Jabiru Prop failure

[Oscar]

RAA-class aircraft (or indeed any aircraft) that require Belleville washer packs to achieve the necessary flange/propellor pressure to keep it within the friction coefficient for operation under normal operating conditions, is information available to the operator. Should you be wrenching any of those, you'd need to do it in accordance with the manufacturer's requirements. The manufacturers do not just derive such arrangements by thrusting a wet thumb in the air, they test (exhaustively, and in a fully documented way) the results of testing in calibrated facilities. You may well be amazed that 'I've always done it this way, and it hasn't failed me yet' is not considered adequate proof for certificating authorities. No doubt they do not have your experience to draw upon. (edited ...moderator)

 

 

[Oscar]

deleted - moderator

 

 

[Dafydd Llewellyn]

Thanks Dafydd - standard engineering practice ...As far as the props go though, selection of timber types to employ, according to those different timbers' varying propensity for water absorption and also proper moisture reduction prior to the use of modern coatings which deeply penetrate the surface of the timber and prevent moisture changes, goes a long way further to preventing these kinds of failures than the use of bolt tension retainers ever will. A now deceased friend (not a prop failure!) was a timber boffin and prop manufacturer and he conducted endless tests on timbers, he mostly chose mountain ash for props, sometimes alternately laminated with birch (I think) but wouldn't ever use hoop pine even though it was more easily available, because he found that hoop pine very readily and rapidly absorbs moisture.

 

The damage seen on the OP's post is typical of crushing resulting in loose bolts once the timber dries and shrinks again, and is unlikely to be found on a prop made of mountain ash, for example. Even though that timber might have a lower crush resistance, it wouldn't absorb moisture as readily in the first place, so it wouldn't swell as much and cause the crush problem in the first place. I think it's important to keep in mind that the vast majority of timber props do not use spring washers of any kind for mounting them. Your Seabird example might well be a good exception to prove the rule, given that the aircraft may be deployed into a humid marine environment but during operations the prop is subject to the drying effects of the engine's cooling exhaust. I would think that attention to the type and condition of the prop's coatings would be especially important for that application.

What is your basis for believing that Mountain Ash would be superior to Hoop Pine? Is it factual or anecdotal?

 

The data I have managed to find on the web show (from the same source) a shrinkage for timber after post-seasoning re-conditioning as follows:

 

Mountain Ash (E. Regnans): Tangential 0.36% Radial 0.23%;

 

Hoop pine (Araucaria Cunninghamii): Tangential 0.23% Radial 0.185%

 

(Most shrinkage data refer to shrinkage from green to 12% MC - and E. Regnans is described as having a notably high shrinkage in that phase - the closest thing I can find to shrinkage post-seasoning is the results above)

 

On this basis, the total dimensional change for Hoop Pine would seem to be about 2/3 that of Mountain Ash; and the ratio of tangential to radial change for Mountain Ash is about 1.26 times that of Hoop Pine. So on this basis (which I admit is not the whole story) it would appear that Hoop Pine is the better of the two.

 

The real question may be the rate at which moisture change occurs - and that is affected by the surface treatment more than by the timber species, I think. The glass sheathing of the Jabiru prop would seem to me to be a pretty good moisture barrier, provided it is intact.

 

 

[Dafydd Llewellyn]

And my Allsize for a Lightwing or Storch ?..................

Do Allsize hold an APMA to manufacture propellers?

 

 

[Dafydd Llewellyn]

Daffydd,thanks for the info and I say you may damn well be correct 'ole bean', the scenario you present of the torque slackening off fits the symptoms to a T. the aircraft has not flown a lot of hours since the prop torque check was c/o at 100 hrly but has sat for a while in inclement weather. The drive face hasn't sanded off but compressed like it swelled and pressed the drive flange into it I am suspect that it has since dried out and shrunk causing the torque loss, if the photos were clear you could see the wood has opened up rather than cracked.

Well, if that's the case, you had better check the security of the attachment of the ring gear assembly to the back end of the crankshaft - because it's likely to have been damaged in the process of trying to act as the flywheel in lieu of the propeller.

 

 

[Head in the clouds]

.... Is it factual or anecdotal?

