Nomadd wrote:

benbeny wrote:

TUGMASTER wrote:

I don’t suppose they ever will....

I think they may not fly that much, but military planes usually stay longer in service than airlines ones. Besides P-8 sometimes fly low flying mission and stay in the air for 8 hours. I wonder will they show premature crack in this area too.

It's cycles they worry about. P-8s don't take off and land six times a day.

Military use is typically more complex than civilian use. More time spent at low altitudes, more maneuvering and higher load maneuvers in normal operations, less avoidance of turbulence, and a lot more training.

The result is different aging patterns in military aircraft than civilian counterparts.

I started looking at this a little bit with regards to the P-3 Orion versus the L-188 Electra it was based on.

Electras served in commercial use for 30,000+ cycles. The P-3 was rated by Lockheed for 7500 flight hours. That right there is potentially telling - that in one use case, the focus is on cycles, and in the other, the focus is on flight hours. The P-3 later underwent several rounds of service life extensions, but also required a fair amount of structural maintenance to do so.

It sounds like for the Navy, the wings were the primary area where fatigue reared its head, not the fuselage.

That experience makes me suspect the pickle forks will indeed need to be addressed for the P-8. However, the P-8 is over a decade newer than the 737NG, and the Navy's ASW flight rate is less than 1000 hours per year, where as commercial airliners operate 3000+ hours per year. I anticipate Boeing will have already developed a long term fix for this by the time the Navy might encounter problems.

mjoelnir wrote:

StTim wrote:

All planes have cracks. Mostly they are stop drilled and monitored. Repairs are done as required.

The slight difference is that there has not been cracking in the pickle forks before so it was unexpected. It is also a complex and highly stressed area so extra care must be taken.

The pickle fork was a component that was not supposed to crack.

No components are supposed to crack, but some do and it’s an accepted procedure to stop drill it.

Trying to think of a none engineers analogy for people to understand.

From a stress point of view the point of a crack is small and the stress is concentrated at that point. A stop drilled hole spreads the load.

So my analogy is trying to stick a sewing needle in your finger, how much pressure would that take? Then imagine how much pressure it would take to push a ball point pen into your finger.

Hope that helps understand how stop drilling reads the load and stops the crack propograting.

889091 wrote:

How are the wings on Boeing's other non CFRP planes mated onto the fuselage? Larger versions of the pickle forks?

Also, why didn't this problem rear its head on the Classics and Jurassics?

The basic large scale geometry of the wing-body joint is similar in that there will typically need to be some sort of large fitting at the front and rear spars, but whatever large fitting is there does not have to be fork shaped.

I found two relevant pictures online. One shows only the center wingbox and the wheel well:
https://www.compositesworld.com/news/mh ... g-capacity

The other shows the entire center fuselage section, including what appears to be large rib-like fitting aft of the rear spar:
https://www.flickr.com/photos/flightblo ... 714902435/

For comparison, here is a photo of the 737 mid fuselage section:
https://www.flickr.com/photos/flightblo ... 367230101/

The NG has a new wing compared to the Classic and the original. It seems likely the wing body joint underwent significant changes at the same time.

I did a bit of digging earlier today.
It appears that the pickle fork used to be made from made from aluminium billet. That material has good uniform properties and you are normally machining the component from a large over size piece of material.

They have changed to an aluminium close to size forging. In my experience the properties are inferior and inconsistent. Now I have no data on how much oversize the forging are, but I have properties to be inferior to billet.

It will be a cost reduction that’s going to bite them on the ar5e.

If they are lucky it’s a batch issue, if not it will be the whole fleet.

Flashy16 wrote:

Sorry, I did not phrase my question right....I am not very good at forums! Not looking to cancel my flight! Just wondering if I should change from a 737-800 to another type leaving at a similar time. Just concerned whether this is a big enough issue to make a change. Sorry again, thank you.

No. Proceed with your flight. Most of the aircraft that would have had issues have already been inspected. Aircraft with issues won't be in service.

And aircraft with issues have been in flight for who knows how long with the cracks before the inspections started.

lowbank wrote:

I did a bit of digging earlier today.
It appears that the pickle fork used to be made from made from aluminium billet. That material has good uniform properties and you are normally machining the component from a large over size piece of material.

They have changed to an aluminium close to size forging. In my experience the properties are inferior and inconsistent. Now I have no data on how much oversize the forging are, but I have properties to be inferior to billet.

It will be a cost reduction that’s going to bite them on the ar5e.

If they are lucky it’s a batch issue, if not it will be the whole fleet.

Forgings are not inherently inferior. They typically have differing strengths in different directions which can be beneficial in some applications and problematic in others. Sometimes forgings are chosen deliberately in order to get directionally optimized strength or better fatigue properties. Bicycle cranks are an example of a part often made from forged aluminum for its mechanical properties.

