Q&A with Bill Gibbs

The friend of the machinist talks about current trends in CAM, an upcoming release, and a new opportunity in the realm of interoperability.

By Ann Mazakas

Bill Gibbs was a manufacturing engineer doing contract programming in the early 1980s. When the Apple Macintosh was first introduced in 1984, it inspired Gibbs to create the first native graphical CAM system. This initial effort has since grown into GibbsCAM, which is developed by Gibbs’s company, Gibbs and Associates of Moorpark, California. Gibbs is president and CEO of the company, continuing his role as an industry leader and innovator. Ever a friend to the machinist, in this interview Gibbs discusses current trends in the CAM industry, an upcoming software release, and a huge new opportunity in the realm of interoperability.

GibbsCAM software is certified to work with current CAD systems like Autodesk Inventor, SolidWorks, and Solid Edge, each with a partnering program. What’s new on that front?

Gibbs: When we talk about the partnering programs that Autodesk, Solid Edge, and SolidWorks offer, I’d also put PTC on the list. PTC is actively supporting interoperability by offering a core library called Granite that’s available for license. Granite is PTC’s own product for reading Pro/Engineer files. This is great because we’re not using reverse engineering to get information from Pro/E; we’re using software produced directly by PTC. By selling the Granite library as an option to our customers, they can read Pro/E files and map them directly into the Parasolid solid modeler we use internally. We see a big demand for stand-alone read capabilities for typical CAD files without buying the actual software. The biggest advance in quality is when the core technology comes from the parent CAD company. For our users, the ability to directly read CATIA, Pro/E, and all the Parasolid derivatives like SolidWorks, Solid Edge, and Unigraphics not only improves quality, it improves utility. Direct read is a big step up in convenience.

In addition to interoperability, people purchasing CAM software place a lot of value on ease of use. How does GibbsCAM respond to that? 

Gibbs: There are different types of ease of use. There is the whole interface ease-of-use aspect that most people focus on: Do I understand what this button does? Does the software do what I expect it to do? Does it follow the Windows guidelines that I’m used to? But there’s a completely different aspect of ease of use: Software is generally easier to use if it has adequate power for the task. When someone buys a system based on superficial qualities like its interface or low cost, they end up wasting hours every day using a tool that’s not designed for the task. At the end of the day they look back and say, “that wasn’t very easy.” What we come back to is a basic question: Can your people program your parts for your machines? If you can solve those issues you’re going to be happy.

If you get sucked into the marketing hype, you end up with products you are not happy with. We know this because replacing other people’s products is a significant part of our business.

Are There Any Hot New Trends in the CAM Industry?

Gibbs: To be honest, the CAM industry is a mature market. There aren’t a lot of frontiers, but there are a lot of opportunities for quality improvements, opportunity to expand the breadth of products and to go beyond standard capability, improving the product based on the quality of the solution provided. Rather than inventing a new mousetrap, we’re focused on improving our capabilities. From my experience, successful companies with very successful products do not stake their fame on a whiz-bang aspect, but base their success on good solid quality, good solid customer service, good solid value, and good solid functionality. That’s more important to our future success as a company than chasing after the latest technology craze.

What Can We Look for in the Next Release of GibbsCAM?

Gibbs: With each release, we focus on developing a collection of functionality with the intent of solving a complete set of machining issues. Our next release, GibbsCAM 2005, is set for release in early summer 2005. This year, we’re focused on cutting parts faster. This is not the same as high-speed machining, although we do address that. What I’m talking about is how fast you can get the part off the machine. When you look at the cost of a job, you’ve got programming time, setup time, and runtime. It’s easy to see that runtime is the biggest cost, which is why we’re focusing on ways to cut the part faster. Reducing programming time is important too, but it doesn’t deliver the same kind of savings.

One of the capabilities we’ve included is a feed and speed optimizer. I want to make it clear that the feed and speed optimizer isn’t limited to specialized machines. Its functionality is designed to work for shops using standard machines, too. What you get is a variety of global editing capabilities and options for speed and feed associativity. If you change your RPM, your feeds will change to match. It also provides the ability to apply a tangential feedrate instructing the machine to keep the cutting point moving at a constant vector velocity, basically, telling the machine to slow down on the inside of a curve and speed up around the outside of a curve. There is also a variety of deceleration options designed primarily for slowing down into corners. You can determine how much you want to slow down, how far away from the corner, and the allowable adjustment to the feedrate. Once you set it up, it runs through the operation, adjusting the feedrates automatically. This gives a better surface finish and cuts the part faster.

We have another related capability called a toolpath optimizer. It looks to modify curvatures and corner radii to produce a toolpath that a machine can follow faster. This is critical in high-speed machining because it puts out a much nicer part. It’s also true for machines you wouldn’t consider as high-speed. There’s less vibration, less chatter, and an overall better quality part—no matter what type of machine you use.

What Are Your Long-range Plans for Development Beyond this Release?

Gibbs: Earlier we talked about interoperability. In addition to CAD, there is CNC interoperability. We’ve identified this level of integration as an important development. In the future we want to let the CNC controller make decisions in a way that’s integrated with the CAM system. The CAM systems today program paths without really knowing enough about the machine tool or controller. So maybe we can hit 75 percent or 80 percent of the optimization advantage, but we can’t hit 99 percent because each model of a machine is different. They each have different performance levels, acceleration curves, harmonics, you name it. All those variations make it a mammoth task for us to reproduce and maintain inside a CAM system. We’re investigating ways where the CAM system dictates the surface finish and tolerance, then communicates that information to the machine in such a way that the controller can make the calculation of the optimal path.

How Close are You to Making CAM/CNC Interoperability a Reality?

Gibbs: Last year we worked on a research project with FANUC. A Gibbs engineer and a FANUC engineer worked together with a customer who needed their parts machined faster. We were able to work up a demonstration of surface-finish/tolerance-parameter modifications in real time on a controller to improve the machine tool performance. We were able to run the part about 45 percent faster without actually programming a different toolpath. CAM interoperability isn’t just the bidirectional communication of a program, it’s also about finding better ways for everyone in manufacturing to work together and find smarter ways to cut parts.

There’s also an opportunity for CAD and CAM systems to help the engineer with the manufacturability of a design. Shallow interoperability is just the bidirectional communication of information. Deep interoperability is where we realize that these discrete operations are not really discrete; that manufacturing affects design and that the machine affects manufacturing. Until we find the optimal ways to share this information, we can do better.

Ann Mazakas is the president of Intelligent Creations LLC. You can reach her regarding this Q&A via e-mail sent to [email protected]. 

Company Highlights

Company: Gibbs and Associates

Headquarters: Moorpark, CA

Founded: 1982

Product: GibbsCAM