Freighters vs. Sailboats

[Dockhead]

Originally Posted by cappy208

Nope. The data derived from the RADARs calculations will be error filled if the vessel does not stay on a course (and IIRC, it is a pretty narrow window that is specified) Of course the ARPA will show A course. It will show A CPA. But the data will be wrong.

Will the data occasionally show the correct CPA, COG and SOG? Yes. Will there be inaccurate display of data. Yes. But it will be just as inaccurate as the relative trail shown on the radar. It will look like a wobbly line, showing a narrower wobble as it gets closer, but an inaccurate wobbe nevertheless.

I am trying to find a course slide show to grab a plate from to show that in the ARPA course one of the facts they drill into you is that if either vessel changes heading it invalidates the arpa info. A small course change does effect it, but a large course change especially screws it up.

A RADAR is range gating when it is in use. It looks for the target in the last
'box' it saw it in. After a few rotations when it has seen the same target it locks onto it. Once it has locked on to it, then it begins the computations on plotting. The rougher the seas the longer it can take. I have seen locks in less than 10 seconds. Some never take?! Alternately, sometimes after a locked target has been tracking, it will loose the lock. This happens when the target is not 'found' in the box again. This is all effected by the heading input.

The accuracy of the ARPA is dependent on 4 things. the target maintaining a steady course and speed. and the ARPA vessel maintaining a steady course and speed.

Course changes simply screw up the data. Does it average out? Yes. Would I trust the data? NO. Feel free though.

Obviously you cannot predict the track of a vessel following a randomly erratic course. This is trivial.

ARPA can, however, make an accurate track prediction of a vessel following a steady course, from a radar located on another vessel following a randomly erratic course, provided it has decent heading data.

That's why ARPA is an order of magnitude more accurate than traditional radar plotting -- you have variations in course of both vessels mixed up in relative motion. ARPA does not use relative motion; it calculates a track over ground using lat long at regular intervals.

That means that ARPA can give you an accurate CPA and TCPA (and course and speed of your target, too) with regard to a target on a steady course no matter what your vessel is doing -- relative to whatever your course is at the moment (and it recalculates after every radar sweep, even recreational ones). That means that if the system has got a good lock on a target on a steady course, all you have to do is hold steady yourself for a sweep or two and you have accurate calculations. Obviously if you change course again, the previous calculations are invalid.

That is all due to the way it works, and again, it works in a fundamentally different way from traditional radar plotting. And due to the way it works, it must have accurate heading data.

The processing power required for this is not great -- a mobile phone's CPU has enough power to do the processing done on your big ship. The big difference between your systems and ours are your giant radar sets with large antennae, giving you incomparably better range and target definition, and your $20,000 gyro compasses.

[foggysail]

Originally Posted by Dockhead

"Just run away", as a collision avoidance technique, if combined with late or inaccurate assessment of the risk of collision, can cause collisions if the stand-on vessel manuevers unexpectedly or in the wrong direction.

But "just running away" by a stand-on vessel is also basically forbidden by the Colregs and can result in liability:

]

NOW REALLY! Please, we are talking/writing about sailboats and freighters. Can we stay with this thread's theme please. So now I presume we must size the freighter so we understand the required distance for the beast to turn. And of course we should size the sailboats also for a better reference to its turning ability. As a practical matter, cruising sailboats for the most part are under 55' so with a functioning rudder, turning........even if you need to drop a sail...........even if you have to start the engine will not be a problem.

So even if it takes the cruiser 200 feet to make an abrupt port or starboard turn, it just might take a mile for the freighter. The idea of a late assessment situation will only occur if the sailboat insisted on following Colregs and not getting the hell out of the way much earlier.

Liability??? For getting out of the way??? For collision avoidance? Really?

Foggy

[goboatingnow]

I'm afraid cappy, I fully agree with Dockhead.


Lets throw out a few myths. Firstly the degree of technical spec between my 4k Simrad yachting nav setup and your 70k one is certainly not 66k of technical difference. You are paying dearly for standards compliance , service networks and big ship pricing. But ultimately for a fraction of the price I can duplicate or easily surpass your big ship technical spec that's a fact.

