Comment by addaon
3 years ago
> and what would that change?
Suitability for purpose. Some obvious ones:
Defined and validated environmentals (temperature, voltage, and in this case pressure).
Qualified components — capacitors chosen for lifetime rather than shaving a cent, perhaps avoidance of MEMS oscillators with helium sensitivity.
Failure analysis. Low and understood probability of fail-unsafe conditions (short circuit), mitigation for those risks, fume-proof and fire-proof PCB materials to protect the sealed environment in case of failure.
Redundancy to handle failures anyway. Multiple independent strings so that single-point failure lead to partial loss of lighting, not all of it.
Load ahedding, eg dropping all but one string at a known voltage above minimum voltage, to save power for other more critical loads during system failure scenarios.
Yes, if one had the budget to do all those things, from scratch, better than an existing component manufacturer.
Not many companies have NASA levels of "throw money at it until it works, and every part has been signed off on five times."
Absent that, I'm having trouble seeing how custom > COTS.
In all probability, anything in-house would have been worse and added new failure modes.
Better to buy, analyze, and adapt as needed.
And if it turns out you don't need to adapt, because failure modes aren't safety-critical or components are viable in the environment, then spend your time on something more useful.
If you can't afford to qualify the components on your 4000m diving vehicle, you can't afford to make a 4000m diving vehicle.
See: the fact that they lost their diving vehicle.
Pressure hull >> ballast control >> thrusters >> everything else
I'm not sure why everyone is taking potshots at a company for trying something crazy with willing passengers.
Everyone involved knew what they were getting into.
Kudos to them for trying, even if they're dead.
> See: the fact that they lost their diving vehicle.
That's an awful lot of keyboard engineering, given nobody knows what happened yet.
45 replies →
> If you can't afford to qualify the components on your 4000m diving vehicle
... which you are taking paying passengers with
This is not necessarily the case. For many people it's worth the risk of death to do cool things (e.g. climb Mount Everest).
1 reply →
Not all COTS are equal. There are plenty of off the shelf controllers built for boats that are designed to handle wet environments such as might be found in an enclosed space where people are exhaling water vapor etc. They don’t however cost 30$ nor do they cost anything close to the R&D required to make an equivalent product.
Of note they might not have condensation in normal conditions, but condensation is exactly the kind of thing that results in cascading failures when just one seemingly minor thing fails.
Absent engineering, an engineered solution is no better than COTS, agreed.
Absent engineering, people die unnecessarily.
Trade offs.
They were charging a quarter million per head. Budget should not have been a concern.
Also using close to $1m in fuel per trip (according to the CEO), not that it changes your point
10 replies →
I get using COTS but the decisions for this submarine would indicate that they have no grasp of the concept of failure modes.
Decisions like using a 3rd party controller (known to be terrible), a wireless controller (introducing a lot of extra risk from batteries to connection problems), and a door that cannot be opened from the inside (what if they get lost but manage to surface?) are all very sus.
NASA gets all that done on $28Bn/year.
There's a huge list of companies that have that much revenue.
In some cases, it doesn't matter, but we shouldn't use cash as an excuse to cut corners with safety and reliability.
NASA isn't producing in-house, they still source from third parties. So, if you want, or need, something from scratch, you pay for the development and industrialisation and then for the parts. And those suppliers are quote often the same ones as they are for the COTS stuff.
It many companies are going places NASA fears to tread. 12000 feet is pretty fucking deep. That’s why the wreck took so long to find in the first place.
You’re conducting a technical analysis that overlooks the legal analysis around fitness for a particular purpose.