Talk – Tuesday 8th December 2015 – ‘The Lighter Side of Defence Engineering’
Perry’s talk covered his career as a ‘hands-on’ engineer and included anecdotes of some of the people he met, as well as some of the technical challenges he encountered.
In 1966 he joined Rolls-Royce Derby as a University Apprentice on a Thick Sandwich course. At his selection interview, he recalled being asked to describe a bicycle to someone who had never seen one before – an interesting challenge which you might like to try. Presumably his answer satisfied the recruitment panel, and his first year in Derby covered a variety of roles, from setter’s mate in a drop forge to assistant to the Works Manager at the Mountsorrel Sheet Metal Works. One of his tasks at Mountsorrel was to assess the value of suggestions put forward through the Suggestions Scheme. If an employee’s suggestion was adopted, he received a percentage of the first year’s savings. Perry remembers trying to assess the value of one suggestion which achieved a significant reduction in the amount of water used for a process. He toured several departments before he found someone who didn’t think he was mad to be asking how much water cost! He also spent a number of weeks on the engine test beds in Derby.
At that time the RN was in the process of buying McDonnell Douglas Phantoms, but the aircraft were being re- engined with Rolls-Royce Speys. For this application Rolls-Royce had to develop a reheated version of the engine to give the short-term increase in thrust required for take-off from a carrier, or for combat manoeuvres. As the picture shows, the sight of an engine running with reheat is pretty impressive, with a sheet of flame 20–30 feet long emerging from the rear of the engine.
Unfortunately, R-R were having problems getting the reheat sysem to light reliably and Perry was helping to run tests to try and sort out the problem. To assist the engineers, a periscope had been installed in the detuner (silencer) downstream of the engine exhaust so that they could see how the lighting was working.
Because the reheat combustion process is relatively inefficient, it produced a large quantity of soot in the exhaust, which meant that the periscope needed cleaning quite frequently. Working with Perry on the test bed was a fellow apprentice, Taffy, who was rather full of himself. Some of the testbed crew decided he needed taking down a peg or two, so they asked him to climb into the detuner tube and clean the periscope lens. While he was in there, they initiated a ‘dry start’ – basically the way you start a gas turbine, except that you don’t inject the fuel. Taffy was a large lad but, when he heard the sound of the engine spooling up behind him, Perry reckoned that he could have won an Olympic medal for the speed with which he extracted himself from the detuner and shot out of the test cell, to be met with hoots of laughter in the control room.
Health and safety was not quite the same in those days! Perry had another interesting experience with the Phantom Spey development – one of the reheat system control pipes was fracturing and no-one could understand why, as it wasn’t very highly stressed. Eventually, one of the engineers decided to strain gauge a pipe as soon as it had been manufactured and monitor the strain on it through assembly and operation. At one point during the assembly process, the strain gauge recorded a massive load. When the team went to see what was happening at that stage, they found that the engine was slung on a cradle and the technician needed to get from one side to the other, so he grabbed hold of the pipe and swung himself under the engine. Probelm solved!
After the first year of his apprenticeship, Perry then went to Leicester University for 3 years to read Engineering. When he applied, he didn’t realise that the engineering building had won a major architectural award for its architects, Stirling and Gowan. Because of this, the building was visited by a stream of architecture students who were told about the design wonders of the building. Perry and his fellow students longed to hijack them and explain some of the multitude of design defects that made the building a nightmare to work in. He noted that, after 50 years, the building had been given Grade 2* listed status, but the complicated glass roof of the workshop area had reached the end of its life and was now in process of being replaced at an estimated cost of £20 million to allow it to last another 50 years. So much for architectural design! He remembered that, during a first year Thermodynamics lecture, the lecturer asked what temperature a turbine operated at. Perry suggested 1100 degrees. The lecturer then asked, “Centigrade or Fahrenheit?” Perry said Centigrade, while another of the students said Fahrenheit. It turned out they were both right. Of course Perry was talking about gas turbines, whereas the other student had come from Parsons and was talking about steam turbines!
One of the subjects Perry studied in his final year (and which was the basis for his final year project) was Tribology. The official definition is ‘The interaction of surfaces in relative motion’ but it is more commonly known as ‘Friction, Lubrication and Wear’ – a topic that would be important later in his career.
After graduating in 1970, he returned to Rolls-Royce for the final year of his apprenticeship. At that time, the key project was the development of the RB211 engine, initially for the Lockheed Tristar aircraft. This was the first triple shaft gas turbine – that is, the engine had three concentric shafts, rotating at different speeds, each carrying a set of compressor and turbine blades. The first (low pressure) stage of the compressor comprised a single set of large diameter blades which initially were intended to be made from carbon fibre. However, although carbon fibre is very strong, it is also brittle, and Rolls-Royce soon realised that the blades would be liable to catastrophic failure in the event of a bird strike. Eventually, the carbon fibre blades were replaced by titanium ones, but these were heavier and the Lockheed contract had severe penalties for failure to meet weight and fuel consumption targets. The net result of this and other development problems was that Rolls-Royce went into receivership in 1971, and all the apprentices were made redundant.