It would also be possible to use a forging for cost savings by forming a part close to net shape to reduce material wasted in the machining step and cost of the machining operation. If this were going to be done to make a drop-in replacement for a part machined from billet, it would be typical to anneal it and heat treat it again after the forging and prior to final machining in order to achieve the same properties and avoid distortion during machining from residual stress. The resulting part should have the same properties as if made from billet, unless an overly aggressive forging operation cause a crack, void, or other defect, or the heat treatment was not properly done.

Catching these kinds of details is part of why it is necessary in aerospace not simply to qualify a part, but also to qualify the process used to make the part. Hence, aircraft designs receive not only a type certificate, but also a production certificate.

Does anybody know, if the 737MAX fuselage went through fatigue testing?

iamlucky13 wrote:

lowbank wrote:

I did a bit of digging earlier today.
It appears that the pickle fork used to be made from made from aluminium billet. That material has good uniform properties and you are normally machining the component from a large over size piece of material.

They have changed to an aluminium close to size forging. In my experience the properties are inferior and inconsistent. Now I have no data on how much oversize the forging are, but I have properties to be inferior to billet.

It will be a cost reduction that’s going to bite them on the ar5e.

If they are lucky it’s a batch issue, if not it will be the whole fleet.

Forgings are not inherently inferior. They typically have differing strengths in different directions which can be beneficial in some applications and problematic in others. Sometimes forgings are chosen deliberately in order to get directionally optimized strength or better fatigue properties. Bicycle cranks are an example of a part often made from forged aluminum for its mechanical properties.

It would also be possible to use a forging for cost savings by forming a part close to net shape to reduce material wasted in the machining step and cost of the machining operation. If this were going to be done to make a drop-in replacement for a part machined from billet, it would be typical to anneal it and heat treat it again after the forging and prior to final machining in order to achieve the same properties and avoid distortion during machining from residual stress. The resulting part should have the same properties as if made from billet, unless an overly aggressive forging operation cause a crack, void, or other defect, or the heat treatment was not properly done.

Catching these kinds of details is part of why it is necessary in aerospace not simply to qualify a part, but also to qualify the process used to make the part. Hence, aircraft designs receive not only a type certificate, but also a production certificate.

A manufacturing process change could explain a lot. As noted, there are times forgings are desired. In fact, I have designed parts where the only way to arrive at reasonable service life is a forging.

It isn't the forging that is the problem necessarily, it is a stress. Perhaps a residual stress not properly heat treated.

Reading about this is why I kept coming back to this thread, finding a root cause. Now for Boeing to determine a solution.

Lightsaber

iamlucky13 wrote:

lowbank wrote:

I did a bit of digging earlier today.
It appears that the pickle fork used to be made from made from aluminium billet. That material has good uniform properties and you are normally machining the component from a large over size piece of material.

They have changed to an aluminium close to size forging. In my experience the properties are inferior and inconsistent. Now I have no data on how much oversize the forging are, but I have properties to be inferior to billet.

It will be a cost reduction that’s going to bite them on the ar5e.

If they are lucky it’s a batch issue, if not it will be the whole fleet.

Forgings are not inherently inferior. They typically have differing strengths in different directions which can be beneficial in some applications and problematic in others. Sometimes forgings are chosen deliberately in order to get directionally optimized strength or better fatigue properties. Bicycle cranks are an example of a part often made from forged aluminum for its mechanical properties.

It would also be possible to use a forging for cost savings by forming a part close to net shape to reduce material wasted in the machining step and cost of the machining operation. If this were going to be done to make a drop-in replacement for a part machined from billet, it would be typical to anneal it and heat treat it again after the forging and prior to final machining in order to achieve the same properties and avoid distortion during machining from residual stress. The resulting part should have the same properties as if made from billet, unless an overly aggressive forging operation cause a crack, void, or other defect, or the heat treatment was not properly done.

Catching these kinds of details is part of why it is necessary in aerospace not simply to qualify a part, but also to qualify the process used to make the part. Hence, aircraft designs receive not only a type certificate, but also a production certificate.

I do understand what your saying and I agree if a part is designed with a forging in mind, you can get improved properties from a forging.
I do understand how aerospace works, I will be signing off a new aluminium forging FAIR next week which will seal the manufacturing process for that part.

Specifically what I have witnessed is moving from billet to close to size forgings (Aluminium) on parts not designed as a forging initially have been problematic.

People get blinded by this large cost saving and it comes back to bite you.

I am sure Boeing are running around trying to understand if it’s a manufacturing issue.

If it’s not and it’s a design issue, it will affect all frames.

Boeing did not need this and I feel for them.

Two from lion air. This time both below 22,000 FC. Very bad news for Boeing. They’re all gonna have to get inspected.

https://www.smh.com.au/business/compani ... 53828.html

Mostly wrote:

Two from lion air. This time both below 22,000 FC. Very bad news for Boeing. They’re all gonna have to get inspected.

https://www.smh.com.au/business/compani ... 53828.html

Oh dear, that is terrible news

nine4nine wrote:

Is the NG pickle fork the same that was used on the classics? If so I can imagine there being issues as the original form was designed for the wing load and stresses of a classic and not the expanded length and weight of an NG wing without any redesign/reinforcement. I’d imagine if it were the same that it’s just another example of a plane being over engineered past its intended design by Boeing.

