Category 8 Cable. Really?

RSS

The need for a higher bandwidth twisted pair cable.

I have always thought that carrying GHz signals over a copper cable just short of a miracle.  With all the resistance, inductance, capacitance (RLC), and other losses you might expect the signal would dissipate to nothing over even the shortest of cables.  But while cable attenuation is great, it is not a knock out factor.  If you can live with a bit of attenuation, copper cables can and do carry GHz signals.  This is especially true of coax cables that deliver RF signals.  You can buy coax good to 40 or 65 GHz if you can live with the 2 to 6 dB/meter attenuation.  You don’t necessarily want to use long runs of it but for short connections between equipment, it is not a problem.  For long runs you need hard line, a large rigid coax cable made of copper and aluminum.  It can be several inches in diameter but the loss at GHz frequencies is minimal even on long runs up a tower to an antenna.

Twisted pair is another whole media altogether.  We already have unshielded twisted pair (UTP) that carries signals up to 500 MHz.  UTP is usually carrying digital data rather than analog or RF signals.  Besides of all the RLC losses there is also the issue of crosstalk between adjacent pairs.  Yet over the years, techniques like equalization, pre-emphasis and de-emphasis and other signal processing techniques have made it possible to maintain signal integrity over many feet.

While most of the world is still living off of Cat 5 and Cat 5e cable, data centers have moved on to available Cat 6A and Cat 7A cables that are good for short distances, usually less than 30 meters.  These are the familiar cables with four 100-ohm twisted pairs and good for up to 500 MHz.  These twisted pair cables are standardized by the Telecommunications Industry Association (TIA) under its TIA-568 category.  Now the TIA is working on the Cat 8 standard which is expected to be finalized by the end of the year.  The 42.7 subcommittee is headed up by Sterling Vaden of the Optical Cable Corporation (OCC), a company that makes both copper and optical cabling systems for data centers.

With data center speed demands increasing almost daily, faster cabling is essential.  While optical cables have been available for years, they are significantly more expensive than copper cables so data centers usually defer to copper if it will do the job.  Cat 6A and Cat 7A cables have almost run their course, thus the push for a Cat 8 version.

Vaden says that it is important for data center managers and suppliers to know what Cat 8 will be like and how it may affect the data center infrastructure.  Overall, Cat 8 cable will contain four shielded twisted pairs and have a diameter about the same as Cat 6a and Cat 7a cables.  The bandwidth is specified to 2 GHz.  RJ-45 connectors will be used.  The goal of the cable is to handle the forthcoming evolution to the IEEE Ethernet standard for 40 Gb/s over copper cables.  At that frequency, the maximum reach is expected to be 30 meters.  I am still not fully buying that but we shall see.

You can expect Vaden’s employer OCC to offer Cat 8 cables as well as the equipment to test them.  For more details contact OCC at www.occfiber.com

It is good to know that copper is not dead.  I am still amazed at how companies have extended the life of copper cables with hard core research and development.  But can there possibly be a Cat 9 in the future?

Discuss this Blog Entry 3

on Jul 6, 2013

Lou Frenzel,

Nice information. I think copper over insulators or nano tubular copper may some time bump up to have low capacitance and low attenuation for high frequencies. As long as one can catch signal and signal integrity is maintained, then attenuation is not a serious problem. In fact signal cable impedance matching and cable termination cause significant attenuation.

Good cable manufacturers are disappearing. People change business overnight. This has done great harm to Physics, Chemistry and Engineering of highly specialized manufacturing. You suddenly find some cheap manufacturer and with useless material printed with standard numbers. I have to hunt several manufacturers and then test their cables to find only one or two good manufacturer and then by next year these companies stopped production so I have to start from square one again.

There are now crazy company buyers or finally they become company product killers. Business is more like Quantum Physics now and you see tunneling effect in business so very often.

Now this not to last over an year type manufacturing is also taking over. earlier I used to plan for years and now I am counting in hours.

I am looking at the evolution of technology, dirty business, and lots of uncertainties. There is this generation gap and perhaps a new way of looking at technologies.

on Dec 16, 2013

i agree with you Lou Frenzel but there are still very good and decent cable manufacturers available who don't cheat and don't change business overnight

on Jul 17, 2013

I always think of optical fiber as the revolution that never quite happened. I recall predictions (by zealous industry types) back in the early 1970s that copper would become obsolete within 10 years! I'm not holding my breath ... The worst aspect of pushing GHz signals down very lossy (often 40 dB or more) twisted pairs is that, while it works when the electrical environment is very benign but, like wireless, is robust and reliable only if you have the luxury of retry after retry. But today's young folks seem to accept that as just matter of fact ... we're all being dumbed down for sure, almost as if to say "That's not a defect, it's a feature!"

Newsletter Signup

Please or Register to post comments.

What's Communiqué?

Blogs on topics such as wired and wireless networking.

Contributors

Lou Frenzel

Lou Frenzel is the Communications Technology Editor for Electronic Design Magazine where he writes articles, columns, blogs, technology reports, and online material on the wireless, communications...
Blog Archive

Sponsored Introduction Continue on to (or wait seconds) ×