Visit our Popular Forums

**Dockhead** · 22-09-2021, 04:26

Some of you may know that I'm on a mission this year to finally get all my instruments properly calibrated.

I installed one of those wonderful new Furuno satellite compasses with 4 different radios and antennas, with 0.5 degrees dynamic heading accuracy, and giving very high quality motion data as well.

But how to calibrate it? I always calibrated previous compasses with the magnetic compass But I'm afraid that's not nearly accurate enough for this.

What if I use radar?

It seems to me that compass data has two critical properties -- accuracy, and stability/repeatibility, where the second thing is critical for autopilot performance, and the first for radar overlay, MARPA performance, etc.

The radar overlay is probably most critical of all, so maybe that should be what I adjust to? So long as the radar is not skewed by more than a degree or so, then I would think that this should be optimal, no?

That will also make MARPA work correctly I think -- because if the compass is off even by a few degrees about where the bow of the boat is, that won't matter (right?). The main thing is that it knows very accurately where North is, and if compass and radar are aligned, the radar will have very accurate bearings.

Not to say that alignment of the compass relative to the boat is not important. I will get erroneous set and drift calculations if this is off. But unless the radar is off by more than a degree or so, I would not think it worth trying to align that too, if I even knew how.

Glad for any tips, as I'm not sure I have a full understanding of all of this.

I use radar a lot for navigation (was my main method in Greenland where the charts are off by miles), and as a check on chart plotter, so really accurate radar overlay would be great.

Another thing I am hoping will happen if I can get the compass well calibrated is that maybe MARPA will work better. The Navico MARPA is famously horrible, and I was using it all summer long as my AIS set blew up and had problems getting a new one shipped to me. I noticed that with the compass better (still not ideally) aligned, it suddenly started working better.

Elleroo · 22-09-2021, 04:57

I do compass adjusting occasionally.

I use an old heavy bronze pelorus, 12inch diameter with dark sight lenses, and came off a scrapped warship.

It takes some preparation, but checking the compass using the sun with the pelorus is the method taught to merchant seamen, so should be good enough for yachts, as I do. I sometimes use known distant objects that I've checked bearings on from a small swinging area, as the sun isn't always shining here in UK.

Graph the sun's bearing from your position for about 60 minutes, and you can interpolate using your time when taking readings.

Make sure there are no underwater cables where you do the swinging, as they can and do affect the magnetic compass.

I can usually guarantee accuracy of the deviation table which results, to be better than 1 degree.

There are smaller pelorus' available commercially, but I can't say that I've used one, so don't know how accurate the results will be.

No electronics to cause problems.

There was an old boy doing adjusting in Fort Lauderdale, and he used a 12v gyro compass, taken off a light aircraft. His results were usually good enough to get a bunch of yachts from Lauderdale to Red Hook St.Thomas without problems.

Bycrick · 22-09-2021, 08:54

Calibrate a satellite compass? I’m not sure this has any meaning. As I understand GPS based compasses, and I have one, they are based on the phase differences of the signals received from several GNSS satellites. Other than making sure that the antenna baseline is physically aligned with the boat's center line, I can’t see that you can do any more precise calibration. I can sit here with a couple of glasses of wine and imagine being able to improve the phasing of the several antenas, or the timing variances of the separate receivers in the GPS compass at the manufacturing level, or tweaking the various algorithms or chips used internally, but I can’t think of much you could do externally.

And I wouldn’t think that radar, which can also be mechanically out of alignment with the center line of the boat combined with.the inherent horizontal beam width would be anywhere near accurate enough to be useful.

**Dockhead** · 22-09-2021, 09:06

Quote:

Originally Posted by Bycrick

Calibrate a satellite compass? I’m not sure this has any meaning. As I understand GPS based compasses, and I have one, they are based on the phase differences of the signals received from several GNSS satellites. Other than making sure that the antenna baseline is physically aligned with the boat's center line, I can’t see that you can do any more precise calibration. I can sit here with a couple of glasses of wine and imagine being able to improve the phasing of the several antenas, or the timing variances of the separate receivers in the GPS compass at the manufacturing level, or tweaking the various algorithms or chips used internally, but I can’t think of much you could do externally.

And I wouldn’t think that radar, which can also be mechanically out of alignment with the center line of the boat combined with.the inherent horizontal beam width would be anywhere near accurate enough to be useful.

I probably used the wrong word. I meant ALIGN the compass to the ship's head.

Naturally you don't swing a sat compass. And your understanding of how they work corresponds to mine.

