I asked a taxi driver the other night why the ball chasing fraternity wait for dark nights and crap weather to indulge their passion. I’ve been asking the same question for years without any satisfactory answer. It’s too hot in the summer, I’m often told, which begs the question why cricketists wear jumpers and long pants. It’s tradition, etc. etc. but not until now did anyone ever explain that chasing a ball in the summer would conflict with the International Champion’s World Cup Challenge League Cup or whatever it’s called and, though some Russian and Norwegian teams do wait for the snow to melt, it ruins their match-fitness at the start of the International Champion’s World Cup Challenge League Cup and, as the England team is made exclusively of English lads (how was I to know that?), it may actually be too hot for them after all. The answer I’d sought for years, and there it was – brilliant! which just goes to show that if you keep asking you’ll get there in the end. On Friday gone (18th November 2011) the team finally raised a pint late in the evening and thought, that’s a job well done and everyone is happy. Continuing to ask and not taking ‘no’ for an answer figured highly in the reason why and at last the relaxation would come. But it didn’t, because a two-year project within a project may have finally been over with a fantastic result but the mental expenditure that evening was massive and coming down took a little longer. It was all to do with Bluebird’s fuel system. What happened is this. The bit of the engine you see, that big tube with a fan up the middle, is the part that actually does the work but it’s a difficult thing to tame with the potential to spew oodles of heat and energy until it explodes or melts so control is everything. To keep it in check and place the pilot in charge there’s a complicated control system bolted to the underside that feeds the beast, or not, depending on what the pilot asks for. So far as we can tell, Donald only ever had one such system despite having two engines so it was on the boat through the whole campaign of 66/67 and eventually landed on the lakebed with everything else. It was this system that acted on Donald’s command and poured kerosene into the Orph’ that fateful morning and we wanted it to do so again on the finished machine should British engineering still be capable of such a thing – a seriously ambitious ask. On the one hand it was a complex piece of machinery that would not take well to neglect and water ingress but on the other its interior remained mostly full of pure jetfuel whilst the outside was smothered in anaerobic mud. Though badly wounded it wasn’t quite dead…
First thing was to find out just exactly what we had so the rebuild manual was acquired from Rolls-Royce who, of course, don’t support legacy engines in any way and could not help at all but if no one wanted these obsolete manuals they were destined for the skip… Reading it in bed was like propping a tombstone on my chest but night after night I waded through it until at last I’d worked out what all the bits did. Sort of. Soon we knew two things. Firstly, the system breaks down into a number of discreet parts, each performing its own function not all of which are strictly necessary for ground-running so that was encouraging and, secondly, it was way too complicated for us to even think about mending in-house so outside help was needed if we were to stand an earthly chance. Having reached some basic conclusions, early research said forget the big boys, they’ll not give you the time of day. Best speak with the industrial gas-turbine world who at least aren’t burdened with the paranoia that surrounds anything that has to fly. The reasoning was sound but no matter what we tried all roads led back to a tiny number of specialist facilities that handled these parts on behalf of the entire planet. One I tried made interested noises as I reeled off the Donald Campbell, Bluebird spiel and I was quickly passed upwards through an increasingly enthusiastic hierarchy repeating my request as I climbed until, at last, I crested the summit to find it ruled by a jobsworth with whom the Donald Campbell story didn’t work at all and who told me that the antiquated junk I had to offer was no longer supported and therefore I hadn’t a hope. I must point out that he exhibited the professionalism that typifies the upper reaches of the aerospace industry; it’s just that duty he did very well but what was needed here was something way beyond the call. At that point it would have been so easy to consign the fuel system to the museum and use another because we’ve oodles of pumps and other stuff. After all, we’ve been collecting spares since 2001 but that felt too much like giving up. Maybe we could rebuild it in-house after all… Out came the manuals again and, after a second round of study, the decision was taken to do some investigative surgery to assess the damage. The filter bowl was easy.
It’s upside down here because it hangs off the underside of the engine normally. Fuel comes in through that severed hose upper-left and out again through the metal tube at the bottom having passed through a filter. There’s a LP warning switch sticking out lower-right with its electrical connector facing upwards. Inside, the filter was in a clean condition with only a small amount of water staining.
It all bode well for the components further upstream and the filter manufacturer still exists today though they don’t seem to list this precise element any more – bet they can make one though. Next in line lies the fuel pump and we hoped the water hadn’t got that far. It turned smoothly and felt mechanically sound but it was not to be. Despite the internals generally being in good condition the water had got in and the disaster was laid bare immediately the end casing came off.