It kinda does not work like that!!
For any design change you have to have specialist reports.
Materials, stress, aero, impact, lifing, and many more. All have to agree in a report that the change has no impact on the components specialists subject.

However, and I don’t know Boeing’s process but normally a change from billet manufacture to close to size forging would not need all those reports. A change control board would decide on what testing was required to make the change.

So replacing the pickle forks once isn't going to be enough. They could have to be replaced multiple times on the same plane.

RickNRoll wrote:

So replacing the pickle forks once isn't going to be enough. They could have to be replaced multiple times on the same plane.

That will be compelling reason to design a replacement part that resolves the issue entirely, if at all possible.

That is, unless it turns out a manufacturing defect is the cause, rather than the design.

Mostly wrote:

Two from lion air. This time both below 22,000 FC. Very bad news for Boeing. They’re all gonna have to get inspected.

https://www.smh.com.au/business/compani ... 53828.html

The article also refers to Virgin Australia checking more aircraft than required, only 19 737's were above the 22,600 cycle mark however VA also checked another 6 aircraft above 18,000 flight cycles

starrion wrote:

Flashy16 wrote:

Sorry, I did not phrase my question right....I am not very good at forums! Not looking to cancel my flight! Just wondering if I should change from a 737-800 to another type leaving at a similar time. Just concerned whether this is a big enough issue to make a change. Sorry again, thank you.

No. Proceed with your flight. Most of the aircraft that would have had issues have already been inspected. Aircraft with issues won't be in service.

The cycles threshold keeps getting lower.

RickNRoll wrote:

starrion wrote:

Flashy16 wrote:

Sorry, I did not phrase my question right....I am not very good at forums! Not looking to cancel my flight! Just wondering if I should change from a 737-800 to another type leaving at a similar time. Just concerned whether this is a big enough issue to make a change. Sorry again, thank you.

No. Proceed with your flight. Most of the aircraft that would have had issues have already been inspected. Aircraft with issues won't be in service.

The cycles threshold keeps getting lower.

No Aircraft has had a failure.... These cracks are being found well before they represent a catastrophic failure.

Keep in mind - with the possible exception of single crystal turbine blades; all metal parts have cracks in them at some level (Microscopic for most new metal parts). The key is to keep the crack from growing large enough to to reach critical length.

Have a great day,

iamlucky13 wrote:

RickNRoll wrote:

So replacing the pickle forks once isn't going to be enough. They could have to be replaced multiple times on the same plane.

That will be compelling reason to design a replacement part that resolves the issue entirely, if at all possible.

That is, unless it turns out a manufacturing defect is the cause, rather than the design.

Or design a reinforcement that is installed before the cracks start forming. As long as to much weight isn't added then it will probably be cheaper than replacement parts.

qf789 wrote:

Mostly wrote:

Two from lion air. This time both below 22,000 FC. Very bad news for Boeing. They’re all gonna have to get inspected.

https://www.smh.com.au/business/compani ... 53828.html

The article also refers to Virgin Australia checking more aircraft than required, only 19 737's were above the 22,600 cycle mark however VA also checked another 6 aircraft above 18,000 flight cycles

Lion Air use their planes in high cycle, low flight hour pattern. That's terrible news for all regional users. 22k is pretty low time if you fly 10 cycles per day.

Does somebody know if the 737NG fuselage and wing were fatigue tested?

I know that I asked this question before, but does anybody know?

2175301 wrote:

RickNRoll wrote:

starrion wrote:

No. Proceed with your flight. Most of the aircraft that would have had issues have already been inspected. Aircraft with issues won't be in service.

The cycles threshold keeps getting lower.

No Aircraft has had a failure.... These cracks are being found well before they represent a catastrophic failure.

Keep in mind - with the possible exception of single crystal turbine blades; all metal parts have cracks in them at some level (Microscopic for most new metal parts). The key is to keep the crack from growing large enough to to reach critical length.

Have a great day,

And Boeing and the users of the 737NG can be thankful for the p2f conversion company having accidentally found this cracks first. Nobody was looking for them in the normal check up system.
If there had been a failure whole fleets would have been grounded, apart from the loss of live at an accident.

2175301 wrote:

RickNRoll wrote:

The cycles threshold keeps getting lower.

No Aircraft has had a failure.... These cracks are being found well before they represent a catastrophic failure.

Keep in mind - with the possible exception of single crystal turbine blades; all metal parts have cracks in them at some level (Microscopic for most new metal parts). The key is to keep the crack from growing large enough to to reach critical length.

The difference being that these are NOT supposed to crack to any extent within the lifetime of the part...