Both I guess, my friend (not Adams) worked closely with CSIRO in Victoria, they documented a lot of his test data for the emerging exotic forestry/timber industries in the 1970s/80s. IIRC he was directly responsible for Paulownia timber becoming a farmed timber in Australia. There is a lot of difference between the suitable applications for timbers that may be suitable and approved for aircraft use in general. I don't know much about timber in the scheme of things but I personally wouldn't use a softwood (hoop pine) for a propellor. It may well have reasonable properties in a static application but softwoods do have quite different properties in terms of rate of moisture change and their failure modes, not to mention the rotting process.

 

The glass sheathing of the Jabiru prop would seem to me to be a pretty good moisture barrier, provided it is intact.

This is a dangerous misconception. Most laminating resins readily allow moisture transfer and at a surprisingly rapid rate, hence the need for gelcoats on boats. I think Jabirus are manufactured using epoxy resins so I'd guess that epoxy is used to sheath the prop and epoxy is one of the worst for moisture transfer. Boats that have been repaired using epoxy resin (for its adhesion qualities) following osmosis damage, or steel/aly boats which are painted with high-build epoxy (also for adhesion) must then be painted with polyurethane as a moisture and anti-chalking barrier. The OP's prop doesn't appear to have been painted ...

 

Well, if that's the case, you had better check the security of the attachment of the ring gear assembly to the back end of the crankshaft - because it's likely to have been damaged in the process of trying to act as the flywheel in lieu of the propeller.

Good point.

 

 

[deadstick]

Th

 

Well, if that's the case, you had better check the security of the attachment of the ring gear assembly to the back end of the crankshaft - because it's likely to have been damaged in the process of trying to act as the flywheel in lieu of the propeller.

Thanks for the advice and I'm already on it. Have new flywheel bolts as well as prop on the way.

 

 

[Dafydd Llewellyn]

Both I guess, my friend (not Adams) worked closely with CSIRO in Victoria, they documented a lot of his test data for the emerging exotic forestry/timber industries in the 1970s/80s. IIRC he was directly responsible for Paulownia timber becoming a farmed timber in Australia. There is a lot of difference between the suitable applications for timbers that may be suitable and approved for aircraft use in general. I don't know much about timber in the scheme of things but I personally wouldn't use a softwood (hoop pine) for a propellor. It may well have reasonable properties in a static application but softwoods do have quite different properties in terms of rate of moisture change and their failure modes, not to mention the rotting process.

OK, Ellis Walker (Perfectus Airscrew) I guess. I knew Ken Adams and Barrie Bishton personally; never met Ellis. They were the only ones holding CAR 30 approval to manufacture propellers in those days - and I recall Ken Adams had a liking for Mountain Ash. I'm aware of the rotting propensities of both Hoop Pine and Coachwood - tho they are both listed in CAO 108.29 (quite a surprising number of species are). Almost all the timbers suitable for aircraft structure are not particularly durable from this standpoint - which is why one should not leave wooden aircraft components standing out in the weather. Hoop Pine has surprisingly good mechanical properties for its weight; it's a permitted substitute for Sitka Spruce for aircraft structure (though I'd never consider Spruce in a propeller application), but has properties similar to Birch. Its high crushing strength means one can use a higher clamping pressure, so it can accept roughly 33% higher peak torque for a given propeller flange size than can Mountain Ash. Also the higher density of the latter makes for a heavier propeller, which is a disadvantage in regard to the gyroscopic loads on the crankshaft, and does nothing to reduce the peak instantaneous torque, which may be why Jabiru use it. I quite agree about the higher porosity of the microstructure of softwoods. I wonder whether Sassafras is classified as a softwood?

 

This is a dangerous misconception. Most laminating resins readily allow moisture transfer and at a surprisingly rapid rate, hence the need for gelcoats on boats. I think Jabirus are manufactured using epoxy resins so I'd guess that epoxy is used to sheath the prop and epoxy is one of the worst for moisture transfer. Boats that have been repaired using epoxy resin (for its adhesion qualities) following osmosis damage, or steel/aly boats which are painted with high-build epoxy (also for adhesion) must then be painted with polyurethane as a moisture and anti-chalking barrier. The OP's prop doesn't appear to have been painted ...

 

Thanks - that's useful information.

 

Good point.

[Guest Andys@coffs]

Gee wiz, and there I was thinking J are at it again, finding the cheapest wood possible, and one that it was infered was a poor choice for a tooth pick let alone a propeller, and before you know it with a few informed posts and it suddenly appears as though they might have actually used some engineering in their choices.......Im astounded, fancy an OEM in Australian, that went to the expense of getting their aircraft "primary category" certification actually using engineering principles......

