In the previous explanation, we introduced the historical context at the start of Comet's development and the Comet's concept.
In the previous episode, the government had actually ordered a jet mail carrier capable of high-speed transatlantic flight, but the development and manufacturing company that received the contract (de Havilland) inexplicably declared that it would develop a large four-engine jet passenger aircraft, and the government accepted this declaration.
The government's actions led to orders for two aircraft from the Ministry of Munitions and seven from the government-owned British Overseas Airways Corporation.
Let's take a look at the development of the Comet from here.
As I've mentioned before, throughout history and across cultures, the typical product development process follows this flow.
First, let's try to imagine what was happening during the conceptualization and layout (rough design) stages.
Concept and layout (rough design) for the Comet development.
First, let's consider the approximate number of engineers needed to develop a massive machine or product like the Comet.
With a machine this complex and massive, at least 300 engineers are involved in the development of the machine itself, and if you include related companies, the development project would involve around 1000 people.
In most cases, aircraft engines are basically purchased from other engine companies, but Rolls-Royce, a prestigious British engine company, was struggling with the development of axial-flow jet engines and was therefore unavailable.
Since there was no other option, they apparently decided to use their own Ghost product, which was among the most powerful centrifugal jet engines at the time and offered flexibility because it was developed and produced in-house.
In that case, at least 500 engineers, including those from affiliated companies, would be involved in the development of the application of the Jet Engine Ghost.
Total 1It will require more than 500 engineers, and with the involvement of management, marketing, and manufacturing departments, it will be a massive project involving a total of 2000 to 2500 people.
Now let's look at the concept and layout.
Comet's concept
Last time, I covered the basics of how to venture into uncharted territory with a new product.
It's unlikely that the Comet was thoroughly considered during its initial planning stages, perhaps due to a sense of urgency.
ProbablySince the crucial jet engine and pressurized cabin technologies were already established to a certain extent, it is highly likely that the concept was conceived using the bottom-up approach described previously.
Given that this was a large-scale, government-led project and a world first, it seems they were in a great hurry, so the period from concept decision to the start of development was probably quite short.
This is where the wasted year of the previous Branzon Committee comes into play.
Furthermore, the government's plan changed from a high-speed mail transport jet capable of transatlantic flights to a large four-engine jet passenger aircraft proposed by the development and manufacturing company.It is likely that most of the two years between receiving the order and starting development were spent on defining the concept.
At this stage, I believe concerns about stress concentration and fatigue failure were raised, albeit tentatively.
Although not entirely sufficient evidence, fatigue tests were conducted on the pressurized chamber during the development of the Comet rocket.
However, I think that because it was already known technology, it wasn't considered a particularly important issue.
The problems up to this point are:This is just my speculation, but it seems that the emphasis was placed more on how quickly they could mass-produce the product, due to their anxiety and urgency, rather than on the technical challenges of venturing into uncharted territory (they were probably quite frightened by America's fierce pursuit).
In other words, there wasn't enough consideration given to what was unknown territory, what technical challenges existed, and how to solve them.
In my own experience, I have seen several projects fail within my company due to various problems during development (technical issues, cost issues, etc.) resulting from insufficient consideration at this stage, and I have also experienced minor issues myself.
The worst case I've ever seen is a project being canceled just before mass production, after all the necessary molds, machine tools, and dedicated production lines had already been prepared.
Despite having several completed new models right in front of them, mass production has been halted.
I cannot disclose the specific amount of the loss.In terms of scale, hundreds of billions of yen and several years of valuable time were wasted.
Mistakes at this stage have a very high chance of being fatal to the project's failure.
hereThe ability to create highly precise concepts and layouts is a measure of a company's true strength that cannot be measured solely by economics.
UnfortunatelyIt is likely that the British government and the development and manufacturing company (de Havilland) at the time had already stumbled significantly due to their impatience.
Next, let's look at the design phase.
Comet Development Design
Based on the concept and layout (rough design) of the Comet described above, the necessary parts for the Comet and the required specifications for each part are roughly determined.
Based on those established requirements, each designer conducts detailed studies and draws up the design drawings for the parts.
Typically, from the stage of drawing the design plans, members of the testing department (testers) are involved in the process, deciding on the specifications of the parts and what will be verified through testing.
