Heads up: Kaurit glue deterioration in wooden aircraft

John Green wrote: Coefficients of expansion and contraction differ with the type of timber and the influence of temperature and humidity. Any adhesive used for joint bonding on an aircraft should possess – among other characteristics – a gap filling and elasticity function. If some of the earlier adhesives tended to lack these qualities, then maybe a repetitive cycle of expansion and contraction could account for the failure of joint bonded elements critical to the integrity of the airframe.

You are so right. Problem is that Kaurit W with the bakelite filler added has a joint filling capacity of up to 2.5 mm. It wouldn’t be certified if it had no gap-filling properties… And when applied correctly it is absorbed by the wooden contact surfaces, and the resulting fully cured joint is quite strong and elastic. So in theory quite suitable. However, the process of storing, mixing, applying, clamping and curing the glue joints is somewhat complex when compared to say Aerodux. And with it it is prone to errors, which may only now show when the glue fails with age. Worrying, but with the correct inspection techniques it is not a problem to catch this in time. Problem is that for a wide range of aircraft there is no mandatory inspection program, and many owners/LAMEs are not aware of it either. I understand some TC-holders in Europe are working on an inspection program, but that will not catch all those experimental aircraft…

Power & Passion,

Re your third paragraph. I smiled when I read it – how right you are ! Four or five years ago I started what seemed like at the time a world wide conflict for daring to suggest that we would have been better served, including the men who flew her, if the beloved Mossie had been built in metal. Merde alors ! Sacrilege ! Hell & Damnation !

It was very obvious to me that if ever the Gods smiled and I was able to invest a significant pile of my pocket money into a new build Mossie, that aircraft would be built in metal and if such was not possible then a wooden airframe cossetted and protected in a climate controlled hangar would be the way forward. The flak that whistled past my ears would have done credit to the Luftwaffe flak crews around Berlin during WW2.

There is one other essential ingredient that has long term stability implications in the wood to wood bonding process. it is the process of expansion and contraction inherent in the use of timber in most environments.

In boats it is less of a problem. Much of the boat frame or skeleton will be fastened with bronze fastenings or copper rivets. The chance of a boat frame constructed according to traditional methods suffering critical joint separation is remote. Perhaps not so with aircraft.

Coefficients of expansion and contraction differ with the type of timber and the influence of temperature and humidity. Any adhesive used for joint bonding on an aircraft should possess – among other characteristics – a gap filling and elasticity function. If some of the earlier adhesives tended to lack these qualities, then maybe a repetitive cycle of expansion and contraction could account for the failure of joint bonded elements critical to the integrity of the airframe

Altho’ these materials will move minutely, I haven’t seen what I would regard as significant movement in manufactured board timber such as ply. On three occasions I’ve seen ceramic kitchen and bathroom floor tiles laid and glued to a timber boarding material, only to see the tiles crack and separate on the join and away from the join. Conclusion: An example of differential separation.

@Powerandpassion: Beetle Cement is the British trade name for Kaurit WHK glue. So material specifications should match between the two. They were both initially known as Klemm-Leim, after the inventor of it, Hanns Klemm. His 1937 dissertation on this glue is here: https://hannsklemm.files.wordpress.c…21-10-1937.pdf.

Re Beetle glue AP.2662A (http://vhjet.com/wp-content/uploads/2017/08/AP-2662A-Standard-Repairs-for-Airframes-OCR.pdf) is helpful, and contains further references.

John, Now thats a first for me, wish I had known that before I bought french paint last year – its about 3 times more expensive than UK general duties paint, but the colour density of french paint is at least two times greater!
I do have a problem though using UK paint as it will not stand up to the extreme climate we have in Deux Sevres. The stuff I bought though as ‘climat extrems’ says it has teflon included as a water repellant, & is guaranteed for 12 years, might not be able to verify that being 80 at the moment! If interested data can be found on v33(dot)fr. Also a useful translation website, if you need it, is: deepl (dot) com.

Does anybody have old technical specifications for U-F glues, specifically Beetle ? The original specification was DTD 335 (Mosquito) and revision DTD448 (Hornet and Vampire).

