Proof of Equivalence
Tube equivalence is established fact. Why is this so?
The chemical and material behavior, plus all mechanical and even electrical properties are permanently built into the alloy when it is originally produced at the mill. After the alloy is created --- the exact alloy carefully specified in ASTM D3241 --- no shaping, grinding, milling, or polishing operation has any chance of effecting the quality or performance of the alloy. We not only employ the same alloy as every competitor; it also comes from the same mill. Non-equivalence is simply a false perception..
Two corollaries arise from this truth. First, since non-equivalence itself is false, it isn't possible to demonstrate non-equivalence using statistics (or, for that matter, by any other method). Secondly, since equivalence is an experimental given established at the mill (rather than a question), the idea of “proving equivalence” is a fools errand. Particularly is this so when tests for equivalence employ an JFTOT rig designed to evaluate fuels --- never intended for adaption to inert tubes.
By extension, the idea of a "tube bias" is also non-existent. Again, equivalence is built into the material at the mill. Also, equivalence was clearly indicated in a tube polishing study, and by every measure – VTR, ETR, and ITR. If a bias could exist, how could this result have been attained?)
JFTOT specifications not only set a unique chemical composition. The heat treatment specified "bakes in" or "sets" a particular, highly desirable, mechanical quality. Nothing a supplier can do changes the essential fact of equivalence.
A review of JFTOT history illustrates how false ideas about equivalence originally developed and were subsequently believed. The idea of "equivalence testing" or "tube performance testing" only evolved around the turn of this century when Standard Heater Tube, Incorporated emerged as a new supplier. Then ASTM created a Research Report RR:D02-1550 or "Tube Protocol;" a procedure for establishing tube equivalence, in 2003. Testing at that time involved pass or fail "specification testing" to determine the fitness for service of fuel batches, and test results were determined visually.
This empirical or visual nature of stability testing made for a subjective test. It is easy to see how a belief arose and was perpetuated that the test, itself, might somehow be influenced by the appearance or "look" of heater tubes. Indeed, that false belief did prevail, and tubes were evaluated visually, subjectively, for a very long time.
In this primitive circumstance, Subcommittee J adopted a false and arbitrary requirement for “visual equivalence.” A second blunder followed this first one. With no possible way to employ an objective measure, authorities resorted to an effort to to “prove” visual equivalence using statistics.
That is how researchers erroneously came to employ "statistics" to evaluate "equivalence,” and formalized the two false requirements in RR:D02-1550.
Fast forward to today, where visual ratings are rapidly giving way to evaluation procedures involving more direct determination of deposit thickness and volume. These current-day evaluation techniques improve test objectivity. Yet, the unfortunatle false notion that "statistics" can be appropriately applied to "tube performance testing" somehow persists. Factually, "tube performance" does not and even cannot exist for a very simple reason: tubes cannot perform. "Tube performance" isn’t an observable phenomena or a measurable parameter. So, “tube performance” can never be proved using statistics, or by any other method.
Both the original JFTOT methodology as well as newer (metrology) testing are designed to evaluate fuel performance, not tube performance.
For reasons given, tube equivalence always exists and is uniformly reflected in all test results. Since this is factual and it is the scientific truth, any and all differences observed in testing differing tubes must currently reflect problems with the test and not tube issues. After all, the existing test lacks any kind of statistical precision.
Further, the exterior of tubes is solid aluminum alloy. Being solid, the alloy is impervious, lacking porosity and with nil internal surface area. Thus, fuel molecules or oxygen atoms cannot penetrate the surface, and the solid alloy presents only a tiny available external area for fuel-associated deterioration products to form. Also since the solid is impervious, insufficient residence time or contact time prohibit tubes from "performing" or for the alloy to ever participate catalytically. This is a second guarantee of the equivalence of inert tubes. Finally, assume that somehow, magically, tubes instantly became highly reactive. We would still need for one tube to be more reactive than another for effects to be observed in a smallish testing difference of, say, fifteen degrees Celsius (with ten degrees Celsius corresponding to an approximate doubling of reaction rate). That presumption defies reason. As Mark Twain is reported to have said: there are three kinds of lies, namely (1) lies, (2) damned lies, and (3) statistics. By contrast, the essential truth about heater tubes and about tube testing is now available, and it doesn't require statistics.
Despite all available evidence, a “tube bias” is thus sometimes falsely assumed, presumed, perceived, argued, or pretended to exist. However, no tube supplier can actually produce a non-equivalent tube; must ultimately be accepted.
You can see evidence that tube finishing treatment has no effect on testing (following alloy formation specified) from the polishing study here. Additional documentation of testing using different suppliers' products – showing the truth of tubes -- is here.
“Tube bias” is mythical. Equivalence is proven.