Autopilot - Heading hold turns 20 degrees to port

[apm42]

I have a new Nimbus 405 Coupe (13m, twin shaft drive) fitted out with all Simrad electronics


I have a problem with the autopilot that has been there since the boat was delivered.



First, it needs to be noted that everything else works fine. Go to waypoint works.
Managing the course with the -1 -10 +10 +1 buttons works fine. The autopilot
can hold a course once given it.


The problem I have is that when I select Heading Hold the boat immediately and obviously makes a 20 degree (guess) to port.


I have an HS60 GPS compass and NSS EVO 3S chartplotter with NAC-3 autopilot.


Can anyone suggest what the problem might be?


My immediate guess would be that the autopilot is being fed a magnetic course which it is treating as a true bearing. Could this be a software bug?


The magnetic declination at my position (New Zealand) is 22.5 degrees.



I am using the latest Simrad software 20.2.1.


I have tried Simrad APAC support but all they send back is a random selection from an FAQ without trying to understand the problem at all. One of the things they sent back was how to commission a Lowrance Outboard Autopilot (I have a Simrad autopilot with twin inboard shaft drive engines).

The marketing for the HS60 GPS compass says:


"The HS60 GPS compass delivers accurate heading and position data for use with Simrad chartplotters, autopilots, radar, and AIS. Ideal for applications where precise heading data is essential, such as autopilot systems, the HS60 provides an all-in-one alternative to a separate compass and GPS receiver."




This cannot be a magnetic compass problem (Simrad support suggested to check this) as it should be using the GPS based 'compass'.


The HS60 is selected as the compass source.


All ideas gratefully received.

[donradcliffe]

I need a better description of the problem, but it could be that the power wires for the autopilot are run close to the heading device. What do you mean by Heading Hold?

[apm42]

Originally Posted by donradcliffe

I need a better description of the problem, but it could be that the power wires for the autopilot are run close to the heading device. What do you mean by Heading Hold?

The HS60 GPS "compass" is mounted on the mast. It is not a magnetic device and is therefore immune to power wires running next to it.


'Heading Hold' is the button you push to engage the autopilot to maintain your current (manually steered) heading.

[donradcliffe]

From the HS60 manual:

PGN 127250 Vessel heading message
Heading sensor value with a flag for True or Magnetic. If the sensor value is Magnetic, the deviation field can be used to produce a Magnetic heading, and the variation field can be used to correct the Magnetic heading to produce a True heading.

It sure sounds like a bug, given that the course change is about the same as the difference between magnetic and true heading. The only question is why does the autopilot shift from magnetic heading before you engage it to true heading after you engage it (or vice versa)?

Do you have a heading readout on the chartplotter, and does it suddenly change when you hit Hold Heading? (probably not)

Fortunately, all the gear is Simrad, so they can't point the finger at the other guy.

[apm42]

Originally Posted by donradcliffe

From the HS60 manual:

PGN 127250 Vessel heading message
Heading sensor value with a flag for True or Magnetic. If the sensor value is Magnetic, the deviation field can be used to produce a Magnetic heading, and the variation field can be used to correct the Magnetic heading to produce a True heading.

It sure sounds like a bug, given that the course change is about the same as the difference between magnetic and true heading. The only question is why does the autopilot shift from magnetic heading before you engage it to true heading after you engage it (or vice versa)?

Do you have a heading readout on the chartplotter, and does it suddenly change when you hit Hold Heading? (probably not)

Fortunately, all the gear is Simrad, so they can't point the finger at the other guy.

Thank you for responding. I am sorry if my previous message was a bit terse I was just trying to be clear and concise.


I am not on the boat at the moment. Yes my plan is to record the heading change when the autopilot is engaged more accurately than a guess. The chartplotter can display your course as magnetic or true. I have selected true but this is just a simple translation of the available NMEA2000 data.


Yes there is some benefit of having all the kit from the same manufacturer.


I am very disappointed in (APAC) Simrad support though. I have sent them two issues and their responses have been purile. All you get is random paragraphs from an FAQ. They show no understanding of how the device works or any form of problem analysis. Clearly, I know more about the systems than at least the first level helpdesk support people.

[goboatingnow]

I would perform the installation calibration and tuning again before doing anything else.

[apm42]

Originally Posted by goboatingnow

I would perform the installation calibration and tuning again before doing anything else.

The autopilot works perfectly apart from this one issue.


It can hold a course. It can follow a route with turns. It can 'goto waypoint'. It follows the course nudge button instructions correctly (-10 -1 +1 +10 degrees).



