CNC Machining & 3D Printing: A Hybrid Approach to Precision Manufacturing

With recent advancements in 3D printing capabilities, it is becoming easier for manufacturers to use additive manufacturing to create parts from a wide variety of materials, including polymers like ABS, TPE, and PLA as well as carbon fiber composites, nylon, and polycarbonates. Even pricey metals like Titanium, Stainless Steel, and Inconel are becoming increasingly common in the world of additive manufacturing as well.

There is no doubt that the additive manufacturing space will continue to develop and grow in the coming years, but will it render subtractive manufacturing methods like CNC Machining obsolete? Absolutely not. In fact, precision CNC machining is likely more important to the additive manufacturing process than you may think, as a new process called “hybrid manufacturing” is quickly taking hold in the industry.

3d printing metal
3D printing of metal parts is becoming more common, but subtractive manufacturing is an important part of manufacturing precision additive parts.

Additive Manufacturing vs. Subtractive Manufacturing

Before implementing a hybrid manufacturing approach, it is important to understand the pros and cons of each method. Here is a quick breakdown of both additive and subtractive manufacturing, and the benefits and drawbacks of each.

Additive ManufacturingSubtractive Manufacturing
Adds material layers to create partsRemoves material layers to create parts
Slower process, better for small production runsFaster process, better for large production runs
Better for smaller partsBetter for larger parts
Rough surface finish that requires significant post-operation finishingMore definied surface finish with minimal post-operation finishing required
Less precise part tolerancesAble to hold extremely precise part tolerances
Cheaper material costsMore expensive material costs
Less material wasteMore material waste
Intricate details easier to createIntricate details can require complex programs and additional capabilities (5 axis)

Using CNC Machining to Create Precise 3D Printed Parts

Looking at the chart above, you will notice that one of the key differences between additive manufacturing and subtractive manufacturing is the surface finish and tolerances that can be achieved with each method. This is where a hybrid approach to additive manufacturing can be extremely beneficial.

As parts come off the printer, they can be quickly moved into a CNC machine with a program designed for part completion. The CNC machine will be able to get 3D printed parts down to the tight tolerances required by many industries and reach the desired surface finish. Advanced finishing tools and long reach, tapered tools from brands like Harvey Tool can easily machine the tight geometries of 3D printed parts, while extremely sharp diamond-coated tooling and material-specific tools designed for plastics and composites can work to create a beautiful, in-tolerance finished part regardless of the material.

long reach end mill
Long reach tools can easily machine hard to reach, intricate part details on 3D printed parts.

By designing a workflow like this in your shop, you can spend less time worrying about the precision of printed parts by adding in subtractive operations to keep material costs low, create less waste, and keep parts in tight tolerances for precision machining excellence.

Using 3D Printing to Increase CNC Machining Efficiency

If your shop is focused completely on subtractive manufacturing methods, you are probably thinking that there is no need for an additive option in your shop. Can’t a CNC machine create everything a 3D printer can, and in less time? Not necessarily. Again, by using the two methods together and taking a hybrid approach, you may be able to lower your manufacturing and material costs.

For example, you could machine the bulk of a part with typical subtractive machines, which would likely take a very long time using additive methods. Then you can go back to that part with a 3D printer to add intricate features to the part that may take complex programming and hours of planning on a subtractive machine. An impeller is a great example, where the bulk of that part can be machined, but the tricky fins and blades could be printed onto the part, and then finished back on the CNC machine.

3d printed metal parts
3D printed impeller waiting for finishing operations

The ability of additive machines to literally “add-on” to a part can also make for a cheaper approach to part design. Instead of using expensive materials like Inconel or Titanium to machine an entire part, portions of the part that do not require extreme heat resistance could be machined out of cheaper steel, while the heat resistant portions using expensive materials can be added later through additive methods.

Hybrid Manufacturing Machines

As hybrid manufacturing workflows become more popular, so do new hybrid manufacturing machines. These hybrid machines are all-in-one machines where both additive and subtractive manufacturing can be performed in a single setup. Many of these machines offer metal 3D printing as well as multi-axis machining capabilities, ready for even the most complex parts thrown their way. With a bit of customization, large-scale 3D printing machines or CNC mills can be retrofit to allow for hybrid manufacturing with add-ons from companies like Hybrid Manufacuring Technologies.

hybrid manufacturing machines
Example of a hybrid machine add-on from Hybrid Manufacturing Technologies, featuring 3D printing spindles and milling tools in the same machine carousel.