Now, let's take a closer look at the design in detail.
Design process
Let's consider the stress concentration at the corners of the pressurized cabin windows and the antenna holes, which was one of the major causes of the Comet crash.
NormallyThe design and testing departments decide together to "pay close attention to and check areas where stress concentration is likely to occur during testing."(Of course, the designers also go to see the actual tests.)
Even in modern design, if calculations show that a component has sufficient strength, we ask the testing department to carefully monitor areas where stress concentration is likely to occur during testing.
While it's possible to roughly predict how much stress will concentrate based on the shape during the design phase, the exact amount of stress (load) generated cannot be determined until the stress is actually measured.
Similarly, when it comes to fatigue failure, we conduct thorough discussions with the testing department during the design phase, deciding what kind of tests to use and which parts to carefully examine.
Unfortunately, the lack of thoroughness in the testing during Comet's development, the results of the RAE reproduction tests suggest a significant lack of attention, consideration, and analysis regarding stress concentration.
Next, let's consider the investigation into fatigue failure, which was the direct cause of the Comet crash.
Even if there was a lack of knowledge about fatigue failure, if the design had been properly considered to some extent, it would be impossible for the actual aircraft to experience fatigue failure after only about 1000 flights.
From the perspective of an average engineer, a difference of approximately 10-20% between the lifespan of a design based on insufficient knowledge of fatigue failure and the actual lifespan of the machine might be possible. However, if the actual lifespan is less than 10% of the design's lifespan, something is clearly fundamentally wrong.
In my opinion, the preconditions for considering the issue were completely wrong, even before the problem of fatigue failure was addressed.
In other words, I believe that the fatigue failure studies were completely meaningless because the way the Comet was used and the environmental changes (such as pressure changes due to altitude changes) were already far removed from reality at the conceptual and layout stage (this highlights the shortcomings of insufficient consideration in the conceptual and layout stages).
This is just my imagination of what the actual design site might look like.
Next, let's consider checking the design drawings and the management system.
Checking and managing design drawings
Regardless of time or place, in machine and product development, design drawings are not officially issued until they have passed numerous checks by people in the testing department, the designer's supervisor, and others.
Furthermore, in large, prestigious companies and large-scale projects, the design drawings are typically checked by the person in charge, then by their immediate supervisor (manager level), and finally by someone from the testing department.
Furthermore, the process may be reviewed by higher-ranking department heads (executives), the overall project manager for Comet, and in some cases, even by the management team.
In addition to design drawings, the designers are also required to submit design review documents and calculation reports, and these are checked just as rigorously as the design drawings.
Moreover, it's common practice to have a system in place where the design drawings and feasibility studies undergo multiple internal reviews (design reviews, engineering reviews), and development cannot move forward unless they pass rigorous checks by many people, including not only the design department but also management.
Furthermore, in the case of a national project, it is highly likely that the design was subject to checks not only by the company but also by the government at key stages.
In other words, under such a rigorous check and management system, it is highly unlikely that all those involved would overlook something as fundamental to mechanical design as stress concentration at window corners and antenna holes.
NormallyAreas where stress concentration can lead to premature fatigue failure can be identified through design drawing checks, management systems, and design reviews.
ましてやThe results of fatigue failure life prediction studies will be scrutinized very rigorously, as this is a critical safety issue.
It is likely that the development and manufacturing company, de Havilland, and the government had ample experience in aircraft development, both military and civilian, even before the Comet, and that the rigorous checks and management systems described above were already in place (establishment of a development system).
Nevertheless, the fact that Comet aircraft continued to crash suggests that the system for checking and managing these design plans was either completely ineffective or completely inadequate.
Even with a lack of knowledge about stress concentration and fatigue failure, I find it inconceivable that all of Britain's top-tier elite engineers and veteran engineers who fought through brutal wars would overlook such a basic fact.
Even in pre-war Japan, a rigorous system for checking and managing blueprints was already in place.
Taking the development of the Zero fighter (around 1940) as an example, the leader, engineer Horikoshi (who was probably a section chief at the time), provided design guidelines, and basically, his younger subordinates drew the blueprints and performed the calculations (even in modern development, the leader doesn't draw the blueprints; the blueprints and detailed calculations are carried out by the person in charge).
