This article is written and contributed to engineering.com by Chandrakant Patel, HP Chief Engineer and Senior Fellow.
The ability of 3D digital manufacturing to create parts on-demand with vivid shapes, and with contours and physical attributes hitherto not possible, heralds the age of intent based systemic design. For the designer, the mantra now becomes the value derived at a system level.
The Age of the Drafting Table
Prior to 1980s, drawings created on a drafting table defined a part by showing the part with orthographic views and auxiliary views. The number of views necessary to define the part in its entirety were determined by imagining a part in a glass box, setting a viewing direction as shown below in Figure 1, and projecting the edges of the part to each of the surfaces of the glass box. Drawings were done on vellum paper, and blueprint copies were shipped to the fabricators of parts. Proper geometric dimensioning and tolerancing techniques applied to the various views were central in assuring accuracy of parts, particularly parts for assemblies. The drawings were done separately from engineering analysis, and often, by dedicated drafting teams. The parts were fabricated in stages – initially with soft tooling to qualify prototypes, and then upon full certification, fabrication was committed to hard tooling. The non-concurrence between design and analysis, and the long cycle prototyping, often meant over provisioning.
The Age of Computer Aided Design (CAD)
Computer Aided Design (CAD) started with 2D, now using a computer instead of a drafting table to draw out multiple views, and printing the drawings. Engineering analysis – stress, structural, thermal, etc - became computer aided as well (CAE). Initially, like the blueprints, the printed drawings were shipped physically to the fabricators. Subsequently, by late 1980s, drawing files were shipped electronically and ported to the fabrication machines.
The advent of 3D CAD by late 1980s led to creation of geometric design of parts in 3D by Boolean operations (adding and subtracting solid bodies), or history based parametric approaches. The coupling of CAD with computer aided engineering analysis (CAE) became stronger and reduced the over provisioning in design of parts. By early 2000s, great improvements in 3D scanning of parts helped further in exploring designs. The part digital files were sent directly to the fabricators. In spite of all these improvements, the exploration of design-fabrication space was limited due to soft tooling/hard tooling conundrum.
21st Century Age of Intent Based Design
The growth of 3D Digital Manufacturing changes the paradigm once again with expansion of design space exploration, and convergence of prototyping and manufacturing. Furthermore, now an engineer can start with the intent of the design. Mass reduction and reduction in the number of components that make up an assembly are oft quoted examples. However, systemic intent driven examples offer even more exciting opportunities. As an example, consider an automobile cooling system design. The intent of the cooling system is to maximize cooling performance while minimizing power required to drive the cooling system. This maximization of cooling system performance per unit power is particularly important for electric vehicles to improve the range of the vehicle by reducing the power draw from the batteries.
In this context, consider a single phase pumped liquid cooling system for a heat dissipating source, such as the battery pack, in an electric automobile. Typically, as showing in Figure 3, the cooling system has a cold plate in which the circulating liquid picks up the heat flux from the heat source and transfers the heat to the heat exchanger that transfers the heat to the ambient. Figure 3 shows a liquid cooling loop, driven by a pump, that transfers the heat flux from the cold plate (the cold plate could be built into the case of the battery pack) to an air cooled heat exchanger. In turn, the air cooled heat exchanger, aided by a fan, transfers the heat to the ambient.
In the “art to part” age of intent based design, the designer of the cooling system starts by thinking holistically with the performance of the cooling system. As shown in Figure 3, she considers the coefficient of performance of the ensemble. The ensemble consists of the cold plate(s), pump(s), piping system, heat exchanger(s), and the fan(s) for the heat exchanger. The Coefficient of Performance of the ensemble (COPensemble) is expressed as the ratio of the amount of heat removed in Watts (Joules/sec) to the sum of all the work required in Watts (joules/sec) by the active components – the pump(s) and fan(s) - that make up the cooling system. She focuses on maximizing the COPensemble.
Maximizing the COPensemble necessitates examining the cooling system input work that is driven by two key elements:
- resistance to coolant flow, quantified as pressure drop (N/m2 or Pascals), in the cold plate and piping due to transitions such as bends, sharp changes in area (contraction and expansion of flow)
- the volume flow rate (m3/s).
The work done by the pump (power draw by the pump) is the product of Volume flow (m3/s) and the pressure prop in the overall piping system (N/m2) divided by the pump efficiency at the operating point Similarly, the work required by the fan(s) in the heat exchanger is the product of Volume flow (m3/s) of air and the heat exchanger pressure prop (N/m2) divided by the fan efficiency at the operating point.
The Magic of 3D Digital Manufacturing
The designer takes the following steps to exploit the magic of 3D digital manufacturing to derive value for the electric car maker, and the user, by improving the COPensemble i.e. maximizing heat transfer while minimizing power draw of the cooling system.
- Designs fluid flow channels and enhanced surface areas, with the range 3D flexibilities, to improve heat transfer in the cold plate and the heat exchanger.
- Designs large sections of the assembly as a single 3D printed part. Conventional methods required the use of multiple components – transitions, elbows, etc - from a catalog to create a cooling piping assembly (figure 4)
- Designs smooth area ratio changes, contractions, expansions, in the piping to reduce pressure drop (see Figure 4). Conventional methods had sharp area ratio changes in flow resulting in higher pressure drop (Figure 4).
- Designs 3D printed impeller for the coolant pump to maximize efficiency. Conventionally, the designer would have to pick a pump from a catalog with a given impeller – or over a period of months iterate with conventional fabrication methods – to create an impeller that provides the right flow rate (m3/s) at a given pressure drop (N/m2) at peak wire to water efficiency (ɳpump).
- Design 3D printed impeller for the air moving system that maximizes the wire to air efficiency (ɳair) for the required air flow and pressure drop. In the past, the designer was bound by the constraints of the fans and blowers in the catalog – or, over a period of months iterate with conventional fabrication methods – to create a right provisioned impeller.
An Example of Intent Based Design
Figure 4 below shows an example of an intent based designed part for use in the HP manufacturing line. The part on the left is the HP 3D MJF part that replaced the machined part on the right. The value proposition comprised of the following salient components: reduction in pressure drop from smoother flow, mass reduction from 575 grams to 53 grams, 91 % cost reduction and part integration from 5 to 1 and lead time reduction of 91%.
3D Digital Manufacturing takes us to an age of Need Based Provisioning of Resources
The systemic intent based design, the close coupling of design and analysis, and the ability to expeditiously produce parts with 3D digital manufacturing heralds the age of “need based provisioning” of resources.