Last year I was in a meeting with M. Macia and his key engineers regarding this issue. What I've understood is that, in Airbus opinion, each cracking of the planet gear (which is coincidentally the outer race of its bearing) starts unavoidably with producing a particle from this outer race. Secondly the crack has a certain progressing time. For the particle they now have a third Full Flow Magnetic Plug with a catch rate of more than 90%. And for the growing of the crack in the rim they have the safety barrier to remove the planet gear every 1000 h (SLL). After all I still have doubts if really each crack will start with a particle. And even if, the particle easily could get lost inside the mgb housing before reaching one of the mag plugs. I'm not convinced of anything until an technically understandable reason for the failure is pinpointed.

Has anyone considered the resulting stresses from combining the bearing and gear components? When one designs a gear one is principally concerned about the tooth bending stress and the contact stress on the flanks of the teeth. I have never designed anything similar to a high speed rolling element bearing but I would think it would be mainly about contact stress but in a different format to the intermittent contact stress on a gear tooth flank.

So if two components are mounted together with a bearing inside a gear there is extra material (and thus weight) and many of the stress influences are isolated within each individual component. If one forms a bearing race as part of the gear, there is less material and fewer stresses are isolated because there is no helpful component boundary. There is a cycle of bending nearer the outside of the component caused by two or more gear engagements and a cycle of contact stresses nearer the inside of the component caused by contact with the rolling elements. The component has surface heat treatment so that material properties on the surfaces of the component are stronger and more brittle and the core remains less strong and more ductile. Everyone hopes that the core ductility of their surface hardened component will deal with the more quirky stresses and development testing should prove it. However, in my head I can visualise the combination of stresses having an effect like peeling up an old piece of lino.

According to Airbus all the different stresses were taken into consideration. Due to the hardened surface of the bearing races and teeth,16NCD13 layer after a nitrid process, and the "weaker" material inside the rim (M50), stress usually peels of the hardened surface. Resulting in flakes, splinters etc. on the mag plug. Which gives an early warning of the degradation. What was new in the G-REDL and LN-OJF case is that only a very limited such surface particle production - if any at all - took place during all the hours where the crack has grown inside. After a lot of time spent trying to understand what happened, my last idea was resonance frequency. In the MGB you have a lot of shafts, gears etc. turning with different speeds, inducing frequencies into the system. What if, in addition to the mentioned stress, the planet gears are exited with a frequency closed to their natural frequency? In combination with a micro pitting in the bearing race, as starting point, this could explain the crack growth inside the rim. Consequently a slightly different rotor speed could lead to a different planet gear behavior. According to this theory a crack would not grow even if initiated by a pitting, if the MR Rpm is e. g. 263 instead of 266 or vice versa.

Resonance is an interesting one. One could have different aircraft seeing different levels of risk based on random variations through acceptable tolerance in avionics settings, engine set-up, gear machining, and so on, plus the variations in load cycles across the fleet. Surely someone has visited these ideas?

All what's said is correct. However, for the gears and bearing materials currently used by the industry (and not only Airbus) it is sheer impossible to improve the products in a way that no spalling or micro pitting will occur anymore. The two main micro pitting drivers are corrosion and the overrun of particles. You couldn't use stainless steel in bearings because it tends to brittle. And you could micro filter the oil as much as you like, you'll never reach 100 % purity, respectively could avoid that particles are overrun in the bearings. Bearing spalling accures since decades on all gearboxes of all manufacturers. What was new is that two planet gears cracked through, most probably initiated by the micro pitting just discussed. So the question is why the planet gears on other h/c models, using the same materials, do not tend to crack through under the same conditions? The only answer which makes sense is that it must be, however, the design. So Airbus had only two options to bring the 225 back in service. Option one would have been a complete new design of the epicyclic, Millions of Euros spent into a product with an unpredictable future. Combined with the admission that the existing design was faulty. Option two was staying with the design, reducing the service life and installing a third mag plug. It's obvious why they have choosen option two.

I think the 225 is out of the medium civil helicopter market for other reasons now: The design looks dated, its too narrow with just three seats in a row, you cannot stand up in it and it's reputation leaves much to be desired.
I flew Pumas and Super Pumas from 1971 to 2008.I have retired and so should the basic design.

Hi Joe .... my comment is about ... "how the crack grows undetected"

I followed this investigation closely (read everything) back in the day ..... a small surface crack may not progress on its own unless there was a flaw in the original metal (rare in this case) ...... the problem is the rolling bearing is compressing oil at a very high pressure that causes hydraulic fracturing into the crack which causes it to keep growing over time.

Same physics that can split a granite rock when a drop of water seeps into a crack and freezes and expands.

Also, the DOC’s are probably comparatively high these days. Only made worse by the reduced TBO on the MRGB

One the other hand, it is the best ride out there. Always my favourite way home. Smoother than all the rest, with those big escape windows, jettisonable big doors, good run-dry spec, good flotation spec. If you know of a large rotorcraft that has done 12 years of intensive CAT without a fatality, please let me know.

PMSL
That’s classic!!!

Thanks - I needed a good laugh.

”Arrrgh, somethings happening - we have some red lights. Do we need to spray Glycol into the MRGB to stop it melting? Can’t really tell - might be a sensor. Spray it in!!? Ok.”
“Did it work? Can’t really tell. Want to wait and see if we get vibrations and/or grinding sounds? No - run the ditching check list.”

Reminds me of the classic “Explosive bolts” scene on “Dr Strangelove” (Peter Sellers 1964)
”Explosive bolts. Negative function”