Engineer Horikoshi meticulously checks the results and design drawings submitted by his subordinates, and in some cases, the review and design process may need to be redone.
From what I've gathered, he was apparently quite strict when it came to drawings and design proposals (though he was generally a kind person).
Furthermore, the military (Navy), which is the client, takes the lead in conducting design reviews at key junctures.
If a soldier notices any shortcomings, they will point them out, and in some cases, the work may need to be redone.
Thus, design drawings are not officially issued until they have undergone numerous rigorous checks.
For more details on this topic, see the excellent book: "Zero Fighter." It covers the youth of respected Japanese predecessors such as Mr. Horikoshi, Mr. Sone, Mr. Honjo, and Mr. Teruo Tojo (brother of Hideki Tojo), the first president of Mitsubishi Motors.
Although the environment may be different today, the feelings of the developers, the atmosphere on the ground, and the methods—their passion—remain completely unchanged.
I highly recommend reading it if you're interested.
It is extremely difficult to understand why Britain around 1947 was unable to do what Japan was able to do before 1940.
The only plausible explanation is that, due to the strict time constraints and pressure of development, while there was a system in place for checking and managing the design drawings, it was likely quite lax.
I myself have seen design drawings submitted for models with extremely short development timelines, even though they would normally be rejected outright, due to the lack of time and pressure.
Naturally, this almost always ends up being a major problem during the test, requiring a redo and taking even more time.
In the worst cases, I've seen products mass-produced with lax testing, leading to problems in the market.
This is just my personal experience, but I once had a terrible time when I was forced to deal with market problems related to a product I had absolutely no involvement in during its development.
For some reason, even though I wasn't involved in the development, I was tasked with summarizing the lessons learned and measures to prevent recurrence during development, and was severely reprimanded by upper management for not being able to detect the problem beforehand (upper management didn't know who developed it, and didn't care).
Well, thanks to that, I was able to thoroughly study, practice, and master reliability engineering and root cause analysis (FTA, FMEA, etc.), so I guess it was all for the best.
Summary of problems encountered during the design of the Comet
Probably something like thisIt is believed that errors occurred simultaneously at the Comet design site, in the checking of design drawings, and in the management system, resulting in the release of immature design drawings.
In other words, what I'm trying to say isI believe the cause of the Comet crash was not only engineering stress concentration and fatigue failure, but also significant flaws in the organizational, management, and development systems of the development and manufacturing company (de Havilland) and the British government, mainly stemming from impatience.
Typically, organizations, management systems, and development systems complement each other to minimize reliance on the technical skill levels of individual employees, allowing companies to maximize their potential.
Considering the technological background at the time, I believe that predicting stress concentration and fatigue failure was not impossible at the design stage, provided that the organization, management, and development systems were functioning properly.
There were at least five opportunities for verification: checks during the conceptual stage, checks by the designer, checks by the testing department, checks by superiors, and checks during the design review (in reality, there were many more).
そのThe probability that none of the checks revealed any concerns about stress concentration or fatigue failure is quite low.
This covers the conceptualization and design phases, and I think you can see that there were already significant problems.
Next time, let's consider what happens after the ground testing of the Comet missile.
Just a little anecdote, but the actual time a designer spends drawing blueprints is less than 10% of their total work time.
My work time mainly consists of meetings with other departments, preparing documents for drawings, and coordinating and reviewing with affiliated companies.
Moreover, project leaders and chief engineers (who don't know the finer technical details) are mostly only concerned with costs and coordinating with others, and are unable to do any technical work (and naturally, they don't have time to draw blueprints).
The happy period of being a designer, where you're always doing calculations and drawing blueprints, that you all imagine, ends after about five years at the company (how much experience you gain during this time will determine your future success).
To those who found this article helpful in understanding design:
Since we're on the subject, I'd like to recommend a book that's essential for mechanical design.
To be honest, the content is extremely unhelpful, but it can be used like a dictionary when you forget the details. If you read this article, you should be able to understand the content and use it effectively. It also includes commonly used standards, making it quite useful.
If you don't already own one, I highly recommend getting one, even though it's a bit pricey. However, new ones are expensive, so if you're considering buying a used one, I strongly recommend checking that the surface roughness conforms to the new JIS standard.
Comment: