Raytheon ST7000 Autopilot goes AWOL??

[Borden]

On a recent long trip in the Med. from Spain to Tunisia our ST7000, for no apparent reason, suddenly decided to change course 245 degrees to stbd of the set heading. This happened four or five times during a five day non stop trip. Battery power was not a problem as due to lack of wind the engine was on most of the time and we carry over 800Ah of batteries. I checked that nothing was interfering with the fluxgate compas and drew a blank. As the contents of the locker close to the compass has not been changed and the Autopilot has been working fine in the past this does'nt seem to be a possibilty. The interesting thing is that on each occasion when it went AWOL it altered course excatly 245 degrees to Stb'd.
Any ideas?

[Pblais]

I checked that nothing was interfering with the fluxgate compass and drew a blank.

Could be the compass and still could be interference. They can go bad and things can change. You can notice errors in the compass by 90 degree quadrant usually if the compass is bad. In flat water do a large slow circle and see what the ships compass heading reads vs. the fluxgate reading. Do it clockwise and counter clockwise several times. Try straight line courses and compare the GPS reading to the fluxgate. It helps to have someone watching one compass as you call out the fluxgate numbers. Watching the two yourself and steering the boat all at the same time can drive you nuts. Make written notes and try to repeat anything you think is a problem.

Another trick is to disconnect the compass from it's mounting and leave it connected and powered. Turn it by hand in small increments going clockwise then counter clockwise while in the slip. It should be possible to see the reading become in error if it is the fluxgate compass. You may need an assistant to do this too.

The last issue is calibration. The calibration of a fluxgate stores a table of degrees and the plus or minus deviations. The calibration process builds this table by making you run in circles. It then knows you are in circle mode then it can compute what should be the headings from what it sees. This accounts for the magnetic deviation on the boat. If that table gets trashed in the memory of the computer that can cause errors like you see too. Re calibration might fix it or there may be an electronic problem and it gets trashed again or never stores properly.

You can't see interference so you can't know for sure it's not there. You have symptoms that could be interference so don't assume it's not. It took me days to find that out. The above will prove it one way or another. There should also be extra cable coiled up on the fluxgate as they don't always cut off the extra. Try moving it and see what happens.

So far none of this will cost you any money, you won't break anything, and you might find the problem with some certainty. After that you need to start swapping components or pull the whole thing out and bench test it. That process is expensive and time consuming. Swapping out parts is expensive since you can't return them if the part you swap is not the problem.

I went through something like this with a Simrad unit in the last boat. In the end it was not the compass. the compass was fine everything was fine. The compass was getting interference and we never found out what it was since there was nothing near by but moving it to a new location worked. The autopilot only ever worked in calm conditions and had been installed improperly by the original owner. On paper the install location for the compass was a great choice but in practice there was interference since we proved it. The problem was masked several ways. The original owner had a defective piston in the hydraulics and the second owner found that. I had a problem with the hydraulic motor and eventually fixed that. That eventually led to a problem where the calibration would fail always. Mechanical problems hid the installation problem for a long time. Everyone always assumed there was no interference. Odd stuff happens. Hope for a cheap solution.

[DeepFrz]

There have been a number of incidents of this type of thing happening. There should be at least a few threads on the subject. I don't know if anyone has ever posted what the final resolution was. It seems that most incidents were Raymarine autopilots.

Paul offers good advice from personal experience. I can only suggest the following.

Have you checked all of your connections?
Checked for any unusual voltage drops in the system?

[Bert]

Deep Frz, you are right! Same experience with the older Raymarine systems. I would highly recommend Simrad's Autopilots, even with a Raymarine enviroment... As an experienced person I have as well no solution for this problem, and nor Raymarine. Remember the cheap fluxgate compasses used by the "cruisers" systems are not a good solution at all, better have a giro system or a GPS compass, both expensive, but worth to think about... considering the use of power of a gyro look for a economic GPS compass, fluxgate as backup.

[psteele235]

Mermaids in the area?

[Sunspot Baby]

Mine jumped 180 deg. Local Raymarine service center says it happens occasionally and to go through the full commissioning calibration (both at dock and underway portions) on the auto helm to fix it. I have yet to do so.

George

[Pblais]

Remember the cheap fluxgate compasses used by the "cruisers" systems are not a good solution at all, better have a giro system or a GPS compass, both expensive, but worth to think about... considering the use of power of a gyro look for a economic GPS compass, fluxgate as backup.

GPS compasses are too slow to capture boat dynamics. I worked with Ring Laser Gyro's. They work best of all. They just cost $100,000. You can take a GPS signal and then lose the GPS signal for a few hours and still know where you are and how you got there with no serious position error works well enough to land aircraft.

Had a Simrad on the last boat and the Ray ST7000 now both about the same age. The Simrad learns. They have the better computer software. There is no sea state settings. It remembers how responsive the boat is and adjusts the amount of force required dynamically. I think the new Ray auto pilots do it now too. They make it sound like they invented it.