MARPA and arpa are in fact identical, MARPA being used early on by raymarine to distinguish its offering from the IMO spec arpa.

Good arpa(marpa) needs three things. (A) a reliable radar contact , hence a narrow beamwidth (B) accurate heading info and (c) high speed processing

Good heading is not damped nor stabilised, good heading means the sensor follows the movement of the vessel accurately and with minimum lag. A good sensor would have a 100hz output with minimal overshoot and a 0.1 degree resolution.

That type of sensor input requires significantly more processing power in order to ensure the plotted vectors remain real time.

In a small boat this means higher end plotters, big radars and heading inputs like GPS compass systems. I have a hemisphere sat compass feeding my plotter it's provides rock solid fast heading absolutely brilliant. ( note this is a two antenna RTK system, it doesn't use cog. ) however while mine is a self development., such units are between 3 & 5 k to buy commercially , add in a nice tight beam radar ( If you have the room) and a good fast processor and for a fraction of the price i can build a system that will outperform any ships system ( except for range as Incan never get the height) the trouble is that that system costs a yachite about 20k and most will never spend that.

So we have compromise systems on leisure boats that range from awful to reasonable to super good.

A modern gyro stabilised compass is a mis-nomer , of course it's not gyro stabilised , it's uses the input from three axis accelerometer to intelligently dampen the underlying compass. Unlike a ships gyro compass , underlying a modern "gyro" stabilised compass is still a basic magnetic flux gate compass sensor providing heading. Where as a proper gyro provides independent heading info,

So stabilisation and dampening are really the same for modern flux gate compasses. Dampening can be regarded as a cruder form of stabilisation. Simple dampening for example is just a low pass filter. But the problem is the system is typically over damped in rough conditions ( resulting in lag and position based errors ) add active dampening using the inputs from a three axis accelerometer and the digital filters can be adjusted to provide better response with lower over or undershoot for a given set of conditions.

Furthermore modern yachts MFDs have a limit to the their processing time, hence there inability to perform real time vector calculations in high rate heading input conditions

All this means the yachite with low cost MARPA gear must be aware of the error that can result. The same is true using a HBC in wild conditions I've seen mine spin +- 20 degrees at times.

All this means , repeatedly telling WAFIs to fix the oncoming target accurately is wasted breath. We do not have stable platform,high out of the water, with high spec equipment. Hence our cone of error can be quite large. Experienced yachites know this and allow accordingly.

Dave

[LJH]

Originally Posted by cappy208

Sorry, Raymarine is garbage. U an referring to the technical specifications that actually determine the qualifications of ARPA. Unfortunately, there is ARPA and there is evrything else! Marpa has degraded ARPA with oversight, and lousy controls. JMHO.

There's yacht quality and there's commercial. N'er the twain shall meet. As I like to call it, D(s)imrad tried to market this type of garbage a couple years ago. They put out all kinds of fancy fliers out, and brochures. But when you actually needed help, or needed someone to talk to the website they pointed you to was simradyachting.bleech. That was ALL I needed to see~~

I have a Raymarine radar and a stand alone AIS. MARPA will only work on a 12 NM scale or lower with my setup, so assuming I am travelling at 6 Kts SOG and a closing target is travelling at 18 Kts SOG, I can acquire the target at a maximum of 30 mins before CPA. Depending upon the sea state it can take 15 mins or more before the MARPA information (CPA and TCPA) starts to be comparable to the AIS information that I have had for the previous 60-90 mins. In fact, most of the time MARPA is becoming accurate at about 5 miles or less before CPA.

I still use my radar to check for non-AIS traffic, but without a doubt the AIS is a far better tool for planning ahead and getting the big picture than my yacht quality radar. I still back it up with a good radar and visual lookout.

[goboatingnow]

Originally Posted by Dockhead

Obviously you cannot predict the track of a vessel following a randomly erratic course. This is trivial.

ARPA can, however, make an accurate track prediction of a vessel following a steady course, from a radar located on another vessel following a randomly erratic course, provided it has decent heading data.

That's why ARPA is an order of magnitude more accurate than traditional radar plotting -- you have variations in course of both vessels mixed up in relative motion. ARPA does not use relative motion; it calculates a track over ground using lat long at regular intervals.

