4
$\begingroup$

How do CAD designers account for the fact that a blank may have to be flipped to a different side in order to allow the piece to be completely milled through a CNC system?

For an example, let's use this car made from a rectangular block of pine:
pinewood derby "leap frog"

It appears that three planes need to be surfaced in order to mill the object.

  • The bottom plane of the block has a slight concave shape
  • The side plane has two hollowed out shapes that would not be accessible from either the top or bottom planes
  • The top plane has a slightly concave inflection applied to the slope

And there are several cuts that go all the way through the blank.

It seems easy enough to mill the bottom shape first, including any through cuts from the bottom. This work on the bottom wouldn't significantly affect the overall shape of the blank, and there shouldn't be any complications moving to the next plane.

But after you complete the bottom, milling either the side or the top would appear to create problems for flipping the blank to the final plane. Specifically, the outline of the original blank would have changed significantly enough that prior mill registrations would be thrown off.

I suspect that this is a somewhat common problem as complex shapes are fairly commonplace. And I'm more interested in the general approach to solving this problem than I am about how the example car would be designed.

So how does a CAD designer account for the fact that the edges of the blank will be changing during milling and that those changes may (will) complicate the milling process?

Disclaimer: My knowledge of CAD design and CNC manufacturing is fairly limited. My knowledge comes primarily from what I picked up at University and has been supplemented by advanced DIY projects.

$\endgroup$
  • $\begingroup$ I am not sure what 'blank' means in this context, anyone can help? $\endgroup$ – Vladimir Cravero Apr 16 '15 at 6:51
  • 1
    $\begingroup$ @VladimirCravero blank in CNC context means the original piece that you put in the CNC and will get worked on. $\endgroup$ – ratchet freak Apr 16 '15 at 9:58
3
$\begingroup$

First, we should recognize that parts aren't just blindly thrown onto a table and a generic mill used to cut whatever shape we want. Each part requires the proper tooling and fixturing. On a production level, the tooling is probably going to be a standard set of tools that you select from for most parts. Depending on what kind of parts you are making, fixturing can shared between parts, but custom parts may require custom fixturing.

For example, at my company, we have different series for our products, and the parts in one series have similar geometry on their key features that allows us to use the same chucks and blocks to hold them during machining processes. However, I recently designed a new part with a new forging that still technically fits into one of our larger series, but will require brand new custom fixturing.

Second, most parts of any complexity won't be able to be fully machined in one operation, and even if it's possible, it doesn't mean it's the best way to get the part to function as intended. This means switching tools while cutting the part from the same direction, and changing the way the part is held to be able to access different sections of the part.

To this end, in the course of designing a part, we try to ensure that the machining operations of the same type and on the same side or in the same section of the part can be done at once, i.e. without having to handle the part. This saves machining time, but it also ensures that we use our fixturing as efficiently as possible.

The last thing to note, more from a manufacturing engineering point of view, is that the parts aren't necessarily machined entirely on one machine. What this really means is that if we are making 100 parts, we can do the first machining operation of the part 100 times. Then, we reset the machine and do the second operation 100 times. We reset again and do the third operation 100 times.

Hopefully, at this point you can see how these things come together. We have a rough intent for the order of operations during design, often executed by creating the features in that order in CAD, so that we have an idea of what the part will look like at each step. These operations are grouped into sets that can be performed with the part chucked in the same manner, and then fixturing is created for each set. The first set of operations may only require simple fixturing because it may use a standard blank like a simple round bar. Subsequent operations may be more complex to implement, but we know what the part will look like, and we can plan the fixturing based on that. Unless we're doing one-piece orders, we can do all of one group for a whole order on one machine, then do the next group on a separate machine, or reconfigure the original machine to do the next operations.

From a workflow perspective, I should note that this process is not entirely handled by the design engineers. As a design engineer, I have a good handle on what is and isn't possible, but for special cases, I'll go to the manufacturing engineers or the machinists and ask for their input, see what will make their job easier, and whether what I want to do can actually be done. They will be the ones to actually plan out the order of operations and design/order fixturing and tooling, but communication is key here, as is often the case.

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy