Dafydd Llewellyn

The impulse coupling is a gadget between the magneto and the engine, containing a powerful spring and a pawl. When the engine is turned slowly, the pawl engages and stops the magneto, about 30 degrees before top dead centre on the firing stroke, so the spring winds up. As the engine passes top dead centre, the pawl is forced out of engagement, so the spring gives the magneto a sharp enough "kick" that it fires. If you are leaning on the propeller when this happens, you're likely to lose at least an arm. As the engine starts to spin, the pawl is held out of engagement by centrifugal force, so the magneto then behaves normally. Jabiru engines do not use magnetos; they have a form of breakerless ignition; however it needs the starter to spin the engine fairly fast.

Required for cars. Not aeroplanes. (yet)

RTFM (for definition, look it up on the web).

What are you actually trying to achieve? Firstly, the indicated airspeed is subject to "position error" (otherwise known as "airspeed system error") - if your aircraft is a certificated type, this error will normally be given in the Flight Manual. Secondly, indicated airspeed is subject to any error in the ASI instrument itself - and the permissible maximum for the instrument to be "servicable" is plus or minus 4 knots. Thirdly, indicated airspeed assumes standard sea-level air density. If you correct for those three things (assuming there is nothing wrong with your airspeed system, like water in the plumbing, leakage, a distorted or misaligned pitot, etc) , you will get True Airspeed. True airspeed differs from groundspeed by the vector sum of the wind velocity. Not by the simple sum, unless you happen to be flying directly upwind or downwind. The net result of all this is that the groundspeeds one calculates using the forecast wind and the normal traditional forms of mechanical computer, is at best a first-order approximation only.

Here are some actual comparative data from aircraft for which I prepared the weight & balance data in the last couple of years. Work it out yourself.

As I said, if people ignore the 3000 microstrain at limit load rule of thumb, carbon fibre does offer considerable weight advantage. From what I see, most modern gliders ignore this; they have restricted fatigue lives accordingly. CF is the darling of the throw-away whitegoods era. The Sirocco, I see, claims compliance with BCAR S - which is one of the design standards that completely ignores fatigue considerations. So, what's its safe life? It may well be a considerable improvement on the original Sapphire, which also ignored fatigue life. The Mk II Ultrabat was almost entirely made from carbon pre-preg; I'm fairly familiar with CF, but I'm NOT into designing throw-way whitegoods. CF can be used - but not in the way you describe - to produce a form of fail-safe structure; I'm using it that way, to comply with FAR 23.573 - but also NOT in the way used in the Beech Starship, which example is rather more relevant than the ones you cite.

Generalisations again! The use of composites - including carbon fibre - in aircraft, started with missiles and gliders, where it was used primarily to increase the structural stiffness to weight ratio; NB stiffness and strength are NOT the same thing. The Pik-20-C was, I think, the first glider to use it, in the main wing spar caps (glider wings are designed more by stiffness than by strength). At that time, there was very little knowledge about its fatigue behaviour, whereas there was a great deal of knowledge about the fatigue behaviour of aluminium alloys. The fatigue life of an airliner structure is a major issue; whereas it is of very little consequence for a missile or a racing car - and in that regard, Oscar is, I think, quite correct; F1 experience has little relevance to an aircraft designer. Gliders were used, to a degree, to obtain some practical experience in its fatigue behaviour in a real-world structure; however by the time a glider wing had sufficient stiffness to avoid aeroelastic problems, it generally had considerable excess strength. That's not the case for short wings on recreational aeroplanes. Its use in recreational aircraft has been facilitated by the guidance material in appendices to the design standard, JAR-VLA (now CS-VLA), which allowed fatigue life testing to be sidestepped provided specified limit-load stress values were not exceeded. However, this basis is not considered to be valid by some fatigue experts, for a number of reasons, not the least being that the standard does not make it clear whether the stress value applies to the laminate as a whole, or to the actual fibres within the laminate. A far more plausible basis is the limit load strain value. (I should perhaps explain - stress is the force per unit of cross-section area of the material; whereas strain is the elongation per unit length of the material that results from the applied stress). A commonly used rule of thumb is that the limit load strain should not exceed 0.003 inches per inch length of the material (3000 microstrain) - that's for E-glass; the corresponding rule of thumb for CF I do not know. The strain relates to the actual fibre stress, not the bulk stress on the laminate as a whole. So, one can consider that recreational aircraft made from CF are in fact still serving as test articles. Not altogether a comforting thought. Alan Kerr did quite a bit of fatigue testing on critical parts of the Jabiru J160 structure, BTW, despite this not being required by the design standard; it contains no CF. The use of a limit-load value for an aircraft relates closely to the flight load envelope. However, the design criterion for an F1 shell, so far as I am aware, is related primarily to impact protection of the driver - and that's largely due to the use of carbon/kevlar mixture, not to straight CF, which tends to disintegrate into lethal shards on impact. This being the case, there is no real correlation between the two. Composite aircraft structure, if designed to keep the limit load strain below 3000 microstrain, does NOT compare particularly favourably with metal structure in regard to its weight; and in any case, the additional safety factor required for composite aircraft structure to cope with its inherent variability, also tends to negate its weight advantage. When fatigue can be ignored, and variability can be dealt with by batch-sampling, as in a missile, CF offers a considerable advantage. Obviously, the major aircraft manufacturers have a great deal of proprietory information on the behaviour of CF composites; and there is some data in Mil-handbook-17, for specific materials, but it is far from being a general cook-book for designing carbon-fibre aircraft structures. The increasing use of it by Boeing etc, presumably reflects more recent knowledge on its fatigue properties - but that ain't in the public domain yet, to my knowledge. The Goulburn STING crash showed just how little occupant protection was provided by the CF structure.