As manufacturing and design techniques get progressively “smarter” with CAM/CAD programs offering generative design and artificial intelligence, these hybrid machines could become a new standard in high-end machine shops working in advanced manufacturing industries like aerospace, medical, defense, and the mold, tool & die market.

Overall, in 2021 we are still early on in this new revolution of hybrid machining and advanced design methods, but it is important to understand the role that adding a CNC machine could have in your additive-focused shop, and vice versa. By combining additive and subtractive together, shops can mitigate the cons of each method and take full advantage of the benefits of having both options available on the shop floor.

How to Optimize Results While Machining With Miniature End Mills

 The machining industry generally considers micromachining and miniature end mills to be any end mill with a diameter under 1/8 of an inch. This is also often the point where tolerances must be held to a tighter window. Because the diameter of a tool is directly related to the strength of a tool, miniature end mills are considerably weaker than their larger counterparts, and therefore, lack of strength must be accounted for when micromachining. If you are using these tools in a repetitive application, then optimization of this process is key.

Key Cutting Differences between Conventional and Miniature End Mills

Runout

Runout during an operation has a much greater effect on miniature tools, as even a very small amount can have a large impact on the tool engagement and cutting forces. Runout causes the cutting forces to increase due to the uneven engagement of the flutes, prompting some flutes to wear faster than others in conventional tools, and breakage in miniature tools. Tool vibration also impacts the tool life, as the intermittent impacts can cause the tool to chip or, in the case of miniature tools, break. It is extremely important to check the runout of a setup before starting an operation. The example below demonstrates how much of a difference .001” of runout is between a .500” diameter tool and a .031” diameter tool.

chart comparing tool diameter for runout in micromachining
The runout of an operation should not exceed 2% of the tool diameter. Excess runout will lead to a poor surface finish.

Chip Thickness

The ratio between the chip thickness and the edge radius (the edge prep) is much smaller for miniature tools. This phenomena is sometimes called “the size effect” and often leads to an error in the prediction of cutting forces. When the chip thickness-to-edge radius ratio is smaller, the cutter will be more or less ploughing the material rather than shearing it. This ploughing effect is essentially due to the negative rake angle created by the edge radius when cutting a chip with a small thickness.

If this thickness is less than a certain value (this value depends of the tool being used), the material will squeeze underneath the tool. Once the tool passes and there is no chip formation, part of the plowed material recovers elastically. This elastic recovery causes there to be higher cutting forces and friction due to the increased contact area between the tool and the workpiece. These two factors ultimately lead to a greater amount of tool wear and surface roughness.

chart of edge radius in relation to chip thickness for micromachining
Figure 1: (A) Miniature tool operation where the edge radius is greater than the chip thickness (B) Conventional operation where the edge radius is small than the chip thickness

Tool Deflection in Conventional vs. Micromachining Applications

Tool deflection has a much greater impact on the formation of chips and accuracy of the operation in micromachining operations, when compared to conventional operations. Cutting forces concentrated on the side of the tool cause it to bend in the direction opposite the feed. The magnitude of this deflection depends upon the rigidity of the tool and its distance extended from the spindle. Small diameter tools are inherently less stiff compared to larger diameter tools because they have much less material holding them in place during the operation. In theory, doubling the length sticking out of the holder will result in 8 times more deflection. Doubling the diameter of an end mill it will result in 16 times less deflection. If a miniature cutting tool breaks on the first pass, it is most likely due to the deflection force overcoming the strength of the carbide. Here are some ways you can minimize tool deflection.

Workpiece Homogeny

Workpiece homogeny becomes a questionable factor with decreasing tool diameter. This means that a material may not have uniform properties at an exceptionally small scale due to a number of factors, such as container surfaces, insoluble impurities, grain boundaries, and dislocations. This assumption is generally saved for tools that have a cutter diameter below .020”, as the cutting system needs to be extremely small in order for the homogeny of the microstructure of the material to be called into question.

Surface Finish

Micromachining may result in an increased amount of burrs and surface roughness when compared to conventional machining. In milling, burring increases as feed increases, and decreases as speed increases. During a machining operation, chips are created by the compression and shearing of the workpiece material along the primary shear zone. This shear zone can be seen in Figure 2 below. As stated before, the chip thickness-to-edge radius ratio is much higher in miniature applications. Therefore, plastic and elastic deformation zones are created during cutting and are located adjacent to the primary shear zone (Figure 2a). Consequently, when the cutting edge is close to the border of the workpiece, the elastic zone also reaches this border (Figure 2b). Plastic deformation spreads into this area as the cutting edge advances, and more plastic deformation forms at the border due to the connecting elastic deformation zones (Figure 2c). A permanent burr begins to form when the plastic deformation zones connect (Figure 2d) and are expanded once a chip cracks along the slip line (Figure 2e). When the chips finally break off from the edge of the workpiece, a burr is left behind (Figure 2f).

