Skip to main content

Routing switchers

It might be hard to get too excited about routing switchers if it weren't for the fact that they often are at the core of interconnection for entire broadcast plants. Routing switchers evolved from good old-fashioned patch panels, probably from telco operations centers just as much as broadcast facilities. However, they provide much more than just the connection to the veins that run to all parts of the facility.

Routing switchers have become the hub of a facility because of the need to build complex and reliable facilities that can evolve over time as the mission of the facility changes. For instance, the types of signals handled in the '90s were often analog, but as the decade came to a close, the signals became increasingly digital. This presented a challenge to the cost of ownership, as a digital level would add significantly to the cost of an installation.

Good engineering has, however, found ways to allow routing systems to do double-duty, with internal conversion both from analog to digital and digital to analog. Implementing such a hybrid system for audio allows a migration from an all-analog audio facility to one that might be largely AES-based long into the future. At the same time, by putting conversion capability into an audio router, the need for stand-alone conversion products is minimized. This reduces the costs of the capital asset, installation and design. Less is done with wires and more is done with the keyboard on the control system, which ought to be a good thing these days.

Several manufacturers have introduced products of all scales that allow this capability — with block sizes as small as 16×16. Small block sizes allow migration plans to be funded efficiently, with expansion of the digital layer as needed. The downside is that as analog crosspoints become less of a necessity, and they surely will, the hardware becomes redundant, and the value of the investment is sometimes reduced.

Embedding conversion inside a routing system is not unique to audio, though with video, it is more difficult to do high-quality conversion inexpensively and in the limited space of a routing system's I/O. At the least, the current crop of converting video routers offer acceptable quality for most purposes, though stand-alone products can do a better job. That presupposes that the ultimate level of quality is important, which in many cases it is not.

Consider that much of the analog video in a plant is rapidly becoming the legacy signal and often is of lower quality. The other use of conversion — digital to analog — allows a router to provide feeds to analog monitors without the expense of external conversion, or at least at a lower cost per port.

When considering digital video routers, in addition to conversion options, it is important to think through the need to have wide bandwidth capability. Though perhaps not a universally valid assumption, it's fair to say that for at least broadcast and production plants, HDTV will replace SD applications over the next decade.

But that doesn't mean a routing switcher needs to be full of wide bandwidth crosspoints and I/O. For example, a 16×16 block of crosspoints might encompass two I/O modules and one crosspoint, which must be capable of 1.485Gb/s. But if you embed the same 16×16 block of signals in a 256×256 router, the number of I/O modules goes up to 32, and the number of crosspoints jumps from 232 to 65,536.

Unfortunately, each crosspoint costs money, and scaling up has a penalty. As with most things, size does matter, and you could be paying for a lot of unused capability. It is entirely valid to pick a total size, which allows for modest short-term growth, a frame size that will accommodate long-term growth projections and a WB count that doesn't break the bank. You can assume that manufacturers charge more for HDTV-capable signal processing than SDTV devices. Over time, the price differential has come down dramatically, and there will be a point in the future when it is more costly to support both SD and HD modules, with SD likely to be the more costly option. Eventually, that will be the case from a manufacturing standpoint.

While signal formats are important, so is the control system. From the standpoint of the operators, as well as the staff who maintains the control system, the panel interfaces and the method of programming the features of the router are critically important. It would be less than useful if the programming took skills beyond those commonly available on staff. As routing control has acquired more features, the programming of the system has become noticeably more complex. Engineers installing new systems today would be well-advised to train on the full features of the system.

Reading a book in the heat of battle when programming needs to be done might not endear the maintenance technician to the news director. Many systems are now interconnected with garden-variety Ethernet carrying TCP/IP protocol control signals.

Installers should be fully familiar with networking. The SBE Broadcast Networking Technologist Certification or other professional training can be a valuable asset. Systems that use other control schemes can be just as valuable, but there is a synergy that can be gained when networking assets are fully used, including connections to automation and facilities management software.

One thing to keep in mind: Insist on a control system that allows the programming to be saved to a file as a protection against failure of a controller. There is no substitute for being able to upload a configuration saved to some non-volatile medium. Routing can be extremely reliable. Keep in mind that the life of the router might be longer than the employees trained to program and maintain it. Recurring training for new staff is good insurance.

Investigate your options carefully. Brand does matter, but less than most salespeople might tell you. Find a company that supports the features you need, can accommodate reasonable growth without breaking the bank and will be in business long enough for the warranty to expire, which can be 10 years these days. Focus on the company's ability to provide backup in the event of card failures or software bugs. You can bet both will happen at the most inopportune time, such as during election night coverage.

John Luff is senior vice president of business development for AZCAR.