 

There's a good post that has just recently been made on the J website that coveres this aspect here:- http://www.jabiru.net.au/images/Blurb_fuel_winter.pdf The file name doesnt seem relevant, but there is a section in the document called "What is an LSA" interesting reading!

 

Andy

 

P.S...not withstanding all that, the engine reliability is still an issue and one that I pressume their liability lawyers are suggesting needs to be fixed, but without disclosures that can be used against it....

 

 

[Oscar]

so I'd guess that epoxy is used to sheath the prop and epoxy is one of the worst for moisture transfer.

May I refer you to: Assessment of Research Needs for Wind Turbine Rotor Materials Technology ( 1991 ) http://www.nap.edu/openbook.php?record_id=1824&page=54 :

 

Moisture effects are modified for wood/epoxy laminate to the extent that the epoxy seals out liquid water entirely and is a fairly effective barrier to the passage of moisture in the form of water vapor. As a result, a large structure will not respond to the short-term moisture fluctuations in its environment, but will instead come to equilibrium with the average humidity over a period of months to years (Figure 3-7)

 

You may also not be aware that gelcoat is water-permeable, and the typical 'blistering' bubbles on polyester hulled boats is almost always a result of the passage of moisture through the gelcoat reacting with uncured areas of resin due to poor mixing techniques / layup in high-humidity conditions ( Swanson yachts were particularly prone to this), with the bad osmosis softening areas being a result of wicking of the water down the matrix/resin interface.

 

 

[Guernsey]

A lot of very good information which ultimately could save lives, well done everyone and thanks for posting.

 

Isn't this a fantastic forum.

 

Alan.

 

 

[facthunter]

YES. Nev

 

 

[Head in the clouds]

May I refer you to: Assessment of Research Needs for Wind Turbine Rotor Materials Technology ( 1991 ) http://www.nap.edu/openbook.php?record_id=1824&page=54 :Moisture effects are modified for wood/epoxy laminate to the extent that the epoxy seals out liquid water entirely and is a fairly effective barrier to the passage of moisture in the form of water vapor. As a result, a large structure will not respond to the short-term moisture fluctuations in its environment, but will instead come to equilibrium with the average humidity over a period of months to years (Figure 3-7)

 

You may also not be aware that gelcoat is water-permeable, and the typical 'blistering' bubbles on polyester hulled boats is almost always a result of the passage of moisture through the gelcoat reacting with uncured areas of resin due to poor mixing techniques / layup in high-humidity conditions ( Swanson yachts were particularly prone to this), with the bad osmosis softening areas being a result of wicking of the water down the matrix/resin interface.

I'm not sure what you're trying to get at. Your pasted statement in italics above states that epoxy seals out liquid water entirely - that is not correct. Take a block of balsa wood, coat it with laminating epoxy and allow it to cure. Weigh it. Submerge it in water for a while and then weigh it again. Then you will have a practical answer rather than a theoretical one.

 

The statement above goes on to agree that water vapor (humidity) does pass through epoxy, so it is clear that an epoxy/glass sheathed prop will vary in moisture content according to changes in humidity. I don't think anyone would classify a prop as a 'large structure' so the concept of it reaching equilibrium with the average humidity over a number of years isn't relevant in that example.

 

Yes, while I was in the north my Company was engaged to repair quite a number of yachts and power vessels with osmosis problems, so I do know that gelcoat is very mildly permeable but nowhere near to the extent that the laminating resins are. For that reason the wise GRP boat owner who mostly keeps his boat in the water will paint his hull with more than just antifoul below the boot-topping.

 

In my experience the blistering damage associated with osmosis has nothing do with any reaction of water with uncured resins. Most boats are built using polyester resin and as long as the catalyst has been added in the first place and the lay-up has then been properly rolled with washer or broached laminating rollers I have never seen areas of uncured resin. Boats with osmosis problems are widespread and I am sure the majority of boat-builders are quite capable of mixing up some resin and catalyst in a bucket for hand lay-ups, or adjusting a chopper gun, so I would think that the industry's generally accepted version of 'dry lay-up' is more likely to be the truth of the matter. Try mixing water and uncured resin, it does not expand so it would not cause blistering. The blistering is a result of pressure caused by gases generated by bacteria which are able to grow within the matrix due to the presence of the water, hence the liquid found within the blisters is dark coloured and the gas has an unpleasant odour. Rather than uncured resin I think you will find that the problem results from insufficiently wetted out lay-ups (aka resin-poor or resin-starved) and that allows the water to wick along the free glass filaments.