These are safe-life parts - so they're designed to a different method and are a) not supposed to fail ever and b) not supposed to be inspected ever. You can't brush off cracking as normal practice when these parts are exceptional to that normal practice.

(N.B. I am not slagging off Boeing or the 737 here, these things happen and it's good that it's been caught - but it is definitely not normal for a safe-life part to "keep the crack from growing large enough to reach critical length" like you imply.)

While I don't doubt this was Boeings thesis - the question is why.

These are seemingly metal parts under variable and high stress conditions. It's known that almost all metal cracks - question is why this was ever defined as safe life

SomebodyInTLS wrote:

(N.B. I am not slagging off Boeing or the 737 here, these things happen and it's good that it's been caught - but it is definitely not normal for a safe-life part to "keep the crack from growing large enough to reach critical length" like you imply.)

Re-reading it some time later, I accept that you could be referring to designing to prevent significant crack growth rather than there being an inspection programme to find such cracks before they become critical. Not sure that's actually what you meant though.

moa999 wrote:

While I don't doubt this was Boeings thesis - the question is why.

These are seemingly metal parts under variable and high stress conditions. It's known that almost all metal cracks - question is why this was ever defined as safe life

Safe-life tends to be reserved for a small number of the heaviest-duty parts such as landing gear, where it's easier just to beef them up so the chance of them failing within a certain time becomes very remote - and they may be replaced after a certain 'guaranteed' lifetime. The pickle-fork probably falls into this category as it's hard to inspect and replace during routine maintenance - hence the safe-life design being for the aircraft lifetime.

Highly-stressed parts which *are* routinely inspected tend to have predictable loading - even when this is variable (gusts etc.), the stress spectrum can be described statistically and used to predict crack growth in terms of flight cycles. The parts would also need to have simple options for repair/replacement, which is where I expect the pickle-fork falls short.

2175301 wrote:

RickNRoll wrote:

starrion wrote:

No. Proceed with your flight. Most of the aircraft that would have had issues have already been inspected. Aircraft with issues won't be in service.

The cycles threshold keeps getting lower.

No Aircraft has had a failure.... These cracks are being found well before they represent a catastrophic failure.

Keep in mind - with the possible exception of single crystal turbine blades; all metal parts have cracks in them at some level (Microscopic for most new metal parts). The key is to keep the crack from growing large enough to to reach critical length.

Have a great day,

Not exactly correct, apologies.

If you can create a part with a widmannstatten structure you can make a part where cracks cannot propagate and a part can have infinite life. I have worked on one part where it was created and fatigue testing was stopped after several million cycles.

If you can work out how to produce that structure, you could be a billionaire.

The billets being discussed were as cast or forged after casting ? I assume forged billets, as a cast billet would not have the required properties.

lowbank wrote:

mjoelnir wrote:

StTim wrote:

All planes have cracks. Mostly they are stop drilled and monitored. Repairs are done as required.

The slight difference is that there has not been cracking in the pickle forks before so it was unexpected. It is also a complex and highly stressed area so extra care must be taken.

The pickle fork was a component that was not supposed to crack.

No components are supposed to crack, but some do and it’s an accepted procedure to stop drill it.

Trying to think of a none engineers analogy for people to understand.

From a stress point of view the point of a crack is small and the stress is concentrated at that point. A stop drilled hole spreads the load.

So my analogy is trying to stick a sewing needle in your finger, how much pressure would that take? Then imagine how much pressure it would take to push a ball point pen into your finger.

Hope that helps understand how stop drilling reads the load and stops the crack propograting.

Saying it from an engineering standpoint. The pickle fork is designed as a life time part, that is not supposed to crack ever. That is why it was not checked upon. That is why cracks were found by accident, not at an regular inspection. Do I have to make it simpler for you as an engineer?

mjoelnir wrote:

Saying it from an engineering standpoint. The pickle fork is designed as a life time part, that is not supposed to crack ever. That is why it was not checked upon. That is why cracks were found by accident, not at an regular inspection. Do I have to make it simpler for you as an engineer?

I'm not sure that is true at all: It may very well be that the design assumption was that over the life of the aircraft that no crack should grow longer than 1/4" (6.35 mm). Then they found cracks that were obviously longer.

I spent many years working with safety critical equipment that was designed and certified to operate for 40 to 60 years continuously and numerous cycles... and there is not a single component that had an assumption of no crack would ever develop. The assumption was always that no crack longer than "X" would develop in various locations ("X being different for different locations). Pressure systems leakage was allowed to a certain point for all systems (and the design of the plant had to assume constant operation with a certain amount of leakage - even of primary reactor coolant and safety shutdown system, and also the worst highest concentrated radioactive waste streams).

That's my experience with engineering. I've never in my life seen a part that had an assumption that it would never have a crack develop in it over its design life.

