The recent spate of Lightning accidents led to a comprehensive review of its accident rate since its entry into service, with particular emphasis on the number of aircraft that were being lost to in-flight fires. As the Lightning was scheduled to remain in service until its gradual replacement in the air defence role by the McDonnell Douglas Phantom in 1974/75, the high loss rate was becoming critical. Although the Phantom was already in service it was being used mainly in the strike/attack role, a task it would continue to fulfil until the arrival of the Anglo-French SEPECAT Jaguar.
The problem of in-flight fires in Lightning aircraft had been known about from the very beginning and had eventually led to a Fire Integrity Programme in the late 1960s, which it was hoped would lessen the risk of aircraft being lost to fire. The design of the Lightning had been optimised for performance with particular emphasis on high levels of acceleration, together with a rapid climb rate and the ability to fly at high supersonic speeds at altitude. To generate the high levels of thrust needed, a twin-engine design had been adopted in which the engines were mounted one above the other and staggered, which had the effect of reducing frontal area, which in turn reduced drag. No.1 engine was mounted low in the fuselage with No.2 mounted above and to the rear of No.1. This meant that No.2 engine was positioned above the No.1 jet pipe so that any fuel or hydraulic leak from this engine was likely to find its way into the hot Fire Zone 3 area of the No.1 jet pipe due to inadequate sealing. To try to prevent this Lightning aircraft were sent to maintenance units in a rolling programme where a product known as Viton was applied to the floor of the No.2 engine bay. This was a synthetic rubber material that gave excellent heat resistance (200 degrees C) and was resistant to fuel and oils. As previously noted, however, Lightnings continued to be lost to in-flight fires even after they had passed through the Fire Integrity Programme.
A report was compiled on the Lightning’s safety record and covered the period from its entry into service to 11 May 1971, with accidents analysed by causes and by marks of aircraft in order to detect any possible trends. Comparisons were also made with a number of other RAF types. In addition fire accidents and incidents were analysed thoroughly for the period January 1970 to 11 May 1971. These dates were used due to a lack of earlier data. An assessment was also made of the risk to the pilot in terms of abandonment and mortality, and the likely level of future write-offs was also assessed.
Since the entry of the Lightning into service there had been a total of 110 accidents, of which forty led to the aircraft being written off. Aircraft fires were the main cause, resulting in 30 per cent of all accidents and 37.5 per cent of all write-offs. Pilot error accounted for 29 per cent and 22.5 per cent respectively and other technical causes accounted for 21.8 per cent and 25 per cent. This left other causes and unknowns for the remaining 19.2 per cent and 15 per cent. A breakdown of accidents by mark showed that, overall, the F.1/F.1A was involved in the most accidents with thirty which represented 30 per cent of the total. Next in line were the F.3 and F.6 which were in practically equal proportions with twenty-five and twenty-four respectively (25 per cent of total accidents). These three marks together accounted for 82 per cent of all accidents due to fire with the F.3 alone being responsible for 33 per cent.
In the case of write-offs, the same three marks accounted for 85 per cent of all cases with the F.1/F.1A contributing the largest proportion with fourteen (35 per cent). Write-offs due to fire did not involve the F.2 and T.4. In this category the largest proportion was represented by the F.3 with six cases (40 per cent). This was followed by the F.1/F.1A with five (33 per cent) and the F.6 with four (27 per cent). It was thus concluded that fires were the largest single cause of all Lightning accidents and write-offs with the F.3 being the mark that was most prone to fire accidents.
In addition to breaking Lightning accidents down by mark, the report also looked at a timeline and here it was noted that fire accidents had become more prominent since 1966, with the exception of 1969 when there were only two fire accidents and no write-offs as a result. Since 1969 there had been five write-offs due to fire, of which three involved the F.3 with the other two being F.6s. It was noted that the same two marks were involved in all eight fire accidents. It was clear that fire accidents and write-offs had been on the increase since 1966 and that there had been a revived upsurge since 1969. When the total number of accidents was broken down into cause groups it was found that in recent years, accidents caused by pilot error or technical causes (other than fire) had shown a downward trend while other causes had remained constant. On a cumulative basis fires were the largest single cause of all accidents, and had been since 1967 in the case of write-offs with a recent upward trend.
At the time of the report, in the case of write-off rates, only the F.1/F.1A and T.5 showed a definite downward trend that had been indicated since 1968 for the F.1/F.1A and 1966 for the T.5, which was very soon after its introduction to service. However, in the case of cumulative write-off and accident rates due to fire, although the F.1/F.1A was now showing a downward trend, its overall rates were still the highest at 1.18 and 2.1 (per 10,000 hours) respectively. The upward trend for the F.3 had been particularly pronounced since 1969 and in the case of fire accidents the F.3 had shown a 50 per cent increase. Although the trend for the F.6 was also upward, the trend for the F.3 was described as alarming. It was concluded that, of the different marks of Lightning, only the F.1/F.1A and T.5 were showing any sign of improvement.
