"An expert is a man who has made all the mistakes, which can be made, in a very narrow field." -Niels Bohr

These words frame the OH CR4P! blog, a place which encourages engineers to discuss, reminisce, and learn about mistakes, failures and mishaps made by those who have become "experts" the hard way.

The Day the U.S. Air Force Dropped a Nuclear Bomb on South Carolina

Posted April 23, 2019 7:34 AM by RSBenner
Pathfinder Tags: bomb Mark 6 Walter Gregg

It was March 11, 1958 and the Cold War was in full swing. A crew from Hunter Air Force Base in Savannah, Georgia, was taking part in Operation Snow Flurry. Their mission was to load a Mark 6 nuclear bomb into a B-47E bomber and fly to England to perform mock bomb drops. All of this would be timed as part of the operation.

Problems with the mission began early as the crew had difficulty loading the bomb onto the plane. The problems concentrated around a locking pin on the shackle mechanism that secured the bomb in place. Being conscious of the time ticking by, a hammer was used to facilitate the pin’s engagement.

Before the plane could depart, the locking pin was disengaged, per regulations, in case of an emergency during takeoff. However, when the bomber reached the proper altitude, the pin would not re-engage because of the same issues that occurred prior to takeoff. The bombardier went to the bomb’s location to manually engage the pin. While reaching for the pin, and to steady himself as he crawled over the top of the bomb, he mistakably grabbed the emergency bomb release lever and, as the name implies, released the bomb from its shackles.

The bomb first hit the closed bomb bay doors, where it hesitated for an instant until the weight of the bomb caused the doors to open and drop out of the plane. Imagine the look on the bombardier’s face as he hung on over the open doors, watching the bomb fall to the Earth.

The bomb fell on the farm of Walter Gregg in the small town of Mars Bluff, South Carolina. The concussion from the blast injured all of his family members, severely damaged their house, destroyed outbuildings and their garden, vaporized chickens and created a 50-foot-diameter, 35-feet-deep crater. Windows were cracked up to 5 miles from the point of impact.

But wasn’t it a nuclear bomb? It was. However, as part of this operation, the nuclear core was not installed in the bomb. The trigger part of the bomb, the part designed to begin the chain reaction of the nuclear explosion, did explode on impact and this is what caused the damage. Had the nuclear core been installed, the story would have been much different.







8 comments; last comment on 04/27/2019
View/add comments

Charles Hatfield, the Rainmaker, and the Hatfield Flood

Posted April 08, 2019 12:00 AM by RSBenner

In 1915, San Diego, California was experiencing a severe drought and officials were desperate to solve this problem. Enter Charles Hatfield, a self-proclaimed “moisture accelerator,” who had assisted 17 communities, from Texas to Alaska, by encouraging rain to fall in their areas. The meeting with Hatfield may have caused in a flood that caused more damage and death than any other in the history of San Diego County.

The city of San Diego made a verbal agreement with Hatfield. If he could make it rain enough to fill the 15 billion-gallon Morena Reservoir within a year, they would pay him $10,000. Hatfield agreed and went to work. He built a 20-foot tower near the reservoir and placed shallow pans on the top. In those pans, Hatfield combined 23 chemicals, his own secret recipe, and set the concoction on fire. The smoke from this fire, Hatfield claimed, would attract the clouds and provide the rain.

Drizzles began on January 1 of 1916. Initial excitement was soon quenched when, during the 5-day span beginning January 15th, approximately 17 inches of rain fell in the areas around San Diego. This amount of rain, after the months of dry conditions, was harmful, not helpful. Rivers overflowed, landslides occurred and the floodwaters washed away homes, bridges and livestock. The entire community of Little Landers, California was destroyed. Despite this destruction, Hatfield continued his work because, after all, the reservoir was not yet filled!

After a brief break, significant rains returned. With no time to recover from the last storm, the area was in trouble. On January 27th, unable to hold back the water anymore, the Otay Dam collapsed, sending a 40-foot-high wall of water down the valley. San Diego found itself cut off from the rest of the world with bridges, communications and roads all destroyed. Almost 30 inches of rain had fallen throughout the month of January and damages were estimated to be approximately $8 million in 1916 dollars (about $200 million today). At least twelve deaths are blamed on the floods, although some think that number to be too low.