 

 

[Oscar]

The point of quoting that paper was that the rate of moisture transmission through an epoxy covering is sufficiently slow that normal prop bolt torque checking should be entirely adequate to prevent excessive crushing of the hub from normal humidity changes.

 

I agree that epoxy is not impervious to moisture; however I have referenced a very thoroughly-researched paper that places the effects in perspective. You stated that 'epoxy is one of the worst' for moisture transmission; I ask you to validate this statement by reference to equivalently authoritative documentation.

 

While it is not germaine to this thread, I have also been involved in osmosis remediation of boat hulls, including for Swanson, Adams, Benetau and Riveria boats. As a result of seeing the osmosis damage to those boats, I would personally never use gel-coat as a water-barrier medium, I would use a 2-pot polyurethane finish every time. The moisture ingress on the Riveria, in particular, through the gel-coat required 2mm of the outer laminate to be mechanically ground off and over 6 months of drying time required before a new outer later of epoxy laminate could be applied. I have personally sheathed a double-diagonal Kauri-planked 48' foot ex-trawler in epoxy laminate ( around 250 litres of WEST 205 required). That boat is still doing just fine, thank you, 15 years later - with its 2-pot poly paint barrier coating in excellent condition.

 

 

[Old Koreelah]

Very interesting discussion, gentlemen. I believe that Queensland Hoop Pine timber is being used in number of countries for aircraft construction, as well as Hoop Pine plywood.

 

 

[djpacro]

I did some engineering approvals for a number of wooden props many years ago, straining the memory without going into my archives for the reports. I did a structural analysis rather than overspeed test. From memory, the hub was typically more critical than the blades in determining prop thickness. Also from memory, we used Qld Maple. Then again, my analysis was somewhat conservative - good in that I am not aware of any issues with their operation from VW to Gypsy to small Continentals - bad in that when I wanted to do a 200 hp Lycoming I realised mine would not be as good as a Hoffman, ran a Sensenich on it for a while but the Hoffman was far better.

 

I knew Ellis pretty well, interesting business to visit (wasn't his props I did approvals for).

 

Managed to catch up with Sensenich prop designer in '96 to compare notes on that prop - short story is that matching the prop to airframe and engine was not easy.

 

Remember when Ralph did his crusade on wood? http://www.casa.gov.au/wcmswr/_assets/main/fsa/1998/jul/34-35.pdf

 

 

[Head in the clouds]

The point of quoting that paper was that the rate of moisture transmission through an epoxy covering is sufficiently slow that normal prop bolt torque checking should be entirely adequate to prevent excessive crushing of the hub from normal humidity changes.

Which is effectively what I said quite a number of posts ago when I mentioned that the use of Belleville washers was not the normal practice and that the vast majority of timber props do not use or need them in their mounting. However I went on to say that if you don't want any significant moisture content change then you need to use a polyurethane coating because epoxy does allow some moisture to pass through. With a decent PU coating the moisture content shouldn't change much and so there shouldn't be a need for Belleville washers. I ran my drifter for years with the same prop and only ever had to re-torque the prop bolts for each engine and gearbox re-build.

 

I agree that epoxy is not impervious to moisture

Then, as I said before, I cannot see what your point is.

 

however I have referenced a very thoroughly-researched paper that places the effects in perspective. You stated that 'epoxy is one of the worst' for moisture transmission; I ask you to validate this statement by reference to equivalently authoritative documentation.

I don't need to go data mining, I have seen the results of quantitative tests. I described it to you before. If you want to see which is worst for yourself, do the balsa test with polyester, vinylester and epoxy (and any other laminating resins you wish). In the tests I saw, the epoxy was about twice as bad as any other resin ... in my book actual practical tests beat the hell out of theory every time. DL's example of the differences between manufacturers' documented figures for deflection and load values of disc-spring washers further demonstrates that. And - it's well known in the boat industry that repairs to polyester hulls are best made using epoxy for better adhesion but coating with PU is essential to prevent moisture transfer, however if you do repair with polyester or vinylester the coating type is not as important.

