Vsplat

Donating Member
  • Content Count

    1,574
  • Joined

  • Last visited

  • Days Won

    51

Vsplat last won the day on March 12

Vsplat had the most liked content!

Community Reputation

405 Excellent

1 Follower

About Vsplat

  • Rank
    5

Recent Profile Visitors

4,143 profile views
  1. Hmmmm.. Maybe, but the logic doesn't really add up. While it is possible the memo you posted is authentic, I have to wonder why the logo is as aliased as it is while the memorandum text is so much crisper. I guess we will see. Vs
  2. I'm not sure of the city, but in our area, Friday was a special case in a couple of places due to reduced lab pickup schedules and no scheduled weekend service. I know a doc in our area that was similarly affected, they were sidelined as a result. Tests on Thursday morning knew by suppertime. I imagine there will be a backlog on testing shortly, if not already, but the test itself doesn't need more than several hours now. FWIW Vs
  3. boestar, I don't know where you got that info. If a lab capable of processing a test is in town, results are available within 6 hours. We are seeing this turnaround in my city. The longer waits are for tests that need to be sent to a centralised lab, still the case in many US cities but I guess the Donald got first class treatment. That said, if indeed the tests were negative, that would be a rare statement of fact from that idiot. Vs
  4. What I would like to know is who among Trump's insider friends did he tell ahead of this announcement. There are going to be big winners and big losers. Vs
  5. Unless that blade buried itself in something that stopped rotation. Vs
  6. This thread reminds me of the one on helicopters and stats. If ever there was a statistics-generator, this might be it. Vs
  7. There are a lot of places in aviation where things become the norm, but don't look so good when the investigators arrive. As for some of the activities you noted, At least some of those are done with long lines specifically to reduce some of the risk as I recall. It may also help to ensure everyone is on the same page when talking about 'low and slow'. If you're in a 212 with two engines and that massive rotor, vice an R22 with something just a bit stiffer than yarn above your head, just how slow is too slow and how low is too low can change quite a bit. Vs
  8. boestar I think you're referring to the height velocity curve. I was thinking instrument error becomes a thing below a certain speed, in certain aircraft. Not really sure that applies to the aircraft in question though, as I expect it had up to date avionics. The only glass in the cockpit of the helicopters I flew was see-through... Vs
  9. It has been far too long since I thought about this stuff, so maybe someone more current in rotary ops can weigh in - that said, on the 'go slow' discussion, I seem to recall there was a minimum speed for many helicopters in IFR, below which you had to be visual. Does anyone else recall this and why that limit existed? Could it be that this machine was below whatever that min speed was when the pilot attempted to penetrate the cloud and that contributed so the loss? Vs
  10. Honestly, unless you are a military communications or drone specialist, I don't think you or I have any idea of the true capacity and risks of military grade datalink, what challenges they face in terms of outage and how they mitigate. There is also no requirement to report drone incidents publically so we can trade opinions on what the military does but that's about it. I'd also be a bit more careful when assessing the limits of another's thoughts. You really have no idea what my background is. As for when my time runs out, I'll be long retired before anything close to this makes it to commercial operations. There will be progress and I for one would welcome increased AI in aviation, but as we are living with the MAX, automation is only as good as the model it's built on. This is not going to ultimately be decided by what can be done, but what should be done and why. Vs
  11. That's an impossible question to answer. Most in flight health events are unreported, as it's often solved with a simple conversation along the lines of 'why don't you take this leg'. Depending on the air operator's safety reporting culture, there might be good reporting of in-flight hand over, but still nothing close to a full statement of risk.
  12. Remote datalink control has a lot of public appeal but is not really an option for this kind of operation. Recent jamming, spoofing and outages of GPS affecting RNP are just one reminder that any worldwide datalink or satellite based solution relies, in the end, on good will from the US and Russia. That is increasingly fragile. Then, of course, there is hacking. Current encryption techniques will be solvable by commercially available computers within a decade. What then? Drone operations are, but their nature, expendable. High cost and undesired, but the reason drones are used often has as much to do with risk as cost. There are also many restrictions on autonomous drone flight over 'populated areas' (where the population are comprised of allies). To be clear, this is not about the capability of the technology when it works. SpaceX, through their recoveries at sea, has demonstrated far more than is needed for a commercial operation. The kicker is what happens to a planeload of passengers, or the subdivision they are flying over, on the bad day when a few things go wrong and the assumptions behind the autonomy aren't met. Vs
  13. certainly understand the proof of concept approach taken here deicer. My primary issue is with the underlying assumption. Whereas the industry was able to design more capable and reliable engines and critical systems such as pressurization and fire suppression, such that extended range on two engines could be clearly proven as safe, short of genetic engineering, there is no similar way to improve human reliability or failure prediction. Put another way - using the same case of engines vs pilots, reducing the number of engines reduced the number of critical failure points - something that could be designed around by improving the remaining systems. Reducing the number of pilots also reduces the number of critical failure points - but without the attendant improvement of the remaining unit. I suspect what will be ultimately be offered is some sort of AI enhanced autopilot that can take over in the event of single pilot incapacitation and Piper has introduced this concept already. On that front, I see promise, as AI can be trained much as humans are. It might serve to have an AI 'training' on flight decks around the world for a few years to see what the crew sees, make decisions independently and compare those decisions to what the humans did. That exercise would, in my view, produce a very capable non-human option, but the training process and data review would be a massive undertaking, as sometimes humans get it wrong and the AI would be just as prone to picking up bad habits during training as we are. That review project would be like FOQA/FDM on crack - it would require a worldwide amnesty agreement on the use of data the likes of which do not exist today. Vs
  14. Well, take a single pilot operation, toss in some food poisoning from that sandwich grabbed in the terminal and viola. Hopefully not something that would affect a takeoff, but could affect the rest of the trip. Vs
  15. Well, I think the optics of operating a Boeing at this stage of things might be difficult for the PMO. I have to wonder why it hasn't gone to a commercial operator. I know there will be numerous mods, but how expensive would it be to return to a commercial standard? Surely this fire sale would make up for it. Vs