deviation chart

[jean1146]

I see huge discrepencies between my autopilot and my compass.I would like to make my own deviation chart but can I use the pilot as a reference at different headings to make a chart?
JC.

[Pelagic]

If you are using a flux-gate compass as a heading sensor for the autopilot they usually have a feature whereby you can put it in auto-correct mode and “swing” the compass to take out most of the deviation. Then make a note of minor deviation as per your deviation card.

[MidLandOne]

I have always corrected the fluxgate first as Pelagic has described and then verified that on courses sailed against the GPS (remembering that gives course over ground not the boat's heading so needs slack water and some checks against charted land features).

I then correct the conventional steering compass against the fluxgate.

We have a steel boat so generally more difficult than a frp or wooden boat but I have always very easily got them all agreeing with the truth within a degree or two (that is just using the North/South and East/West correctors on the steering compass, even though the boat is steel have not found the need for quadrantial soft iron balls as often fitted - the few other steel boats I have done have also been the same).

Now, I cannot steer within a degree or two in a sail boat so I have never found the need for a deviation card. I always have a giggle when I see small sail boat owners have a compass adjuster on board, who corrects the compass to a half a degree and produce a deviation card with only a max of a degree or two deviation.

I suspect that if you cannot set your compasses up within such agreement then you likely have something else going on - eg a damaged or faulty compass, some overwhelming deviating factor by one compass - and I would be inclined to sort that out first.

[Ex-Calif]

Pretty much as others have said. Our club pier is aligned magnetic north (for now - as mag north changes slowly over time) so I stood off, aligned with the pier and adjusted the error out of the autopilot fluxgate compass.

For the wet compass one procedure is to shoot a known line from a mark or buoy to a feature on the shore. It's best if you can get a north-south line and an east-west line.

Approach on the north line, note the error on the compass (say 10 degrees west). Approach on the south line note the error (say zero for simplicity). Take out half the north south error. Repeat.

Do the same thing for the east west line. Check all 4 headings.

Some notes -

The compass will swing simply with acceleration so the speed should be stabilized with flat seas if possible. Remember ANDS - Accelerate north, decelerate south.

You should swing the compass with the "normal" gear running. Depending on proximity to metal and magnetic field deviation can be influenced. Some will say all gear off. I would sit the boat somewhere, note the heading, turn on all your gear and see if the compass is inlfuenced.

Unless you have a big, precise compass, this is more of a comfort than a necessity as MidLandOne says. Few people can hold a degree with a 5 inch compass and with currents, drift and changing sea state one rarely steers for long on a ded reckoning course. You will usually get a gps location or shoot a celestial reading (if that's your bag).

PS - After I swung my compass I checked the cardinal headings against the autopilot compass and it was "pretty "close.

PPS - Some may wonder why bother. We often get thunderstorms here that drop visibility to 100m or less. We could stay ashore but we are now used to reefing the sails, shooting a heading and plowing through. It's important to sail a heading so as not to go ashore during the 30-40 minute storm. We also improved in a race this way last year. We sailed a course and when the storm lifted 4 boats had fallen off and were headed for the wrong (leeward) mark. We went from last to 3rd inside the storm by sailing in the right direction during the low vis.

[delmarrey]

Originally Posted by jean1146

I see huge discrepencies between my autopilot and my compass.I would like to make my own deviation chart but can I use the pilot as a reference at different headings to make a chart?
JC.

The first question is; are your charts variation up to date in accordance with the latest weekly issue of the "Notice to Mariners"???

Second, has your compass been swung with a recorded deviation table??

Without these two factors everything would be a guess where you would be heading. Then, you would have to calculate for "set & drift".

Your compass may have one deviation factor and your fluxgate another just because of their location on the vessel.

One guess is as good as another if instruments are not tied in together thru a NEMA system.

[jean1146]

thank you all for yor input.I will check a few things this week end and will report.JC.

[David M]

Not to be argumentative, but some of the previous information is in error.

You cannot use a GPS to create a Deviation Table because a GPS gives you course over ground and not heading. A deviation table is all in reference to the lubbers line on your compass which of course is your heading. The "self-correcting" feature that some fluxgate compases come with can be in error as well.

It's really easy to create your own deviation table, it does not take much time and it is kind of fun, so I would just get underway some sunny morning with your sails down and do it.

The most accurate way to create a deviation table is to range off two things that are stuck in the Earth that are also plotted on a chart...such as a dayshape and a radio tower in the background, for example.

You will need to make one table for each compass and not one table with respect to another compass because BOTH compasses can and probably will have deviation error. Referencing one compass to another compass would sort of be like the blind leading the blind. This includes your autopilot compass.

A general rule of thumb is that if you have a greater then 10 degree compass error in your deviation table then your compass needs adjusting. Your compass will have instructions for doing this which is done by moving the adjusting magnets inside the compass. There are also professional compass adjusters who can do this for you.

Here is a good reference for creating a deviation table.

Boater's Bowditch: The Small Craft ... - Google Book Search

[jean1146]

Thanks David. I was thinking about going out this week end and just do what you are talking about. thanks again.JC.

[MidLandOne]

Originally Posted by David M

Not to be argumentative, but some of the previous information is in error.

Well as I was the one that mentioned GPS it must be me that you don't want to be argumentative with and who you are accusing of providing erroneous information.

Now, most people find no need to be argumentative with me because they generally realise that I do know what I am taking about - I assume that because you say I am wrong, you have some other reason not to want to argue with me.

So -

Originally Posted by David M

You cannot use a GPS to create a Deviation Table because a GPS gives you course over ground and not heading.

I did not say to use the GPS to produce a deviation card, I said that it can be used as a tool in correcting the compasses. In case not realised one corrects the compasses and then produces a deviation card - it is not possible to do both at the same time.

I also did, in fact, point out that the GPS gives course over ground, not boat heading. That does not reduce its usefulness as a tool however, as long as that point is recognised so checks are made in still water or with heading reasonably close to being directly into or directly with any current (if you do the maths you will see that it only has to be "reasonably close").

Originally Posted by David M

The "self-correcting" feature that some fluxgate compases come with can be in error as well.

Contrary to your own, I am assuming, experience it would be most unusual in my experience to find the fluxgate more than a degree or so out after perfoming the correction routine for it, certainly within helming error, unless there is some gross deviating problem or fault somewhere which should itself be addressed. Perhaps there are "junk" fluxgates out there that have the problem you mention, but I would suggest not using them on a sail boat if its navigation is important.

Originally Posted by David M

Referencing one compass to another compass would sort of be like the blind leading the blind. This includes your autopilot compass.

If you were to try it you would find that it is in fact dead easy and will within a couple of minutes, unless one has some gross influence somewhere which itself should be corrected, get the compasses well within helming error.

You seem to not recognise that what I am saying is the fluxgate is first corrected (in my experience, even on steel vessels and properly located that will be well within helming error) and the steering compass adjusted to that. The steering compass is just corrected to the fluxgate as near as it is possible to do so and which I have always found to result in good agreement.

Of course, they may both then be in error (unlikely, as would be unusual for the fluxgate to be far out if properly installed) so I did say that some (only need be a few) checks need to be made using the GPS as a tool and/or charted features (and as another has said one of the easiest of those is ones own berth, assuming it is accurately charted as ours is).

Those may or may not show that the fluxgate has an inportant error in which case the cause of the error should be removed through relocation.

Originally Posted by David M

A general rule of thumb is that if you have a greater then 10 degree compass error in your deviation table then your compass needs adjusting.

Frankly, if you cannot get your compasses within a degree or two then you have some bigger problem somewhere that needs addressing. The deviation card is not a cheap fix for gross errors.

You also mention having a deviation card for each compass - I would respectfully suggest that most of us (including myself) have enough of a problem always getting the correct compass course worked out allowing for variation and deviation in our heads that adding two different cards into the picture is just more room for error. Far better in my mind to have all compasses just corrected within helming error and forget deviation cards. In fact, for small commercial vessels which may be required by regulation to carry a deviation card, that in the countries I have worked has only been required for the main steering compass - it is that compass that is the primary one and if a course is locked on the pilot the resulting course is checked against the steering compass and the locked heading then adjusted if necessary.

I mention helming error -

It is important to remember that we are are not swinging a ship here but a small boat which is impossible to manually helm, or for the pilot to helm, accurately in anything other than smooth seas. Furthermore, the point of sail the boat is on, the sail combination used and the particular helmsman at the wheel will all affect accuracy of steering as will the sea state and its direction relative to the boat's course.

