Cable for GigE Vision -- Can Traditional Category Cable Do

GigE Vision™ Interface Technology

The new GigE Vision™ standard being finalized by an AIA committee provides a unique solution that optimizes system performance while maintaining cost levels that are recognized as being more attractive than USB 2.0, 1394b, and Camera Link. With full resolution camera frame rates up to 200 fps, GigE Vision not only breaks the bandwidth barrier of Firewire and approaches the performance of Camera Link, it does it at a lower cost than either LVDS or Camera Link by using field installable cabling and connectors that are available virtually everywhere.

Are There Limitations to Category Cable?

Depending on the application and installation length, GigE Vision will work with Cat5, Cat5e, Cat6, or Cat6a cable. That's the positive side of the GigE Vision standard. The potential negative is that these cables, which are mostly produced with solid copper wire using plenum or riser type construction, were not designed to be used in a factory, or any environment, with apparatus that is in motion. So the question becomes: when GigE Vision is applied in installations for industries such as food processing, medical imaging, pharmaceuticals, security (PTZ), transportation, and machine vision and control, will the intrinsically stiff and designed-for-static-application category cables meet the requirements or fall short?

What Is the Ideal Cable Design?

Numerous interviews with camera manufacturers and probable users of GigE Vision have helped to define some of the ideal GigE Vision cable attributes:

Meet signal performance and emission requirements of GigE Vision at lengths up to 100 meters.
Have an OD of 6 millimeters or less.
Interface successfully with the Honda GigE Vision connector and standard RJ45 connectors.
Have an industrial flex life of at least 1 million cycles.
Be soft with little or no 'memory.'
Have black matte jacketing that complies with RoHS requirements and is suitable for traditional factory environments.
Incorporate specialty materials capable of withstanding the abuse of moving factory equipment and robots.
Similar cost to Cat5, Cat5e, Cat6, and Cat6a cabling.

Can One Type of Cable Meet All Needs?

Questions arise regarding the signal, emissions, and physical attributes of the 'ideal' cable. When the topic of a 'silver bullet' or a cable as described previously is argued, the engineering response will undoubtedly be "No one cable can be all things."

Even an informal assessment of the potential future of GigE Vision indicates that traditional category cable will be inadequate when endeavoring to meet the often demanding mechanical requirements of real-world applications in the factory, medical facility, or food processing plant. If one is to presume that all applications incorporating GigE Vision are not going to be static, then it would seem that the potential future success of the specification depends in significant part on the ability of the cable, or cables, to support the standard as it evolves.

So the question then becomes, "If not traditional solid wire category cable, then what?" The answer can be found in the requirements of each specific application. For example, if a task requires mechanical parameters similar to ITC/PLTC exposed run-rated with an oil and weld-slag resistant jacket and the available category cables didn't meet this requirement, should the project be abandoned for GigE Vision because of the cable limitations? If equipment in motion operated by either robotics or humans is the ideal target for GigE Vision, should the mechanical limitations of category cable cause the rejection of the GigE Vision concept so that a standard with known high flex cables such as Camera Link or 1394 can be incorporated? If there is an application in medical imaging that requires soft or crush-resistant cables, should it be ignored because the standard category cables can't meet the requirements? In the long run, in order for GigE Vision to mature, the answer to each must be, no!

Certainly there will be applications that are ideally suited to conventional category cable and traditional RJ45 connectors. But as the standard continues to evolve there will be abundant opportunities to service requirements in which higher demands are placed on the cable and connector. This is precisely this reason that Honda has designed and put into production an industrial Gigabit Ethernet RJ45 interface connector with thumbscrew locking -- the premise behind this design being that the traditional RJ45 connector was not going to be robust enough for a number of potentially demanding applications for GigE Vision. Using this same theory, a more adaptable cable or even a variety of cables will be required to fully meet the needs of this emerging market.

Some currently defined areas of non-traditional cable needs are:

High-Flex of 1 million cycles plus.
Continuous Flex with constant motion requiring at least 10 million cycles.
Cut resistant jacketing.
Oil and chemical resistant jacketing.
Underwater cabling.
High and/or low temperature cable.
Crush resistant cable construction.
Outdoor weather resistant cable.
Small O.D. cable.

Is 100 Meters of 'High-Flex' Cable Possible?

There are a number of important factors to consider when designing a high-flex cable product. As any cable design engineer will tell you, it is certainly as much of an art as it is a science. A precise mix of costly high tensile strength alloys, smaller component and cable O.D., durability, toughness, and high flexibility are required to produce an extraordinary cable. The basic properties of flex cable design might form a highly desirable product from a mechanical standpoint but conversely these same desirable characteristics could potentially limit the signal performance of the final creation.

In conductor design, the proper stranding and alloy selection combined with the proper torsional twist will produce an effective end result but might limit the performance of the finished product when compared with cable produced with solid wire. Conductor insulation type and thickness also play a significant role in the development of high-flex cable but could also create latent issues with signal performance. Even the shielding and cable jacketing that is ideal for flex cable may not offer the optimal performance parameters found in less flexible cable designs.

What's the Solution?

The answer would appear to be the incorporation of a number of cable types to the GigE Vision standard to support all potential installation requirements.

Traditional category cable is perhaps best for the longest runs, in static applications, and where field installation and lowest cost are of higher importance than other and often more expensive mechanical attributes.
High-Flex and Continuous-Flex cabling, due to the type of materials and construction methods employed, may not function in all applications at 100 meters, but will have usage in shorter runs where motion historically takes place.
The majority of applications requiring 'flex' cable are estimated to be less than 10 meters.
Custom cable for specific markets can be built in small runs as needed.
Soft, flexible, and small O.D. cable construction could adversely influence length but most requirements for this type are expected to be shorter runs.