Although the Lightning’s recent problems were giving cause for concern, overall its accident rate compared favourably with other types including the Javelin, Hunter and Buccaneer in terms of accidents and write-offs. However, in some cases the causes of these accidents varied considerably. A comparison was made of the write-off rates of the Lightning, Hunter and Javelin based on cumulative flying hours in which it was discovered that the proportion of write-offs due to fire for the Javelin (36.1 per cent) was actually very similar to that of the Lightning at 37.5 per cent. In the case of the Hunter the figure was only 9.7 per cent. A comparison of write-offs due to pilot error showed the Lightning in the best light with 22.5 per cent, which compared with 27.8 per cent for the Javelin and 40 per cent for the Hunter. Write-offs caused by other technical issues were similar for all three aircraft. The comparison with the Hunter therefore showed the greatest differences, the Lightning being nearly twice as good from the pilot error aspect, but four times worse in terms of fires.
With regard to the Lightning’s recent accident history, the most important part of the report was its findings in relation to fire hazards. The breakdown of fire accidents looked at the marks involved, whether the occurrence took place on the ground or in the air, the type of fire, stage of flight, the use of reheat and whether the aircraft had passed through the Fire Integrity Programme. In all marks of Lightning most of the fire incidents occurred during the climb or when under cruise combat conditions when high power settings were being used with corresponding higher temperatures within the fire zones. However, no link could be established between fire incidents and the use of reheat due to the fact that in 66 per cent of in-flight fires the use of reheat was not recorded.
Of all the real fire warnings that had occurred, 75 per cent involved No.1 engine and the prevalence of Fire 1 and Reheat 1 warnings was common to all marks of Lightning. With regard to hot gas leaks, however, these were recorded in equal numbers for each engine installation with three quarters resulting in either a Fire 1 or a Fire 2 warning. As well as real fire warnings, there were numerous spurious fire warnings on the Lightning and 75 per cent of these resulted in Fire 2 and Reheat 2 warnings. The marks that were most likely to receive spurious warnings from the No.2 engine were the F.3 and T.5.
Not surprisingly the most common causes of actual fires were fuel leaks and hydraulic oil leaks with each contributing 30 per cent to the total of all fires. These two causes accounted for all fires in the F.6 and 45 per cent of those in the F.3. The only actual fire to occur in a T.4 also resulted from the burning of hydraulic oil. The F.3 and F.6 were the only marks to have suffered from engine titanium fires. Of particular concern was the fact that of the nineteen airborne fire accidents and incidents in 1971, fourteen occurred in aircraft that had been through the latest Fire Integrity Programme and two thirds of the aircraft involved in actual fires in the air had had the fire integrity modifications embodied. The only mark of Lightning that did not have an accident or incident to a fire modified aircraft was the F.2/F.2A.
As has already been recorded, fires also occurred on the ground and of these 75 per cent occurring during start-up. Once again they were most prominent in the F.3 and F.6. During the period under review there were two cases of hot gas leaks on the ground, one occurring during taxying in a T.4 and the other during start-up in an F.6. There were also six spurious warnings that were evenly split between start-up and taxying. The main causes of fires on the ground were AVPIN (58 per cent) and fuel venting (33 per cent) with all of the AVPIN fires occurring on start-up. The mark that was most affected by fuel venting was the F.6 with three cases out of four resulting in fires. The occurrence of fires on the ground in aircraft that had had the fire integrity modifications fitted was less than in the air at 33 per cent.
When the analysis of fire incidents was widened to include those from January 1970 it was found that there had been twenty-six in all of which eighteen had been through the Fire Integrity Programme. Having tabulated the figures for this period an estimate was then made of likely aircraft losses in the future. Assuming that the write-off rate remained the same it was calculated that by 1976 a further twenty-six would have been lost of which half would have been F.3 and F.6 aircraft. Thereafter it was estimated that losses would have been two aircraft per year. The risk to the pilot was also calculated. Assuming a four-year tour on the Lightning force, the chances of a typical pilot, flying some 320 sorties a year (20 hours per month), being involved in an aircraft fire leading to abandonment were estimated at 6 per cent with a subsequent risk of mortality of 1.5 per cent. When all accidents were taken into consideration there was a 17 per cent chance of having to eject during a typical tour with a 4.25 per cent risk of mortality.