Looking for a new job was made more difficult because the postmen had decided to go on strike at that time. However, eventually Perry was invited for interview at the Admiralty Oil Laboratory in Cobham. This was a standard Civil Service interview, and the panel included a Principal Scientific Officer from the RAE at Farnborough, who he recalls asking him at great length to describe the problems of lubrication in outer space. The subject obviously interested him, though what relevance it had to the problems faced by the Royal Navy was less clear to Perry! Fortunately, Perry’s answers seemed to satisfy him and, after the inevitable delay for the security clearance process, he received a letter saying that he had been appointed a Scientific Officer of the Royal Naval Scientific Service at Her Majesty’s pleasure.
In those days, Terms and Conditions of employment were not exactly clear, and he could only assume that, if Her Majesty became displeased, he would be disappointed! Perry joined AOL in June 1971 – a very different environment from Derby. He had his own office, with French doors opening onto the rose garden; he could use the putting green at lunch time … He was put in charge of the Mechanical Evaluation of Lubricants Laboratory – his main interests were in the performance of gear oils and hydraulic fluids for use by the RN.
The principles of gear lubrication are quite complicated but two factors are particularly relevant when trying to assess whether a lubricant will be able to protect the gears satisfactorily: the first is that the gear teeth are only in contact along a line so, with a reasonable load, the pressure on the contact area is high, even allowing for elastic deformation of the metal in the contact area to spread the load slightly. The second factor is that, especially as the teeth come into or out of the contact area, there is a significant amount of sliding between the teeth. The combination of high contact pressures (which can mean that the high spots on the gear surface penetrate the lubricant film) and high sliding speeds generates frictional heating which, if the loads are high enough, can cause the high spots to weld together and then get torn apart as the gears slide out of contact, leading to scuffing failure.
The picture of a scuffed gear shows the failure on one of the standard test machines used in Perry’s lab. The load on the test gears is increased in stages until this failure occurs – the failure load is then an indicator of the ability of the oil under test to resist scuffing. Of course, for test results to be repeatable, the gears have to be made extremely accurately, which makes the test expensive to run. Because of this, initial tests are usually carried out on a cheaper (but less representative) test rig such as the 4 ball machine. In this test, three 1⁄2” diameter ball bearings are locked together in a cup, which also holds the test lubricant. The fourth ball is held in a collet and rotated at 1800 rpm. The test is run for a set time and repeated at increasing loads, until eventually the load is so high that the frictional heat causes the balls to weld together. One can plot the size of the wear scar against load, and this gives an indication of the lubricant’s anti-wear and anti- scuffing capability.
Perry used to give welded ball pyramids to visitors as souvenirs. One in particular was memorable – the photograph shows Perry explaining the risk of welded balls to Admiral Dorian Dymock, who was obviously somewhat perturbed by the idea! Also worth mentioning is Rex Clutton, Perry’s head of engineering at the time, who has his back to us in the photo. He was an amateur racing driver in his spare time, racing a vintage GP Amilcar. His everyday transport when Perry first knew him was a Ford Escort, with a 3 litre V6 shoehorned under the bonnet! He swapped that for a Reliant Scimitar GTE (also driven by Princess Anne). You may recall that they had fibreglass bodies.
Perry remembers him driving them down to Bath once and, as they headed down the M4, explaining that he didn’t do more than 110mph as, at higher speeds, the doors lifted away from their seals at the top, and things got rather noisy. Despite that, Perry reckoned he was a very safe driver, no doubt because his racing experience had given him exceptional anticipation as well as good reflexes.
The work on gear lubrication became topical at the time of the ‘Cod Wars’ in the 1970s. Because the RN was at high alert, trying to protect British trawlers from the Icelandic patrol vessels, some of our ships were at sea almost continuously. After a few months, AOL heard that the RN dockyards were running short of white metal bearings used in ships’ main gearboxes. It seemed that the bearings were failing because water was getting into the gearboxes and reacting with the chlorine based anti-scuffing additive in the gear oil, causing the white metal to oxidise. The question was, could the RN use a different oil without the chlorine-based additive? The problem was that RN gearing was more highly loaded than other gearbox designs, and there were no other commercially available lubricants that offered the same level of scuffing protection as the OEP-69 used by the RN.