 

While it is not germaine to this thread, I have also been involved in osmosis remediation of boat hulls, including for Swanson, Adams, Benetau and Riveria boats. As a result of seeing the osmosis damage to those boats, I would personally never use gel-coat as a water-barrier medium, I would use a 2-pot polyurethane finish every time. The moisture ingress on the Riveria, in particular, through the gel-coat required 2mm of the outer laminate to be mechanically ground off and over 6 months of drying time required before a new outer later of epoxy laminate could be applied. I have personally sheathed a double-diagonal Kauri-planked 48' foot ex-trawler in epoxy laminate ( around 250 litres of WEST 205 required). That boat is still doing just fine, thank you, 15 years later - with its 2-pot poly paint barrier coating in excellent condition.

Gelcoat isn't specifically used as a water barrier although it serves the purpose for boats that aren't kept in the water full-time. Mainly gelcoat is used for its colour, appearance, prevention of surface blemishes when molding, and that it stops a weave (or CSM) pattern being visible on the surface. Again, as I said before, the wise GRP boat owner who keeps his boat afloat will coat the hull below the boot-topping with coatings other than just antifouling i.e. 2 pack PU.

 

It sounds like the osmosis affected hulls you were involved in weren't much affected at all. Unless the problem is detected very early on, and that's rare, they generally require almost all of the original laminates to be removed, in some cases/areas more than 25mm thickness rather than the 2mm you describe. And even in the tropics it does take months of alternately fresh water washing and re-drying with heat lamps before rebuilding and fairing the hull from the outside. Perhaps the uncured resin thing you described wasn't osmosis?

 

Anyway it's been a nice discussion but we're off topic so I'm out now, cheers.

 

 

[Oscar]

I don't need to go data mining, I have seen the results of quantitative tests.

For the sake of information of value to forum members, can you please provide the results? I personally rely on scientifically justifiable testing, rather than unsupported assertions.

 

 

[Dafydd Llewellyn]

I'd certainly like to get to the bottom of the epoxy permeability discussion. I'd always thought epoxy was much better than polyester, but I've never studied the problem in detail. I regard 'Gell coat' as two four-letter words. I know that polyurethane is better than any of them; in fact it's so good that people have to go to considerable trouble to dry out airliner structures before painting them with polyurethane, or the water it traps under the paint causes trouble (corrosion in skin laps & the like). The water leaches through the paint and escapes into the stratosphere, with all other types of paint - but not with polyurethane.

 

However, I think the situation with timber propellers is that any finish process simply reduces the rate of change of moisture content; it does not do anything much to alter the end state, just the time it takes to get there. A good finish reduces local stresses due to uneven moisture distribution within the timber whilst moisture is moving into or out of the wood (a major source of warpage and "air cracking") so a good finish process is valuable for that reason. Timber shrinks as it dries and swells as the moisture content increases, and short of making propellers with densified impregnated timber in the hub region, that's the fundamental characteristic of the material and there is no practical way to alter that. Nominally seasoned timber contains about 8% ~ 12% of water, and that water is essential to the structural integrity of the lignin - so there must always be some water within the timber. The problem is to keep the % as constant as possible.

 

No form of finish alters the need for the hub attachment design to be able to cater for swelling and shrinkage; nor does it reduce the amount of swelling or shrinkage that must be catered-for - and neither does a change in timber species; the one with the least total shrinkage is the best, in the long haul - assuming a finish that keeps the rate of change down to something that does not generate too much internal stress.

 

If you do not cater for this fundamental characteristic of the timber, then there is no escaping the fact that you will be re-tightening the prop bolts every now and then - but you will never be loosening them to allow the timber to swell when the humidity is high. So the timber gets a bit more crushed every time it's exposed to steamy conditions for a while, and the slack gets taken up by somebody diligently re-tensioning the bolts - so the flange and the clamp plate become progressively ever more deeply embedded in the timber as the total distance between them progressively decreases - until somebody says "enough" and the propeller gets a clock in its navel. If you fail to re-tighten the bolts at the right point in the moisture cycle, you risk losing the propeller. So either you ruin the propeller gradually by diligently re-tensioning the bolts - or catastrophically, by allowing them to become loose.

 

This is a fundamentally unsatisfactory situation; we cater for ignition failure and fuel pump failure by duplication of those systems. We cater for induction system icing by having an alternate-air/hot air system. Why do we tolerate a half-baked design in regard to the propeller attachment? (Because we're fundamentally lazy - but that's no excuse).

