Autopilot Oscillates

[amxr39]

This is a new installation of a Garmin Autopilot. The rudder oscillates all the time regardless of responce setting. The video is on “hold heading “ on a straight course. Unfortunately the video file was unable to attach.
Any experts out there that can help?
Thank you.

 

[DavidM]

Some autopilots require “tuning”. Does your and have you done it?

David

 

[twistedtree]

The video would be great, if you can find a way to post it somewhere and provide a link.

When you say the rudder oscillates, does it hunt back and forth like it keeps overshooting the desired position, corrects and overshoots again, etc. with an oscillation period of around a second? Or does the heading oscillate, which is slower and a more common complaint?

Rudder oscillation can be cause by a bunch fo things. First to check is all the steering gear to be certain there is no play in any of the various joints. You will need someone to rock the steering wheel back and forth while you inspect joints. There should be no play anywhere. Even the smallest amounts can lead to rudder hunting. Also check the APs rudder feedback linkage to be sure there is zero play.

Air in the steering system can cause this too. If the pump was added/replaced as part of the installation, air in the system is a distinct possibility. Bleeding air out of a steering system can be maddening difficult for what should be a very simple task. If you have groaning sounds from the steering when the pilot is operating, it's likely air in the system.

I'm not familiar with the Garmin pilots, but the others I have dealt with have a setting for Rudder Deadband which is the smallest amount that the pump can reliably move the rudder. If the deadband is set too low, the rudder will hunt. The pilots I'm familiar with automatically set deadband as part of the commissioning process, and usually get it wrong - sometimes quite wrong - so you may need to adjust it manually. 0.75 to 1 deg is a good range. If you have to set it higher to eliminate hunting, then you probably have play or air in the system. If you can get it below 0.75, power too you. Lower is better and reflects finer rudder position control.

Another factor directly linked to deadband is the steering pump capacity, which translates into how fast it can move the rudder. Fast is good when maneuvering, but fast also increases deadband. With too fast a pump, but the time the pilot has commanded a rudder movement, the pump has pushed the rudder too far, and that leads directly to hunting. A slower pump allows fro lower deadband. You need to find an acceptable middle ground between rudder movement that is fast enough for maneuvering, and slow enough to get acceptable deadband. I have found around 12 seconds stop to stop rudder movement is at least a good starting point.

 

[SteveK]

If you have hydraulic then the air pressure must be correct, not enough and it will wander/oversteer.
Single or twin engines? I adjusted my rudder toes and that made a difference. I have a 32 year old Benmar that steers a straight course. Oh I also adjusted the gain on the circuit board in case any with Benmar did not know that adjustment is available.

 

[amxr39]

The video would be great, if you can find a way to post it somewhere and provide a link.

When you say the rudder oscillates, does it hunt back and forth like it keeps overshooting the desired position, corrects and overshoots again, etc. with an oscillation period of around a second? Or does the heading oscillate, which is slower and a more common complaint?

Rudder oscillation can be cause by a bunch fo things. First to check is all the steering gear to be certain there is no play in any of the various joints. You will need someone to rock the steering wheel back and forth while you inspect joints. There should be no play anywhere. Even the smallest amounts can lead to rudder hunting. Also check the APs rudder feedback linkage to be sure there is zero play.

Air in the steering system can cause this too. If the pump was added/replaced as part of the installation, air in the system is a distinct possibility. Bleeding air out of a steering system can be maddening difficult for what should be a very simple task. If you have groaning sounds from the steering when the pilot is operating, it's likely air in the system.

I'm not familiar with the Garmin pilots, but the others I have dealt with have a setting for Rudder Deadband which is the smallest amount that the pump can reliably move the rudder. If the deadband is set too low, the rudder will hunt. The pilots I'm familiar with automatically set deadband as part of the commissioning process, and usually get it wrong - sometimes quite wrong - so you may need to adjust it manually. 0.75 to 1 deg is a good range. If you have to set it higher to eliminate hunting, then you probably have play or air in the system. If you can get it below 0.75, power too you. Lower is better and reflects finer rudder position control.

Another factor directly linked to deadband is the steering pump capacity, which translates into how fast it can move the rudder. Fast is good when maneuvering, but fast also increases deadband. With too fast a pump, but the time the pilot has commanded a rudder movement, the pump has pushed the rudder too far, and that leads directly to hunting. A slower pump allows fro lower deadband. You need to find an acceptable middle ground between rudder movement that is fast enough for maneuvering, and slow enough to get acceptable deadband. I have found around 12 seconds stop to stop rudder movement is at least a good starting point.

This is a new boat. There should be no play in any of the steering linkage. It is about a one second repetition rate. I’m not sure of the piston size. But I can find out.
I can post the video on your or my Facebook page if you’d like it.
You know your stuff! Thank you.

 

[twistedtree]

This is a new boat. There should be no play in any of the steering linkage. It is about a one second repetition rate. I’m not sure of the piston size. But I can find out.
I can post the video on your or my Facebook page if you’d like it.
You know your stuff! Thank you.

I would still give everything a check over. You would be amazed at what can work loose, or never be fully tightened in the first place on a new boat. But air in the steering system would seem most likely on a new boat.

Have you talked to the builder or whoever did the AP install? It seem they aren't quite done with teh job yet.