But, the Morena Reservoir was now filled. Having fulfilled the specified goal, Hatfield dismantled his tower and went to the city for payment. However, having gotten much more than they bargained for, and since their contract was never put into writing, they refused to pay. Hatfield sued the city, but, after 20 years of litigation, the lawsuit was dismissed.

Did Charles Hatfield actually cause it to rain? There is much debate about the usefulness of his mysterious, 23-chemical formula or if he was a just sharp talking con man who had knowledge about weather patterns. Either way, many believe his experiments were the beginning of the modern practice of cloud seeding.



11 comments; last comment on 04/10/2019
View/add comments

Overweight Spanish Submarines

Posted February 18, 2019 12:00 AM by RSBenner
Pathfinder Tags: S-80 S-80 Plus spain submarine

The Spanish submarine navy is small – currently operating only three French-built, Agosta-class, S-70 submarines. During the 1980’s, when these boats were commissioned, Spain began planning for their successor, the S-80. Developed by the Spanish shipbuilder Navantia, one of the major enhancements of this new design was the incorporation of a bio-ethanol air independent propulsions (AIP) system which would enable the boats to be submerged for longer spans of time than their diesel-electric predecessors. Additionally, this new design touted highly automated systems allowing for a smaller crew, advanced armament capabilities and a significant decrease in the boats’ magnetic, infrared, visual and radar signatures. The first boat was scheduled to enter naval service in 2015.

However, the S-80 development came to a screeching halt in 2013 when, during testing, it was determined that weight imbalances caused the boat to lack proper buoyancy. In short, the submarine could submerge without a problem, but may not be able to resurface afterwards! During a review, it was determined that the 2300-ton displacement of the original design was overshot by more than 70 tons, thus causing the buoyancy problem. But, how did this happen? A former Spanish official stated that a decimal point was placed in the wrong spot early on during the engineering design phase of the project and 'nobody paid attention to review the calculations'.

General Dynamics, Electric Boat Division in Groton, CT was hired to find an answer to the buoyancy problem. The proposed solution was to lengthen the S-80 submarine from 71 to 81 meters in order to increase the buoyancy.

But wait, there is more.

In July of 2018, it was announced that the length added to the S-80 submarine, sometimes now called the S-80 Plus, caused the boat to be too long to fit into the 78-meter-long docks at the Cartagena Naval Base in Spain where it will be stationed. The port will need to be dredged and reshaped in order to properly dock the S-80.

As a result of all these changes, the per boat cost of the S-80 submarines increased from an originally estimated €439 million ($503 million US) to €978 million ($1.12 billion US). General Dynamics was paid €14 million ($16 million US) for their consulting fees which necessitated an estimated €16 million ($18 million US) expansion project at the naval yard. And the 2015 delivery date for the first boat of this class has been rescheduled to 2022.

Always check your work - especially your decimal points!

Note: As if the above problems were not enough, the AIP design will not be ready for installation until the third boat is deployed in 2026, with the first two boats receiving refits in 2032.





Image: https://www.elsnorkel.com/2017/07/el-sistema-aip-de-los-s-80-plus.html

23 comments; last comment on 02/22/2019
View/add comments

“Watch Your Units!” Part 3 – Space Mountain and the Gimli Glider

Posted November 19, 2018 12:00 AM by RSBenner

“Watch your units!” Unit conversion errors can be found most anywhere, even in Japan and Canada. For Part 3 of this series, we will look at Tokyo Disneyland and Air Canada.

Tokyo Disneyland's Space Mountain Derailment

On December 5, 2003, one of the rocket vehicles of the Space Mountain attraction at Tokyo Disneyland derailed just before the end of the ride, forcing it to come to a sudden stop. Luckily, no one was injured due to the derailment. However, investigation proved the problem was caused by unit conversions.

The Space Mountain attraction at Tokyo Disneyland opened with the park on April 15, 1983. Based off the original, which opened in Florida in 1975, the coaster was built and maintained using US customary units (USC). In 1995, design specifications for the axles and bearings were updated to reflect the International System of Units (SI). However, the original drawings were never purged. So, when axles were ordered in August of 2002, the incorrect (USC) drawings were used, resulting in incorrectly sized axles. This resulted in a gap between the axle and its bearing to be over 1 mm, while the design specified a gap of 0.2 mm. Excessive vibration and stress occurring during operation due to this large gap and caused the rear wheel axle to break and the vehicle to derail.

