Skip to main content

OM4+: A case of the emperor's new clothes?

Over the last year, you may have heard proponents espouse the benefits of OM4+ multimode fibre. For instance, they claim it is better for single-wavelength applications, and it’s comparable to OM5 for multiwavelength application support. While OM4+ may have some useful attributes, which vary from company to company, that variation is a clue to the fact that in all cases it is proprietary, meaning it is not standardised. Essentially, the public is being asked to take OM4+ proponents at their word. As New York Yankees legend Yogi Berra once said, ‘It’s like déjà vu all over again’. Here’s why.

In 2000, the ANSI/TIA/EIA-568-B.2-1 standard defining Category 6 network cabling was still in development when talk of a ‘new and improved’ version, Cat 6e, began to emerge. The performance claims from manufacturers had a familiar ring: ‘Category 6e performs better than Category 6 specifications at frequencies well above the 250-MHz limit, so it must be better. That’s why we’re calling it 6e (enhanced).”

However, at the time there was no standard defining or referencing Cat 6e. This prompted Cabling Installation & Maintenance to caution network operators: ‘…please know, for your own purposes in choosing products for your upcoming projects, that the TIA currently has no plans to pursue anything that it will call Category 6e’. How prophetic those words proved, because even now, 18 years later, no standard has emerged.

Shortly after the article was published, the ANSI/TIA/EIA-568-B.2-1 standard for Cat 6 was adopted and published. Yet, Cat 6e claims persisted. The CompTIA A+ Complete Review Guide, published in December 2015, made a point to remind readers: ‘While not an official standard, many vendors now offer a Cat 6e cable and market them as providing better performance, which may be true, but whatever improvement there is will vary from vendor to vendor, since there is no official standard.’

This is exactly where we are today with OM4+ fibre. While OM4 and OM5 are widely referenced in Ethernet and Fibre Channel applications, there are no application standards that reference OM4+. How could there be? To be referenced in an application standard, an optical fibre standard must first exist. That’s because application standards bodies do not define optical fibre and cables; rather, they reference the work done by the appropriate standards bodies, in this case TIA TR-42 and IEC 86A.

While testing has been performed to support OM4+ claims, it has been carried out privately. Further clouding the issue, in stating the bandwidth available for each wavelength supported, some OM4+ proponents have starting using the non-standard metric ‘effective bandwidth’ (EB) instead of the standard ‘effective modal bandwidth’ (EMB). These are, in fact, two quite different metrics that are not directly comparable. EB is a combination of modal bandwidth and chromatic bandwidth, based on assumptions about the spectral characteristics of laser source which may or may not be true. In contrast, EMB is a metric defined in standards for OM3, OM4 and OM5 fibres for which standardised test procedures exist.

In other cases, OM4+ specifications claim EMB ranging from 5000 to 5640 MHz·km (at 850nm). While this is indeed higher than the accepted 4700 MHz·km minimum (at 850nm) for OM5, studies show that values above 4700 MHz·km result in small and diminishing improvement in application performance. Thus, even though claims of higher EMB can be made, there appears to be little benefit.

This is where industry standards organisations play such an important role. Their work is highly methodical because, in many ways, they are charged with plotting the course for the industry. For any new technology to be incorporated into the standards, it must prove that it provides significant and meaningful advantages over alternatives. The validity of the proposal is judged by a collection of companies (in TIA) or national committees (in IEC) that constitute the standards body. Standards are developed using a process of iterative refinement until the result satisfies the committee. Only then does it become a standard.

Such was the path OM5 took. In June 2016, OM5 was standardised in TIA 492AAAE as the first multimode fibre specified for wavelength multiplexing (i.e. wideband) applications. The path to publication consisted of six face-to-face meetings, thirteen teleconferences, dozens of contributions by many companies, and three ballots over twenty months. Since then, it has been incorporated in emerging standards for Ethernet at 50Gb/s, 100Gb/s, 200Gb/s and 400Gb/s, as well as 64G Fibre Channel.

The meaningful difference between OM4+ and OM5 – or most proprietary versus standardised technologies for that matter – is OM5’s recognition within application standards and its industry-vetted performance requirements. So, when it comes time for your next cabling project, which option will you choose – one making proprietary claims, or one following proven methodologies that allow it to be referenced by the applications you want to deploy?

Paul Kolesar is an Engineering Fellow in the Enterprise Solutions division of CommScope in Richardson, TX

Media Partners