You're correct about apples & oranges; a sailcloth wing is a very different proposition to covering a conventional airframe with Ceconite, with rib-stitching. I wasn't talking about trikes; they're another world. There is a place for all forms of construction; it's hard to beat welded steel tube for the cockpit "crash cell"; weld is the fastest-setting glue of them all. Sheet metal is best for energy-absorbing structure, and where fatigue is not an issue, usually lighter than composite, if they are designed to the same code, because composite has to have an ultimate safety factor about 50 % larger than metal. The article in the latest RAA magazine on carbon fibre adds a new constraint to its use. Composites are best where complex curvature is an advantage (rather than a styling feature). But all generalisations will inevitably have exceptions.

The usable CG range of a simple aircraft is usually around 12% or so of the mean aerodynamic chord of the wing. The MAC of , say, an early Jabiru, is around 1 metre, so the CG range is of the order of 120mm. If the fuel is in the wing and the seating is side by side, the variable load items can be kept inside this range. An Auster J5G has an unusually large CG range, around 9 inches (229 mm) if my memory serves me correctly, but it has a much larger chord. There's no earthly way to keep the major variable loads inside such a range, with a tandem layout, but you can put the optional occupant on the CG and use the essential occupant as a balance mass for the engine etc - which is what De Havilland did for the Moth series. However, that gives the pilot a very poor forward view, except in a glider, where he can be in front, so not popular.

Hmmm. We seem to be drifting away from the Sapphire, onto "how to design a minimim-cost ultralight". I'm not going to go there, but there's one comment I'd like to make: A tandem seating layout, in an aircraft of 540 Kg MTOW, usually demands a larger CG range than a side-by-side layout, unless it's designed to be like a Tiger Moth, i.e. solo from the rear seat. Such an increased CG range necessitates considerably more sophisticated aerodynamic design, especially of the control system, in order to provide acceptable handling at the aft CG limit. It therefore becomes a certification risk item. If you're talking about PA18 MTOWs, it becomes feasible, but the cockpit layout may still be a bit too cramped to meet FAR 23.562 head-strike criteria. There's a reason why all the littlies are side-by-side.

I don't want to try to write a condensed book on aircraft structural fatigue; and no, wood does not fatigue; the principal issue with wood is the reliability of the glue - but if there are metal fittings in the critical load path areas, that's another story. Aluminium alloys fatigue. Most other materials have some sort of "fatigue limit" - either a stress value or a strain value - below which, for practical purposes, they can be assumed to not fatigue. Welded steel structures need to be treated with some caution in areas such as the lift-strut carry-through, where they were not designed by buckling. Brazing, silver soldering or bronze "welding" of aircraft steel structures is a well-known cause of fatigue problems. If you look at most conventional aircraft structures carefully, the likely fatigue spots are generally not too difficult to at least make an intelligent guess at; aluminium-alloy lift-strut end fittings, for example, and undercarriage attachment bolts, and control-system parts that are subject to reversing bending loads, or cyclic loads. Against that, it is all too easy to come up with a form of structure whose fatigue properties are impossible to analyse - and which is likely to be extremely variable in its structural reliability. Such structures attract requirements to prove their long-term integrity, in design standards such as FAR Part 23; but FAR 23 is not used for recreational aircraft, so potentially unreliable forms of structure can creep in. For aircraft aluminium structures, the methods for estimating the safe life are quite thoroughly documented; see FAA AC 23.13A . However, the life can be severely reduced by poor detail design or poor installation practices or by corrosion or weathering. Why do people persist in building aircraft from aluminium alloys? Because they are predictable, to a greater degree than many others. The fact remains that almost all small GA and recreational aircraft are single-load-path structures in the critical areas - so turning a blind eye to fatigue issues is fundamentally stupid.