The attraction was closed for two months during which all axles were inspected and (I assume) old drawings were purged.



The Gimli Glider

On July 23, 1983, Air Canada flight 143 departed Montreal for a transcontinental flight to Edmonton. However, the flight never made it there due to faulty fuel gauges and an incorrect unit conversion.

About one hour after departure, while cruising at 41,000 feet, both engines stopped operating. In addition to flight power, these engines also supply electrical power to the instrumentation and supply power to the hydraulic system. With the engines off, only a few battery-powered emergency flight instruments were operational and very limited hydraulic power was available. Despite these difficulties, the crew glided the aircraft and successfully landed it at a closed air force base, Station Gimli in Manitoba, with only minor injuries to the 69 people on-board.

Investigation into the “Gimli Glider,” as it became known, determined that a major factor in this near disaster centered around a unit conversion issue. In 1983, Canada was in the process of converting to SI units, and their new fleet of Boeing 767 jets would be the first to completely use this system. Because of the inoperative fuel gauge on Flight 143 the fuel level had to be checked manually. Calculations were then performed in order to determine how much fuel needed to be added to the plane before take-off. During this calculation, the ground crew, less familiar with the SI system, used a USC conversion factor in error. This resulted in the addition of 4,917 liters of fuel rather than the required 20,088 liters. Therefore, the reason for the engine failure was determined to be lack of proper fuel supply due to a faulty unit conversion.

What can I say except, “Watch your units!”




8 comments; last comment on 12/10/2018
View/add comments

“Watch Your Units!” Part 2 – The Space Program

Posted October 22, 2018 12:00 AM by RSBenner

“Watch Your Units!” Part 2 – The Space Program

“Watch your units!” There is that phrase again! I can hear it in my sleep.

The most common place modern engineering students will run into unit conversion problems is when dealing with the International System of Units (SI) versus US customary units (USC). For Part 2 of this series, we will look at these types of unit conversion challenges from the US space program.

The Mars Climate Orbiter

Launched on December 11, 1998, the mission of the Mars Climate Orbiter was to maintain an orbit around Mars and study the Martian atmosphere and climate. This was not to be. Instead we have the most famous, and most expensive, example of a unit conversion error.

On September 23, 1999, as the Mars Climate Orbiter attempted to insert itself into its first orbit around Mars, communication was lost and was never regained. Subsequent investigation found that USC units were used in the ground software system while all other systems operated in SI units. This caused the trajectory figures to be off by a factor of 4.45 and resulted in a closer approach to Mars than expected. It is assumed that, since the orbiter was too close to the surface of the planet, heat and drag from the atmosphere destroyed the Mars Climate Orbiter.

“People make errors. The problem here was not the error. It was the failure of us to look at it end-to-end and find it. It’s unfair to rely on any one person.”

-Tom Gavin, NASA Jet Propulsion Laboratory

They failed to ‘watch their units.’ It was a $125 million mistake.



Constellation Program

Who would think that unit conversions could be so expensive? In the case of the Constellation Program, it contributed to the cancellation of the program.

Begun in 2005, the Constellation Program was developed as a replacement of the aging space shuttle program. The focus of the program would be on manned flights with plans to return to the moon and an ultimate goal of a trip to Mars.

One of the underlying objectives of this program was the implementation of SI units across the entire program. In addition to new designs developed for the program, the ground and mission infrastructure (launch pads, test stands, etc.) that was largely developed in the 1960s for the Apollo Program would have to be updated to achieve this goal. However, it was estimated that the unit conversion costs would be approximately $370 million! In a futile attempt to support the program’s budget, it was ultimately decided to drop this plan and retain the USC units. Unfortunately, this did not save the program.

Although overall schedule and financial problems ultimately led to its cancellation in 2010, addressing units of measure is listed as one of the “Lessons Learned” in a NASA publication about the cancellation of the Constellation Program.




20 comments; last comment on 11/08/2018
View/add comments

Previous in Blog: “Watch Your Units!” Part 1 – Even Columbus Had Trouble With Units  
Show all Blog Entries in this Blog