Also, as alluded to by another, most small boat compasses are not graduated in single degrees, 5 degree increments being usual which infers a resolution of only 2.5 degrees - which is not a problem because no one can reliably helm a small boat to that accurate a course anyway.

Most helmsmen will have a bias in their steering - for example, I tend to end up sailing higher than the compass course when sailing on the wind as I instinctively try to make gains to windward in lifts, probably because of years spent racing dinghies (and also in heavy gusty winds luffing to reduce heel). Helmsman alertness (eg tiredness due to time on helm or conditions) will also have a big affect on the accuracy of the course steered.

This also applies to ones pilot. According to the sea state, the sail combination and point of sail the accuracy of the course sailed will vary. For example, on my own boat sailing on a broad reach in heavy conditions under a foresail alone on the outer fore stay (ie no main) the pilot will sail the boat on average lower than the locked course.

Then, when we set a compass course to be steered we make guesses at leeway, tide, etc all of which add to the error in the actual track sailed regardless of the accuracy of the compass.

So, while it may sound like very fine seamanship to swing ones compasses just as they do on the big ships, it is all rather pointless really. In fact, as David says, it really is just "fun" - that is I assume it is fun playing like real seamen. Nothing wrong with having "fun" if so inclined but it in fact ignores the practicalities of the small boat navigation situation.

The method I gave is tried and proven as a quick and easy one (by myself and others) for setting up a small sail boat's compass well within helming error and only takes a few minutes. One should get an accuracy making a deviation card redundant (unless one really believes one can steer to within a degree or two). If one cannot achieve that accuracy then one has some gross error somewhere and it should be removed, not "fixed" by use of a deviation card covering large unresolved deviations.

Not only is it useful for correcting the compasses it introduces a habit of doing exactly the same from time to time when voyaging - making spot checks of the agreement between the compasses and with the GPS for the course one is on should be second nature. If they are not in agreement, then find out why - for comparison with the GPS is may just be tide that creates the discrepancy, but it may not be either.

I, of course, assume that you have actually tried the method I set out and have based your dismissal of it as erroneous because of difficulties you found in its execution. If that is so, perhaps you can outline the specific difficulties that you experienced so that we can perhaps identify their cause.

Whew, time for a beer :-).

[Ex-Calif]

Originally Posted by MidLandOne

You seem to not recognise that what I am saying is the fluxgate is first corrected (in my experience, even on steel vessels and properly located that will be well within helming error) and the steering compass adjusted to that. The steering compass is just corrected to the fluxgate as near as it is possible to do so and which I have always found to result in good agreement.

Wow - Long post! I assume that you align the fluxgate using a bearing line defined by two stationary objects?

Something has to be aligned first and there is only one way to get a true alignment - fixed reference heading.

Everything else you said is just noise.

And yes I have swung hundreds of compasses.

[MidLandOne]

Originally Posted by Ex-Calif

I assume that you align the fluxgate using a bearing line defined by two stationary objects?

Something has to be aligned first and there is only one way to get a true alignment - fixed reference heading.

I thought that was obvious and assumed it was already correctly installed and the alignment done as the subject was correction not installation. But it only has to be done on one heading - and that can be in ones own berth using the berth orientation itself if it is accurately charted (as our own one is).

Of course given that you raise that you could have also mentioned that the steering compass should be correctly installed too, as far as longitudinal alignment with the boat is concerned.

Originally Posted by Ex-Calif

Everything else you said is just noise.

Thanks for the simple dismissal of the contents of my post. I assume you disagree with all that you regard as just noise? Perhaps you would like to elaborate based on your own sailing experience?

Originally Posted by Ex-Calif

Wow - Long post!

It was worth it, if only for the sake of the cold beer afterwards :-).

John

[Ex-Calif]

Originally Posted by MidLandOne

Thanks for the simple dismissal of the contents of my post. I assume you disagree with all that you regard as just noise? Perhaps you would like to elaborate based on your own sailing experience?

John - Sorry to have been rude. My observation was that your post contained a lot of information about sailing the compass rather than aligning it.

This is a simple question of aligning the compass and preparing a deviation card. As you now point out and the crux of the entire question of this thread is -

You must start with a fixed reference in order to correclty align a compass.