No facebook here, sorry.

 

[amxr39]

I would still give everything a check over. You would be amazed at what can work loose, or never be fully tightened in the first place on a new boat. But air in the steering system would seem most likely on a new boat.

Have you talked to the builder or whoever did the AP install? It seem they aren't quite done with teh job yet.

No facebook here, sorry.

The so installer used a “power bleeder” with a hose running to the vent at the fly bridge. I want to check for free play in the wheel and the rudder when I get in this evening. I would think air in the system would show when manually turning the wheel.

 

[FWT]

I have Simrad. Not familiar with Garmin.

But on the theory of simple things first to rule them out:

Should have a routine to calibrate the system. Involving heading into free water, pressing a button to start it, and let it take over to do a circle or whatever pattern it uses to calibrate. Its easy. Takes 15 minutes tops. Check your manual for what is involved and how to do it.

Simple things like software updates can throw it out of whack. I've done it twice, and its no big deal.

Might have nothing to do with your issue but its easy and no harm / no foul.

 

[Circumnavigator]

If the boat is new, then this should be a warranty fix and your working on it may be a warranty problem.

There can be air in the power loop but not in the manual steering.

Jim

 

[EngNate]

No one has questioned if the dockside and sea trial setups have been done in accordance with the manual....

I believe that response setting is seastate filter. You've got too much rudder gain. I've setup a couple Garmin's and they needed no adjustments following the setup and auto tune procedures. A pump of too high capacity can be the issue. At dockside using power steer mode check hardover to hardover time, should be not less than 10 seconds or more than 15 seconds. Improper rudder sensor geometry or wrong calibration could cause your issue.

I want to clarify on twistedtree, deadband is the amount of course error the pilot allows to occur before any response is output. This is incorporated into the seastate filter.

 

[twistedtree]

I want to clarify on twistedtree, deadband is the amount of course error the pilot allows to occur before any response is output. This is incorporated into the seastate filter.

That may be on some pilots, and I think I'd call that heading deadband. At least on the Furuno and Simrad pilots, the manuals are clear that it's rudder deadband, and it is the minimum change to the commanded rudder angle. Until that much or more change is needed in the rudder angle, the pilot won't command any movement.

I believe heading deadband issues will cause weaving while underway, but won't cause rudder position hunting when docked. Rudder deadband will cause rudder position hunting as well as weaving. There is typically more hunting when not moving because there is less stabilizing water pressure against the rudder.

 

[Totto99]

I had the same issue with my Raymarine Autopilot. The boat swung from one side to the other on strait course. The solution was easy. I changed the setting from 'work boat' to 'power boat' and the problem was gone. I have no idea what the pre set is doing but the issue is gone.
I was the meaning that for a trawler the setting 'work boat' would be appropriate. It was not. Maybe there is a similar setting for Garmin.
Kind regards Thorsten

 

[Bmarler]

There may also be a speed setting for rudder response. Mine was improperly set up by the PO and hunted horribly. Maybe that’s the work boat/power boat setting referred to?
Again though, a thorough read through the setup process and commissioning instructions should cure it unless there’s a failing sensor or conflicting heading data.

 

[Nepidae]

There could be issues with your placement of the compass.
It could be a number of things.
The best advice I can offer based on my use of a Garmin autopilot (AP) is to get on your boat and call Garmin (800-800-1020) and the support person will walk you through calibrating the AP. It isn't difficult.
We did an 8yr Loop and I used the AP 95% of the time, and she was always right on.
Let me know how it works out.

 

[mvweebles]

Assuming the mechanical linkages are sound as suggested by TwistedTree, you may want to start calibration from scratch. I'm not familiar with Garmin, but usually calibration includes turning rudder stop-to-stop; and underway doing very slow circles. Once that baseline is set, you can begin tweaking the AP for your particular boat

Each manufacturer may use different terminology, but there are really only three adjustments:

• Gain (or "Rudder"): How quickly and to what extent rudder angle is adjusted when off-course is detected.
• Counter-rudder: If you notice when you're hand-steering, your inclination when returning to center is to correct a bit in the opposite direction. This is called counter-rudder and is programmed into the AP to more quickly halt the turn and return to course.
• XTE Error (or "Deadband" or "Yaw"): As mentioned by TwistedTree, this setting tells the AP when to ignore course deviations. This is especially helpful in heavier seas to prevent the AP from being over-worked.

Natural inclination is to increase sensitivity - increase gain so it responds more quickly; increase counter-rudder so it returns to dead-ahead more quickly; or reduce the deadband so it tolerates less deviation before corrective actions. In my experience, doing the opposite - reducing sensitivity - delivers better results four times out of five.

The hard part is there are three parameters so it's hard to adjust one and evaluate the overall affect. On my new Simrad, adjustments are 0-100 so it's really hard to figure out where to start (would be nice is they gave a base setting for a displacement speed boat vs a go-fast sport fisher). Definitely helps to start in flat water/calm conditions. Unfortunately, even with experience, setting up an AP is a bit of trial and error. If I had to guess - and it's only a guess, sounds like all of your settings are too tight and the AP is fighting itself. Reduce all settings, then increase as needed. The diagram below must be 30-years old but I still find it intuitive.

Good luck -

Peter