It doesn't, of course. So this is the way most people will enter recreational flying. This should have been, but was not, recognised by those who wrote the original rules; as a result, the cheap, second-hand aircraft will not be as good as they should have been. Therefore, it's not a good place for impulse buying; get a really thorough inspection of the aircraft by somebody you can trust.

Actually, rag and tube (rag and bone, as George Markey called it) is very labour-intensive. It's a good way for a one-off amateur-builder, because it requires little in the way of tooling; but it's a mistake to try to go that way for a bottom-end production aircraft. For production, it needs good tooling - and the tooling has to be designed to achieve a balance between the cost of production man-hours and the cost of the tooling itself. As a production exercise, the Jabiru is a brilliant balance between these two. The investment in tooling is considerable. For a kit-marketing exercise, the RV matched-hole technique is very clever; but of course the software for the NC machinery necessary to cut, form and drill the sheets is a substantial development cost. So, unless people are prepared to clear out a shed and roll up their sleeves, the bottom line - for a new aircraft - is at this level. If $25K is your limit, and you can't DIY, then go look at sailing dinghies. Sorry, but that's the truth of it.

Well yes, I suppose I am a cranky bugger - especially when I've repeatedly provided that link on this website. It would be good if there were an index to subjects explained somewhere on the website, so people can find these answers, instead of having to ask the same questions again and again. When you read that CASA advisory circular - and I hope a fair few RAA members will also read it - what you will see is that there are a number of purposes for which an experimental certificate can be issued. Let us suppose you have a Cessna with a standard C of A, and you want to modify it. Since it's NOT an RAA aircraft, CAR 42U normally applies, and the mod. must be done in a CASA approved workshop, in accordance with approved data. If it's a new minor mod., you would go to a CASR Part 21M Approved Person (used to be a CAR 35 engineer), and organise with him to supply an approved Engineering Order. If the approval process requires flight testing, you would apply for an Experimental Certificate for the purposes of showing compliance. If it needs some development, you would apply (on the same EC) for a Research & Development permission also. ECs for these purposes are normally limited to 12 months duration. Whilst the aircraft is flying for these purposes, its normal COA is deemed to be suspended. When the testing process is completed, the Engineering Order approved, and the job signed off by the LAME, the experimental certificates expire and the original COA is then back in force. The procedure is also used for a major mod, but in that case the process has to be done as a Supplemental Type Certificate, which brings CASA into the act from the beginning, and normally requires a Certification Plan - see CASR part 21 subpart D. This procedure is also used for the certification flight testing of new aircraft types. You cannot get an "amateur built" EC for a factory-built aircraft, because it will not satisfy the 51% rule. So no, you cannot simply convert a factory-built aircraft to an experimental aircraft in order to sidestep the normal requirements for approval of a modification. As you will see, there are some ten purposes for which an EC may be granted, and nine of them are identical with the FAA Experimental purposes. I suggest you study the Advisory Circular, and then ask me if you still have a question.

For the 427 th (approximately) time, see http://www.casa.gov.au/wcmswr/_assets/main/rules/1998casr/021/021c10.pdf And you do NOT register a VH aeroplane as anything but a VH aeroplane. RAA does not have certificates of airworthiness, so it uses the registration to denote the airworthiness status. CASA does not do this; VH-XYZ remains VH-XYZ whether it is operating on a standard CoA, a Special CoA or an experimental certificate.

See http://www-das.uwyo.edu/~geerts/cwx/notes/chap14/foehn_gap.html

Yes, the Mk II 'bat was all pre-preg, mostly carbon, cured in an autoclave. A revision using vacuum infusion is being considered, I understand, but I don't know how far that's got; I suspect it will depend on the level of interest.

