Question for expert navigators

[Phoenix of Hamble]

Re: Ok - lets go one step further ....

Steady on Nigel....

I read your post very carefully, and indeed the whole thread.....

I too know how to navigate... you make it sound like you are the only one who has ever calculated an EP....!

[ QUOTE ]
Nav is simple ... finding out where you were ... and then the art of extrapolating to say where you probably will be in x time ... That's it. Plain and simple.

[/ QUOTE ]

So you are telling me from your above posts that you are still talking about DR, and not EP?

If so, then your comment i've quoted above is meaningless...... DR, as you well know doesn't tell you where you are at all......

Let me now ask you a question..... so you've run up your DR..... calculated your EP..... do you now decide that as you can't determine the errors, you assume there aren't any...? of course you don't.... and I don't believe someone of your experience would.....

And I therefore stand firmly by my statement... only a poor navigator doesn't take into account the possibilities of error in their EP, and analyse what risk that might present them.... and to Searush's comment.... i'm suprised about your comment "how much accuracy do you expect?"... surely, I was arguing just the opposite?...... I expect poor accuracy, and therefore like to think about what the implications of my EP being wrong are....

 

[DeeGee]

Re: Ok - lets go one step further ....

Here's my spanner.

Mathematically speaking, I think the error is pretty constant UNDER CONSTANT CONDITIONS. Constant in percentage terms.

If you have gone 100m since last fix, and your typical error under the given conditons (different factor for different conditions, moderate vs strong winds, mild seastate vs confused...) is 10%, then you can expect the circle of confusion to be 10 miles (let's keep away from normal distribution curves and such). If you had no fix for 10 days, and had done a 1000miles, and conditions had been constant, then your error is likely to be 100miles out. We can slip normal distributions in and establish confidence limits...

What the hell, the reality is that it is anyone's guess, as we all know. So let's stop messing about, and get one of our multiple GPS's out......... /forums/images/graemlins/tongue.gif

 

[BlueSkyNick]

Re: It definitely seems you didn\'t read properly ...

"So all the years and examinations / tickets etc. I gained in The Merchant Navy are wrong ??? "

i'm struggling to believe that even you wrote such an arrogant statement as this, Nigel.

 

[bilbobaggins]

Gunfleet,

This is a question that has taxed the best of professional navigators for as long as records go back. One can manufacture and calibrate each instrument meticulously, check 'em daily for 'odd behaviour' and damage, double and treble-check one's calculations - and still have a number of factors where one has to estimate an 'allowance', to work up an EP.

Traditionally, years of 'reflection' on one's Residual Errors when making landfall was the only way to refine and calibrate one's judgment of 'how much and in which direction' to allow the allowances and margins. Those who were quite good at this got more respect from their peers than t'others. And promotion.

We are doubly fortunate today, for we have the means to hand to sharpen our DR skills to a degree that the old shipmasters would have given their pensions for..... Except few bother to make use of it.

The GPS.

If you run a thoughtful manual plot on passage, as you have described, keeping a log-note for yourself of the allowances and margins you permit, run up to a succession of EPs - and also log ( and later plot ) the GPS fix for the same time as the EPs, you have a swift and accurate 'calibration' of the accuracy of your EP process. You can then inspect, compare, and determine whether you e.g. over-estimated leeway, assumed a higher mean course steered than actual, etc., and swiftly come to refine your 'allowances' in various conditions.

After a season of this, you will have 'tweaked' your Deduced Reckoning, in your own boat, to a degree that no wooden shipmaster was able to, and when the fog and drizzle sets in, the battery packs up and the GPS stops blinking, you will *know* exactly how much confidence you can put on your DR/EP - and that's a valuable confidence!

This is exactly how RAF air navigators were trained; despite the array of sophisticated electronic fixing devices, regular and frequent nav training flights were/are flown with simulated 'restricted aids' for much of the route, so that *when* kit actually failed, it wasn't a crisis....

'Train hard, fight easy!'

