Hacker News Reader

Comment by wly_cdgr | Hacker News Reader

"Ms Thomas suggested that in some cases she gave metal positive results because she thought it was "stupid" that the Navy required the tests to be conducted at -100F (-70C)" - was she right, or was there a sound rationale for this?

View on HN

>was she right, or was there a sound rationale for this?

It literally doesn't matter. You've been hired to do a job, or contracted to do a job. You decide on the front end whether it is a job you can perform or not. Not once the contract is signed.

Or, you use official channels to express your disagreement with the requirement and up the price by whatever it needs to be upped by.

You don't make arbitrary changes to expectations based on your gut feeling without telling anyone. That's called fraud at worst, and just plain dumb at best.

If it results in some one death, it could be considered manslaughter.

I’m curious about the requirement too but I agree that it doesn’t matter. The client wants their steel tested at -100 deg F, so do it. Someone who is a Director of Metallurgy should take pride in delivering the product that was asked for, not in think they know better than the Navy.

The Navy has a mandate to operate their subs anywhere on earth. Global lowest temperatures on record are around -90C, with the lows (in [0]) outside of Antarctica still hitting -68C. Being in the ocean will mediate some of those lows, but there's still the potential for exposure to very low temperatures on the surface.

[0] https://en.wikipedia.org/wiki/Lowest_temperature_recorded_on...

In general, testing systems outside of their normal operational profile is a risk mitigator. Because that black swan event can happen.

Also, there are lives ate stake--not just reputations.

It's a submarine, -70C means it's been frozen into a colder iceberg than exists on this planet.

I am curious if this is a mitigation for repeated stress cycles or an analogue of metallurgical properties in deep ocean water.

... or it's sitting at the surface at a very cold place. (yes, it obviously won't completely cool down to that temperature, but is exposed to temperatures at least close to it)

My assumption is that very cold temperature conditions may be a proxy for some other set of conditions which are classified. Do cold temperatures mimic something like neutron bombardment, in a way that would be metallurgically useful?

The short version is, no, there is almost no chance that this is related to anything exposed to a relevant radiation flux.

Steels can undergo a transition to becoming brittle when they get cold (called a “ductile-to-brittle” transition). It’s important to know what the properties would be like in this regime and -70C is enough to get there (even 0C can be enough, depending on the alloy).

The reason this person may have thought the -70C test eqs stupid is because a sub will never be working in conditions much colder than the freezing temperature of water (which is not strongly pressure dependent, btw), since the water would want to freeze - not good for the boat.

I was talking to an engineer from Raytheon about a project and one thing came up I found interesting. When a system is developed the prototypes are built using lab conditions meaning very tight tolerances often coupled with equally ridiculous tests. This ensures the thing works reliably in any situation it might encounter and possibly beyond. Once you have a working prototype then the manufacturing design phase begins where you then look to relax the stringent lab standards. This means lower tolerances that still allow the thing to be reliable while now making it more easily manufactured at volume.

BUT That isn't always the case because of deadlines and and some parts or the entire project never move past lab spec and retain insane tolerances. So yes, sometimes tests are "stupid" but that doesn't mean you ignore them and lie because of "feelings".

In the case presented here I think the accused wasn't very forward thinking. She only thought of the sub as a purely water born ship but fact is they operate both at and below the surface. They also need to operate near the earths poles where they might surface for hours exposing the hull to temperatures damn near -100F.

I have no idea if there are systems on a submarine that might experience extreme temperatures. But for the hull, since it is constantly surrounded by liquid water, I would expect that it does not regularly experience temperatures less than ~0C. Maybe there's a concern when surfacing in the arctic? That said she obviously shouldn't have faked data whether it was stupid or not.

The sound rationale for the tests to be conducted at -100F is that the Navy contracted for the tests to be conducted at -100F.

If someone, and particularly a military organization, asks that something be tested according to a certain spec, then you do it. Who's to say the steel was even actually being used for submarines?

Having worked for a test lab, I can say that she should have just followed the test procedure.

At the same time, we don't know her qualifications other than that she is a 'metallurgist'. We do know that the Navy assessed no impact in the end. Maybe she used her knowledge of metallurgy to determine that the 'stupid' requirement was not needed, and maybe she was right. But she obviously handled it the wrong way.

-70C can be expected in Arctic.

Not in sea water:

> Deep ocean water has a very uniform temperature, around 0-3 °C

https://en.wikipedia.org/wiki/Deep_ocean_water

Subs come to the surface though.

If it's on the surface for an extended period of time, presumably upper elements of the hull could reach that temp?

Even exposed portions of the hull can be expected to not reach equilibrium with surface air. There will still be liquid water splashing around and also conduction from the warmer parts of the hull. Maybe the sail surface could get much colder but I doubt that it's subject to impact hazard under these conditions.

Possible, but that is also the time when the hull is under the least amount of stress.

So maybe they are worried about sitting on the surface of the Arctic in the dead of winter for an extended period of time and then crash diving?

Would you want your unit test to not run because it thinks "there's no way this code could be broken"?

Whether it's stupid or not is not her call to make.

This is how we get user-facing error dialogs with the text “This error will never happen!”

Perhaps the flow of water or air nearby could lower the temperature? Maybe pressure changes? Maybe coolant systems next to or integrated into the part? Hers was not to reason why…

Reply on news.ycombinator.com