Have a great day,

ps: of course, if a crack does develop... there are various ways to address the situation short of part replacement in most cases. Stop drilling only works for some situations. I've seen "dogbone" shaped holes with machined high alloy dogbones inserted (heated to lengthen them for insertion) that then took the strain and stopped the crack from growing. That's what you do on heavy machine casings and heavy wall pipe where you cannot have a hole in them as they are a pressure boundry. There are other interesting and very creative ways to handle the stresses and stop (or greatly reduce crack growth) as well... Depends on the location, the equipment, and the significance. Of course, sometimes replacement is required. But, that is rare in my experience unless its just a small minor piece of equipment or part.

2175301 wrote:

mjoelnir wrote:

Saying it from an engineering standpoint. The pickle fork is designed as a life time part, that is not supposed to crack ever. That is why it was not checked upon. That is why cracks were found by accident, not at an regular inspection. Do I have to make it simpler for you as an engineer?

I'm not sure that is true at all: It may very well be that the design assumption was that over the life of the aircraft that no crack should grow longer than 1/4" (6.35 mm). Then they found cracks that were obviously longer.

mjoelnir may be correct. You may be correct. You may both be correct.

mjoelnir may be correct that when designed for the 100/200, the expectation was zero cracks.

You may be correct that the Boeing expectation changed as model weights, engines thrusts, etc increased, and cracking over time was to be expected.

The acid test though surely, is if cracks up to a certain length are acceptable and expected, is that the model inspection regime, A requires inspection, and B details acceptable cracking based on hours and cycles. As it appears this was not the case, surely the inference is Boeing expected zero cracks, not cracks up to a certain size are acceptable.

2175301 wrote:

mjoelnir wrote:

Saying it from an engineering standpoint. The pickle fork is designed as a life time part, that is not supposed to crack ever. That is why it was not checked upon. That is why cracks were found by accident, not at an regular inspection. Do I have to make it simpler for you as an engineer?

I'm not sure that is true at all: It may very well be that the design assumption was that over the life of the aircraft that no crack should grow longer than 1/4" (6.35 mm). Then they found cracks that were obviously longer.

I spent many years working with safety critical equipment that was designed and certified to operate for 40 to 60 years continuously and numerous cycles... and there is not a single component that had an assumption of no crack would ever develop. The assumption was always that no crack longer than "X" would develop in various locations ("X being different for different locations). Pressure systems leakage was allowed to a certain point for all systems (and the design of the plant had to assume constant operation with a certain amount of leakage - even of primary reactor coolant and safety shutdown system, and also the worst highest concentrated radioactive waste streams).

That's my experience with engineering. I've never in my life seen a part that had an assumption that it would never have a crack develop in it over its design life.

Have a great day,

ps: of course, if a crack does develop... there are various ways to address the situation short of part replacement in most cases. Stop drilling only works for some situations. I've seen "dogbone" shaped holes with machined high alloy dogbones inserted (heated to lengthen them for insertion) that then took the strain and stopped the crack from growing. That's what you do on heavy machine casings and heavy wall pipe where you cannot have a hole in them as they are a pressure boundry. There are other interesting and very creative ways to handle the stresses and stop (or greatly reduce crack growth) as well... Depends on the location, the equipment, and the significance. Of course, sometimes replacement is required. But, that is rare in my experience unless its just a small minor piece of equipment or part.

If a crack was expected over time as you suggest, why would there be no inspection requirement?

Far more likely that no crack was expected, don’t you agree?

Have a totally fabulous, bodacious, awesome day.

2175301 wrote:

mjoelnir wrote:

Saying it from an engineering standpoint. The pickle fork is designed as a life time part, that is not supposed to crack ever. That is why it was not checked upon. That is why cracks were found by accident, not at an regular inspection. Do I have to make it simpler for you as an engineer?

I'm not sure that is true at all: It may very well be that the design assumption was that over the life of the aircraft that no crack should grow longer than 1/4" (6.35 mm). Then they found cracks that were obviously longer.

I spent many years working with safety critical equipment that was designed and certified to operate for 40 to 60 years continuously and numerous cycles... and there is not a single component that had an assumption of no crack would ever develop. The assumption was always that no crack longer than "X" would develop in various locations ("X being different for different locations). Pressure systems leakage was allowed to a certain point for all systems (and the design of the plant had to assume constant operation with a certain amount of leakage - even of primary reactor coolant and safety shutdown system, and also the worst highest concentrated radioactive waste streams).

That's my experience with engineering. I've never in my life seen a part that had an assumption that it would never have a crack develop in it over its design life.

Have a great day,

ps: of course, if a crack does develop... there are various ways to address the situation short of part replacement in most cases. Stop drilling only works for some situations. I've seen "dogbone" shaped holes with machined high alloy dogbones inserted (heated to lengthen them for insertion) that then took the strain and stopped the crack from growing. That's what you do on heavy machine casings and heavy wall pipe where you cannot have a hole in them as they are a pressure boundry. There are other interesting and very creative ways to handle the stresses and stop (or greatly reduce crack growth) as well... Depends on the location, the equipment, and the significance. Of course, sometimes replacement is required. But, that is rare in my experience unless its just a small minor piece of equipment or part.