/forums/images/graemlins/grin.gif

 

[Cruiser2B]

http://www.irbs.com/bowditch/pdf/chapt07.pdf

Bowditch describes "fix expansion" which is a bit over-simplified. Obviously some errors (steering and log) errors have a +/- component, but others are vectors (current, leeway, effect of swell). Applying all of these errors as an expanding radius around your DR is over-conservative IMHO.

 

[Guest]

You are still basically saying same as me !!

If you read carefully - I have said all along that a Navigator would use his experience to assess his position accuracy.

The original post is about DR not EP .....

My comment about the art of Navigation is a comment about the Art of Navigation - not DR ...

Sorry but yoy are reading to suit an argument that does not exist.

Second - I take umbrage at the comment that "you make it sound like you are the only one who has ever calculated an EP" .... I would never be so stupid - I am friends with and also know quite a few on these forums who have come from same MN background as myself.

Basically I don't understand why you create a dispute where there is no dispute.

Let me say it again so there is no confusion.

A DR is a DR ... a simple non-accurate position based on course steered and speed by log.
An EP is a DR corrected by tide / current and estimated Leeway.
A fix is a position derived by such means to be accepted as reasonably accurate.

A navigator will assess the likelihood of error based on dangers, vessels run up to that time .. any factors that are applicable.
Let me go further .... I have met and chatted with many boat-owners at club-bars ... on boats etc. etc. And often the EP is confused with DR ... Second that if a navigator has information at hand that can affect the position that he decides as "DR" - then that technically is no longer a DR by definition ... as he has used extra information to try and reduce its error.

If I still had my Admiralty Manuals of Navigation .... which my Brother "knicked" to take his Masters .... I would extract sections out for postimg ... /forums/images/graemlins/grin.gif /forums/images/graemlins/grin.gif /forums/images/graemlins/grin.gif

 

[jimbaerselman]

Re: You are still basically saying same as me !!

A simple definition question, probably not applicable in the merchant navy, but definitely applicable on a yacht:

Does DR take account of steering error or log error? Or does it assume they don't exist or have been removed?

 

[Phoenix of Hamble]

Re: You are still basically saying same as me !!

And you wonder why I look to "pick an argument"?..

Nigel.... your post is patronising.

 

[jimi]

Re: Ok - lets go one step further ....

I would tend to the view thatsuch error would be systematic and therefor would not cancel out (ie consistent under/over speed, compass error, leeway etc etc)

 

[Gunfleet]

Re: Ok - lets go one step further ....

I wrote DR in the original question because I thought that would keep things simple. What a mug!

 

[Skyva_2]

Position Probability Area!

Admiralty Manual of Navigation Vol 1 sheds light on the subject. Somehow I picked this up second hand for £4, it's the best produced book on my shelves.

They cover errors in some detail but still don't answer the original question, probably because it's not answerable.

Link here: PPA

I particularly like the technical term used for a basic navigation error - it's a 'blunder'!

 

[DeeGee]

Re: Position Probability Area!

See my posting under Gunfleet's Question. Look again at your AMN Vol 1 (your assessment of which I agree with). After the pages I put on my site, there are pages 477-496 which enlarge on the mathematical basis of error treatment. It is all TOTALLY academic, but that was the nature of Gunfleet's original question. Good one, too, it makes one think about things (that don't really matter?) /forums/images/graemlins/wink.gif

 

[jimi]

Re: Position Probability Area!

have'nt read it but surely mathematical treatment of error assumes random error around a mean whereas on a boat error is u nlikely to be random but "systematic" ie underestimating leeway by 4%, unknown compass error of 2% etc etc so that the error will tend to accummulate.

Opposed to "errors" in helming around a course which do tend to cancel

 

[DeeGee]

Re: Position Probability Area!

Ah, lazy bugg*r... yes, the treatment differentiates systematic, semi-systematic, and random. Systematic or constant errors e.g. lubber line on compass not aligned with fore and aft line of ship, are additive. The semi-systematic, e.g. residual error in gyrocompass after correction, are treated as random. It is all terribly boring, but pp 460 and 461 treat... /forums/images/graemlins/crazy.gif

 

[Guest]

Not my problem ....

