Collision Avoidance, Cones of Uncertainty, and Appropriate CPA

[Paul Elliott]

Originally Posted by ramblinrod

We just witnessed an experienced navigator, who had this been real life, would have pasted his vessel to the front of a 600 ft ship travelling at 20 knots.

No we didn't, and no he wouldn't. He was replying to your hypothetical situation challenge. In real life he wouldn't have tried that crossing.

[ramblinrod]

Originally Posted by Dockhead

But in any case, you never calculated the CPA with the ship behind correctly, and you keep repeating these incorrect calculations over and over again.

There is no error in the calcs of post # 591, that accurately indicates what would happen with a 180 ft vs 2.5 cable CPA.

I don't need to do the math or any plot to know that a 3 cable CPA from the transom of the lead ship, plasters you on the bow of the following ship.

To say after you're dead that it doesn't matter because the whole thing was unsafe is ludicrous. If it did play out, as proposed, I would sail on, and you would die.

The point is, if the ships didn't change course or speed, the 180 ft CPA I proposed would work and the 3 cable CPA you proposed wouldn't. Period.

[conachair]

Originally Posted by ramblinrod

Of course this would not play out in the real world; that isn't the point at all.

The point I was making, was how important it is to get one's head out of their arse and electronics.

Seemed to be a lot of effort put in to defending a course of action which you now say doesn't play out in the real world.

[evm1024]

Originally Posted by ramblinrod

There is no error in the calcs of post # 591, that accurately indicates what would happen with a 180 ft vs 2.5 cable CPA.

I don't need to do the math or any plot to know that a 3 cable CPA from the transom of the lead ship, plasters you on the bow of the following ship.

To say after you're dead that it doesn't matter because the whole thing was unsafe is ludicrous. If it did play out, as proposed, I would sail on, and you would die.

The point is, if the ships didn't change course or speed, the 180 ft CPA I proposed would work and the 3 cable CPA you proposed wouldn't. Period.

The whole point of your 50 ships in a row is to set up a defense for your 180' CPA behind a single ship in open water. Attempting to cross 180 CPA astern of a single ship is indefensible and thus the string of ships is a strawman.

I and others have shown that you could open your CPA (to 600' in my case) and still have some margin for error when the follow on ship comes along. But you are so wedded to your 180' assertion that you cannot even take that at face value.

When asked if opening your 180' CPA to 200' would provide a decrease of risk you evade answering the question. You evade because any number other than 180' would be unacceptable to you.

This is a serious error in your outlook and judgement.

[ramblinrod]

Originally Posted by conachair

Seemed to be a lot of effort put in to defending a course of action which you now say doesn't play out in the real world.

Please, a line of 50 ships exactly 1-1/4 nm apart?

Anyone under the perception this was a real world scenario needs their head examined.

But you are correct, I have spent way, way, way, too much time on this, but I think it was important to demonstrate the need to get ones eyes off the screens now and again.

PS, I note you are still following and posting. ;-)

[evm1024]

Originally Posted by ramblinrod

Please, a line of 50 ships exactly 1-1/4 nm apart?

Anyone under the perception this was a real world scenario needs their head examined.

But you are correct, I have spent way, way, way, too much time on this, but I think it was important to demonstrate the need to get ones eyes off the screens now and again.

PS, I note you are still following and posting. ;-)

So you are saying that you have been trolling!

[Dockhead]

[QUOTE=ramblinrod;2507122]

Originally Posted by Dockhead

2 or 3 cables of CPA is enough that you are clearly passing behind.
QUOTE]

Nope.

I can't believe you are still supporting your 3 cable CPA.

From a distance out, a 180 ft and 3 cable CPA, both look very similar; the angles simply aren't that different. Negligible.

If you use a 3 cable CPA, your set-up still appears as a collision course, and in fact you will run into the following ship.

There will be no easy landing for you sir if you attempt this and the ships don't move.

What does the following ship have to do with it? I really don't understand the argument you are making. Are you saying that if a safe pass with the ship ahead creates a dangerous pass with the ship behind, that a dangerous pass with the ship ahead magically becomes a safe one?

I've said maybe a dozen times by now, that 2 cables behind the ship ahead, the least which is more or less doable, results in an unacceptably dangerous pass with the ship behind -- did you miss that?

The pass with the ship behind is a separate issue which does not change anything about the pass with the ship ahead. 2 cables is bloody close, and the bridge crew will not allow it unless you agree with them -- they will be expecting a mile. But it can be done by two cables in good conditions -- barely. It will show on ARPA and AIS as a pass behind with a bloody tight but meaningful positive CPA.