Interesting, never heard of allowable cracks, with specified lengths. You learn something new everyday.
Only allowable cracks in my experience were surface cracks, which didn't propagate to any depth. But even these were ground out at the first opportunity. If a non critical equipment developed a crack, drilling a hole at the crack end was one of the options, normally in hollow sections or load carrying cylindrical components. Never for pressure vessels. For pressure vessels, clamping till proper maintenance like welding, etc was possible.

Maybe what he’s saying that if you zoom in far enough with an electron microscope you can always find tiny cracks within the crystal structure.

But it’s perfectly reasonable to design something well enough that won’t visibly crack in its service life. To imply that literally everything cracks is wrong.

My plane was made in 1966 and the landing gear hasn’t cracked. And if it does, I’m not going to call Cessna to ask them what the crack length limit is. They’d laugh me out of town and tell me to replace the damn part (or more likely, the whole plane).

mjoelnir wrote:

lowbank wrote:

mjoelnir wrote:

The pickle fork was a component that was not supposed to crack.

No components are supposed to crack, but some do and it’s an accepted procedure to stop drill it.

Trying to think of a none engineers analogy for people to understand.

From a stress point of view the point of a crack is small and the stress is concentrated at that point. A stop drilled hole spreads the load.

So my analogy is trying to stick a sewing needle in your finger, how much pressure would that take? Then imagine how much pressure it would take to push a ball point pen into your finger.

Hope that helps understand how stop drilling reads the load and stops the crack propograting.

Saying it from an engineering standpoint. The pickle fork is designed as a life time part, that is not supposed to crack ever. That is why it was not checked upon. That is why cracks were found by accident, not at an regular inspection. Do I have to make it simpler for you as an engineer?

Sir I respect your posts.

However, all parts are inspected at intervals, whilst we design parts for life of the aircraft, experience has shown that sh1t happens.

Do you think we just wait till a wing falls off and think “oh shit how did that happen”

maint123 wrote:

2175301 wrote:

mjoelnir wrote:

Saying it from an engineering standpoint. The pickle fork is designed as a life time part, that is not supposed to crack ever. That is why it was not checked upon. That is why cracks were found by accident, not at an regular inspection. Do I have to make it simpler for you as an engineer?

I'm not sure that is true at all: It may very well be that the design assumption was that over the life of the aircraft that no crack should grow longer than 1/4" (6.35 mm). Then they found cracks that were obviously longer.

I spent many years working with safety critical equipment that was designed and certified to operate for 40 to 60 years continuously and numerous cycles... and there is not a single component that had an assumption of no crack would ever develop. The assumption was always that no crack longer than "X" would develop in various locations ("X being different for different locations). Pressure systems leakage was allowed to a certain point for all systems (and the design of the plant had to assume constant operation with a certain amount of leakage - even of primary reactor coolant and safety shutdown system, and also the worst highest concentrated radioactive waste streams).

That's my experience with engineering. I've never in my life seen a part that had an assumption that it would never have a crack develop in it over its design life.



Have a great day,

ps: of course, if a crack does develop... there are various ways to address the situation short of part replacement in most cases. Stop drilling only works for some situations. I've seen "dogbone" shaped holes with machined high alloy dogbones inserted (heated to lengthen them for insertion) that then took the strain and stopped the crack from growing. That's what you do on heavy machine casings and heavy wall pipe where you cannot have a hole in them as they are a pressure boundry. There are other interesting and very creative ways to handle the stresses and stop (or greatly reduce crack growth) as well... Depends on the location, the equipment, and the significance. Of course, sometimes replacement is required. But, that is rare in my experience unless its just a small minor piece of equipment or part.

Interesting, never heard of allowable cracks, with specified lengths. You learn something new everyday.
Only allowable cracks in my experience were surface cracks, which didn't propagate to any depth. But even these were ground out at the first opportunity. If a non critical equipment developed a crack, drilling a hole at the crack end was one of the options, normally in hollow sections or load carrying cylindrical components. Never for pressure vessels. For pressure vessels, clamping till proper maintenance like welding, etc was possible.

Sorry to put this out there.

But we do work with crack propagation limits.

We use low circle fatigue and high cycle fatigue testing to understand how fast cracks propagate.

If a part is found to have cracked that shouldn’t then the crack propagation data will determine the inspection intervals.

lowbank wrote:

mjoelnir wrote:

lowbank wrote:

No components are supposed to crack, but some do and it’s an accepted procedure to stop drill it.

Trying to think of a none engineers analogy for people to understand.

From a stress point of view the point of a crack is small and the stress is concentrated at that point. A stop drilled hole spreads the load.

So my analogy is trying to stick a sewing needle in your finger, how much pressure would that take? Then imagine how much pressure it would take to push a ball point pen into your finger.

Hope that helps understand how stop drilling reads the load and stops the crack propograting.

