We are saddened to report that Professor Dornfeld passed away in March, 2016. If you enjoyed his blog, please consider making a contribution to either of two funds at UC-Berkeley that have been established in his memory.

David A. Dornfeld Graduate Fellowship
David A. Dornfeld Scholarship

Friday, April 27, 2012

Leveraging Manufacturing, Part 2


Some details on processing

The last posting began to dig into the leveraging discussion and started to elaborate on this topic using an example. The example was  from a recent paper from our research group at Berkeley and focussed on an important aspect of vehicles and transportation - the gear train.

The efficiency of gear systems was described as deriving from a variety of factors including the surface roughness of the mating surfaces.  Other studies have shown an even greater dependence on the surface roughness of the mating surfaces for hypoid gear pairs, which are found in automotive differentials. And because the vast majority of environmental impacts of an automobile occur during the use phase  the impact of increased manufacturing precision through better surface finish on the final drive reduction of an automotive manual transmission drivetrain makes this an ideal example of leveraging.

The gear manufacturing process chain is relatively complex with several options available to the manufacturer at each fabrication stage. It is assumed here that the main process chain would be unchanged and that only gear finishing would need to be altered to produce gears with higher surface finish.

It might be helpful to digress a bit (I enjoy digressing!) to talk about this important manufacturing process that underlies the efficient operation of most machines and transportation. Gear finishing, essentially abrasive machining, is one of those seemingly small and innocuous steps in manufacturing that "gets no respect" (to quote Rodney Dangerfield). In the world of machining with hard tooling (meaning not using lasers or some other type flow process) cutting processes are categorized by the geometry of the tool used and according to whether or not the tool is stationary relative to the workpiece or if the tool rotates. There is a logical division of these processes that, at the highest level, distinguishes between cutting tools that have a “defined geometry” (meaning specific dimensions that determine the shape of the tool) and cutting tools that are not defined (meaning the shape is more random)—having an “undefined geometry.”

Grinding uses undefined geometries - that is, the "tool" or abrasive doing the cutting does not have defined edges at all. Abrasive processes (grinding, sanding, polishing, etc.) use abrasive particles that are natural materials like sand, aluminum oxide, and so on, or materials made to appear as natural shapes. In typical grinding operations several abrasive grains (usually referred to as “grits”) are held together by a bonding material, as they would be in a grinding wheel or for abrasive (sand) paper. The shape of the grains is not defined but “random” depending on how the grain of abrasive was crushed to get the desirable size. If you've every sanded wood or other material or used an emory board on your finger nails you've been abrasively machining.

Importantly, as you observed when standing and noted that small grits gave you a better surface finish (meaning smoother or lower roughness), we control the desired process output by controlling the grain size and the way it moves through the material surface during grinding. Chip formation with an abrasive grain is illustrated in the figure below and shows the grit displacing/removing


workpiece material. The "v" is the velocity of the grit over the work and the arrow shows the relative movement between the grit and the work - the speed and direction. In grinding there would be hundreds of thousands of grits coming into contact with the work and each grit removing a small chip of material.

Thanks to many decades of research on grinding by engineers and academics (like Professor Steve Malkin of University of Massachusetts-Amherst) the relationship between grinding process parameters and material removal and, finally, surface finish is well characterized. We can summarize the gist of this rather simply as follows. We can describe a general empirical relationship that links the achieved average height surface roughness of a grinding process to the process specific volumetric
removal rate and the grinding wheel speed, the v in the above diagram, by assuming a direct correlation between the surface roughness and undeformed chip thickness.

Let me explain.

The undeformed chip thickness is the depth of cut of the grain into the workpiece. Seen in the figure above it would be the difference between the bottom of the grit in the work and the top of the work surface. Specific volumetric removal rate is the volume per unit time of material removed by the process - here dependent on the number of grains moving over the surface at the undeformed depth.  Each grain removes a small volume and the grains pass the surface at a specific rate. Surface roughness is a measure of the variation of the surface of a workpiece at a very fine scale, usually microinches or micrometers. The smaller the variation the smoother the surface. So small is good in the world of surface roughness!

OK … still with me?!

Then we can move on to the leveraging part. The figure below shows the link we are trying to quantify. This figure illustrates the discussion above about removal and surface effects but also shows where




the impact comes in. This figure is motivated by the work of another researcher at the University of Kentucky - Professor I. Jawahir. He studies the connection between process parameters, surface integrity and part function. The trade off between volumetric removal rate and surface roughness must be done understanding that if we adjust the grinding (or finishing) process to create a better surface finish it will cost us something. Here the cost is likely to be time (as finer finishing processes often take longer - remember how much time you need to sand with fine paper as apposed to rough sand paper to achieve a certain surface?). It will also cost us energy - we'll see why next time.

The "leveraging" comes in with the expected fuel savings due to the better efficiency of the gear operation due to the better surface finish. We need to determine if the increased consumption of energy in finishing is paid back in the improvement in the operation of the gear train and accompanying reduction in fuel use.

We'll "do the numbers" next time.