 

So, whilst the debate about finishes is most interesting - and I want to get to the bottom of it - it is peripheral to the fundamental issue, which is that wood propellers really, really need elastic hubs that can allow them to swell & shrink as they must, whilst maintaining the clamping force necessary to transmit the torque. It should have become standard prectice (nearly did, in fact) by 1940 - but since then, people have forgotten. We're re-inventing the wheel; and therefore it's time to complete the development of wood propeller technology by addressing this issue. Making propellers using HYDULIGNUM in the hub region is one answer - but proper use of Belleville washers is a far less costly one.

 

Wood is, for practical purposes, unaffected by structural fatigue. It has high internal damping, so it's pretty resistant to resonant vibration modes. Wood propellers are both affordable and reliable, and can be treated as fundamentally a consumable, so putting the effort into the hub rather than the propeller makes sense. Belleville washers are ten bob a half-bushel case, so why not get serious about them?

 

 

[facthunter]

They must have been pretty good at one point as the Schneider trophy engines and even the early Hurricanes ran wood props and they had well over 1,000 HP, so it might be worth your while to revisit the techniques employed back then.

 

You can remove moisture from wood by heating and subjecting it to a vacuum. Model aeroplane props we used to use were done that way and had plastic impregnated into them afterwards. Nev

 

 

[Dafydd Llewellyn]

They must have been pretty good at one point as the Schneider trophy engines and even the early Hurricanes ran wood props and they had well over 1,000 HP, so it might be worth your while to revisit the techniques employed back then.You can remove moisture from wood by heating and subjecting it to a vacuum. Model aeroplane props we used to use were done that way and had plastic impregnated into them afterwards. Nev

Yes, the vacuum impregnation process is something I want to experiment with - but it adds to the cost of the propeller, probably more than adding some Belleville washers to the hub. Also, it is needed mainly in the hub region, it is disadvantageous to increase the density of the timber further outboard. As far as I can ascertain, the penetration into the timber is still fairly limited, so either you get an epoxy/timber "crust" around unmodified wood in each laminate, or you have to build-up the propeller from peeled veneer laminate thicknesses. It's questionable whether you have a "wood" propeller after that; under FAR 35 it would be classified as a form of composite propeller. If you are going to certificate a composite propeller, you may as well start with a quality-controlled man-made fibre. So you lose the fundamental advantages that allow wood propellers to be affordable.

 

In a temperate climate, the natural elasticity of the propeller timber provides some "spring" and you can bet people were keeping a close eye on prop bolt tensions for the Schneider Trophy aircraft, as well as early Hurricanes, Spitfires, and ME 109s, all of which used FP wood propellers. The propeller finish most in use in England was called "chrystofin" and I think it was originally a German invention - some sort of cellulose nitrate brew, I suspect - Ken Adams was still using it in the 1960s; so they certainly understood the need for a moisture-resistant finish - especially so because the propellers were mainly glued using casien glue.

 

I want to see FP wood propeller stop being a maintenance-intensive item, in regard to their bolt tensions. I set out to achieve that, using Belleville washers - and it worked. Take it or leave it.

 

 

[Head in the clouds]

For the sake of information of value to forum members, can you please provide the results? I personally rely on scientifically justifiable testing, rather than unsupported assertions.

I think you miss the point again - I saw the test performed, it was not a document, it was a real, practical test conducted in Berwick, Victoria, and for a completely unrelated purpose, interestingly. As far as I know it was not documented it was done to prove a point, just like this one. I have told you how to conduct this very simple test and I have told you the results that I saw from the test I saw conducted. You say "I personally rely on scientifically justifiable testing" when in fact you quote a paper related to theory of 'large structures'. If you want results from testing, then test it! It's very simple to do. If you want to accelerate the test then heat the water. If you want to accelerate the (semi-permeable membrane) test further, then use rock salt instead of balsa to increase the concentration difference across the membrane. This is basic schoolboy stuff that yields accurate real-world results Oscar, not 'unsupported assertions'.

 

... there is no escaping the fact that you will be re-tightening the prop bolts every now and then - but you will never be loosening them to allow the timber to swell when the humidity is high. So the timber gets a bit more crushed every time it's exposed to steamy conditions for a while, and the slack gets taken up by somebody diligently re-tensioning the bolts - so the flange and the clamp plate become progressively ever more deeply embedded in the timber as the total distance between them progressively decreases - until somebody says "enough" and the propeller gets a clock in its navel. If you fail to re-tighten the bolts at the right point in the moisture cycle, you risk losing the propeller. So either you ruin the propeller gradually by diligently re-tensioning the bolts - or catastrophically, by allowing them to become loose....

