Computer Comtrolled Machining

In this unit, we are going to "build something big". Fair enough... So my first idea was to build an Adirondack chair which would be very useful and is a very attractive object. I have some left-over Ipe ( a tropical hardware) from a deck I recovered. Upon further examination and given the time constraints and complexity, I decided to proceed with a more modest project which has several of the technical aspects but is less challenging from a design perspective. These are the principle skills that I wanted to acquire in this unit:

  • Translate a design into format that V-carve/Aspire can use
  • Setup up precision milling to a relatively high tolerance
  • Use different types of tool heads for different types of cutting/engraving
  • Understand how to mill dimensional (rough cut) lumber into planed, prepared wood
  • Understand how to mill both sides of lumber accurately




Bees, bees, bees...

Part of what is keeping me busy this time of year is getting my bees ready for the spring. I use Langstroth hives and apply the Rose method to managing colony growth during the season. One thing you can never have enough is hive boxes since they provide space for the bees but also are great places to store frames (especially prepared frames). Boxes are actually very precisely build objects as the inside dimensions are precisely designed to accommodate the bees - commonly referred to as "Bee Space". On the right is an image of a standard Langstroth box and for those familiar with beekeeping - you'll recognize that it uses the standard frames and not the Hoffman frame used in the Rose method. Also note that these are not Rose Hives since I don't have the resources (yet) to start building my own boxes. I also have some issues with the practicality of the Rose Box but I digress...

The box itself is straight forward with the sides being simple rectangles which are relatively straight forward to mill/cut. The ends are more complicated as construction requires offsets and on the top and sides for construction and the placement of a frame holder that precisely positions the standard frames with respect to each other - something that isn't necessary when using Hoffman frames which is practical but significantly more expensive.




Designing the pieces

An existing Langstroth box was measured - it was interesting to note that there is a non-trivial deviation in what are considered standard dimensions. This website has a table showing measurement of the "standard" Langstroth box from across the globe. Since boxes need to fit precisely when stacked, existing boxes from my collection were used to ensure conformity with existing hives.

Solidworks was fired up and sketches were made. One thing that I wasn't able to do in Solidworks was sketch the indentation for the handle. We'll come back to this in the next paragraph. The pieces were all sketch/extrude/cut operations and presented no difficulty.

The outside handles a more complicated. First the handle has a lot of Bezier curves that intersect with flat surfaces and I had never done this type of geometrical modelling in SWs. Secondly, handles are traditionally cut into boxes using either a table router or tablesaw. Since we are interested in using the ShopBot, we need to come with a different design since the geometry of such cuts cannot be replicated by the ShopBot. Interestingly, the V-carve software has a handle like object already available in the modelling library so we are going to try and use this for the box.




The good, the bad and the worst

Boxes are traditional made from pine, fir or spruce. Trouble is that getting wood that is of decent quality is either time-consuming, expensive or both! For the purposes of the prototype, we're going to use dimensional lumber used for cement framing. This is not the type of material I would use in a finished box since it is full of knots which may eventually fall out creating a new ventilation opportunities in the hive. But for the purposes of this exercise, it will have to do. The first thing that needs to be done is to plane both surfaces of the wood to (a) get the desired width and (2) smooth the surface so it can be properly finished. This requires a surfacing bit which some of you might have seen if you were involved in surfacing the sacrificial surface of your ShopBot. V-graph was used to generate a simple rectangle slightly larger then the actual piece of lumber being used. The workpiece was fixed on the table surface, zeroed on the upper surface and a 2mm run was made. The board was flipped, fixed and the same job rerun. Thickness was measure and a final run was made so wood was of appropriate thickness.

Unfortunately, the previous paragraph is not how it actually worked out since we had a lot of problems designing the handle. Usually, the handles are build using an angled table saw - many videos exist on Youtube showing this type of construction. Since we wanted to build it with a CNC this was not possible due to cut geometries that don't permit making an angled cut into a plane. We used Aspire to make a 3D design but unfortunately, the version of V-carve on our ShopBot doesn't have 3D design functionality. We tried several other means of getting the design data into the systems none of which worked. It was very disappointing that something as simple as a handle in a box could not be reproduced without a tremendous effort and even when the design was done it was not compatable with the software running on the machine.

Two steps back

We finally did a very simple plaque so we could practice using the Shopbot.

Material Management

Boxes are traditional made from pine, fir or spruce. Trouble is that getting wood that is of decent quality is either time-consuming, expensive or both! For the purposes of the prototype, we're going to use dimensional lumber used for cement framing. This is not the type of material I would use in a finished box since it is full of knots which may eventually fall out creating a new ventilation opportunities in the hive. But for the purposes of this exercise, it will have to do. The first thing that needs to be done is to plane both surfaces of the wood to (a) get the desired width and (2) smooth the surface so it can be properly finished. This requires a surfacing bit which some of you might have seen if you were involved in surfacing the sacrificial surface of your ShopBot. V-graph was used to generate a simple rectangle slightly larger then the actual piece of lumber being used. The workpiece was fixed on the table surface, zeroed on the upper surface and a 2mm run was made. The board was flipped, fixed and the same job rerun. Thickness was measure and a final run was made so wood was of appropriate thickness.

Unfortunately, the previous paragraph is not how it actually worked out since we had a lot of problems designing the handle. Usually, the handles are build using an angled table saw - many videos exist on Youtube showing this type of construction. Since we wanted to build it with a CNC this was not possible due to cut geometries that don't permit making an angled cut into a plane. We used Aspire to make a 3D design but unfortunately, the version of V-carve on our ShopBot doesn't have 3D design functionality. We tried several other means of getting the design data into the systems none of which worked. It was very disappointing that something as simple as a handle in a box could not be reproduced without a tremendous effort and even when the design was done it was not compatable with the software running on the machine.




To-do

That Adirondack chair in Ipe would still be nice to have. And based on our experience with the bee box, technically much easier to realise since the chair is a collection of flat pieces with rounded edges.




Bibliography

N/A