Autopilot trying to kill me?

[Refueler]

I had an ongoing problem for ages with my AH800 on the 25 .... the power plug and socket .... what we call a 'Aldis Plug' ... the two pin brass job with one pin bigger than other.
I could knock the plug or lead and BEEP - AP would stop or go to STBY having power feed broken for an instant ..

Finally I lost my rag with it ... took a Dremel to the pins and cut slots into each pin ... then carefully opened the slot with thin blade .. so that the pins made better contact ... just needs an occasional tweak to keep pins opened enough to make good contact.

 

[johnalison]

I'd agree with those who suggest that the conditions were far beyond anything an autopilot could be expected to handle. I almost always steer manually under sail, for several reasons:

1. I do a better job of reacting to wind shifts, gusts and waves than the autopilot.
2. I have limited battery capacity, and the autopilot uses a lot of power.
3. I enjoy sailing the boat! I only use the autopilot under sail to allow me to take short breaks.
4. I mainly use autopilot when under engine, because it's boring to steer under engine!

Snap. I am always mystified by those who buy a sailing boat and then choose not to sail it, though on anything of a passage I will have the autopilot set up ready for use if I need it.

 

[jbweston]

I think it's important for you to establish whether the autopilot was at fault - in other words would an averagely competent helmsman have had the same problem while hand steering because you had too much sail for the wind strength, or the boat is hard to steer downwind in a lively sea state because of a small rudder or because it's generally skittish, etc.

Few autopilots are going to prevent a broach if a good helmsman who's concentrating on the job couldn't prevent it. If you're satisfied that a human helm wouldn't have had any problem, then either the autopilot isn't capable of steering in those conditions even when it's working as well as it ever does, or it was a new problem (either a fault or something causing a malfunction, of the type suggested in the posts above).

In my previous boat I found the autopilot generally steered better than I did in most conditions, I think because it was better at keeping up its concentration and didn't get tired. But I had to make sure I wasn't giving it an impossible job with sail trim or too much sail.

 

[Daydream believer]

How I love my Aries when the going gets tough. Steers me all over the place in light airs but just gets better as the wind gets up & the Raymarine death trap begins to fail

 

[Daydream believer]

Snap. I am always mystified by those who buy a sailing boat and then choose not to sail it, though on anything of a passage I will have the autopilot set up ready for use if I need it.

But there is something magical about watching the Aries do all the work

 

[thinwater]

• Intermittent power. Tiller pilots are famous for losing fix if the plug is jostled. But it could be anything.
• Once on a long cruise the pilot started acting weird. I wasn't doing anything ... other than removing some old adhesive with a metal scraper near the autohelm compass, under the deck. It took me a few minutes to put it together, because when I stopped what I was doing to investigate, it straightened out.

 

[sailboatliving]

That would have taken you 10 days to hit (-;

No, I was going 7 knots+ at 32 knots of wind, I had under 12 minutes and the boat wouldn't turn back. I was under 1.5 kNm from land

 

[Mark-1]

I was sailing downwind first the wind was 8-10 knots and later on the forecast was 15 knots but the wind built up to 28-32 knots. Apparent wind was probably 22-26 I was sailing downwind (30-40 degrees from run) not particularly fast (6 knots) with dirty boat bottom then the autopilot went on a 90 degrees course change to close haul and heading straight towards land(under 1.5 kNm away). The toe rail was under water and the rudder didn't respond, so I had to turn all around which put a lot of extra stress on the sails.


Then it happens again a few minutes later

My autopilot is a Raymarine type 1 on a 37 foot sailboat at around 8 tonn.

The autopilot is on leisure setting.

It happened a few days day before as well in strong wind.

I furled in the Genoa and it behaved better.

I haven't moved anything close to the sensor thing.

What can I do in the future to avoid the autopilot losing control of the boat? I could have reefed but the speed was not even 7 knots and I was in a hurry to arrive before dark.

Was it steering to wind angle, compass or COG? Were you following a route and accidentally hit 'next waypoint'?

If it was steering to compass could something ferrous have momentarily got near the fluxgate compass? (eg Metal object rolling in a locker.) If it was wind angle did something obstruct the wind transducer or could movement of the mast have significantly affected the apparent wind direction at the masthead?

 

[sailboatliving]

My first thought is that if the rudder wasn't responding to you after it happened, how would the autopilot be able to do any better? That fits with the boat being overcanvassed, the autopilot not being able to react quickly enough, and then it not being able to get back on course (I doubt it's programmed to turn 270 degrees if it can't do 90).

Anything in common with what you were doing at the precise moment it happened on the other two occasions?

It was probably due to over canvassed both days. The wind built both days. As you can see at the picture it happened twice in a few minutes and stopped when I rolled in the Genoa.

 

[sailboatliving]

Was it steering to wind angle, compass or COG? Were you following a route and accidentally hit 'next waypoint'?

If it was steering to compass could something ferrous have momentarily got near the fluxgate compass? (eg Metal object rolling in a locker.) If it was wind angle did something obstruct the wind transducer or could movement of the mast have significantly affected the apparent wind direction at the masthead?

Steering to compass heading

 

[Wandering Star]

I believe as Paul Rainbow has already pointed out, the “type 1’ refers to a plumbed in drive unit so it’s not going to be anything to do with plugs or sockets. My own view is the boat probably had too much sail up front and as someone else says - it effectively broached.

