4 ms·

Why has this sub been so hard to find? I keep seeing reports that they've search an area "the size of" Massachusetts or Connecticut. Shouldn't OceanGate have

by tivert 3y ago

Why has this sub been so hard to find? I keep seeing reports that they've search an area "the size of" Massachusetts or Connecticut. Shouldn't OceanGate have recorded the exact coordinates where it was released and then been able to localize its possible location to a small area around that by modeling ocean currents or something? My understanding the Titan was designed to sink to the bottom and could only move very slowly under its own power. I know OceanGate was cocky and cut a lot of corners, but I just can't believe they wouldn't have the exact release location recorded somewhere, even if it was just an automatic track log on their ship's GPS navigation system.

eigenspace 3y ago
> Shouldn't OceanGate have recorded the exact coordinates where it was released and then been able to localize its possible location to a small area around that by modeling ocean currents or something? Modelling and knowing surface currents is one thing, but this submersible was thousands of meters deep. Deep ocean currents can be very fast, change often, and we have way less data on them.
- anigbrowl 3y ago
  'Telemetry? who needs it' - Oceangate, probably
  - BurningFrog 3y ago
    There is no way to transmit telemetry to the surface anyway.
    - meepmorp 3y ago
      Sound. There are commercially available acoustic modems for underwater use.
    - dingaling 3y ago
      ROVs do so routinely, at greater depths.
  - bg46z 3y ago
    Telemetry doesn’t work very well under water. There’s work to be done on telemetry that is capable underwater, but it seems a long ways off.
    - nameless912 3y ago
      Not to be crass, but wouldn't any low bandwidth sound-based protocol be easy to rig up for underwater? Obviously there's a bit of adaptation required, but it shouldn't be _that_ hard to set up a 28.8k modem with a directional speaker and microphone rigged up to point vaguely toward each other (i.e. at the surface pointing down and on the sub pointing up). Hell, even automated Morse code sending literal text messages at 3 sps is better than nothing.
    - anigbrowl 3y ago
      Why are you assuming it's wireless? ROVs use data tethers with the spool kept on the base ship. Or you could have a black box recorder attached to the sub by a cable that blows its connection and floats to the surface if it loses contact for more than a few minutes. I can think of 10 different ways to maintain some contact with the surface and store some data about dive conditions. Honestly perplexed at all the people who are still buying into Oceangate's demonstrably bad way of doing things and just saying it's impossible to do any better.
  - gregoriol 3y ago
    When communication has been lost, what kind of telemetry would you expect?
somenameforme 3y ago
This is something I'd also like to know. I'd assume they were assuming a worst case scenario of something like a ballast malfunction paired with a propulsion malfunction, so you have a partially buoyant (enough to stay off the ground, not enough to surface) craft being pushed both by currents, and possibly its own propulsion, in an unknown direction in some radius outside of the Titanic. I've no idea of the depth sonar can accurately ping, so they may also have been considering the 3-d area around the Titanic in which case you're hitting radius cubed, and so even a "small" max distance could be huge.
- pizzafeelsright 3y ago
  Ever use a fish finder / depth finder? A narrow beam is sent in one direction, generally down, and reflections back are calculated. I would imagine that was the start of their search. After enough time or support: military grade sonar? https://man.fas.org/dod-101/navy/docs/es310/SNR_PROP/snr_prop.htm https://man.fas.org/dod-101/navy/docs/es310/SNR_PROP/snr_pro...
  - somenameforme 3y ago
    Yip, but the issue is the depth. The Titanic's at 3800 meters. An average military sub isn't going to hit 1000 meters. So I'm reluctant to make any assumptions about deep search systems.
    - civilitty 3y ago
      Most work class ROVs top out around 3,000 meters but there are plenty of ultra-deepwater ROVs that go to 4,000 meters and beyond that are specialized for search and rescue operations. The usual ROV players like Oceaneering International, Saab SeaEye, TechnipFMC, etc. all make them. The Navy & Coast Guard would definitely have access to their own fleets
LeifCarrotson 3y ago
> by modeling ocean currents or something This sounds a lot like https://xkcd.com/793/ https://xkcd.com/793/ - "and then add some secondary terms to account for <complications I just thought of>." They lost contact 105 minutes into a 120 minute, 2.5 mile descent. The release point for that descent was well known, and currents estimated closely enough to allow the sub to descend close enough to the shipwreck that the submersible's thrusters could move it very slowly to viewing locations. They don't know what happened to cause it to lose contact more than a quarter mile above the ocean floor. They didn't know whether it went neutrally buoyant at that point, whether it ascended quickly, or slowly, or stayed near the bottom of the ocean and continued looking at the Titanic and only later drifted off course - they've done that before. Those ocean currents, unknowns, and distances are large; merely pulling the release point from a GPS track does not suddenly make the search point tiny.
MallocVoidstar 3y ago
They looked on the surface because they didn't have anything that could go down to the Titanic. A ship with an ROV that could go to the Titanic arrived and they found the debris on the sea floor.
- ericbarrett 3y ago
  Right. It's my understanding that two remote submersibles that arrived earlier were lost trying to reach the sea floor—they weren't rated for the depth but the attempts were made anyway. (Don't have a link, I think it was nytimes)
paulpauper 3y ago
because the distance is so far that the sub even departing a few degrees from its intended destination means it can be anywhere along a huge swath of the sea floor, plus it's pitch dark and very far and cold and inhospitable to both human and machines. drop a penny into a swimming pool vs drop it into the ocean.
cobaltoxide 3y ago
> I keep seeing reports that they've search an area "the size of" Massachusetts or Connecticut This was the size of the search _on the surface_.
taco_emoji 3y ago
the ocean is three dimensional

by tivert 3y ago

eigenspace 3y ago

> Shouldn't OceanGate have recorded the exact coordinates where it was released and then been able to localize its possible location to a small area around that by modeling ocean currents or something? Modelling and knowing surface currents is one thing, but this submersible was thousands of meters deep. Deep ocean currents can be very fast, change often, and we have way less data on them.