Bycrick · 22-09-2021, 09:17

OK, now we’re on the same page. I’m not sure there is a "good" way to do it. My GPS compass has a molded-in arrow on one end that’s supposed to point "forward," but that’s as close as they get to talking about how to orient it. I just sort of guessed as mine is only speced to +/- 2 degrees. It will be interesting to see what better ideas come along.

smac999 · 22-09-2021, 19:35

I use radar overlay to adjust all my compass installs. you need to make sure the radar alignment is good first. on the radar only page (not overlay) steer the boat straight at a bouy, and make sure it's on the radar heading line dead ahead. adjust radar install offset if not.

once the radar is straight use it to set compass offset with overlay.

Nicholson58 · 22-09-2021, 20:00

You could do what they taught in class. Line up range markers and check the heading.

**Dockhead** · 23-09-2021, 09:24

Quote:

Originally Posted by Nicholson58

You could do what they taught in class. Line up range markers and check the heading.

I was never taught that. Sounds like a great idea.

I wonder however how accurate the ranges are around here.

Nicholson58 · 23-09-2021, 09:35

Quote:

Originally Posted by Dockhead

I was never taught that. Sounds like a great idea.

I wonder however how accurate the ranges are around here.

I don’t recall if the range bearings are magnetic or true. Check your charts for confirmation. Drive toward the range marker and also the reciprocal @ 180 degrees. Our rate compass and GPS data can be selected on the chart plotter as true or magnetic. Variation is shown on my paper charts. Technically, the CG inspector can ask to see the calibration data you have collected in this way to handle the magnetic deviation your boat inflicts on the compass. Never happens, I think they have bigger fish to fry.

requiem · 23-09-2021, 09:50

For aligning the heading, what comes to mind is to find a tall mountain or other object with a well-surveyed position, then get as far from it as you can whilst still being able to see it. Point your bow at the object and then adjust your heading sensor to match the object's bearing.

Assuming you can get a few miles away, and your boat's position error is no more than a few meters, the error should be quite small. An exercise for the reader...

Thinking a bit more, simply find a north-facing slip and align using Polaris. Check your almanac to determine what time it's precisely due north.

Bycrick · 23-09-2021, 10:01

The problem isn’t that I can’t get a known range line on the "ground." With my sat-compass there’s no line on it. With a stated accuracy of +/- 2 degrees, I might get it close by eyeball. But otherwise, the orientation of the device becomes just a best-guess.

**Dockhead** · 23-09-2021, 10:12

Quote:

Originally Posted by Nicholson58

I don’t recall if the range bearings are magnetic or true. Check your charts for confirmation. Drive toward the range marker and also the reciprocal @ 180 degrees. Our rate compass and GPS data can be selected on the chart plotter as true or magnetic. Variation is shown on my paper charts. Technically, the CG inspector can ask to see the calibration data you have collected in this way to handle the magnetic deviation your boat inflicts on the compass. Never happens, I think they have bigger fish to fry.

GPS compasses are natively in true so I don't care about deviation in this process.

Now that I think about this method, however, how accurate can it be? If you drive down the range, you will be aligning COG, not heading with the range, right?

toddster8 · 23-09-2021, 10:14

As the other folks have said, "Ground Truth" has to come from, well, the ground. Calibrating one electronic thingy against another is always going to be iffy.

But with overlay function, it's obvious at a glance if radar and electronic compass don't agree and something needs adjustment. I've found that most often when that happens, it's because some metallic or magnetic object has gotten too close to my fluxgate compass. (Not a very big boat so it's hard to find an isolated spot and maintain magnetic discipline around it.) I suspect that the most recent error was introduced by loading 150 feet of chain into the bow, which suggests that the compass might read differently when the anchor is run out.

There is a big steel bridge that lands right next to my marina entrance, so when calibration is off, the angle between the radar and plotted position is very clear.

**Dockhead** · 23-09-2021, 10:49

Quote:

Originally Posted by toddster8

As the other folks have said, "Ground Truth" has to come from, well, the ground. Calibrating one electronic thingy against another is always going to be iffy.

But with overlay function, it's obvious at a glance if radar and electronic compass don't agree and something needs adjustment. I've found that most often when that happens, it's because some metallic or magnetic object has gotten too close to my fluxgate compass. (Not a very big boat so it's hard to find an isolated spot and maintain magnetic discipline around it.) I suspect that the most recent error was introduced by loading 150 feet of chain into the bow, which suggests that the compass might read differently when the anchor is run out.

There is a big steel bridge that lands right next to my marina entrance, so when calibration is off, the angle between the radar and plotted position is very clear.

You have enumerated several strong arguments in favor of satellite compasses.