From the report into the Lightning’s safety record it was clear that the marks that were most at risk from in-flight fires were the F.3 and F.6. Although five F.1/F.1A aircraft had been lost as a result of fires, this mark had shown a significant improvement since 1968 whereas the figures for the F.6, and particularly the F.3, had shown a worsening trend. The most likely reasons for this situation were given in an Air Ministry memo dated 23 September 1971. It was believed that there were several engineering features of the F.3 and F.6 that were responsible, at least in part, for their poor fire record.
The F.3 and F.6 were both powered by Rolls Royce Avon Mk.300 series engines which, when operating at ground level and full military power (without reheat), ran at carcass temperatures about 20 degrees C higher than the Mk.200 series engines as fitted in other marks of Lightning. It was also noted that the surface temperatures of some parts of the Mk.300 series engine were higher than the spontaneous ignition temperatures of fluids passing through the engine bays. Another difference of the Mk.300 series Avon compared to the Mk.200 series was the fact that the former employed a high pressure fuel-draulic system that had been prone to defects that had caused fuel to leak into the engine bays. This was mainly due to the very high fuel pressures involved. The engines of the F.3 and F.6 were also fitted with a Hot-Streak reheat initiator unit that had also been susceptible to faults in which fuel had been spilt into the engine bay areas. Paradoxically, although theses three features were also incorporated in the two-seat T.5, the fire record of this mark was the complete opposite of the F.3 and F.6. From an engineering point of view this was impossible to explain, although it was thought that the T.5’s role as a trainer may have accounted for its better record as regards in-flight engine fires.
During the official correspondence on this matter it was noted that Coltishall’s record of engine fires in Lightning aircraft was considerably better than all other Lightning stations. This was put down to the particular mix of aircraft that was flown from the Norfolk base. Although its complement of aircraft amounted to thirty-seven, at the time of the report it only operated four F.3s. As has already been noted, this was the variant that was most prone to engine fires. Although Coltishall still operated ten F.1As, this mark had shown a considerable improvement in its fire record in recent years.
The success, or otherwise, of the Fire Integrity Programme was a matter of considerable debate, although it appeared to have been effective regarding the earlier marks of Lightning, clearly there was still a way to go with the F.3 and F.6. As such it was agreed that further work should concentrate on the differences between the two sets of marks, i.e. the higher engine bay temperatures, the fuel-draulic system and the Hot-Streak reheat ignition, and that weaknesses that could not be readily cured should be contained.
The weaknesses in the fuel-draulic system were identified as the repeated breaking of a flange joint, the inadequacy of a gasket that had been employed, and the susceptibility of a flexible pipe to damage through bending and its liability to fatigue as a result of large ripples in pressure. By mid-1971 action was in hand to bring about improvement on all three counts. A problem had also been identified on the low-pressure fuel system on a F.6 aircraft and improved seals were to be fitted that were capable of withstanding temperatures of 200 degrees C. A method to box in this part of the fuel system was also proposed to act as a heat shield and to contain and drain away any leakage. Also at this time large-bore overboard fuel drains were due to be flight-tested, which were intended to drain fuel should a massive leak occur. Attention was also focused on the hydraulic system which had been the cause of a number of in-flight fires. Hitherto the system had not been subject to periodic replacement, but it was now clear that bonded seals in the hydraulic system, deteriorated due to heat. Because of this the existing seals were scheduled for replacement by improved seals within a year. In the meantime Viton tape was used to minimise spray from leaks.
A number of recommendations were made with regard to the engines and these included examination of the jet pipe areas to check for cracks following a recent hot gas leak in a F.6. Ultimately it was proposed that the intermediate jet pipe be replaced by one of a new material. The programme for substituting steel outlet guide vanes for titanium to prevent titanium fires was underway and a cyclic fatigue test of the Hot-Streak reheat ignition system was being carried out. Flight trials were also due to start with fuel additives that were intended to show up fuel and hot gas leaks. Servicing was also being reviewed and a new system of service training had recently been set up so that technicians who had not previously worked on the Lightning were given the best possible background to ensure fire integrity. This knowledge was also to be passed to the squadrons. In addition, there had been a review of all outstanding modifications and these were placed in strict order of priority in the fire integrity context.
The Fire Integrity Programme was thus continuing and in time, with a greater knowledge of the problems that affected the Lightning, the number of fire incidents was gradually reduced. Had the number of accidents as a result of fire continued at the levels recorded in 1970/71 however, as one writer put it, it would probably have resulted in the aircraft’s premature demise. This would have been an extremely serious situation and may well have resulted in a gap in air defence capability before this role was taken on in the main by the Phantom. In the early 1970s the version of the Phantom as flown by the RAF also had major problems relating to its Rolls Royce Spey engines.