 

[jamie N]

As others have stated, and the OP now knows, the boat was over canvassed and overcame the abilities of the unit.
This tale did remind me of a near disaster that I had whilst leaving Inverness marina and going under the Kessock bridge. The TP10 was doing the steering whilst I tidied the fenders/lines/stuff away. As we passed beneath the bridge, maybe 5 metres away from the bridge pontoon (steel protection), the TP10 made a hard stbd, and to my horror, aimed at the bridge! I just managed to disconnect, and correct the manoeuvre, missing the bridge by a red c**t hair, and carried on.
It's obvious of course that I was too near to the bridge, and was very lucky to 'Just' get away with it. Coming back in from a pretty good sail around the Firth, I engaged the TP10 without it connected and went close to the steel channel markers in the area, and was surprised to find that it was affected out to about 7metres by the field from these piled structures.
From my ROV work I know that piles have a very strong magnetic field after being hammered, which doesn't diminish and nearly caught me out. Lessons learnt.

 

[AntarcticPilot]

No, I was going 7 knots+ at 32 knots of wind, I had under 12 minutes and the boat wouldn't turn back. I was under 1.5 kNm from land

kNm would be kilo Nautical Miles, I.e 1,500 nautical miles. I think you mean 1.5 Nm.

 

[Mark-1]

As others have stated, and the OP now knows, the boat was over canvassed and overcame the abilities of the unit.

Well yeah, but I'd have thought it would be difficult to mistake that mode of failure for a rudder generated turn.

Clearly not.

 

[flaming]

Well yeah, but I'd have thought it would be difficult to mistake that mode of failure for a rudder generated turn.

Clearly not.

If you've never experienced a broach before, and the first was under autohelm, you probably would wonder what on earth was going on.

 

[sailboatliving]

I did change it from leisure mode to cruiser mode now, so the autopilot will keep a more strict course. The autopilot is ACU 200 btw. There was no joystick options.

Broach was exactly what happened twice in a row. I started sailing full time 6-7 months ago and it was my first broach. Scary s... In 32 knots wind

 

[DrSpock]

I did change it from leisure mode to cruiser mode now, so the autopilot will keep a more strict course. The autopilot is ACU 200 btw. There was no joystick options.

Broach was exactly what happened twice in a row. I started sailing full time 6-7 months ago and it was my first broach. Scary s... In 32 knots wind

The ACU is the actuator - it has no course decision making in it. The autopilot processor is part of the EV-1 or EV-2 sensor core. This is why the control head will show No Pilot if the EV is disconnected.

 

[Roberto]

kNm would be kilo Nautical Miles, I.e 1,500 nautical miles. I think you mean 1.5 Nm.

Or a hell of a torque

 

[johnalison]

As others have stated, and the OP now knows, the boat was over canvassed and overcame the abilities of the unit.
This tale did remind me of a near disaster that I had whilst leaving Inverness marina and going under the Kessock bridge. The TP10 was doing the steering whilst I tidied the fenders/lines/stuff away. As we passed beneath the bridge, maybe 5 metres away from the bridge pontoon (steel protection), the TP10 made a hard stbd, and to my horror, aimed at the bridge! I just managed to disconnect, and correct the manoeuvre, missing the bridge by a red c**t hair, and carried on.
It's obvious of course that I was too near to the bridge, and was very lucky to 'Just' get away with it. Coming back in from a pretty good sail around the Firth, I engaged the TP10 without it connected and went close to the steel channel markers in the area, and was surprised to find that it was affected out to about 7metres by the field from these piled structures.
From my ROV work I know that piles have a very strong magnetic field after being hammered, which doesn't diminish and nearly caught me out. Lessons learnt.

This happened to me yonks ago, in the Caen canal. Quite why I chose to use the autopilot in a canal is lost in the mists of time, but it took a couple of bridges before I worked out what was happening. When I checked the anomaly it was quite considerable.

 

[Daydream believer]

As others have stated, and the OP now knows, the boat was over canvassed and overcame the abilities of the unit.
This tale did remind me of a near disaster that I had whilst leaving Inverness marina and going under the Kessock bridge. The TP10 was doing the steering whilst I tidied the fenders/lines/stuff away. As we passed beneath the bridge, maybe 5 metres away from the bridge pontoon (steel protection), the TP10 made a hard stbd, and to my horror, aimed at the bridge! I just managed to disconnect, and correct the manoeuvre, missing the bridge by a red c**t hair, and carried on.
It's obvious of course that I was too near to the bridge, and was very lucky to 'Just' get away with it. Coming back in from a pretty good sail around the Firth, I engaged the TP10 without it connected and went close to the steel channel markers in the area, and was surprised to find that it was affected out to about 7metres by the field from these piled structures.
From my ROV work I know that piles have a very strong magnetic field after being hammered, which doesn't diminish and nearly caught me out. Lessons learnt.

Funny that. My Raymarine AV100 went wrong - veering 30-40 degrees when I was below deck & under power- just near the Kessok bridge in 2014. I did not notice course change. I ran aground & inspite of it being flat calm & only on the bank for an hour I lost my rudder on a rock or small boulder. A dolphin swam round & round taking the p..ss. I was towed back to Inverness marina by the inshore RNLI, who happened to be under the bridge on a training run. So they used me as a real life training exercise.