180 feet cannot. 0.3 cables is not within any meaningful precision of any method we have of determining CPA. It's a collision course. As has been explained in a lot of different ways, and by a lot of different people.

[ramblinrod]

Originally Posted by evm1024

The whole point of your 50 ships in a row is to set up a defense for your 180' CPA behind a single ship in open water.

Incorrect.

The scenario was designed to require the development of a solution to cross large, fast moving ships in close quarters.

Attempting to cross 180 CPA astern of a single ship is indefensible and thus the string of ships is a strawman.

Incorrect, a "strawman argument" was defined quite well previously. I did not attempt to alter someone else's argument to make it easier to shoot down. It was simply a hypothetical scenario.

I and others have shown that you could open your CPA (to 600' in my case) and still have some margin for error when the follow on ship comes along.

Yes, but at 180 ft CPA, you are already going to be perilously close to the ship behind. Extending the CPA by any amount increases the risk of being hit by the following ship. The whole scenario is based on the ships holding course and speed, so increasing CPA from the lead ship does not help you.

[Dockhead]

Folks, could we please be respectful towards each other? No one is obligated to change his mind about anything. Let's please discuss the ideas -- which I think are really interesting -- and not each other. OK?

[ramblinrod]

[QUOTE=Dockhead;2507238]

Originally Posted by ramblinrod

What does the following ship have to do with it? I really don't understand the argument you are making.

What does the following ship have to do with the cross between two ships at high speed?

Seriously?

Obviously, there is a need to miss hitting the lead ship AND the following ship.

Are you saying that if a safe pass with the ship ahead creates a dangerous pass with the ship behind, that a dangerous pass with the ship ahead magically becomes a safe one?

Wow, you are still trying to use strawman arguments. Amazing.

I never ever ever said it was "safe" to attempt this crossing at all. And of course, what is "safe" is subjective. There's a big difference between "safe" and "successful".

It is highly unlikely anyone would consider this maneuver "safe". However, I certainly consider the one that gets across the other side (me) is much safer, and more "successful" than the one that gets nailed by the following ship (you) if the ships maintain course and speed.

I've said maybe a dozen times by now, that 2 cables behind the ship ahead, the least which is more or less doable, results in an unacceptably dangerous pass with the ship behind -- did you miss that?

Nope.

And not the point at all. The point is that the 3 cable CPA you proposed behind the lead ship was not safe, AND you got nailed by the following ship. Did you not get that?

The pass with the ship behind is a separate issue which does not change anything about the pass with the ship ahead.

Seriously?

You are crossing between two ships. Anything you do to affect distance from one, also affects distance from the other.

If you navigate in congested waters, only looking at how your maneuvers affect your relationship with one vessel, well I'm surprised as heck you are still here.

2 cables is bloody close, and the bridge crew will not allow it unless you agree with them -- they will be expecting a mile. But it can be done by two cables in good conditions -- barely. It will show on ARPA and AIS as a pass behind with a bloody tight but meaningful positive CPA.

First of all, the scenario was based on the ships not responding to communications. There is nothing magic that makes the 180 CPA a "collision course" and the 3 cable CPA not. In either case, the bow of the sailboat is pointing ahead of the bow of the ship (initially) and the distance between is closing rapidly. In the end, your 3 cable CPA leaves you running into the following ship.

Bang, Glug Glug Glug.

180 feet cannot. 0.3 cables is not within any meaningful precision of any method we have of determining CPA. It's a collision course. As has been explained in a lot of different ways, and by a lot of different people.

You can setup a CPA of 0 with the stern of the lead ship if you wish.

180 ft CPA is not a "collision course". If the CPA is maintained and the ships maintain course and speed, there is no collision.

The 3 cable CPA you proposed is a "collision course". You will most definitely hit the vessel following, even if the plan is executed flawlessly.

[evm1024]

A few years ago I was participating in a Power Squadron ACN training session where we took a bunch of "students" out on a night crossing from Poulsbo to Shilshole Bay (Seattle area). To do this you need to cross a traffic separation zone so any ship there must not be impeded. The ships are often doing 13 kts but some are doing 20 kts.

Our speed was about 8 kts heading roughly SE.

Anyway, At some point (might have been 1nm from the edge of the VTS) I throttled back from 8 kts to about 5 kts. Others asked why I was slowing. I pointed out that there was a Tacoma (south) bound ship doing about 20 kts over 5 miles away that we would get very close to if we kept on out course and speed.

By slowing for 10 minutes we allowed the ship to move ahead and our potential collision scenario became an obvious crossing more than 1 nm behind the ship.