I was not correct. There are other questions raised in your post as related to rigging that deserved to be explored.

Significantly the question about whether consider the deviaiton problem at the time of compass installation (location selection) or to consider relocating the compass if there is a large deviation present.

However - there are times where the optimum location of the compass to reduce deviation cannot be achieved for various reasons and then a deviaton card with substantial corrections may be the only option.

[David M]

John,
What I said works and is correct. What I described is the most accurate way of determining deviation error. Perhaps I should have phrased your way as being less accurate instead of being inaccurate...for that I apologize. If however you want to guestimate deviation error by using a GPS or use the "self correcting" error of a fluxgate compass then have at it. There is no way of positively determining if the fluxgate deviation correction was done correctly. The GPS it has its own problems such as the necessity to be moving through the water to get a COG and any lateral vector to your heading such as that caused by current.

The ships Standard Compass obviously needs its own card as well as the Steering Compass because what happens if the ships Standard Compass is damaged? Yes, I realize that most yachts have no dedicated Standard Compass.

For even the best compases to be adjusted down to a degree or two of deviation error is either dumb luck or the boat has no iron onboard.

To use a GPS for determining deviation, one would need to know current, the boats leeway, the winds effect on the vessel and any other factor that might make a difference between COG and HDG. How many yachties have ever done a current vector?...probably not many. You already discussed the importance of keeping things simple by mentioning the need for only one deviation card at most. For many, doing an accurate current vector every time a major C/C is made to factor out deviation would be rather complex and a big pain in the rear quite frankly.

Yes, the helmsman can and does induce error into an intended trackline by being off course. So why aggravate this error by introducing a second error such as an inaccurate deviation correction?

How you choose to correct and use a compass is up to you of course and in all probability works for you as well. I have no argument with your way. I just like my way better because I know with 100% confidence that it is accurate. There is usually more than one way to do something.

Again, my apologies for offending you.

David

[MidLandOne]

Dan, perhaps some room for misunderstanding.

Originally Posted by Ex-Calif

This is a simple question of aligning the compass and preparing a deviation card.

An essential part of correcting the compass and whether a deviation card is needed or not, is how accurately one can steer the boat given all the outside influences on the helmsman's or autopilot's ability to maintain a set compass course. I have found that is not a very well understood point but is part of the simple question of how diligently one should go about correcting the compass and preparing (or not) a deviation card.

While most seem to understand that one does not need a micrometer to measure the length of their boat, it seems to be a common belief that it is essential that a sailboat's compass be corrected to be as accurate as possible and what little bit that cannot be corrected out be displayed on a deviation card and applied to all course computations. But there is not much point sweating and cursing to do those things if we have no hope of maintaining an accurate course when using them. Yet, I see (and read in forums ) people who do just that.

I was endeavouring to point out what many of those influences are while trying to sail a set compass course and therefore why we cannot maintain a very accurate compass course at all - our track will deviate significantly from it apart from in benign conditions.

I do not know what the typical error in helming is but many experienced sail boat sailors I have discussed this with volunteer somewhere around 4-5 degrees averaged over all conditions ie sometimes can do better, at other can only do worse. Obviously, the bigger the boat the better and if one typically sails in benign conditions then I would also expect the average to be better.

Without getting buried in the statistics of experimental errors that would suggest to me that a compass that is corrected within 2 degrees is amply accurate enough and if correction is attained to within 2-3 degrees a deviation card is likely superfluous.

David - Yes the way I describe is not as accurate as a formal correction and swinging to prepare a table of deviations would produce (even on small sailboats I have seen compass adjusters working to half a degree whereas I am talking of a degree or two). Hence my lengthy explanations justifying why just accurate enuff is good enuff - well in my opinion .

And no offence taken - we all have something to contribute.

John

[Ex-Calif]

John,

I have absolutely no disagreement with the steering errors, the possible lack of need to do a deviation card, especially on a compass graduated in 5 degree increments or anything else you've said about steering the boat.

In full disclosure mode and self flagelation I have swung our compass and correct out as much error as possible (5 inch compass) - Making a deviation card? Of course not - LOL

PS - I'd be lucky to hold +-2 degree steering - LOL That's what Otto is for.