To answer your question: Because it didn't have one. If you bother to look up CAO 95.25 - there used to be a copy in the RAA Technical Manual, if I recall correctly, but I don't know if it's still there - the basis for acceptance of the CAO 95.25 aircraft was a declaration by a CAR 35 engineer that (a) it complied with the very rudimentary requirements of CAO 95.25, and (b) that it contained "No unsatisfactory feature" (or words to that effect). I was the CAR 35 engineer concerned, in the case of the Sapphire, and whilst I certified that it complied with CAO 95.25, I flatly refused to certify to part (b) - and post # 32 shows why. The main fuselage tube on Drifters and Thrusters is also prone to this; it's virtually impossible to avoid it in that form of structure. It was inevitable that this would eventually occur - but CAO 95.25 - in common with a number of other design standards for ultralight aircraft - contained no requirements whatsoever in relation to long-term structural reliability; the people who wrote them considered that aircraft of that ilk would fly no more than about 40 hours a year and would be scrapped for other reasons before they would suffer a fatigue failure. Time has shown that this is not always valid. The fatigue cracking of that sort of rear fuselage normally occurs inside the rear attachment collar from which the tubular tail boom is cantilevered, and so is almost impossible to detect by inspection, before the tube actually fails. The only saving feature is that it normally fails under the upward landing load case, rather than falling off in flight under the download from the tailplane. Yes, I liked Scott Winton; but his designs contained a number of - shall we say, "unconventional" bits of detail design - about which he was very stubborn. And as history has shown, not always with justification. The Sapphire, like the original Drifter and Thruster, was structurally adequate when it was new. How long it remains adequate is completely unknown. That is not what I consider "containing no unsatisfactory feature". These aircraft served their purpose at the time, and we can remember them with some pleasure; but time has moved on, and we need to be realistic about their limitations. TOSG has addressed most of such features in the Thruster, to some extent; and the strut-braced Drifter did similar - though the standard at the time still did not address long-term structural reliability; and indeed most standards for recreational aircraft still do not. The moral of this story? If there is one, it is that people need to have a much better understanding of the way airworthiness requirements are applied to the aircraft that they fly.

That was the Mk 1 Ultrabat in its original form. I flew it in that form, for the purposes of allowing an extension of its Permit to Fly (that was before we had an experimental category). The 532 had a problem with its rotary disc valve, and was later replaced by a hotted-up 583, same as the Mk 2. So the 532 is part of the development history, not part of the Mk 2 aeroplane, which is the version being considered.

See post # 26. I am not sure whether putting the original 95 - 25 version of the Sapphire back into production would be allowable, any more than the wire-braced Drifter. If it IS allowable - which I doubt - the Rotax 447 is no longer available - so the design would have to be significantly changed - which I think would eliminate any possibility of a "grandfather clause". Even more so if it went to electric power. Unless there is a "grandfather clause" available, then there are only two ways to manufacture a recreational aircraft as a factory-built product: Firstly, by gaining a Production Certificate (see CASR Part 21 subpart G) - for which the pre-requisite is a Type Certificate. So the aircraft would have to undergo a process that was not required of the original - and which it would not have been able to pass. The other option is LSA - but that would require compliance with the ASTM standard; and unless the manufacturer already held a production certificate, that requires that he has the services of competent professionals. Either way costs a lot of money; the certification process (or the ASTM compliance process) does not cost half as much if the aeroplane is a single-seater - it costs much the same for one seat as it would for anything up to about six seats. So either way is, for all practical purposes, not financially viable. The only way would be for somebody to put out kits for home building under CAO 95.55 - 19 registration. So this thread is really all wishful thinking.

Non-viable commercial product. It's not sufficient for it to be a damn nice little aeroplane, unfortunately, or there would be a lot more of that ilk around. For exactly the same reason, nobody is going to set up a factory to build a $50K aeroplane as a certificated factory-built product.

Nearly; because they couldn't sell them at a price that made the effort worth-while. I have no idea whether they marketed them at Natfly.

Think, now - Why do you imagine people stopped building them? To correct another misapprehension - the Sapphire was one of the four original CAO 95.25 aircraft (the others were: Lightwing, Thruster, Drifter). So it does NOT have a Type certificate.

What does the Rotax 532 have to do with it? The Sapphire, as I recall, had a Rotax 447. The Ultrabat Mk II had a Rotax 583 (yes, five eight three).

You may as well get the story right; See http://theultrabat.com/