The calibration is to do with how the autopilot works the steering pumps in order to effect a course hold or change. If this was not set up properly it would not be able to
do the above or to do it effectively.



As far as I can see it is being commanded to follow the wrong course.

[HeywoodJ]

Does the boat turn 20 degrees and stay there? Or does it (or would it if given enough time) turn back to the desired heading after "a while"?

We don't have a Simrad, but with our AP, when engaging the "Auto" button (same as your "heading hold") the AP always sets the rudder angle to zero. Does your installation include a rudder angle sensor connected to the AP? This was a problem for us when we had an unbalanced helm, we'd get a big turn on going to Auto, followed by eventual correction when the darn thing figured out that rudder angle zero ≠ straight course. Very poor control algorithm programming where the designers completely failed to understand the concept of "bumpless transfer". Don't know if that's your issue, but it is one we had (have) with identical symptoms.

[apm42]

Originally Posted by HeywoodJ

Does the boat turn 20 degrees and stay there? Or does it (or would it if given enough time) turn back to the desired heading after "a while"?

We don't have a Simrad, but with our AP, when engaging the "Auto" button (same as your "heading hold") the AP always sets the rudder angle to zero. Does your installation include a rudder angle sensor connected to the AP? This was a problem for us when we had an unbalanced helm, we'd get a big turn on going to Auto, followed by eventual correction when the darn thing figured out that rudder angle zero ≠ straight course. Very poor control algorithm programming where the designers completely failed to understand the concept of "bumpless transfer". Don't know if that's your issue, but it is one we had (have) with identical symptoms.

A good point (does it turn back). I will investigate. So far my reaction has just been why are we heading in that direction and correcting it immediately with the buttons.


I do have a rudder feedback sensor. The rudder angle is displayed on the chartplotter. I will monitor this once I can get out on the boat again and run a test.

[Kastaway]

I would suggest changing (in setup on the chart plotter) from magnetic course to true, or vise versa and see if it makes a difference when engaging auto to hold the current course.
Also if using an external GPS you could first check that it is set as the default and if so try changing in the set up to use the GPS built in to the MFD.(use this device, in setup - from memory) If any of these ideas are successful it may help pinpoint where the issue is.

[JebLostInSpace]

I don't have a solution for you, but mine also does the same thing. Mine is also a Simrad system. When I first engage the autopilot to maintain my current heading, it routinely turns about 8 degrees to port. I know it is 8 degrees because I always hit buttons to turn 8 degrees starboard immediately after engaging the system. I don't even wait for the boat to turn off course, I just engage the system then ask for 8 degrees starboard.

I have also suspected an issue with magnetic vs. true north, but have never been able to confirm or find anything useful in the settings.

I can say that in my case, I don't believe it is a rudder angle resetting to centered. This issue occurs when motoring with the rudder centered as well. I engage the autopilot, it provides a rudder input moving it off of center and turning us to port.

[svfinlandia]

Originally Posted by JebLostInSpace

I don't have a solution for you, but mine also does the same thing. Mine is also a Simrad system. When I first engage the autopilot to maintain my current heading, it routinely turns about 8 degrees to port. I know it is 8 degrees because I always hit buttons to turn 8 degrees starboard immediately after engaging the system. I don't even wait for the boat to turn off course, I just engage the system then ask for 8 degrees starboard.

I have also suspected an issue with magnetic vs. true north, but have never been able to confirm or find anything useful in the settings.

I can say that in my case, I don't believe it is a rudder angle resetting to centered. This issue occurs when motoring with the rudder centered as well. I engage the autopilot, it provides a rudder input moving it off of center and turning us to port.

I’m no help, but just followIng this thread to see what the solution is.
I have a Raymarine system and a few years ago I bought a new chart plotter and had a similar problem. Mine finally healed itself after a while all by itself.

Al, S/V Finlandia

[SaltyMetals]

You have had lots of good suggestions but let me throw in something completely different.

I had a similar problem with my Raymarine system, did the usual 2-3 circles to re-callibrate but the compass heading on the AP kept on dropping back by about 20-30 deg. I eventually tracked down the problem. A crew member had put her phone on charge on the top of the cupboard, directly underneath was the giro compass ! I have now added a warning note on that socket.
]Andrew

[Quebramar]

I incurred a similar problem in a chartered boat. I changed the AP response level from low to high and it actually solved it. The low level used to change course on a 2-3 minutes basis and probably the system anticipated a drift… not sure though but might be worth checking the setting.

[Howler]

If you still need help, try N. America service (early morning NZ time to catch them in the prior day’s afternoon). I had problems with a B&G autopilot and they were great at patiently helping to troubleshoot.