Episode 19: Shownotes & Transcript

Welcome to The STEM Sessions Podcast.  I am your host, Jarl Cody.

Don’t usually discuss current events on this podcast

  • Typically don’t have anything to add that hasn’t already been offered by countless other authors and analysts
  • Though mostly my production of this podcast is simply too slow to discuss the news
  • So decided not to fall into the cycle of relying on current events commentary to drive listeners
  • Then again, I haven’t really done anything else to lure listeners to this podcast
  • And to be fair, this news is a month old at this point – does it still count as a current event?

That said, the current event I discuss here really got under my skin – both as an engineer and a tax payer

  • I suppose this is a rant of sorts
  • Though to balance the rant, I provide other research
  • Not a technical analysis of the event, mind you – that’s been done elsewhere
  • Instead, I share what my teams have done when faced with similar choices

I also discuss the relationship between engineers, our customers, and the requirements they give us

  • And why even though we engineers are typically smart, we don’t always know better

This is The STEM Sessions Podcast Episode 19 –  Inconvenient Requirements

Nov 2021, Elaine Thomas pleaded guilty in U.S. District Court to major fraud on the United States

  • Ms. Thomas is the former Director of Metallurgy at Bradken Inc.
  • Bradken is the leading supplier of cast high-yield steel for U.S. Navy submarines
  • Ms. Thomas will be sentenced in February 2022, facing up to a year in prison and a fine of up to $1 million
  • Bradken, for its part, has entered a settlement of nearly $11 million

So what is this major fraud that has a company paying $11 million and a retired employee facing a year in prison?

  • Bradken produces steel to be used in the hulls of U.S. Naval submarines
  • As part of the certification process, the steel is tested to ensure it meets minimum strength and toughness requirements so that it the submarine will maintain structural integrity during collisions and other wartime events
  • Ms. Thomas, in her role as Director of Metallurgy, falsified the results of these tests – changing the results from failing to passing
  • Not just once or twice, but over 240 production lots of steel, and over the course of 30 years
  • In all cases, the steel was delivered to the Naval contractors as if nothing was wrong

In the company’s defense, it appears no one else was involved

  • Fraud was discovered in May 2017 when an lab employee was reviewing legacy test cards and noted they had been altered
  • Bradken is still on the hook, of course, but the company’s openness when the test cards were discovered is likely why its settlement was civil and not criminal
  • Judgement on Ms. Thomas, however, is criminal

There’s no justification for falsifying test results once, let alone 240 times across three decades

  • But I can almost understand the reasoning for doing so once
  • Let’s say you have a long history of your product always passing its qualification tests, then you randomly experience a failure
  • Let’s say the failure is only by a few percentage points
  • You double check your processes, your equipment, your raw materials… everything is as it should be
  • You suspect the product is good and it’s most likely a testing error
  • Correct action would be to run the test again, but schedule is tight, so you fudge the results and push out the product
  • Promising yourself if it happens again, you’ll do a deeper investigation

Like I said, while that reasoning makes some sense, the scenario I just laid out is still completely unacceptable

  • Instead, the only course of action is to bring up the issue with your team lead, the program lead, subject matter experts – whoever is needed to be in the loop
  • They may decide the program should eat the delay in schedule and retest the product
  • They may decide to ask the customer for a waiver or other authorization to ship
  • Either way, the situation is above board, and everyone agrees with the path forward
  • I’ve worked on numerous programs where last minute gotchas pop up – they’re not uncommon
  • But we’re always open and work through it as a team

So what was the motivation behind falsifying test reports for 30 years?

  • According to indictment, the Navy believes Ms. Thomas “knowingly devised and executed a scheme with the intent to defraud the United States Navy, and to obtain money and property by means of materially false and fraudulent pretenses and representations”
  • Countering this claim, Ms. Thomas, through her lawyer, explained while she “took shortcuts and made material misrepresentations… [she] never intended to compromise the integrity of any material… [and] this offense is unique in that it was neither motivated by greed nor any desire for personal enrichment”

For the time being, let’s take Ms. Thomas at her word and assume neither greed nor personal enrichment, such as bonuses for beating schedule, was the motivation

  • That leaves laziness and arrogance
  • Laziness because she continued to falsify test results for 30 years instead of fixing the problem – she was the director of metallurgy, a position with pull, a position that could have easily enacted change
  • And arrogance because she felt the requirement her steel was failing – testing the material down to -100F (-73C) – was “stupid” – that’s the word she chose to use it in court

Speaking as a mechanical engineer in the aerospace and defense industry, and one who takes a lot of pride in the hardware we produce, this brush off of a requirement really pisses me off

Worked on defense contracts for over 20 years, and for half I’ve been directly involved in negotiating, defining, and allocating requirements

  • And that experience has shown me many of requirements flowed to use are stupid and overly constraining and make our job more difficult
  • But you know what? We don’t get to ignore them just because we think we know better!