The staining was only light but these machines are exceptionally intolerant of any imperfection so the game was over so far as running it again without a rebuild was concerned.
This is a spherically-ground bearing face on which seven small pistons ride round and round like a merry-go-round, or at least it used to be. It’s called a cam plate and the parts of the pistons that rub on the cam plate are called slippers and those circles on its face are where the slippers festered against the face for forty years. The cam plate rocks like a see-saw (lots of children’s play park analogies going on here) on those journals you see in the one o’clock and seven o’clock positions to increase or decrease the pump stroke and thereby govern the amount of fuel shifted but the face should be mirror smooth otherwise it will heat up and wear itself out along with the slippers in very short order. It’s all lubricated by the fuel that it pumps, you see, and it’s not a great lubricant; hence the mirror finish. The slippers look like this.
This is actually the complete pump rotor assembly and you can see the pistons below the slippers. There’s also a water damaged bearing band around the bottom that runs in a carbon bearing on a thin film of Avtur. That wouldn’t last five minutes either because the carbon bearing includes lead and other metals that heat up and melt rapidly if the bearing band isn’t perfect, then the whole thing fails in the most dramatic way. Inside each piston is a spring to force it up its bore against the cam plate as the rotor spins, drawing fuel through an open port beneath, syringe-like, after which it’s immediately pushed back down by the canted over cam plate forcing the fuel back out through a different port as it goes. It’s a brilliantly simple concept but the engineering to make it work reliably is somewhat more involved. Even the springs are an adventure – but more of those in a while. The rest of the pump is devoted to controlling itself because otherwise it would enter a self-destructive spiral. With rotor speed being directly proportional to engine speed the faster the engine turned the more fuel it would pump into the combustors and you can guess how that would end. Therefore there’s another part of the pump that exists purely to govern itself. It generates its own pressure, servo pressure, which has only one purpose, to push on a piston that de-strokes the pump and accurately managing servo pressure is the Holy Grail of not blowing yourself up! The way it works is that dumping servo pressure calms the engine down, the theory being that all your control system has to do is let the servo pressure go at the critical moment and you’ll not end up picking red hot fragments of engine casing out of your body..
This is the servo piston and rod and, as you may imagine, the lip seal around it isn’t available at your local Bristol Siddeley stockist. And that’s only a tiny part of the problem. Look to the right and you can see that something has gone rusty. That’s called an amplifier valve, or ‘amp-valve’ for short and it helps control the pressure either side of the piston. It’s a really trick design that uses what amounts to half a ball bearing with a ground, flat face swivelling in a hemispherical cup to press squarely over a small hole in a thing called an orifice plate to cut off the flow of kerosene from beneath. The cup is held in the grip of four tiny leaf springs to be sure it has no fulcrum position so the flat of the half-ball always lands flat against the orifice plate to form a good seal and when you read the book it soon becomes damned obvious that setting it up is well beyond what can be done on the average kitchen worktop. From here things only get worse because the pump is relatively straightforward in what it does – making pump and servo pressure and that’s about it – but the CCU (Combined Control Unit), on the other hand, lies at the heart of the system and basically runs the show. It takes fuel from the pump and shares it out in two ways. Some is distributed to the burners at constant pressure to keep the fire going while a second circuit is controlled by a throttle valve so the engine can be made to spool up and down as required. Other modules in the control loop are also tapped into it as fuel pressure provides opposing forces to springs and diaphragms all over the place. Then there’s the barometric pressure control unit stuck to the CCU like a conjoined twin where, on an aircraft at least, it spends its time watching the compressor inlet pressure so it knows what it’s doing whilst messing with your servo pressure and the fuel delivery rate depending on how high your plane is. Flying shouldn’t be much of a concern in our case but the BPC still has to be calibrated if only to effectively remove it from the control loop and ensure it has no ill effects on the big picture. There’s a thing down there called a pressure ratio limiter too… its job is to keep an eye on the pressure in the front of the engine because as your plane flies higher the air gets colder and thinner and the ram effect and high compressor speeds begin to upset the ratio between the pressure in the compressor and that in the hot end. It can get too high and heat up the compressor – something you don’t want if you don’t want your engine to give up prematurely. Or it can get too low and your engine can belch fire in the wrong direction. Either of the above might befall you if the PRL didn’t bleed away some more servo pressure at the crucial moment. We’re not going flying but no matter, the pressures have to be right for ground running too so at least with the PRL working and calibrated we’re safe in knowing that we’re not going to have problems and another box is ticked. Sounds most awfully complicated, doesn’t it. Well it is, because next you have the air/fuel ratio controller and the pressure ratio switch. The AFRC again watches the compressor pressures but this member of the ensemble is devoted to keeping things in check when you slam the throttle open. What you don’t want is to throw a big slug of fuel into the engine only to have it accelerate away like a runaway train and blow its own fire out so the AFRC fiddles with the servo pressure just like everything else to make the engine spin up without drama no matter how hard you hit the loud pedal. Of course, you really ought to have the PRL in the loop while all of this is going on if you’re to do the job properly. Oh, and let’s not forget the pressure ratio switch. OK, it’s another altitude compensating thingamabob but who knows what horrible effects it might have if not working properly on the ground. Essentially what it does is use a barometric capsule to close a valve and introduce a step in the control curve of the AFRC when a certain altitude and therefore engine intake pressure is reached so you still won’t blow the fire out even if you slam the throttle open at a zillion feet. Right, there’ll be a test later so have you got all that? OK – so now maybe you can appreciate that what you don’t do next is marinate the whole system in a mixture of mud, kerosene and fresh water for three and a half decades then drain and store in a dry place for another ten years before declaring, ‘It’ll fix’. Rachel had a quick read of the above at this point and added the following.
* (Any Bluebird widows, Rule 1: learn the jargon then repeat the three last words of every sentence and nod in appropriate places. It has worked for me for 11 or so years!) *
Thank you for that – may we move on? The system is an intricate hydro-mechanical computer that makes Babbage’s Difference Engine look like an abacus and has so many close tolerance and, let’s face it, obsolete parts that it really was going to take something special to get it going again but who to turn to? We thought we’d start small so as not to seem too frightening and the pump was chosen as a place to begin. What happened once we’d abandoned the industrial gas turbine world and gone back to aviation is that we’d be told, go talk to such and such, with the rider that those recommended either wouldn’t touch it or it would cost the earth. In this instance I tended to say let me worry about that and so by this process we were relentlessly shepherded to the door of Goodrich in Birmingham. The word was that if mending our pump was possible only they could do it – but they never would… Now then, companies of such stature rarely grant an audience to commoners like us because random members of the public do not, as a general rule, rock up with a heap of junk wanting it mended so a stranger on the phone who doesn’t speak fluent aerospace faces something of an uphill struggle. Add to that the number of aeroplane nutters attempting to take to the skies in something quite rightly grounded at the end of the Cold War and the path to getting your pump mended becomes properly treacherous. Having gained some understanding of the obstacles in our way a plan was needed and a multi-pronged one at that. First of all it seemed disrespectful to show up with a muddy box of broken engine parts asking for a fix so the very least we could do was get them clean. Over several weeks each module was prized apart, scrubbed and sizzled in the ultrasound bath, dried, then threads were lightly oiled and everything loosely reassembled. It still looked like something Granddad brought back from the war but at least it wasn’t going to vomit rusty, kerosene flavoured mud-water over the silk tie of some aerospace suit during its begrudged inspection.
Here’s a cleaned up rocker arm from the BPC with one of those half-ball things in it. First time one of those came apart and the half-ball skipped away under the bench such a panic ensued until it was recaptured and put back. We got wise to them after that, which of course was the second part of the plan. In the unlikely even of us being given the time of day by a pump-whisperer the outcome might depend on our being able to head off some of the objections at the pass so as each unit came apart it was compared to the rebuild manual and the names of all the parts learned along with their function – if we could work it out, which wasn’t always possible. We took plenty of pic’s and made notes about which bits were good and what had gone rotten, how far the water had penetrated and where we had donor parts of a similar type. All this would stand us in good stead were the pump gurus to say, the cordwangle that acts against the bogling-fork is bound to be seized, to which we could reply that actually the cordwangle is perfect because the water didn’t get that far and there’s a good bogling-fork in this donor unit from our stores… Conversely, were they to make the point that it really was knackered at least we’d have a chance of understanding why and accepting their verdict, but this didn’t solve the problem of who to ask. And then came the real breakthrough. You see, around this time various parts of the problem were being mentioned on the forum and in the diary and in due course we were quietly contacted by someone who actually worked in the piston pump rebuild game. Someone who preferred to remain nameless but who had access to drawings and repair schemes for the various knackered bits and when we saw how complex the engineering was we realised just how far we were out of our depth. Mending the likes of this needs a factory and people who do this every day because no matter how much info you have there’s always the human factor and inevitably the complex setting up of the worzel sprocket will fall to Old Bob in the corner because he’s been doing them for thirty years and has a real feel for what’s to be done.