Tool Path Best Practices for Miniature End Mills

Because of the fragility of miniature tools, the tool path must be programmed in such a way as to avoid a sudden amount of cutting force, as well as permit the distribution of cutting forces along multiple axes. For these reasons, the following practices should be considered when writing a program for a miniature tool path:

Ramping Into a Part

Circular ramping is the best practice for moving down axially into a part, as it evenly distributes cutting forces along the x, y, and z planes. If you have to move into a part radially at a certain depth of cut, consider an arching tool path as this gradually loads cutting forces onto the tool instead of all at once.

Micromachining in Circular Paths

You should not use the same speeds and feed for a circular path as you would for a linear path. This is because of an effect called compounded angular velocity. Each tooth on a cutting tool has its own angular velocity when it is active in the spindle. When a circular tool path is used, another angular velocity component is added to the system and, therefore, the teeth on the outer portion of tool path are traveling at a substantially different speed than expected. The feed of the tool must be adjusted depending on whether it is an internal or external circular operation. To find out how to adjust your feed, check out this article on running in circles.

Slotting with a Miniature Tool

Do not approach a miniature slot the same way as you would a larger slot. With a miniature slot, you want as many flutes on the tool as possible, as this increases the rigidity of the tool through a larger core. This decreases the possibility of the tool breaking due to deflection. Because there is less room for chips to evacuate with a higher number of flutes, the axial engagement must be decreased. With larger diameter tools you may be stepping down 50% – 100% of the tool diameter. But when using miniature end mills with a higher flute count, only step down between 5% – 15%, depending on the size of the diameter and risk of deflection. The feed rate should be increased to compensate for the decreased axial engagement. The feed can be increased even high when using a ball nose end mill as chip thinning occurs at these light depths of cut and begins to act like a high feed mill.

Slowing Down Your Feed Around Corners

Corners of a part create an additional amount of cutting forces as more of the tool becomes engaged with the part. For this reason it is beneficial to slow down your feed when machining around corners to gradually introduce the tool to these forces.

Climb Milling vs. Conventional Milling in Micromachining Applications

This is somewhat of a tricky question to answer when it comes to micromachining. Climb milling should be utilized whenever a quality surface finish is called for on the part print. This type of tool path ultimately leads to more predictable/lower cutting forces and therefore higher quality surface finish. In climb milling, the cutter engages the maximum chip thickness at the beginning of the cut, giving it a tendency to push away from the workpiece. This can potentially cause chatter issues if the setup does not have enough rigidity.  In conventional milling, as the cutter rotates back into the cut it pulls itself into the material and increases cutting forces. Conventional milling should be utilized for parts with long thin walls as well as delicate operations.

Combined Roughing and Finishing Operations

These operations should be considered when micromachining tall thin walled parts as in some cases there is not sufficient support for the part for a finishing pass.

Helpful Tips for Achieving Successful Micromachining Operations

Try to minimize runout and deflection as much as possible when micromachining. This can be achieved by using a shrink-fit or press-fit tool holder. Maximize the amount of shank contact with the collet while minimizing the amount of stick-out during an operation. Double check your print and make sure that you have the largest possible end mill because bigger tools mean less deflection.

  • Choose an appropriate depth of cut so that the chip thickness to edge radius ratio is not too small as this will cause a ploughing effect.
  • If possible, test the hardness of the workpiece before machining to confirm the mechanical properties of the material advertised by the vender. This gives the operator an idea of the quality of the material.
  • Use a coated tool if possible when working in ferrous materials due to the excess amount of heat that is generated when machining these types of metals. Tool coatings can increase tool life between 30%-200% and allows for higher speeds, which is key in micro-machining.
  • Consider using a support material to control the advent of burrs during a micromachining application. The support material is deposited on the workpiece surface to provide auxiliary support force as well as increase the stiffness of the original edge of the workpiece. During the operation, the support material burrs and is plastically deformed rather than the workpiece.
  • Use flood coolant to lower cutting forces and a greater surface finish.
  • Scrutinize the tool path that is to be applied as a few adjustments can go a long way in extending the life of a miniature tool.
  • Double-check tool geometry to make sure it is appropriate for the material you are machining. When available, use variable pitch and variable helix tools as this will reduce harmonics at the exceptionally high RPMs that miniature tools are typically run at.
variable pitch versus non-variable pitch
Figure 3: Variable pitch tool (yellow) vs. a non-variable pitch tool (black)