When a customer, the U.S. Department of Defense in these cases, sends a Request for Quote (RFQ) to its contractors, the RFQ is accompanied by a requirements document

  • Level of detail and definition varies RFQ to RFQ
  • Contractor reviews the requirements, and develops a bid based on the requirements and follow-up questions

After winning the contract, systems engineers review the requirements in detail and begin allocating those requirements to the pertinent subsystem teams

  • During requirements review, it is common to find missing, incomplete, confusing requirements
  • Engineers job to work with customer to resolve these issues
  • May also find requirements that are over constraining or contradicting
  • Also the job of engineering team to resolve these issues with customer
  • In the case of over constraining requirements, we’ll ask for a change using reduced schedule, reduced cost, simplified testing, cross-program compatibility as justification

An example might be the operating or storage temperature range is so wide it precludes the use of industrial rated electrical components (resistors, capacitors, relays, etc), and instead need military rated components

  • Industrial rated components are widely available, and comparatively inexpensive
  • Military rated components are often times made to order and expensive because there is no continuous demand for them
  • We present the customer with the opportunity to reduce cost and risk to schedule by modifying the temperature ranges
  • Customer either agrees to change the requirement, or says industrial components are acceptable but the requirement needs to remain for other aspects of the design, or tells us no, the requirement needs to stay as is
  • They don’t need to give justification for their answer

Additionally, Requirements and results of verification testing are the first artifacts you look during a failure analysis or when determining if a legacy system will work on other platforms or future programs

  • Accurate traceability is therefore vital
  • Poor traceability could have negative ripple effects down the line after the program’s engineers have all moved on to other work

Why does the Navy invoke seemingly over the top, “stupid” requirements?

  • April 1963, the USS Thresher, a nuclear submarine, was lost during deep dive testing east of Cape Cod, MA
  • After experiencing a mechanical failure, it was unable to surface
  • Wreckage found 8000 feet below the surface
  • All 129 crew members and shipyard personnel aboard the Thresher dies
  • Second deadliest sub incident on record
  • Subsequent analysis, determined implosion occurred at 2400 feet (730 m) below the surface
  • As is navy tradition, Thresher was never decommissioned and remains on “eternal patrol.”

Navy created a set of standards, named SUBSAFE, to make sure this loss would never be repeated

  • SUBSAFE has a simple doctrine: In event of any problem, the submarine must be able to reach the surface
  • Impacts all areas of the submarine life cycle: Design, Material selection, Fabrication, Testing, and Maintenance
  • Certification gets as specific as recording the source of the alloy for every piece of equipment that’s SUBSAFE compliant
  • Prior to the implementation of SUBSAFE, the U.S. Navy lost 16 subs (nuclear and otherwise) including the Thresher
  • Since implementation of SUBSAFE, no submarine built to those standards has been lost

Similar to difference between airborne and space in aerospace engineering

  • We have typical ariborne requirements and typical space requirements
  • Compared to each other, space seems overdesigned
  • Space requirements are much stricter because you only get one shot
  • If it doesn’t survive launch, you’ve wasted years of money and caused significant setbacks in technology development

Requirements are written and flowed down for specific reasons, and those reasons do not need to be divulged by the customer

  • As the engineer you may not always understand those reasons or agree with them
  • There may not even be a technical reason the requirement is written as it is
  • Requirement may be out of date or it may be a carry-over from another program and not directly applicable to the new one or its written out of an abundance of caution
  • In this case of the submarine steel, it’s very possible the -100F requirement includes significant margin of safety which is why no failures in the field were reported
  • And the Navy confirmed this during follow up testing which showed the structural integrity of vessels aren’t compromised

But you can’t rely on dumb luck to save your butt

  • As an engineer, even if you know that a requirement has a significant built in margin of safety, to the point you feel the requirement is stupid, it is not your call to ignore it or falsify test results
  • That stupid requirement may be in place to prevent catastrophes and loss of life

Thank you for listening to The STEM Sessions Podcast. 

This episode was researched, written, and produced by Jarl Cody.

Here at The STEM Sessions, we strive to share accurate and complete information, but we also encourage you to do your own research on the topic we discussed to confirm the accuracy of what we’ve presented.  Corrections are always welcome.

Shownotes, contact information, and details of our other activities can be found on our website thestemsessions.com

If you received value from this episode, and wish to give some back, please visit thestemsessions.com/valueforvalue for ways to support the podcast.

Finally, please remember STEM is not a tool exclusive to experts, policy makers, and talking heads.  Every presenter is susceptible to unconscious and, sometimes, deliberate bias, so always verify what you read and what you’re told. 

Until the next one, stay curious.


“Metallurgist admits faking steel test results for US Navy subs”

“Former metallurgist lab director pleads guilty to major fraud on USA”

“Metallurgist Admits She Falsified Test Results for Steel Used in Navy Submarines”

“Metallurgist admits faking steel-test results for Navy subs”

“Navy Has ‘Mitigated’ Risk of Suspect Steel From Company in Federal Fraud Case”

“Declassified documents shed new light on notorious sinking of USS Thresher”


“USS Thresher (SSN-593)

Leave a Reply