I have a few documents based on CSIR studies of Mosquito and Anson wings, that show that moisture could lead to a 30% reduction in strength of a timber based structure, so the thing that will actually make a Hornet crack up under chocks and full engine power might be a hot and humid day. These structures should really be stored in a humidity controlled environment, if they are to last 100 years.

In 1928 the RAF stipulated that all service aircraft should henceforth be made of metal, due to the rapid deterioration of timber structures, shrinking and swelling in the weather. To assist manufacturers to convert from carpentry to metal work, the R100 and R101 airship projects were brought on in 1928 to help subsidise investments in metalwork tooling and capacity. In 1928 Bristols embarked on the steel strip Bulldog design and Hawkers on the steel Hart design, and the world moved away from wood, except for those stick in the mud folks at Hatfield.

It would be interesting to compare the mechanical specifications of DTD 335 and DTD 448 with some of the modern adhesive systems available today. Perhaps a timber aircraft should have a humidity recorder fixed to it, to gauge cycles of stress induced by swelling and shrinking, which would no doubt affect glue bonds.

Within the original specification, like original paint specifications, there may be the ‘bend test’. In the case of paint a painted metal panel would be bent backwards and forwards to demonstrate elasticity. I am sure a chemist could explain what happens to ageing U-F compositions in the presence of air and water and stress.

Somewhere I remember reading about Cellophane, where German cellulose starter material was sourced from pine trees, and British cellulose starter was sourced from Cotton. Cotton is a purer form of celluse and wood has lignins and all sorts of stuff in it. I wonder if this same issue resolves out of the starter materials used for U-F ? Where different processes used in different countries to make U-F? Do these determine ‘national characteristics’ for the same compositions that may explain what happens after 50 years ?

Hypothetically timber has an indefinite life, not subject to fatiguing like metal. But what is really under consideration is a ‘system’ of timber and adhesive, largely affected by humidity.
I would park my Hornet in a coolstore, keep it at steady humidity and temperature, keep my apples fresh too.

Dev One

One other useful feature of PVA is mixing it with exterior grade paint in the proportion of about one part PVA to five parts paint produces a paint much more durable and longer lasting. Most of us who are DIYers know how poor is the durability of many exterior grade paints. Spoon in some PVA and the result will be a longer lasting paint finish without affecting at all the colour and finished appearance.

Typical, when one wants to find a reference…….from my memory David Gearings book ‘On the wings of a Gull’ stated that Percivals were one of the earliest to use CIBA Aerolite & did so with Luton produced aircraft before WWII.
My fathers Proctor I (‘HNA) was the second Proctor produced by Hills & I think did not suffer badly from glue problems & was probably built using Aerolite, mind you as a licensed engineer he kept it well for 18 years!
HTE I think was partially restored before Tango Charlie acquired her ( at one time she was used by Derby Airways & flown by my Uncle). I saw HTE about 3 years ago alongside ‘KEX/LZ791, but that was only the fuselage, with ‘KEX almost finished rebuild.

From memory (and I might be wrong) I believe casein was used in the construction of my old Proctor V (G-AHTE). It soon became apparent, following the swap with the Welsh Museum for my old Provost (WW388) that it was literally falling apart. My Father worked in an adjacent workshop to where I dry stored it. He regularly heard parts fall off onto the floor! I had the entire aircraft assessed by PPS at Booker and they helped to compose a repair scheme and restoration schedule which was completely beyond me. Essentially, it required that every joint needed to be remade and the work projected was colossal. That was approximately 30 years ago. I understand that it is being restored to fly which must be a massive undertaking – given the glue issues on those aircraft.

On the same subject, I recall working for a company repairing Sailplanes at Booker during the early 80’s. Some of the glue in the vintage gliders that came through the door for their C of A’s were only fit for grounding on the spot. Pieces came off in your hand!