A very good point and on the surface it is rather a worry. But is that the difference between theory and practice? In the real world props last for many years - I mean many, many years. Example - A well known Lightwing with a Rotax 912 80hp engine hit 15yrs old recently and because it is sometimes used for training ops it had to have a new engine. The engine was about 100hrs short of TBR anyway, so is that about 1900hrs? Maybe 1400hrs. Still had the original prop which is in fine condition. The prop plate and hub show the normal impression marks but they haven't noticeably got any deeper since the first time the prop was removed. That plane has operated in the far north coastal, inland in the desert and right down into Tasmania, so has experienced the full gamut of humidity range.

 

My Drifter was similar, timber prop 2000hrs/eight years old, no cracks, no deep impressions of the prop plate or hub. No charring from burnishing either - and no friction plate. Maybe that's the smaller horsepower we're dealing with. But isn't that compensated for by smaller bearing surface area, smaller bolts and less overall tension?

 

I also remember a discussion with a Tiger Moth operator who was devastated when he lost his prop (the engine of the crankshaft broke off, not the prop hub) because he'd been flying the same prop for half his life.

 

A mate had a seaplane (constantly getting wet and dry, salt and fresh) with an 0-320 and a timber prop - no compression washers, no splits, no working or deep impressions.

 

It would be logical that the only way you'll get deep impressions in the timber hub is if the wood is allowed to swell and contract regularly, and that swelling causing crushing as you mentioned. So since we're not actually seeing that crushing/impressions perhaps it doesn't actually swell as much as might be being suggested. You mentioned earlier that you couldn't actually see any impression in the OP's hub of the failed prop, saying it might have been sanded down. By your own example if the bolts are sufficiently torqued when the prop is first mounted (and let's say that's done when the timber is at a low moisture content) then the clamping force itself should prevent the wood from taking up moisture in the hub region when the ambient humidity increases, and so there shouldn't actually be any crushing ... I think that testing wood's moisture absorption characteristics while under pressure could be a valuable exercise for this discussion.

 

Having given this much more thought, I am now beginning to wonder whether the OP prop was perhaps originally sheathed when not sufficiently low in moisture content and then installed. Subsequently the wood may have lost moisture and contracted and the (axial) edges of the sheathing could have prevented proper clamping force being applied to the timber heart of the hub when/if the bolts were re-tensioned. That would depend on how large the drive flange and prop plate were in relation to the hub diameter of course. If the flange and plate were significantly smaller than the hub OD then it'd seem more like the bolts didn't actually get re-tensioned early in the piece, when the prop first contracted after the initial installation. Just my 2c.

 

 

[jetjr]

I had the of the Jabiru props with little success. 3300 is pretty smooth running and prop added vibration not absorbed it. 2200 is a different story

 

Warping and vibration were major problems, had to be tracked every 50 hrs or so to keep running true.

 

Jabiru looked after me well but I went to composite type and wont look back.

 

I do believe the humidity is a factor and we saw seasonal changes in prop so moisture was moving in and out. Coming from the coast to dry inland was my take on what was going on in addition to highly variable humidity we see away from sea.

 

Just before one went back it was accurately checked for warping and it was seen one tip was 5-10 mm out from the other. I doubt it would have left maker this poorly made.

 

My understanding is the belleville stack keeps prop tensioned until next adjustment is done. The point being to stop fretting on the drive lugs which was an old problem I believe. Rusty looking prop faces indicated heat and movement.

 

 

[Dafydd Llewellyn]

I think you miss the point again - I saw the test performed, it was not a document, it was a real, practical test conducted in Berwick, Victoria, and for a completely unrelated purpose, interestingly. As far as I know it was not documented it was done to prove a point, just like this one. I have told you how to conduct this very simple test and I have told you the results that I saw from the test I saw conducted. You say "I personally rely on scientifically justifiable testing" when in fact you quote a paper related to theory of 'large structures'. If you want results from testing, then test it! It's very simple to do. If you want to accelerate the test then heat the water. If you want to accelerate the (semi-permeable membrane) test further, then use rock salt instead of balsa to increase the concentration difference across the membrane. This is basic schoolboy stuff that yields accurate real-world results Oscar, not 'unsupported assertions'.