Saying it from an engineering standpoint. The pickle fork is designed as a life time part, that is not supposed to crack ever. That is why it was not checked upon. That is why cracks were found by accident, not at an regular inspection. Do I have to make it simpler for you as an engineer?

Sir I respect your posts.

However, all parts are inspected at intervals, whilst we design parts for life of the aircraft, experience has shown that sh1t happens.

Do you think we just wait till a wing falls off and think “oh shit how did that happen”

So what was the inspection interval for the pickle forks and why were the cracks not found by regulated inspections?

mjoelnir wrote:

lowbank wrote:

mjoelnir wrote:

Saying it from an engineering standpoint. The pickle fork is designed as a life time part, that is not supposed to crack ever. That is why it was not checked upon. That is why cracks were found by accident, not at an regular inspection. Do I have to make it simpler for you as an engineer?

Sir I respect your posts.

However, all parts are inspected at intervals, whilst we design parts for life of the aircraft, experience has shown that sh1t happens.

Do you think we just wait till a wing falls off and think “oh shit how did that happen”

So what was the inspection interval for the pickle forks and why were the cracks not found by regulated inspections?

I thought it wasn't regularly inspected until now?

lowbank wrote:

I did a bit of digging earlier today.
It appears that the pickle fork used to be made from made from aluminium billet. That material has good uniform properties and you are normally machining the component from a large over size piece of material.

They have changed to an aluminium close to size forging. In my experience the properties are inferior and inconsistent. Now I have no data on how much oversize the forging are, but I have properties to be inferior to billet.

It will be a cost reduction that’s going to bite them on the ar5e.

If they are lucky it’s a batch issue, if not it will be the whole fleet.

It's costing airlines, but does Boeing have to compensate them ?

lowbank wrote:

maint123 wrote:

2175301 wrote:

I'm not sure that is true at all: It may very well be that the design assumption was that over the life of the aircraft that no crack should grow longer than 1/4" (6.35 mm). Then they found cracks that were obviously longer.

I spent many years working with safety critical equipment that was designed and certified to operate for 40 to 60 years continuously and numerous cycles... and there is not a single component that had an assumption of no crack would ever develop. The assumption was always that no crack longer than "X" would develop in various locations ("X being different for different locations). Pressure systems leakage was allowed to a certain point for all systems (and the design of the plant had to assume constant operation with a certain amount of leakage - even of primary reactor coolant and safety shutdown system, and also the worst highest concentrated radioactive waste streams).

That's my experience with engineering. I've never in my life seen a part that had an assumption that it would never have a crack develop in it over its design life.



Have a great day,

ps: of course, if a crack does develop... there are various ways to address the situation short of part replacement in most cases. Stop drilling only works for some situations. I've seen "dogbone" shaped holes with machined high alloy dogbones inserted (heated to lengthen them for insertion) that then took the strain and stopped the crack from growing. That's what you do on heavy machine casings and heavy wall pipe where you cannot have a hole in them as they are a pressure boundry. There are other interesting and very creative ways to handle the stresses and stop (or greatly reduce crack growth) as well... Depends on the location, the equipment, and the significance. Of course, sometimes replacement is required. But, that is rare in my experience unless its just a small minor piece of equipment or part.

Interesting, never heard of allowable cracks, with specified lengths. You learn something new everyday.
Only allowable cracks in my experience were surface cracks, which didn't propagate to any depth. But even these were ground out at the first opportunity. If a non critical equipment developed a crack, drilling a hole at the crack end was one of the options, normally in hollow sections or load carrying cylindrical components. Never for pressure vessels. For pressure vessels, clamping till proper maintenance like welding, etc was possible.

Sorry to put this out there.

But we do work with crack propagation limits.

We use low circle fatigue and high cycle fatigue testing to understand how fast cracks propagate.

If a part is found to have cracked that shouldn’t then the crack propagation data will determine the inspection intervals.

Why sorry? I had never used or heard of such techniques in the heavy industry I work in. Its new for me. But on searching the net it seems these techniques are quite prevelant in the nuclear, aerospace and shipbuilding industries. Most of the studies on the net are quite theoretical.
Our exposure to crack detection was in inspection of inbound material for cracks, especially forgings. Or checking of cracks in running equipment. And surface cracks detection in ground surfaces. But was never a exact, systematic process. Mostly experience based, using DP testing or ultrasound. Also for high pressure welded joints.
But no comparative charts to check crack propagation limits, etc., to base decisions on.

If you look at https://en.wikipedia.org/wiki/EPR_(nuclear_reactor)#Flamanville_3_(France) bad welds were discovered in steam transfer pipes and that will cost a couple billion euros to repair, even though no nuclear fuel has been loaded yet, so there is no radioactivity to deal with.

mjoelnir wrote:

lowbank wrote:

mjoelnir wrote:

Saying it from an engineering standpoint. The pickle fork is designed as a life time part, that is not supposed to crack ever. That is why it was not checked upon. That is why cracks were found by accident, not at an regular inspection. Do I have to make it simpler for you as an engineer?