Tuesday, April 3, 2012

Leveraging Manufacturing, Part 1


First, some background

Where did March go?!

We finished a long set of postings on the power of the digital age in the form of software to connect the designer to the process with an eye to achieving all the normal requirements of a product but, in addition, incorporating measures to drive sustainable product design and green manufacturing. There is certainly more that can be said about that.

But, not now.

I'd like to get back to a subject that was mentioned first about one and a half years ago in an earlier posting - leveraging.

This was a follow on to a discussion centering on the "buy to fly" ratio concept used in the aerospace industry and discussed in another posting in November 2010 covering the impact of manufacturing on product performance.

That posting cited some results from VW on the role of manufacturing in the life cycle impacts of a particular VW automobile. It turned out that for a VW Golf example the data showed that there was a 20% manufacturing phase  versus 80% use phase contribution to the life cycle impact of the vehicle. I then did some simple calculations about the effect of some significant savings in one phase of manufacturing due to some "greening" efforts (like using lower energy machine tools, for example) and it turned out that, when this ripples through the production and use phases of the vehicle, we get, at most, single digit improvements in the lifecycle impact.

So, the question was, why bother?!

We rationalized that if you are paying the electricity bill for the factory and this small technology wedge improvement is added to a lot of others in machine operation it can add up to real savings. But, maybe still not impressive compared to the full life cycle of the auto.

But, I reasoned, if we follow that logic we are leaving a lot of potential impact reduction from manufacturing "on the table."

Then I gave the example of something discussed in another prior posting on precision manufacturing about a major German auto manufacturer (but not VW in this case) who has been working to improve the "power density" of some of its diesel engines over the past years and has seen an improvement of almost a factor of 3 in power per unit of displacement. That means, for the same engine size (displacement) they have managed to squeeze three times as much power out. Coupled with advanced fuel injector systems operating at very high pressures (once thought absurd) they see enhanced performance in a small engine - increased fuel economy, improved acceleration (due to reduced mass), and reduced emissions. And this was due to advanced manufacturing.

When we ripple that effect through the life cycle of the vehicle, the impact is enormous. Since most of the life cycle impact (think CO2, for example) is due to operation of the vehicle and the production of the fuel to consume in it, savings due to engine efficiency are highly leveraged.

And, the savings are attributable in large part to manufacturing.

There had been a similar example with respect to improved machining tolerances for airframe structural components in aircraft. Tighter tolerances due to improved machine tool control lead to less weight for the structural components (since we can still meet size/strength/performance requirements without "overbuilding" the component) and that means either more cargo per flight or lower fuel consumption due to decreased aircraft weight. Either one improves the performance of the aircraft. Another example of leveraging.

So, we need to look at this in more detail!

The example I'd like to use to illustrate the fundamentals of leveraging is from a recent paper from our research group at Berkeley. The full citation is “Evaluating the relationship between use phase environmental impacts and manufacturing process precision,” CIRP Annals, 60, 1, 2011, pp. 49-52 and I encourage you to look this up (or contact me and I'll send a copy) for all the details. It is co-authored by two of my research students Moneer Helu and Athulan Vijayaraghavan (now with System Insights).

The example focusses on another aspect of vehicles and transportation - the gear train.

We saw the example for the German auto maker how manufacturing precision can have a strong effect on the operational efficiency of an automotive engine. The operational efficiency of an automobile can be generally measured based on its fuel economy. The fuel economy is strongly influenced by the construction of the powertrain, where tight tolerances and high quality surfaces in the camshaft and crankshaft bearings are required to ensure relatively low losses. Tight tolerances are also required between the piston, piston ring, and cylinder surfaces to enable the use of lower viscosity oils that reduce frictional losses in the engine. In addition to the powertrain, the drivetrain is another component of automobiles that is vital to fuel economy.

Recent work has shown that the efficiency of gear systems is due to a variety of factors including the surface roughness of the mating surfaces, assembly errors (e.g. shaft misalignments), and other manufacturing errors (e.g. form errors). Because the vast majority of environmental impacts of an automobile occur during the use phase as we saw illustrated in the VW example, the impact of increased manufacturing precision through better surface finish on the final drive reduction of an automotive manual transmission drivetrain presents the ideal case study for this investigation.

For a good review of the terms powertrain and drivetrain I suggest any book in automotive engineering or our friend Wikipedia! The term powertrain usually includes the engine, transmission, drive shafts, differentials, and the final drive (drive wheels) but it sometimes refers only to the engine and transmission. Wikipedia sums up the case well-

"Competitiveness drives companies to engineer and produce powertrain systems that over time are more economical to manufacture, higher in product quality and reliability, higher in performance, more fuel efficient, less polluting, and longer in life expectancy. In turn these requirements have led to designs involving higher internal pressures, greater instantaneous forces, and increased complexity of design and mechanical operation. The resulting designs in turn impose significantly more severe requirements on parts shape and dimension; and material surface flatness, waviness, roughness, and porosity."

It's this last bit - about imposing stricter requirements on, among other things, surface features including waviness and roughness - that we are going to focus on here.

But, we'll start that in the next posting. Soon!