anigbrowl 3y ago
'Telemetry? who needs it' - Oceangate, probably
- BurningFrog 3y ago
  There is no way to transmit telemetry to the surface anyway.
  - meepmorp 3y ago
    Sound. There are commercially available acoustic modems for underwater use.
  - dingaling 3y ago
    ROVs do so routinely, at greater depths.
- bg46z 3y ago
  Telemetry doesn’t work very well under water. There’s work to be done on telemetry that is capable underwater, but it seems a long ways off.
  - nameless912 3y ago
    Not to be crass, but wouldn't any low bandwidth sound-based protocol be easy to rig up for underwater? Obviously there's a bit of adaptation required, but it shouldn't be _that_ hard to set up a 28.8k modem with a directional speaker and microphone rigged up to point vaguely toward each other (i.e. at the surface pointing down and on the sub pointing up). Hell, even automated Morse code sending literal text messages at 3 sps is better than nothing.
  - anigbrowl 3y ago
    Why are you assuming it's wireless? ROVs use data tethers with the spool kept on the base ship. Or you could have a black box recorder attached to the sub by a cable that blows its connection and floats to the surface if it loses contact for more than a few minutes. I can think of 10 different ways to maintain some contact with the surface and store some data about dive conditions. Honestly perplexed at all the people who are still buying into Oceangate's demonstrably bad way of doing things and just saying it's impossible to do any better.
- gregoriol 3y ago
  When communication has been lost, what kind of telemetry would you expect?

somenameforme 3y ago

This is something I'd also like to know. I'd assume they were assuming a worst case scenario of something like a ballast malfunction paired with a propulsion malfunction, so you have a partially buoyant (enough to stay off the ground, not enough to surface) craft being pushed both by currents, and possibly its own propulsion, in an unknown direction in some radius outside of the Titanic. I've no idea of the depth sonar can accurately ping, so they may also have been considering the 3-d area around the Titanic in which case you're hitting radius cubed, and so even a "small" max distance could be huge.

pizzafeelsright 3y ago
Ever use a fish finder / depth finder? A narrow beam is sent in one direction, generally down, and reflections back are calculated. I would imagine that was the start of their search. After enough time or support: military grade sonar? https://man.fas.org/dod-101/navy/docs/es310/SNR_PROP/snr_prop.htm https://man.fas.org/dod-101/navy/docs/es310/SNR_PROP/snr_pro...
- somenameforme 3y ago
  Yip, but the issue is the depth. The Titanic's at 3800 meters. An average military sub isn't going to hit 1000 meters. So I'm reluctant to make any assumptions about deep search systems.
  - civilitty 3y ago
    Most work class ROVs top out around 3,000 meters but there are plenty of ultra-deepwater ROVs that go to 4,000 meters and beyond that are specialized for search and rescue operations. The usual ROV players like Oceaneering International, Saab SeaEye, TechnipFMC, etc. all make them. The Navy & Coast Guard would definitely have access to their own fleets

LeifCarrotson 3y ago

> by modeling ocean currents or something This sounds a lot like https://xkcd.com/793/ https://xkcd.com/793/ - "and then add some secondary terms to account for <complications I just thought of>." They lost contact 105 minutes into a 120 minute, 2.5 mile descent. The release point for that descent was well known, and currents estimated closely enough to allow the sub to descend close enough to the shipwreck that the submersible's thrusters could move it very slowly to viewing locations. They don't know what happened to cause it to lose contact more than a quarter mile above the ocean floor. They didn't know whether it went neutrally buoyant at that point, whether it ascended quickly, or slowly, or stayed near the bottom of the ocean and continued looking at the Titanic and only later drifted off course - they've done that before. Those ocean currents, unknowns, and distances are large; merely pulling the release point from a GPS track does not suddenly make the search point tiny.

MallocVoidstar 3y ago

They looked on the surface because they didn't have anything that could go down to the Titanic. A ship with an ROV that could go to the Titanic arrived and they found the debris on the sea floor.

ericbarrett 3y ago
Right. It's my understanding that two remote submersibles that arrived earlier were lost trying to reach the sea floor—they weren't rated for the depth but the attempts were made anyway. (Don't have a link, I think it was nytimes)

paulpauper 3y ago

because the distance is so far that the sub even departing a few degrees from its intended destination means it can be anywhere along a huge swath of the sea floor, plus it's pitch dark and very far and cold and inhospitable to both human and machines. drop a penny into a swimming pool vs drop it into the ocean.

cobaltoxide 3y ago

> I keep seeing reports that they've search an area "the size of" Massachusetts or Connecticut This was the size of the search _on the surface_.

taco_emoji 3y ago

the ocean is three dimensional