For your interest, and I think I’ve mentioned somewhere before on a forum, that my Father was based in Malaya on National Service during the 1950’s. A DH Hornet was running up its engines at RAF Changi whereupon both engines simultaneously detached from the aircraft and bounced down the runway. The ‘word on the ground’ was that glue failure was the root cause. Shortly afterwards, he witnessed teams turn up at RAF Seletar who cut the tails off all their squadron planes and break them all up.

John, Thank you for your reply regarding PVA. Off topic I suppose, but I have used it of course on household woodwork & found it stronger than most wood when breaking things up! As for mould I have not seen any, but I think silicone is very prone to growing it – only have to look in one’s bathroom! Talking of diluting it, yes, very good for adding to cement mixture to create a ‘waterproof’ cement & one can buy it as a sealant. I have used it recently on some windows where the timber has suffered some dry rot, drilling some small holes into the affected area & injecting dilute PVA into them so that the fibrous area absorbs the PVA & will then solidify & bond those fibres similar to GRP. Must be dilute PVA as normal viscosity is too thick to be absorbed.
Polysulfides are (or were used on TSR2 & Hawk to my knowledge) used in aviation as a sealant & used with a separator whilst curing will create a very good gasket for removable access panels. One disadvantage though is using it to seal laminated glass as it infiltrates the interlayer causing the cloudiness growth from the edges. I would have thought though that the joint would be very flexible compared to the other glues and whether that material would break down in shear under variable loading as in wooden aircraft structures. Might be OK for GRP aircraft though?

Eric is correct. PV acetate is a water soluble polymer with middling to good adhesive qualities under certain conditions. Diluted at about a three or four to one proportion it makes a fine sealant when applied to brickwork, tiling, or to prevent concrete dusting.

One group of very effective adhesives that should be mentioned tho’ I do not know about their acceptability for aviation use, are the sealants; polyurethane, polysulfide and silicone. The first two are very powerful adhesives with differing qualities. For wood to wood bonding, the strength quality of these two is described as excellent.

I have no information that gives the comparative adhesive values in respect of; compression, deflection, tension and shear. I’m fairly certain that this information exists.

In the marine environment, I’ve used all three of these capable sealant adhesives – they would always be my first choice. Anecdotally, I had once a vey torrid time trying to separate a bond between a stainless steel fitting and grp !

In the absence of any specific prohibitions, if I were building a wood airframe, two of the three would be my bonding weapons of choice.
If anyone is interested, I have a sealant/adhesive bonding table offering bonding comparisons between types of material and grading them from excellent/very good/good/ not recommended. If anyone wants to see it, I’ll put it on the forum.

Regarding PVA glues, as fsr as I know these are not certified for aviation use. They are certainly not approved for the types I worked on. Apart from that, I would imagine that they are not water resistant and susceptible to moulding…

(Would be great if you can share the results of the spar tests when they are available!)

I think your American friend may be right. I too have long suspected the glue skin issue, it is indeed mentioned in the manual. Problem is that if you stir the skin back in, glue joints appear really solid. But not maybe so 50 years down the road as they degrade chemically. The D-noses and spars seem to have been built with much more care regarding glue handling. I have rarely come across problems with Kaurit-WHK joints in D-noses (and then only in ribs) of wings. Never in spars of wings.

Regarding ribs, the issue may well be in the way they were produced. Schleicher had several local contractors make these, based on prices per rib. This of course promotes working too fast by sometimes unskilled labour not used to working with synthetic glues.

Thanks Eric, that’s a really interesting read. I’ve repaired two, scrapped two and have one currently on the go, which will hopefully fly at the end of March.

We’ve undertaken a number of tensile coupon tests on glue scarf coupons taken from a spar of a wing wiith significant trailing edge glue failure and this looks quite encouraging that the spars are safe. Hopefully the results will be published shortly.

I’ve had some correspondence with a long time American inspector who believes the basic problem was in the management of the kaurit at the factory. ie When mixed, it has a pot life but only when kept in an air tight container. If not stored in an air tight container, a skin would form on the surface and it was common practice to stir this back into the liquid. Although an instruction was produced not to do this, it was not particularly well respected.

The whole issue certainly has a strange randomness.

As for interior inspections, I’m now using a small CCD camera with a built in light, attached to the end of a fishing pole. The camera has a long lead hooked up to a laptop. It works a charm.

I’m also an Aerodux fan but recently I’ve done a bit of work with West Epoxy on a US home build , and it’s very impressive.

Curiosity I suppose, but what if any are the merits of using PVA adhesives? They seem to be good for domestic use.

https://www.researchgate.net/publication/239450519_Durability_of_wood_Adhesives_in_50_year_old_aircraft_and_glider_Constructions/download

is also very nice reading material.

Good to see you know your glues John!

It is always good to store wooden aircraft in an environment which has a constant temperature and humidity. Some of the aircraft were actually stored under cover in a warm dry place for most of the 50+ years they have been around, yet still they had issues. I have worked with a fair number of glues in structures (casein, Aerodux 185, Araldit, Aerolite, and the likes), but have never come across a problem so consistent as age-related glue-joint failure in Kaurit-WHK. It is not a new problem; I have seen references to it in 1960s research and technical literature. It was indeed the reason some national authorities put a moratorium in Kaurit-WHK for new-built aircraft, with manufacturers using Aerolite, Araldit and Aerodux instead, in the early 1960s already.

With regards to ageing, there is no ureaformaldehyde glue that ages gracefully. Kaurit-WHK can be very strong after decades, but I have seen some shocking failures, as mentioned above. The glue just becomes brittle and turns to powder on the touch. Aerolite, another UF-glue, does not behave that way, but tends to glass out and becomes prone to microcracking following peak loads (such as heavy landings). Aerolite however holds up much much better than Kaurit. My all-time favourite is still Aerodux. Waterproof, practically uv-resistant,easy to use. Just takes some care to only use it on wood with the correct moisture content, and above a certain temperature. I use it for almost all repairs – except to repair in casein-glued aircraft as casein residue and Aerodux don’t go well together.

Re inspection holes: couldn’t agree more. The problems in Kaurit-WHK glued structures are not detectable using for instance an endoscope or mirror. They require applying gentle force to structure (gussets, ribs, ply joints), which requires access holes to be cut. A more permanent access would be nice…

Ex Brat,

Many thanks for that suggestion. I will follow it up. I have previously read largely anecdotal reports concerning problems applicable to the glued joints on Mosquito’s used in the tropics.

Eric,

Do you think that storage with the convenience of temperature and humidity would have delayed or prevented degradation ?

Having been imprinted at an early age with the caveat: – among others, “Don’t use a chemical solution to solve a mechanical problem”, I became interested enough to build and repair wood boats in the size range 20 – 70ft. The odd aircraft inserted itself at regular intervals.

Some forty or more years ago, the principal tools in use were Cascamite, Aerodux and Aerolite and in the marine environment I can’t recall a single failure, admittedly within a contracted timescale.eg. if a mistake had been made resulting in a partial or total failure, we’d long departed the scene !

I do recall that at critical points in the boat skeleton, accumulative weight not being a consideration, half, butt, scarf and tenon joints would be supported by the use of either galvanised or aluminium metal flitch plates of varying shapes. The idea being that if the glue failed, the joint would still maintain at least some integrity until the next inspection – if ever.

Which brings me neatly to my next point : Inspection. I’ve never yet seen any light aircraft which to my eye featured an adequate number of inspection ports giving ready access to all parts of the interior. Even allowing for the use of extended inspection tools, as far as I am aware, it is not possible to examine the interior of most GA a/c without creating significant additional reinstatement work.

It seems that over the years, there was some questioning of the use in the marine environment of the glues mentioned above. Notwithstanding, urea formaldehyde is still a favourite. Resorcinols are top of the list being waterproof and unaffected by UV, tho’ Aerodux and Cascophen are temperature sensitive when curing. Epoxies are well up the list tho’ water resistant not waterproof and liable to degrade when exposed to UV.

I’ve always thought that thorough maintenance of wood aircraft requires more frequent intrusive internal inspection. I don’t know whether there is any clear statistical information that supports.