That means that ARPA can give you an accurate CPA and TCPA (and course and speed of your target, too) with regard to a target on a steady course no matter what your vessel is doing -- relative to whatever your course is at the moment (and it recalculates after every radar sweep, even recreational ones). That means that if the system has got a good lock on a target on a steady course, all you have to do is hold steady yourself for a sweep or two and you have accurate calculations. Obviously if you change course again, the previous calculations are invalid.

That is all due to the way it works, and again, it works in a fundamentally different way from traditional radar plotting. And due to the way it works, it must have accurate heading data.

The processing power required for this is not great -- a mobile phone's CPU has enough power to do the processing done on your big ship. The big difference between your systems and ours are your giant radar sets with large antennae, giving you incomparably better range and target definition, and your $20,000 gyro compasses.

Dockhead arpa nor MARPA does not need GPS input. It needs heading and speed , and range and bearing that's all. GPS is a nice to have arpa existed long before GPS was common.

Dave

[MacG]

I am talking about a military type doppler radar that measures and locks on any subject.

A simplified version is available for non-military use. It works with a computer, a rangefinder with integrated compass and a mil spec gps. It is not specifically for the small yachtmarket but could be used.

[Lodesman]

Originally Posted by Dockhead

That's why ARPA is an order of magnitude more accurate than traditional radar plotting -- you have variations in course of both vessels mixed up in relative motion. ARPA does not use relative motion; it calculates a track over ground using lat long at regular intervals.

That means that ARPA can give you an accurate CPA and TCPA (and course and speed of your target, too) with regard to a target on a steady course no matter what your vessel is doing -- relative to whatever your course is at the moment (and it recalculates after every radar sweep, even recreational ones).

On my last ship we had an old-school plotting head display beside the (then) state-of-the-art Pathfinder ARPA. For kicks I pulled out my grease pencil and went manual, while the OOW used the ARPA - I was consistently faster and more accurate at determining CPA/TCPA. One of the drawbacks of the ARPA, was that any own co/sp changes required a re-think by the processor (iirc, it was 8 sweeps for rough calc, then another 8 to refine it), whereas I could do the relvel for our new course and replot the relative tracks while we were in the turn. The other thing working against ARPA is how the tracking gate is weighted - whereas I could assess and mark the corner of the paint that most accurately corresponded with the target position. BTW, we had shipboard inertial navigation system, which is a more accurate gyro-compass, so no issue there. Where ARPA shines is track management - I think it could track something like 50 targets, and slow as it was could give reasonably accurate and complete info on all of them - the best I could manage would be maybe half-dozen most dangerous targets with all track info and relative tracking on perhaps another dozen, but I'd be fixed at one range scale with periodic long-range checks.

If you have that much difficulty with your marpa, then go here The Navigation-L Page and print off the dma5090 (it's a zip jpeg); laminate it and tape it to a flat surface near your helm/nav station. Take a bearing with a HBC, range from radar and mark a dot or an X on the board (use a china marker/grease pencil or dry-erase marker); write the time next to it. 3 minutes later do the same, then use a straight edge (ruler) join the dots and continue on past the centre of the board. There you have CPA. Measure distance between the dots - calculate over 3 minutes for relative speed; measure distance to CPA; apply rel sp to calculate TCPA. Occasionally check range/bearing to see if it's following the plotted rel track; refining as necessary.

[cappy208]

@ Lodesman (iirc, it was 8 sweeps for rough calc, then another 8 to refine it),

That is the technical specification I am referring to. Due to advanced age my old brain just can't remember all this trivia at a moments notice!!!!!

Any vessel using ARPA is subject to the same constraints. A constantly changing course will keep the processor re thinking, and starting over in the initial level of calculations.

ARPA has a two fold processing process. By statue, it must develop a crude solution within one minute. But the accurate solution has up to 3 minutes to show. Sometimes (actually almost all the time) it shows up within a minute or two. But the integral ARPA function senses the course and heading change and goes back into the one minute function

Wouldn't it be funny if a lot of this issue of degraded ARPA, MARPA, ARP (or whatever the different companies call (less) than full arpa ) due to heading slop that is contributing to the freighter Vs SV story!