More than one of the participants indicated that this was a very good training session as that they would not have even thought to look 5 miles out for ships (1 nm from the VTS) but now realized that the simple action of slowing opened out CPA form near 0 to over 1 nm.

It was about midnight when we arrived at Shilshole by the way.

[Exile]

Originally Posted by ramblinrod

180 ft CPA is not a "collision course". If the CPA is maintained and the ships maintain course and speed, there is no collision.

Merely stating the obvious here, but a "collision course" is different from a "collision." It's already been explained how you would need to be heading for the bow of the big ship to pull off your 180' pass behind its stern, that in doing so you would likely disappear out of sight of the big ship's crew, and how in that scenario the big ship would be obliged to take evasive maneuvers that could kill you.

Since we are apparently back to the concept that mere avoidance of collision on its own doesn't equate to Colregs compliance, basic safety, or common sense, how about you explain your rationale for waiting to pass a single ship in open water as close as 180' when all other alternatives were open to you?

Is this not what you posed before introducing your hypothetical of a long line of closely spaced ships, or am I introducing another "straw man" argument?

[carstenb]

<I'm enjoying parts of this, but let's try to look at this crossing in a truly practical way.
For the sake of argument, a nautical mile will be 6000 feet(after all, we have our heads out of the elctronics and besides the n aked eye can't see any better than that and the round number makes the math easier to do in my head)

A 600 foot freighter is traveling along at 20 knots (20 x 6000 = 120,000 feet per hour). Rod wants to be 180 feet from his transom at crossing. (here again, I'm going to round off and say 200 feet - because the math is easier to do).

So the basic question is - how far way from the ship will ROd's boat be so he can zip in and be exactly 200 feet from the ships transom quarter?

Rod says his boat is going 5 knots (5 x 6000 = 30,000 feet per hour)

So we have point on our chart where Rod wants his boat to be and exactly 200 feet ahead of this is the transom quarter of the ship. The ship is moving with 120,000 feet per hour or 120,000/60 = 2,000 feet per minute or 2,000/60 = 33 feet per second.

therefore the ships transom will move from the spot Rod wants to occupy to 200 feet away in (roughly) less than 6 seconds.

Rod is moving at 30,000 feet per hour or 30,000/60 =500 feet per minute or 500/60 = 8 feet per second.

In the 6 seconds from when the ship passes the imaginary CPA point to when Rod has to be there, his boat will move 48 feet.

This then begs the question - how long is Rod's boat? And when saying CPA is he talking about his bows?

Regardless of anything else - the above little math shows that his bows must be <50 feet from the freighter as it passes.

All of the above naturally assumes that the ship is on rails and the officer of the watch is brain dead or else out in the head putting one through the hoop.

I'd have to say that if i were in the cockpit of the sailboat and someone tried this, I'd mutiny and take control of the vessel and move ti to safety.

The above feat would require something akin to superhuman powers to accomplish.

[evm1024]

Back to the theme of cones of uncertainty....

Let's just suppose that we are trying to get a CPA of 180' behind a single ship going at right angles to us. We are doing 5 kts the ship 20.

At 1 nm out from the crossing the ship is between 4 and 5 miles away.... We know that due to our speeds the ship is 14 degrees forward of our beam.

So the actual point of crossing is a mile ahead of us and 4 miles ahead of the ship. Just a random piece of ocean that looks like all the others.

Off we go.

We are all on rails and Poof we cross behind the ship with a CPA of 180' All is golden.

But what if ??????

What if our speed is not exactly 5 kts? What is our speed is 1% slower or faster?

As it turns out that plus or minus 1% in speed means that our boat 7.1 sec early or 7.76 sec late. This places us 60' further or 65' behind to make our intended 180' CPA.

Oh our speed effect the ships position as well. That same 1% variation in our speed moves the ship 239.7' back on his path or 262' forward.

The results is that a +- 1% variation in our speed defines an uncertainty box that is 125' wide and 500' long.

How about crossing angle. What if we are are not crossing at exactly 90 degrees? At 1nm a +- 0.5 degree difference in our heading moves us port or starboard 53'.

We are not counting any speed differences or angle differences for the ship.

You might say but I see the ship and I'm adjusting my angle - OK good on you. You will need to be roughly 150' from the ship when its bow is dead ahead of you. Make that 17 seconds from the ship. Miss that and you are toast.

[Dockhead]

Originally Posted by Exile

Merely stating the obvious here, but a "collision course" is different from a "collision." . . .

Exactly. And we've been through this before.

A 1 foot CPA is also not a collision, in case you get away with it, but that doesn't mean that it's not a collision course. A "collision course" is any course without a meaningful CPA -- the classic definition is decreasing range and without an appreciable change of bearing. This is very basic seamanship.