Up to this point we were still considering reworking the pump in-house and, though the drawing didn’t come with the message spelt out, what it seemed to say was, don’t even think about it. But what was more important was that at last we discovered where both the engineering excellence and the willingness to make this happen was to be found.
Remember when you had to fix your old Mini or Ford Escort because the winter just killed it and the alternator or starter motor had ‘Lucas’ written on it? Well those parts were made on a huge site in Birmingham that once employed thousands of people. Sadly it’s a weed-strewn wasteland nowadays with only the remnants of the car park and heaps of crushed bricks testifying mutely to the vast industry that used to be, but alongside what was once its big brother remains a slender, slightly art-deco building where Joseph Lucas used to make gas turbine machinery. Once the poor relation, it’s now all that remains of an empire but inside persists a miracle.
Behind that ageing façade you will find ‘British engineering’. The real deal, the thing that people shake their heads at sadly and say is gone forever. Oh how wrong they are. Don’t ever mourn the passing of British engineering because not only has it gone precisely nowhere, (Doesn’t Red Bull Racing hail from Milton Keynes or somewhere equally nondescript?), British engineering is fitter and healthier than it ever was. Behind those walls are things that would astonish you but it’s a commercially sensitive site so that’s all you’re getting. But how to get inside the fancy offices and in front of the decision makers? It took a bit of telephoning but eventually we were granted an audience with two of the top engineers. “You realise you’re asking us to hand you the opportunity to blow yourself up.” The first said concernedly during our first ever conversation. That’s a tough one to talk your way out of but the meeting went ahead and we discussed the finer details of the pump. I explained that we were trying to use as much original as possible and at some point the piston springs were mentioned. Thinking a spring is pretty much a spring I suggested that ours seemed OK and might live to fight another day (I was actually trying to make their job easier, cheaper and hopefully more palatable). “You don’t know what you’re talking about, do you.” The other engineer said with a grin that masked the tiniest hint of dismay. What could I say? He was right, because it turns out the springs are made of wire that has already been shot-peened to form a pre-stressed skin on the metal so the resulting spring will perform beyond what the metal is theoretically capable of doing and they’re so highly stressed that they really ought to be replaced at overhaul. The big problem if one breaks is with those ports mentioned earlier through which fuel is sucked up into the space under the pistons.
This is the port plate and it doesn’t turn and lives at the base of the rotor which, of course spins at high speeds to open and close the inlet and outlet ports beneath each piston one after another. The bottom of the rotor looks like this.
Now think about what would happen if a spring broke and dropped down one of those holes to where the spinning rotor and the fixed port plate meet. Yep – you got it – sliced and diced then pumped around the entire engine by the six good pistons. We agreed that new springs were a good idea. By the way, you can also see the damaged bearing band around the base of the rotor. A similar question arose over the elastomers – the diaphragms in the pump, all of which looked OK.
They had to be replaced too because, contrary to my naïve belief that failure of one would merely cause our engine to stop, it turns out that quite the opposite was far more likely with fuel pressure winning the fight over servo pressure and the engine running away with itself until it exploded. We decided new elastomers were also an excellent idea. By now we’d been joined by some of the bosses and the big company politics began to kick in. New parts were going to have to be re-manufactured to the original drawings and a test rig snaffled to test things and this would impact on production at the other plant where spanners were wielded and which subsidiary would handle the complex machining work and how could it all be snuck beneath the bean-counters’ radar? “What would you like us to do?” I was asked eventually after we’d batted the problems back and forth for an hour or so. Three plastic boxes containing our precious fuel system stood stacked at the side while the ravaged fuel pump held centre stage like some forlorn paperweight. It was time for the big pitch; this would likely be our one big chance. “In a perfect world…” I began, “we’d like you to take these parts and give us them back rebuilt and working.” No harm in aiming high but they didn’t look too enthused so I hurried on. “If that’s not possible maybe we could have some good, second-hand spares like the ones Kearsly Airways gave us to rebuild our second LP boost pump and we’ll have a crack at it ourselves.” There still didn’t seem too much excitement so I offered third place then shut up. “If we could only have access to the technical data…” Contemplative silence hung for a beat during which glances shot this way and that and then, to my unspeakable joy, I was invited to hand over the boxes. We were on a roll. That was two years ago and since then the entire system has undergone a rebuild in keeping with the standards usually applied to military and commercial aircraft but with conserveering in mind. Wherever an original part could be saved it was cleaned and put back but the process wasn’t without its difficulties. The pump – being the easy part (relatively speaking) was dealt with swiftly and enthusiastically.