Your post annoys me - but life is life.

 

[Guest]

Estimated or not ...

In terms of Lubber lines not aligned and other such non-random errors - these should have been picked up in calibration checks ....

Compass can easily be checked for accurate fore-aft alignment by pointing boat at transit line or by even tieing up to a dock that is properly charted and checking compass against charted dock alignment !! (assuming no untoward Jimi's have left some girt big magnet around to affect your scientific data !!)
Speed can be checked against the measured mile runs etc.

So generally on a boat it is the random that remains that cannot be checked and lead to estimates ...

I like the % of course .... in speed - 5% is calculable ... but how do you calculate 5% of a course ?

 

[jimi]

Re: Estimated or not ...

.. in a perfect world there are no errors in calibration, however unlike you, I do not inhabit such a place!

 

[Guest]

Why arrogant ????

The guy argued with a stand that in fact was not so different from his fundamentally ... but then went on to imply I don't know EP from DR etc.

Sorry - you may think arrogant - to me it is simple statement to support my view/

Like it lump it - It is my opinion. In this NOW I will be arrogant .....

To be honest sometimes some people go overboard on rubbish ... making issues where issues don't exist .....
/forums/images/graemlins/tongue.gif /forums/images/graemlins/tongue.gif /forums/images/graemlins/tongue.gif /forums/images/graemlins/tongue.gif

 

[Ships_Cat]

Re: Estimated or not ...

I think that unless very gross systemic errors would be very minor in comparison to others. For DR we are only interested in measuring time, speed (from the log) and compass.

Can assume that no error in measuring time as timekeeping is very accurate and the magnitude of any error is zilch over relatively short periods. For the log we are measuring speed through the water but then using that (and time) to compute a distance across ground on the chart so will be subject to considerable error (on most occassions) due to tide, etc (the error in the log is also likely to vary according to boat speed, sea conditions, boat attitude, etc). For compass it is unlikely that a helmsman can over time steer a course anything like as accurately as the compass (and error in the compass varies according to heading).

Even under autopilot set to sail a compass heading under sail the pilot may sail the boat high or low as it corrects for wind effects (eg the boat rounds up slightly in a gust and pilot aplies rudder to correct back to the compass heading, so over time the boat sails higher than the compass heading). Similar effects with a helmsman, very difficult to steer within 5 degrees over time unless calmish conditions and alot of close attention, and most helmsmen will tend to sail high (eg trying to gain to windward or pinching/rounding up in gusts) or low compared to the heading set.

I think all the above "rubbish" /forums/images/graemlins/smile.gif means that systemic errors in the inputs to computing DR can be ignored in general. But once one tries to take account of any other error I believe, as I think Nigel is saying, you no longer (by definition) have a DR position. Therefore to talk about error for DR, at least on a yacht where I am suggesting that error in compass, log and timekeeping are likely to be small compared to other errors in any computation (which other error we can't take account of as then would not be a DR position), is meaningless.

Moving on to EP and using the above comments again then the accuracy of compass, log and timekeeping are largely irrelevant (unless gross) compared to other errors. Those other errors are not systemic (at at least not purely so as while they may be constant for one point of sail, one helmsman, one sea condition, etc) but I would claim that they are biased that is while they cannot be disposed of by calibration (so not systemic) they will be much the same over a period of time in similar sea, helm, wind, etc conditions - perhaps over the times that several EP computations are made.

Therefore, over normal sailing times between a number of fixes the errors are not random, so not likely to cancel, and therefore the estimated (assumed?) error in the computed result must be added.

Outcomes (and talking yachts only) -

1. To talk of error in DR is meaningless
2. Error in EP is not random so should be added from EP to EP (and even if it were random they should still be added as random outcomes, by definition, cannot be predicted so for safe clearance from dangers one must assume that the "dice' falls the same way each time).

I appreciate the fortitude of those who may work their way through all of that /forums/images/graemlins/smile.gif.

John

 

[BlueSkyNick]

Re: Why arrogant ????

/forums/images/graemlins/smile.gif /forums/images/graemlins/smile.gif

Best of luck with the boat move.