[jacquesm]

Not really. I work with a company that does this kind of stuff (deep see exploration vehicle design), materials science is a tricky but very interesting subject. When someone seemingly does 'the impossible', subjects paying passengers to extreme risks and throws the rulebook on safe design out of the porthole you can bet there will be an entry in 'Risks Digest' at some point or another.

Carbon fiber is a great material, if used where it shines. For this application I don't think it is a good choice, even if it worked a few times. What I didn't understand is the focus on the electronics and the control system, those are irrelevant if the hull isn't going to work out.

[jcranmer]

> What I didn't understand is the focus on the electronics and the control system, those are irrelevant if the hull isn't going to work out.

Two reasons, I think. First, the (apparent) shoddiness of a control mechanism is easily understood, whereas the reasons why carbon fiber makes a poor choice for a pressure hull are more difficult to understand. Consequently, the use of a game controller is a more salient example of a company cutting corners than their choice of material in the pressure hull. (This is somewhat undermined by the fact that a game controller actually isn't necessarily a problem--Ars Technica ended up running articles on consecutive days with the first one being a tee-hee-they're-using-a-game-controller article and the second one being a why-the-military-uses-game-controllers article.)

The second reason may be that some people would rather focus on the potential failure that admits a possibility that the inhabitants are alive instead of potential the failure that guaranteed their death before the search started.

[ansible]

> Carbon fiber is a great material, if used where it shines. For this application I don't think it is a good choice, even if it worked a few times. What I didn't understand is the focus on the electronics and the control system, those are irrelevant if the hull isn't going to work out.

That's all true, but the use of the Logitech game controller and and other consumer-grade electronics rightfully drew scorn of its own.

I deal with automotive electronics, which is less rigorous than aerospace and/or military applications, never mind deep sea operation. Even so, in the automotive space, there is considerable attention paid to making the housings, cables and the electronics themselves robust with regards to the environmental conditions. Heat, cold, moisture, dirt, vibration, salt, etc. The idea of using non-redundant, consumer grade electronics to control a deep sea vessel is ridiculous. Just being regularly exposed to salty air could induce corrosion in critical components.

[guax]

I think you're under estimating the controller and over estimating your understanding of the conditions it would live on the vessel. Why it's non redundant? There were likely spares around, if it misbehaves you can get a new one in seconds.

[ansible]

> Why it's non redundant? There were likely spares around, if it misbehaves you can get a new one in seconds.

So you are saying they were carrying around a spare controller, and a spare everything else inside the sub? At all times?

[jacquesm]

They in fact did say somewhere that they carry 'a couple of spares'.

[HPsquared]

It really depends on the environmental conditions. In the case of a deep sea submersible you might expect that the controller will always be in controlled indoor conditions, depending on when/where they open the hatch. In normal operation it's going to be dry, and if there's a leak the controller probably isn't going to be of much use anyway. (same goes with the rest of the electronics).

A car is of course much different, most components have to operate for years without maintenance in a very uncontrolled, varied set of environments.

[ansible]

Salty sea air and sea spray are all around you on a ocean-going ship. I'm sure they weren't intentionally dunking the Logitech controller into sea water, but that doesn't mean that it couldn't possibly be effected by the environment.

https://pomametals.com/salt-air-inland-distance-for-metal/

The Logitech controller isn't sealed for moisture. A single drop of sea water on it can slide inside and start the corrosion process. It looks fine on the outside, because the plastic isn't as affected.

Most of the buttons on a commercial grade PC input device use conductive carbon pads pressing against bare metal contacts on a PCB. Yes, the contacts may be gold plated, but there is still exposed copper on the sides.

Heck, even ordinary contamination from skin oil has caused one my keyboard's keys to stop working. I was able to take the keyboard apart and clean it, fixing the problem. The keyboard was never exposed to anything beyond a normal office environment, and never had liquid spilled on it.

Marine grade switches are waterproof and dustproof:

https://newwiremarine.com/marine-grade-switches/

If you take care and seal the soldered wires to the switch electrical contacts, it should be good for years and years of use, even at sea.

[WORMS_EAT_WORMS]

My point was highlighting how strange it has been reading commentary of the event online. I mean no offense at all (!) but even your comment plays into it perfectly in a way.

[jacquesm]

If this is the first time you observe this then you're going to be pretty bored soon because most discussion about such subjects follow that exact same pattern.