Before we knew it, it was on the rig spinning madly and moving fuel again after over forty years.
But the CCU was a different matter.
If the pump is the highly strung chef of the system then the CCU is maître d and perfect waiter rolled into one. It doesn’t spin, nor does it make a noise. Its actions are measured and precise and in that sense it’s way more sophisticated than any pump. Ours didn’t look pretty when we found it but, like the pump, most of it could be reused and soon it lay on the bench in a million bits awaiting painstaking reassembly.
It went back together beautifully too…
Shortly afterwards it was plumbed up in a thousand ways with juice flowing through its veins again.
Those test cell doors are closed when there’s something violent going on. They’re tough and keep the racket inside… I was once in the building when a main fuel pump for a big Trent engine was being put through its paces. The high pressure hose feeding it convulsed angrily, the ground shook and it screamed as though in agony. I remember wondering whether those big engines run silently and the noise you hear when one flies overhead is in fact the pump. No such drama with our little CCU though but the prognosis was bad. Even on shutdown the throttle valve was leaking four and a half gallons of fuel an hour and this, apparently, was a total disaster. I immediately had one of those, ‘don’t know what you’re talking about’ moments because I really couldn’t see the problem but I wasn’t daft enough to say so this time. After all, the engine wouldn’t run on such a small amount of fuel and if it wouldn’t run the fire would go out and the pump would stop and then it wouldn’t leak anything at all. Simple. I nonchalantly dropped the question into a conversation months later to find out what all the fuss had been about to be told that it was all to do with start-up because the leaky CCU would let fuel into the engine immediately it started cranking and this would cause hot starts and overspeeding of the turbine and all sorts of undesirable symptoms so a new throttle valve was needed. This was what stopped the job.
Yep… those minuscule scars meant that the throttle valve was knackered. Where it got interesting is that no one really knew what difference it would make, if any, because in the aerospace world even the tiniest imperfection would have had that part flung skipwards without so much as a backwards glance but as we were trying to save as much as possible and, seeing as the valve lies at the heart of the CCU and is the very one that fuelled Donald’s last gallant effort, it was worth finding out if it would perform despite the damage. But no joy, I’m afraid – it was scrap. Nor was the decision to retire it taken lightly because all sorts of options were explored first. Everything from grinding the sleeve oversize and making a new valve to electroplating and re-grinding both components but none were considered sufficiently reliable. The problem is that the valve and sleeve are match-ground to 0.0001 of an inch. That’s one ten-thousandth of an inch and when you consider that a human hair works out about four thousandths thick you get the idea… The gap (or not) between the sleeve and valve is small enough to force a molecule of jetfuel to turn back. The next part of the story, as I understand it, is that we were heading for the rocks here because having something like this manufactured is a serious business and beyond the usual scope of the corporate freebie department even if you do happen to be an aerospace giant. But by now that indomitable, British determination had taken hold and no one was going to be beaten so the drawing was pulled and to everyone’s amazement it turned out that it had last been visited and updated in 2007. This could only mean one thing… the valve and sleeve had been remanufactured recently and in this case it turned out to have been done for the Indian air force and, better still, the parts had been made by a company in Liverpool called Hartley Precision. Result! But did they have a spare one? Sadly, the answer was no, but they quickly came to our rescue by agreeing to make up a new valve free of charge. There’s just no way to put a value on this sort of support.
Here it is, you could hold these parts in the palm of your hand but if you wanted one, one of the few places on the planet you could try is Hartley Precision and, because of their selfless support, we now had a working CCU – or so we thought. The throttle valve had one more trick up its sleeve. The rebuilt unit went back on test as soon as a rig came available and even that was a case of waiting for a legitimate window in the normal production of systems to keep the world’s air forces and airlines airborne; but to the engineers’ dismay its flow figures only matched the projected data over the first 30% of its travel. A new throttle valve remanufactured to the proper standard yet it wasn’t working, what had gone wrong? Nothing, as it happened, because the gremlin was diagnosed as a numbers problem. The flow rates given in the test schedule were for a pump different to the one fitted to the rig and once this was realised and put right the valve performed faultlessly – phew! In fact, K7’s original pump was fitted to the test rig following an archaeological dig though fifty years of adapter plates back to the one that mounted it and the two parts of the system once more sang in harmony. That was the big bits dealt with and after that making the smaller modules work threw up no problems that I was ever told about. The AFRC…