Yeah, the pathfinder was a great unit. I have been amazed that Raytheon entered the small boat market with the stuff it did. Sad.

[cappy208]

Originally Posted by goboatingnow

I'm afraid cappy, I fully agree with Dockhead.


Lets throw out a few myths. You are paying dearly for standards compliance , service networks and big ship pricing.
Dave

To say nothing of the 17" screen with a 12" display! Oh yeah the ability to overlay and interface all AIS, plotting, GPS, and depth data on all the surrounding screens!

Nothing to worry about. If we have stirred up some brain cells here, all is good. When I find the technical specs I will post them.

But as another said, lets get this off radar theory and back to how to avoid ships. I know I would appreciate that!

[Dockhead]

Originally Posted by cappy208

To say nothing of the 17" screen with a 12" display! Oh yeah the ability to overlay and interface all AIS, plotting, GPS, and depth data on all the surrounding screens!

Nothing to worry about. If we have stirred up some brain cells here, all is good. When I find the technical specs I will post them.

But as another said, lets get this off radar theory and back to how to avoid ships. I know I would appreciate that!

Indeed! We've wandered far off topic. Radar is very cool -- one of the coolest toys we have to play with. Maybe I'll take a refresher course when I get my new toys this winter.

As to collision avoidance, I think we've also stirred up a few brain cells. I have learned a lot from this thread, although I'm afraid I don't feel safer for it

[cappy208]

Originally Posted by Dockhead

Indeed! We've wandered far off topic. Radar is very cool -- one of the coolest toys we have to play with. Maybe I'll take a refresher course when I get my new toys this winter.

As to collision avoidance, I think we've also stirred up a few brain cells. I have learned a lot from this thread, although I'm afraid I don't feel safer for it

This reminded me of an old line from 'Cool Hand Luke': "A Mans GOT to know his limitations"

The best thing about this forum is it gets many people to actually take out the Colregs and READ them.

Colregs is a dry and boring book. NOONE reads it for fun. But it sure does make (wake) the gray matter up!

[nigel1]

Originally Posted by goboatingnow

I take that comment with a grain of salt, in my experience, having contacted ships, aside from good bright calm days, they mostly dont see me to actually navigate around me. That is a failure under the COLREGS, while I do what I can to be visible they must also maintain active lookouts, Ive been on the bridges of many ships, the quality of lookouts is often very poor and many are just "radar jockeys". The COLREGS are not optional, the ship , under current regs, MUST maneuver ( and must continue to maneouver even if the risk of collision continues). COLREGS nevers allows the give-way vessel the option to "do nothing", yet thats what mostly actually happens


Sorry, we disagree here big time, Neither stand on or give way is absolved of responsibilities, Hence the ship MUST maneovuer, thats is what expected of him under the COLREGS. The problem is far too many merchant vessels ignore sailing yachts either through lack of identification or deliberate decision.


The reality is commercial shipping either through systems neglect or deliberate action ignores sailboats. Lets realize that and change the COLREGS.



Dave

Dave, I cannot disagree with you view if that is the situations you have encountered. I assume that you sail the UK south coast which has far more traffic than the area I sail (Irish Sea).
At work, we can usually detect yachts at about 8 miles radar range (providing the yacht has bothered to fit a decent reflector), and I require my OOW's to plot all targets, and if the CPA is less than 1 mile, to take avoiding action (with respect to a sailing vessel), and in open waters, this will be about 4 to 5 miles distance.
Obviously, you are encountering vessels which don't take this approach.
Have you considered reporting the offending vessels. I assume that you are aware of the CHIRP program which encourages reporting of unsafe acts at sea.
Reading many of their reports, the organisers do take action in that they will bring to the attention of the ship operator a situation that has been reported, and ask for a response from the vessel concerned.
At the very least, it might make the operator tighten up watch keeping.

[Lodesman]

Originally Posted by cappy208

The best thing about this forum is it gets many people to actually take out the Colregs and READ them.

All that only matters if they actually follow the rules.

We seem to have come full circle with posters again recommending a re-write of the colregs. I respectfully disagree with them - if the problem with the colregs is that too many watchkeepers aren't looking out the windows, then we have the technology to deal with that. I'm taking the advice and fitting an AIS transceiver. I expect it will reduce much of the anxiety that is present in being the stand-on to a much larger vessel.

