Knomentous Customer Login Cable Tracking System

40-Year Life

August 3, 2010

40-Year Life


What tests have you performed to verify the 40 years of additional life, thus being able to extend a 40 year warranty?

I have a similar question for the global warming crowd. Why when weather is so difficult to predict ten days out do some have such confidence in computer models that presume to predict the weather decades into the future. Those guys should focus on next weekend, because I want to know if it is going to rain on my picnic. Predicting the future performance of cables is much easier than predicting long-term weather trends, and much less controversial too. First the common threads—both predictions utilize finite volume modeling, a very computationally intensive task. Second, many initial conditions can be established with decent engineering precision.

Now the five biggest differences: (1) Nobody has a complete model of the weather, and you can’t measure what you don’t understand. A patented model (See U.S. Patent 7,643,977) of cable rejuvenation is available and has been vetted against publically available data. (Editor’s note: I know this is what the global warming prognosticators said too, but they did not get a patent on their model. There is a legal requirement to disclose the best mode … in other words, you can’t fudge the data with the USPTO.) (2) The ultimate value of any weather parameter at very large times is unknown (indeterminate) and therefore can only be known by model extrapolation or by waiting for a very long time. Neither is a satisfying prospect. The condition at very large time for rejuvenation is unambiguous—the concentration of the treatment fluids will be zero and the impact on reliability will be zero. (3) Weather is perturbed in unpredictable ways by cosmic radiation, sunspot activity, changes in the earth’s magnetic fields, and others. The only significant perturbation to the rejuvenation model is the fluctuating operating temperature of the cable. Historical weather patterns, historical load patterns, and load growth can generally be estimated with engineering accuracy. (4) The size of a finite volume used by the weather guys is a cube on the order of 83 km3 and the Δt (time) is measured in fractional hours. In contrast, the volume of a single finite volume in our work is 1018 times smaller and the Δt is often as small as six seconds! Smaller time slices, much much smaller finite volumes yields more accurate results. (5) The weather is a three-dimensional problem; a cable collapses to a two-dimensional problem.

There are two papers published in 2005 and available in the Novinium Library, which describe the challenges of extrapolating cable life. They are available at:

Accelerated Aging of Rejuvenated Cables-Part I

Accelerated Aging of Rejuvenated Cables-Part II

As mentioned previously U.S. Patent 7,643,977 published in January 2010 lays out the details of how the method introduced in the second paper is implemented. We call the ‘977 model “MFlux,” which is short for “Mass Flux.” A friend of mine, Dr. Dan Scott, a Stanford trained Operation Research mathematician, continues to advance the implementation of MFlux at Novinium. The model has demonstrated that it can predict actual performance within about a plus-or-minus 10% error band. The key paradigm shift required to understand how such extrapolations can be made requires you to abandon traditional accelerated life studies, which are loaded with inherent compromises between the effects of different accelerating variables. For example, the permeation of fluid along the radius of a cable is profoundly impacted by the temperature. In a traditional experiment it is difficult to balance the accelerating effects of temperature and voltage. If one variable is accelerated by 10X and another variable is accelerated 100X, the results are not likely to be easily extrapolated. MFlux solves that problem. Finally a direct answer to the first part of your question, “What tests have you performed?” We have done hundreds, perhaps thousands of tests of six primary types:

  1. We measure the permeation properties of individual components.
  2. We measure the permeation properties of component pairs.
  3. We measure the reaction kinetics of our reactive species.
  4. We measure the permeation and reaction results in one-sixth scale model cables and full size cables.
  5. We measure the dielectric impact of fluid components on cable polymers.
  6. We measure the dielectric properties of fluids on full size cables.

Most of this work is carried out at the Michigan laboratory of Dr. David Busby. The work on full size cables is done only to validate the results of the MFlux model. Once the permeation properties and reaction kinetics are determined for a fluid formulation we can perform virtual experiments without any compromises to accelerating factors. We use the e-field environment, geometry, and temperature/loading environment of the actual cable in question. No acceleration is required—time is virtual. The time it takes to perform such a simulation is limited only by the speed of the computer and the efficiency of Dr. Scott’s MFlux algorthim.

Now for the second part of your question, “What tests have you performed to verify the 40 years of additional life, thus being able to extend a 40-year warranty?” In every actual and virtual experiment that we have ever run, Cablecure 732/733 [Ultrinium] fluid technology outperforms the technology that has been in use for over 20 years by at least a factor of 2 and usually more like a factor of 3. Now the old technology is good, after all it was invented by Novinium founders. In fact it typically provides about 20 years of additional life in non-demanding applications. Being twice or three times better than good provides us the confidence to offer a 40-year warranty.