 

 

A very good point and on the surface it is rather a worry. But is that the difference between theory and practice? In the real world props last for many years - I mean many, many years. Example - A well known Lightwing with a Rotax 912 80hp engine hit 15yrs old recently and because it is sometimes used for training ops it had to have a new engine. The engine was about 100hrs short of TBR anyway, so is that about 1900hrs? Maybe 1400hrs. Still had the original prop which is in fine condition. The prop plate and hub show the normal impression marks but they haven't noticeably got any deeper since the first time the prop was removed. That plane has operated in the far north coastal, inland in the desert and right down into Tasmania, so has experienced the full gamut of humidity range.

 

My Drifter was similar, timber prop 2000hrs/eight years old, no cracks, no deep impressions of the prop plate or hub. No charring from burnishing either - and no friction plate. Maybe that's the smaller horsepower we're dealing with. But isn't that compensated for by smaller bearing surface area, smaller bolts and less overall tension?

 

I also remember a discussion with a Tiger Moth operator who was devastated when he lost his prop (the engine of the crankshaft broke off, not the prop hub) because he'd been flying the same prop for half his life.

 

A mate had a seaplane (constantly getting wet and dry, salt and fresh) with an 0-320 and a timber prop - no compression washers, no splits, no working or deep impressions.

 

It would be logical that the only way you'll get deep impressions in the timber hub is if the wood is allowed to swell and contract regularly, and that swelling causing crushing as you mentioned. So since we're not actually seeing that crushing/impressions perhaps it doesn't actually swell as much as might be being suggested. You mentioned earlier that you couldn't actually see any impression in the OP's hub of the failed prop, saying it might have been sanded down. By your own example if the bolts are sufficiently torqued when the prop is first mounted (and let's say that's done when the timber is at a low moisture content) then the clamping force itself should prevent the wood from taking up moisture in the hub region when the ambient humidity increases, and so there shouldn't actually be any crushing ... I think that testing wood's moisture absorption characteristics while under pressure could be a valuable exercise for this discussion.

 

Having given this much more thought, I am now beginning to wonder whether the OP prop was perhaps originally sheathed when not sufficiently low in moisture content and then installed. Subsequently the wood may have lost moisture and contracted and the (axial) edges of the sheathing could have prevented proper clamping force being applied to the timber heart of the hub when/if the bolts were re-tensioned. That would depend on how large the drive flange and prop plate were in relation to the hub diameter of course. If the flange and plate were significantly smaller than the hub OD then it'd seem more like the bolts didn't actually get re-tensioned early in the piece, when the prop first contracted after the initial installation. Just my 2c.

Lot of "Ifs" and "might bes" there; we could postulate forever and it would achieve nothing useful.

 

You are quite correct that an aircraft that is (a) not based too close to the sea and (b) not left outside when not in use and © not taken from its base to a location where the average humidity is much lower, can survive indefinitely on the elastic compression capability of the timber. However if you happen, for example, to be based near Darwin and decide to escape the wet season by flying to Alice or Adelaide, or expose the propeller to a similar change in environment, you are likely to exceed the capability of the elastic compression of the timber, and that's when things are likely to result in permanent compression damage to the propeller. I've operated an Auster as a glider tug, out of Bathurst, and whilst it was a Gipsy major - which does have belleville washers - the propeller never gave any trouble. The LAME that did the maintenance would have checked the bolt tension every 100 hours, as required by CAR Schedule 5 Part 2 para 12©.

 

My point is that whilst the traditional wood propeller installation works most of the time, it does not work all the time. This can be fixed quite simply. Propeller failure is an ugly thing, and I see this as a completely unnecessary defect in aircraft design. I've pointed out how it can easily be fixed. If you do not want to do so, don't bother justifying that to me.

 

 

[Head in the clouds]

This can be fixed quite simply... I've pointed out how it can easily be fixed....It's a fairly critical piece of engineering design, and each installation has to be treated separately. NOT something to be designed in the backyard

Your words - so is it simple or not? I side with Ed Heinemann, particularly the 'simplicate' bit, I've never found good reason to complicate something that works just fine as it is. And I spend my days rectifying structural engineers' unnecessary over-complication of just about everything they put their mind to, so that certainly keeps me skeptical.

 

If you do not want to do so, don't bother justifying that to me.

Most people mellow as they age Dafydd.