After discussion with MoD, AOL decided to run trials on the nearest commercial equivalent, an oil used by the US Navy. It was put into 2 County Class Guided Missile Destroyers, and Perry spent many happy hours taking impressions of gear teeth (oddly, the best material they found for this was something called Xantopren, more normally used by dentists for taking impressions of human teeth). He did have one major dilemma, though. When he visited a ship for one of these inspections, he used to have lunch in the Officers’ Mess. RN puddings were always a treat, and their bread and butter pudding in particular was really special. His dilemma was: should he forego the delights of the Naval puds and make life easy for himself when bending over the gearboxes in the afternoon, or enjoy the puds and suffer the consequences? As you may have guessed from his description, he enjoyed the puds.
One visit, to a ship in Rosyth dockyard, was memorable for a different reason. The ship concerned was due to do a high power run across from Germany and Perry was to inspect it after the run to see what the effect on the gearing had been. He was told that the ship would be available for him to inspect first thing on a Monday morning, which meant flying up to Edinburgh on a Sunday. He was asked if he would take a young chemist who had recently started at the lab with him, so he was joined on this trip by an attractive blonde by the name of Joanna Pedley. On Monday morning they turned up at the ship and asked for the Marine Engineering Officer. He arrived and apologised profusely: the weather had been so bad that they had been unable to carry out the high power run. That meant there was no point in carrying out an inspection, but Perry thought he might salvage something, so asked the MEO if he’d mind showing Joanna the machinery spaces since they were there. He was quite happy to do that. As they donned their white overalls, he explained that the ship had held an open evening for families the night before and some of the sailors were suffering from the after-effects. As the three of them went down to the machinery spaces, the MEO leading and Perry bringing up the rear, he saw a series of somewhat hung-over sailors take a casual glance at 3 guys in white overalls passing by, then the double takes as they spotted the long blonde hair and realised that the middle one wasn’t a guy at all. It almost made the wasted trip worthwhile!
Perry’s other key responsibility was the effect of lubricants on ball and roller bearings. One key point is that, because the contact area between ball or roller and running track is small, the contact stresses are high. Unlike gears, the motion is pure rolling, so scuffing isn’t a problem, but the high cyclic stresses lead to fatigue failure. Using the right lubricant can help maximise the life of bearings; equally, some lubricant additives can reduce bearing life. This is complicated by the fact that fatigue is a random phenomenon so, no matter how good the bearings (or lab test pieces) are, their life varies widely. His lab had several tests to evaluate the effect of lubricants on fatigue life but, because of this random effect, they had to carry out a number of repeat tests on each to get some feel for the relative performance of different fluids.
Why does all this matter, you may ask? Well, the RN’s nuclear submarine fleet used a special hydraulic fluid which was able to form a stable emulsion with sea water if it got into the hydraulic system. While the oil/ water emulsion was not a brilliant lubricant, it was a lot better than having the sea water being carried around as a slug of neat water which, if it went through a hydraulic pump or motor, could cause immediate catastrophic failure. In those days, there was only one supplier of this fluid and, around 1972, the Venezualan crude oil they used as the base ran out, and they had to reformulate the fluid with a different base oil. Perry had been working at AOL for just over a year when, as he was running some tests on these fluids, he found that, while the new formulation had a similar effect on fatigue life to the old one, a 50/50 mix of the two fluids seemed to reduce bearing life dramatically. This had some interesting implications.
One option was to dock the nuclear submarine fleet, flush all the old fluid out of their hydraulic systems and re-fill with the new formulation – not likely to be a terribly popular move, especially as there was a possibility that, even though the tests were based on a statistical analysis of the results of 24 runs with each type of fluid, they might still be rogue results. His first step was to ask his long-suffering technicians to re-run the tests, working 24/7 to complete them as soon as possible. At the same time, they decided to put a fluorescent dye marker into the new formulation of the fluid so that it would be obvious which submarines were using each type of fluid.
Eventually the repeat tests were completed and, fortunately, he was able to show that the initial results had been a rogue set, so there was no need to dock the nuclear fleet. The lab used to host regular visits by Chief Petty Officers to explain what it did and, on one of these, Perry talked about these tests and said that, now it had been decided that the two formulations were compatible, there would be no need to keep putting the dye into the new formulation. At this point, the Submariner CPOs got rather agitated and said that the dye was the best thing for a long time – now, whenever there was a leak, it was so much easier to trace the source! So the dye stayed in the fluid – a happy conclusion to a worrying time.
As someone employed in an R&D lab, it was important for Perry to keep his knowledge up to date, and he used to attend IMechE Tribology conferences regularly. Tribology has been something of a niche subject, and has often confused non- specialists. At one conference in Glasgow, the organisers had arranged for a well-known Scots singer to entertain the delegates one evening. She appeared rather red-faced, and explained that she had put together a selection of songs from around the world, as she thought the delegates studied tribes! Nevertheless, Perry and the others enjoyed her singing. That conference was also memorable because it was sponsored by Seagrams, who at that time produced 100 Pipers whisky and Chivas Regal. The Tribologists were certainly well lubricated! Another Tribology conference, in Leeds, had a rather more erudite after-dinner entertainment. Prof Dowson there was one of the pioneers of designs for hip replacements, and delegates were treated to a film of an operation in glorious technicolour. Perry still remembers the commentator saying “… and now the surgeon taps the replacement ball joint into the femur ..” as the film showed him hammering like mad to get it into place. The patient must have had a very sore hip!
