Quote of the Day
Great leaders are almost always great simplifiers, who cut through argument, debate and doubt to offer a solution everybody can understand and remember.
— Michael Korda
Introduction
I have been talking to contractors who install Fiber-To-The-Home (FTTH) products and they frequently have more questions about the electrical cabling than the fiber cabling. Today, I had an installer ask me a good question about coaxial cable (example shown in Figure 1). His question was:
How much does the length of coaxial cable change with temperature?
This is an interesting question because I have seen deployments run into reliability issues because the length changes were not accounted for when the cable was deployed. The solution is to include expansion loops when using coaxial cables. I consider these loops similar to what others call service loops.
The answer depends on how you define the length of the cable. The discussion was interesting and I thought worth documenting here. It will be similar in tone to this post, which discusses the thermal expansion of PVC. I have also discussed how the length of fiber optic cable varies with temperature in this post.
Background
My focus in this post will be on deriving a useful rule of thumb on the amount of length change an outdoor-mounted, unburied, coaxial cable will experience based on air temperature changes. Buried coaxial cable experiences minimal length changes because the temperature variation of the soil is less than that in air.
Coaxial Cable Construction
Figure 2 shows a cutaway drawing of a typical coaxial cable. There are many kinds of coaxial cable. For this post, I will focus on RG-6 cable because it is the most deployed cable in modern residential deployments.
As shown in Figure 2, coaxial cable is composed of the following parts:
- center core
This a round, 18 AWG, copper wire.
- delectric insulator
This is normally a plastic called PTFE, one formulation of which is commonly known as Teflon. As we all know, Teflon(TM) does not stick to much. Since copper and the PTFE have different coefficients of thermal expansion (CTE), this means the wire moves relative to the dielectric.
- metallic shield
This is usually an aluminum braid. Aluminum also has a different coefficient of thermal expansion than PTFE or copper. In general, this material does not cause movement issue because the braid is fabric-like and, therefore, flexible.
- plastic jacket
This can be composed of any number of plastic materials. It generally is very thin and flexible, which means that its movement does not really cause any issues.
My discussion will focus on the expansion of the copper and dielectric materials because they are quite rigid and tend to push and pull on whatever they are mounted to.
Why do we care how much the length changes?
There good and bad aspects to the fact that the length of coaxial cable changes with temperature. Here is a brief discussion.
Good Aspects
The core copper conductor in coaxial cable will move independently of the rest of the cable. This relative movement can cause the contact area between the wire and its termination to rub, which can be useful in keeping the electrical contact area clean. Here is an excerpt from this reference that discusses the movement.
The movement of the center conductor occurs when the cable heats up, and cools down with normal temperature changes. The coefficient of expansion is different for the center conductor verses the shield of the cable and that difference keeps the center conductor moving back and forth slowly to keep the contact area clean.
Bad Aspects
The fact that the length changes means that the installers must account for this movement. Because RG-6 can be quite stiff (especially the quad-shield version), the cable may push and pull hard on the electronics it is connected to. This movement can cause the electronics to flex, which may result in reliability problems over time. A good installation will ensure that cable movement will not put stress on the electronics it is connected to.
Analysis
My focus in this post is to create rule of thumb for installers that will help them understand the amount of cable movement they can expect to see. Since the copper and dielectric in coaxial cable move independently, it is important to understand how each material's length changes with respect to temperature.
Coefficients of Thermal Expansion
Table 1 shows typical expansion coefficients for materials that comprise a coaxial cable. Since these materials are not rigidly attached to one another, they can move independently.
Material | Value | Units | |
---|---|---|---|
Copper | 16.5 | μm/ (m⋅K) | |
PTFE | 135.0 | μm/ (m⋅K) | |
Aluminum | 23.1 | μm/ (m⋅K) |
While I list the CTE for aluminum in Table 1, it really does not come into play for installation issues. The aluminum is woven in the form of a braid and it is flexible. So my discussion below will focus on the copper and PTFE CTEs.
Expansion Rules of Thumb
Figure 3 summarizes my thermal expansion calculations. Based on these calculations I can create the following rules of thumb:
- The copper core can move ~0.25 inches for every 25 feet of RG-6 cable during a typical year (i.e. 100 °F temperature)
- The dielectric can move ~2.5 inches for every 25 feet of RG-6 cable during a typical year (i.e. 100 °F temperature).
Conclusion
Engineers often have to prepare installation guidelines for the folks who install the gear they design. It is always important to make sure the installers understand why they are doing things. This will give you a much greater likelihood of having your products installed correctly.
That is very interesting, You are a very skilled blogger.
I've joined your rss feed and look ahead to searching for more of your fantastic post.
Also, I have shared your site in my social networks
Sorry, but this analysis fails to take in to account the real world...namely friction between the inner, conductor, dielectric, shield and cover. These cause extension and compression of the components due to force, not just temperature. So the components don't expand and contract as calculated above without this consideration. We all wish the world were that simple!
I agree with you. I have seen examples of calculating the movement of composite materials, but the methods are so complex that it is better to just measure it. My objective here was to provide some insight for installers on the movement that they are seeing in coaxial cable. I have seen this movement cause damage to the electronics it was connected to. I was working to encourage them to allow some slack on long cable runs.
Thanks for commenting.
mark