add usually 0.003 in (in both directions) offset, and it is usually a trial and error. I’ve seen your video in which you do that manually when you fit finger joints if I remember. Also, you’ll have to remember in this case that plywood behaves differently than hardwood, and 0.005 offset maybe too much for mahogany. You’ll have to test in the same species unfortunately. If the bit diameter was wrong, then when we cut a large pocket with the CNC the error would multiply and the result will be way off in an inlay for example, but that is never the case, the added offset remains the same (0.003 for example) no matter how large the female inlay pocket. I hope this helps.

[@DarrellPeartFurnitureMaker]

Actually the bit was an Amana Spetkra bit.
I respectively disagree. The diameter is the starting point. If its not correct and the tolerances are close then you will have problems. While the bit should have measured .375 ( no run out or other issues) the slot I cut measured .368. You are right about the .003 offset for mating parts - I called that out wrong in the video. If you recall in the video the pieces that mated correctly were .365 which was .003 off from the measured slot.

[@JohanLarssonKiruna]

I guess a deep curve, U-shaped male/female, would be easy and accurate to test fit with? Also, it would be interesting to see if multiple cuts actually multiply any errors or not, like Fadi mentioned previously. Maybe I'm wrong, but it seems to me that's the only way, or at least the best way, to confirm where the errors are.

[@DarrellPeartFurnitureMaker]

the shape is easy to test fit as you can see in the video - a difference of .001 made a noticeable difference. Multiple cuts do magnify any error - but the combine error is not enough in this case to make things go awry.

[@rme5830]

way above me like the space station, were the parts cut out in 1 pass, as a tool and die maker before the calculator was invented and dial calipers were big $ a good mic. was the key along with gage blocks and .0001 indicator was used. could heat from cutting change the diameter. of the bit .001 or 2?

[@DarrellPeartFurnitureMaker]

the parts were cut out in several passes then a .01 clean up pass was taken. I'm not scientific to know how much heat would change the diameter.

[@garycaruso4883]

I wonder if you have any tool offset or backlash compensation in your CAM that is playing with this?

[@DarrellPeartFurnitureMaker]

Thanks Gary, I should look into that. I didn't even know there was such a thing as backlash compensation. I love what CNC can do but I only dive into the details when I have a problem - so some of my knowledge is lacking.