Sir I respect your posts.

However, all parts are inspected at intervals, whilst we design parts for life of the aircraft, experience has shown that sh1t happens.

Do you think we just wait till a wing falls off and think “oh shit how did that happen”

So what was the inspection interval for the pickle forks and why were the cracks not found by regulated inspections?

I'm guessing grandfathering. The pickle forks in previous versions had been milled from solid ingots. More expensive but stronger and, apparently, still going strong. The manufacturing process has since changed to a cheaper one.

benbeny wrote:

mjoelnir wrote:

lowbank wrote:

Sir I respect your posts.

However, all parts are inspected at intervals, whilst we design parts for life of the aircraft, experience has shown that sh1t happens.

Do you think we just wait till a wing falls off and think “oh shit how did that happen”

So what was the inspection interval for the pickle forks and why were the cracks not found by regulated inspections?

I thought it wasn't regularly inspected until now?

I know that there was no inspection regime, but that poster claims it is. So I am asking him to back up his claim.

Aesma wrote:

If you look at https://en.wikipedia.org/wiki/EPR_(nuclear_reactor)#Flamanville_3_(France) bad welds were discovered in steam transfer pipes and that will cost a couple billion euros to repair, even though no nuclear fuel has been loaded yet, so there is no radioactivity to deal with.

This is from my industry and I know a lot more about these cracked welds that what is likely in Wiki. It's far cheaper to fix them now before things get radioactive.... but, overall.... If you welding and heat treatment process is so bad to create such cracks in the first place in a new joint in those locations... There's a lot worse coming down the line from those welds.

The real question is why they ever used people and procedures to create such poor welds in the first place. It raises a lot of questions about quality control right down to the welder level (and I've known too many good welders who will jump up and admit that the weld is not right before they progress very far - and then grind it out (or re-machine depending on the location), possibly heat treat, and restart welding. It's sure faster than completing the weld and then finding problems later.

Even good people running automatic welders and following the weld bead and process know when things are going bad - and stop the process.

I assure you that the debate in the US Nuclear industry over the Flamanville reactor welds and flaws is how could that have been allowed to happen.

Have a great day,

Wouldn’t any change of manufacturing process need to be certified?

StTim wrote:

Wouldn’t any change of manufacturing process need to be certified?

These days, wouldn't that be self-certification anyway?

Phosphorus wrote:

StTim wrote:

Wouldn’t any change of manufacturing process need to be certified?

These days, wouldn't that be self-certification anyway?

The whole non-av industry is mostly „self certified“. This is not the issue!
The issue is that the internal change management and process qualification seems to be in a very poor state.

Aesma wrote:

lowbank wrote:

I did a bit of digging earlier today.
It appears that the pickle fork used to be made from made from aluminium billet. That material has good uniform properties and you are normally machining the component from a large over size piece of material.

They have changed to an aluminium close to size forging. In my experience the properties are inferior and inconsistent. Now I have no data on how much oversize the forging are, but I have properties to be inferior to billet.

It will be a cost reduction that’s going to bite them on the ar5e.

If they are lucky it’s a batch issue, if not it will be the whole fleet.

It's costing airlines, but does Boeing have to compensate them ?

One of the big claims of Boeing in regards to the 737NG was lower maintenance life cost than the A320. One or two pickle fork repairs and the cost of checking for it, with the associated down time, can have an influence on the cost of maintenance over the lifetime of a 737.

Phosphorus wrote:

StTim wrote:

Wouldn’t any change of manufacturing process need to be certified?

These days, wouldn't that be self-certification anyway?

That’s done through an internal change process.

And this is where training and experience is really needed.

A couple of examples, your drilling an 8 mm hole, your using a normal carbide drill. The way it works, your do a material assessment cut up on a part with the drills firsts use. You predict with experience it will drill 150 holes. You make 149 parts and hold them, drill the 150th and cut it up again to do a material assessment. That’s good. You set your tool life at 150, it’s then mandatory to change the drill every 150 parts. A new thru coolant drill comes on the market which could last longer and therefore be a cost reduction. A change control document is submitted which is assessed by a local control authority, they decide a cut up assessment is going to be an acceptable method of validation. The manufacturer states this drill can drill 500 holes. The control authority is going to ask that every 100th part is cut up allowing the previous 99 to be released if the cut up assessment is good. Controlled by a local team of experts.
Let’s say the 500th fails, you then cut up the 450th, if that’s good you release up to 449. If you cut up 460th and it fails you throw 50 parts in the bin and set tool life at 450.

If your going to change from billet to forging, in my company that would be submitted to the local authority. They will suggest validation requirements and crucially then send that on to a higher level of control. That level of control specialists will assess and decide if LCF or HCF testing is required or it could ground or flight testing.

I sit on a local control board about twice a month, it’s normally easy to know which you deal with and deal with and you escalate.