Perry was also involved in Government to Government Information Exchange meetings with the Americans, Canadians and Australians which were very good value. In the early days, they were flown out to the North America courtesy of the RAF in VC10s. There were two main differences from civil airlines: the first was that they were dry – no doubt good for fighting jet lag, but it didn’t help the time pass! The second was that the seats faced backwards (these were basically troop carriers, and rearward facing seats were safer in an emergency landing). This was not a problem in normal flight but the VC10 had a very good rate of climb and, after take-off it felt as though the passengers were in danger of being tipped out of their seats. He remembers one return flight from Washington particularly. He arrived at Dulles Airport (not exactly the centre of the universe) to discover that the take-off had been delayed by 4 hours. Apparently a message should have gone to his hotel, but he hadn’t received it. Eventually, the flight was called and as they were being ferried out to the VC10 by bus, the RAF Loadmaster said over the PA system “Ladies and Gentlemen, apologies for the delay, but our navigation system has failed and we’re flying overnight so we can navigate by the stars.” The passengers all laughed dutifully, at which point he said “Don’t laugh – it’s absolutely true.”
In the event, their astral navigation was pretty good, as the flight arrived on time, and the landing was as gentle as usual – RAF pilots’ ability to land their aircraft with the minimum of noise and bounce makes most civil pilots look very amateur.
By 1983, AOL had become part of the National Gas Turbine Establishment as part of a Defence Rationalisation programme, as had the former Admiralty Engineering Laboratory at West Drayton. One day, Perry was called in by the Head of Naval Engineering Department and told that they wanted him to take on the job of Head of the Machinery Controls Section (MACO) at West Drayton. Perry did point out that, while he had a pretty good understanding of Tribology by then, his knowledge of control theory had been one of his weaker subjects at university! This didn’t seem to worry the Head NED so, like a good civil servant, he agreed t do as he was told, with one proviso: he pointed out that plans were well advanced for the next Information Exchange meeting in Australia and he had a key role to play at the meeting. He didn’t want to leave his colleagues in the lurch by abandoning them at the last minute. The fact that he had never been to Australia and was keen to see the continent was of course entirely irrelevant. The Head NED graciously agreed to this, so Perry prepared to change jobs. The key challenge facing the MACO team was to develop a real-time simulation of the proposed Single Role Minehunter. They would use it both to assess how the ship was likely to operate under manual control, and to examine the performance of automatic ship position control systems. The Naval Staff Requirement for the design stated that the ship had to be able to circle a suspected mine under automatic control with a course accuracy of X metres RMS. Perry can’t remember what X was and, even if he could, he couldn’t tell you!
Anyway, they produced an initial algorithm and tried it on the simulation. The result was not an astounding success – as the ship moved off its initial hover point to start circling, the effect of wind and tide was to take it almost over the suspected mine. The control system did recover fairly quickly and, after that, the ship followed the desired course reasonably well. The interesting thing was that, even though in real life the initial excursion would probably have caused the mine to detonate, analysis of the track between the two points showed that it met the accuracy specified in the NSR – writing specifications needs a lot of care! As well as using the MACO simulations, their customers in MoD decided that it would be a good idea for a small team to visit a ship using a commercial automatic position control system for 24 hours to see how it operated. Initially, the plan was that Perry’s deputy, the technical lead in the section, would represent them.
However, he came up with some weak excuse (such as his wife was having a baby) so Perry agreed to go instead. So it was that, one January morning, a small group assembled on the dockside at Aberdeen to board the Stena Seaspread, which was a Rapid Intervention Vessel used to support the North Sea oil rigs. As the ship left Aberdeen, the captain welcomed the MoD team on board and explained that they were going out on a one month tour of duty! The team would be taken off by helicopter and flown back at the end of their visit.
As you will see from the photograph, the heli-deck was above the bridge – almost the highest point on the ship! If you can imagine the way the ship rolled in the Force 9-10 winds that were blowing at that time, it’s not surprising that it was nearly a week before the weather moderated enough for a brave pilot to slam his helicopter onto the deck for just long enough for the team to scramble in, before taking off again: truly a white knuckle ride for all concerned.
Having said that, Perry was impressed with the ability of the ship to ‘hover’ under automatic control about 100m from an oil rig , holding position to an accuracy of about 3m in that foul weather. There is more to Perry’s career, taking him through the Royal Aircraft Establishment and on to QinetiQ, but that’s another story.