[cappy208]

Originally Posted by Lodesman

All that only matters if they actually follow the rules.

For me it is a question of not so much if 'THEY' are actually following the rules, but knowing 'IF' they are following the rules. This is a larger issue than can be admitted I believe. By and large, I do believe the problem lies NOT from those reading this board, but those who are NOT! How do you fix that?

By my count, there seems to be a lack of discussion about when, how to decide extremis. It is this point that is being ignored by WAFI's and not understood.

In order to follow the rules one must know how to interpret the outside world. This seems to be a difficult task. Some want to change the Colregs. Some want to enforce the Colregs already on the books. From my limited reading skills it would appear that the concept of a Power Driven Vessel vs a Sailing vessel is already covered by the Regs.

How does one (or a whole body) regulate the diverse amount of knowledge and experience (Or account for the the LACK of knowledge and the INexperience.)

Several posters make note of the high speed of the vessels involved should warrant the give way vessel to slow down. I can address that directly. I am almost the same speed on my lovely 500' tug and barge. I make around 10 knots. Most SVs do around 9 when really moving along. I have had numerous close calls, just as described here. It is NOT the 'high speed vessels that are causing this issue. It is the inability to judge risk of collision!Or conversely, the inability to judge that definitely there is NO existing risk.

This reminds me of the sign posted over the time clock in the office.
"Piss poor planning on your part does not constitute an emergency on mine!"

If a WAFI cannot judge what is or is not a collision course or what is or is not a risk of collision, I would say that WAFI should NOT be aboard a vessel in those conditions.

At some time a WAFI has to take control of the situation and stay safe. I see some excellent sailors out there. I also see WAFI's. Until I see proof, I assume you to be a WAFI. You never hear from other sailors about NON incidents do you?

[Dockhead]

Originally Posted by nigel1

Dave, I cannot disagree with you view if that is the situations you have encountered. I assume that you sail the UK south coast which has far more traffic than the area I sail (Irish Sea).
At work, we can usually detect yachts at about 8 miles radar range (providing the yacht has bothered to fit a decent reflector), and I require my OOW's to plot all targets, and if the CPA is less than 1 mile, to take avoiding action (with respect to a sailing vessel), and in open waters, this will be about 4 to 5 miles distance.
Obviously, you are encountering vessels which don't take this approach.
Have you considered reporting the offending vessels. I assume that you are aware of the CHIRP program which encourages reporting of unsafe acts at sea.
Reading many of their reports, the organisers do take action in that they will bring to the attention of the ship operator a situation that has been reported, and ask for a response from the vessel concerned.
At the very least, it might make the operator tighten up watch keeping.

My experience off the English South Coast (the English Channel) has been different.

I do a lot of crossing of the channel at right angles to the flow of traffic, so I've had scores of big ship encounters over the last few years, day, night, various conditions of visibility. I am almost always under sail so mostly I'm the stand-on vessel.

In my experience, at least 80% of the ships I encounter are well run, see me, make a clear maneuver 5 or 6 miles out to pass behind. The other 20% either don't see me, or prefer to hold course and speed and let me maneuver, which I do at a decision point of 2 miles out when it can still be done without drama.

If both SV and MV are keeping a good watch and identify the risk of collision at a safe distance, then it's not rocket science to keep out of each other's way.

The problem begins when you have a risk of collision with two or, God forbid, three vessels at the same time. This has happened to me more than once!

The scariest moment I ever had was with three vessels fairly close together but not on parallel courses. Here is where ARPA would have been a life-saver -- it can do three targets at once with ease, while the HBC and grease pencil system I use starts to break down because of the lack of processing power in the wetware . None of these vessels maneuvered -- and it may well be that it may have been difficult for any of them to calculate a maneuver which would avoid me but not put them in danger of one of the others. So I was dodging all of them, and there was no realistic 2 miles decision point, I'll tell you. I avoided the last one with a fast tack just 3 or 4 cables away -- good thing I wasn't wearing white pants!!

All this about 02:00 on a moonless night and one of my first crossings, when my technique was less refined than now.