Blewdy Yaink, glad to hear you have first hand knowledge. Can you fill us in on the number of fiberglass motorcycle tank failures that led to the ban on the tanks? Was this a UK ban, US, or both? Any ideas why there was never a recall, if the tanks were so dangerous?
OK, two different threads here.
One as it was then ...
There was not a statutory requirement for recalls in the UK at the time. There was a safety office within Ministry of Transport and the procedure was that an authorized officer in the various police authorities within the UK would place notices, reports or recommendations with the safety officials; there was no fixed format for these notices and they could be in any of a number forms or as reports, questions, suggestions, etc. When I began going to the MoT meetings, the safety office had received a number of notices from around the country (I was not present for the first of these meetings, they were in process when I came along). Some of these notices were in the form of "Is anyone else seeing an increase in motorcycle fires", others were more dry accident reports (some of which were disturbing and graphic), others were "would it be proper for MoT to conduct technical testing of glassfibre tanks" but the authorities were taking it as an issue to be investigated. If the MoT safety management decided that there was a useful measure that could be implemented and needed to be done, they could either amend the MoT regulations or, for more important matters, refer the issues to Parliament for Parliamentary committees to recommend changes to be legislation passed (either new or as amendments to existing law) in Parliament.
A few of the early questions -- How often is fire involved in motorcycle crashes? If there is fire, how much does risk to persons or property increase? What's the relative contribution of different fire sources in motorcycle crashes -- tank structure, ancilliaries (filler caps, valves, and pipes), methods of fixing the tanks, leaks from carburetors, etc.? Are there dynamics involved such as acceleration/decelleration "G-loads" in crashes? Are different types of tanks showing different characteristics in crashes involving fire?
(There was an additional factor from other countries in Europe that we considering similar concerns. i.e. if the UK decided not to regulate fuel tanks but other markets did, how would we handle that?)
But there was a clear mood within Mot that "something must be done" and "something must be seen to be done", so there was some pressure to get on with "something". Technical experts were asked to take a general look into differences in crashworthiness -- if the tank was hit by a shart pointed object, what was the difference between metal and glassfibre? (Answer, little difference, glassfibre and metal tanks could be compromised). If the tank was hit by a large object more rounded or flat - or by hitting the road surface after an accident, what was the difference? (Answer, not much except for the crushing factor). How did the tanks differ in respect to crushing? (Answer - this seemed to be a factor against glassfibre; info from testing by one of the Norton suppliers showed that while the glassfibre tanks seemed to hold up well against breaking open [or at least as well as metal], they appeared more likely to collapse and allow fuel to be pressurized. John Nelson, Norton International Service Manager described it as "a glassfibre tank being crushed looks like a bagpipe being squeezed, with fuel being ejected from any available opening".) Was there a difference in resistance to abrasive damage? (Answer, glassfibre tanks show surface or visual damage more readily than metal but they tend to have thicker surfaces so they might be less likely to be compromised by abrasion damage resulting in fuel spills.) Are there other factors? (Answer: Police authorities had reported that if tanks were physically torn from their fixings, then dramatically large fires seemed to result. No data on how frequent this was, if glassfibre tanks were more likely to be detached, or if there were fire-behavior differences between the types of tanks.) Were any difference noted in the likelihood of one type of tank being more likely to cause sparks or other ignition sources in a crash? (Answer, that's one in favour of glassfibre tanks.) What about "dynamics"? (Answer, there seemed to have been a pattern of tanks being struck by riders if the motorcycle is in a frontal collision i.e. the riders pelvis slams into the rear of the tank putting loads onto the tank. Similarly to above, there's no data on whether this was a larger factor between the types of tanks.)
The situation was summed up by MoT as that they were under the pressure of some publicity to "do something" although the data seemed to be mixed. It was acknowledged that there might be some part of the factor that a few glassfibre tanks were more involved in dramatic fires, possible made worse by the exceptional cases where a tank is completely torn open with large fuel loss or the "bagpipe" effect where large quantities of fuel can be expressed in a short time.
Our tank suppliers were curiously ambiguous. The Roadster tanks were beginning to need renewed tooling (for glassfibre production) anyway and prices for plastic and other petroleum products were going through the roof ("international oil embargo" time for readers as old as I am) so they were being squeezed with their prices needing to go up to the level of metal tanks. Since there was need for new tooling, that cost could be carried over to metal tanks and the resulting financial burden to Norton (at least to Norton, I'm not sure about other manufacturers) was minimal.
So the decision was made that the motorcycle manufacturers would not object to a phase-in of a new requirement for metal tanks. Some important questions about tank safety were raised but few solid answers were obtained. But, at that point, the decision was settled.
(My own personal take -- there was not a lot of overall difference between metal and glassfibre tanks at the time, except maybe if you were in a Really BIG One involving big inertial loads or vectors and you were relying on the tank to retain fuel and prevent or reduce the danger of fire, then a metal tank *might* give you a margin of protection.) That's where we were, then.
Two - as it is now.
The glassfibre tanks that were built into Nortons in the early 1970s were made of epoxy plastic strengthened with some kind of flexible glass fibres. The basic epoxy was strong and resisted damage but when it was put under load or pressure, it tended to crack or split in a dramatic fashion. A tank's full strength as a molded part was in the interface between the plastic and the glass fibres.
If you've lived with 70s plastics, you can understand when I say that the stuff is almost organic. Like wood, it has a grain and variations between the components that make it up; plus epoxies from the time were brittle when overstressed; the strength of the final molded item was mainly in the interface between the glass material and the epoxy. If you will think of an organic material like wood, you'll understand that it's characteristics depend on its environment. Wood that is kept in dry and stable conditions will last a long time, but if its subjected to stress, it will deteriorate and change in ways that reduce its strength. Besides the fact that the plastic materials from that time will continue to harden and stiffen as they age, if there's thermal stress or interactions with chemicals, the plastic tends to lose its contact with the glass reinforcement material. Once this happens, there's more space for contaminants to reach the "grain" in the molded structure and further attack the strength of the entire tank.
What might have been a component with strength and resistance to fatigue 50 years ago is likely to have compromises in its strength today. At worst, it might have areas where fuel can enter the structure of the plastic composite and either deteriorate composite further or find its way through the structure and leak. Also, there's the issue of ethanol in fuel that is known to dissolve the chemical bonds in the plastic and either melt the plastic away into the fuel inside the tank or seep into existing micro-passageways in the structure and bread down the plastic/glass bond. Not knowing if a tank was exposed to ethanol or other chemicals (even in vapor form) could be a factor in whether an old tank is safe today.
I would not be surprised to find NOS tanks from the 70s that were in perfect condition for safe and effective use; on the other hand, a tank that looks good on the outside might very easily have deteriorated "honeycomb" structure that might prove to either leak soon after fuel has been put into it or be a fragile and unsafe product that can not be reliably expected to retain its strength in use. The problem is that we don't know but the odds are that there has been some change to the overall component or its constituate materials and the tank might fail to hold the fuel either in ordinary use or when exposed to extraordinary stresses.
.
