Auto Pilot ST60 Help Please

[Gardan]

Hi I have a Bavaria 38 1999 and last year had to change my rudder bearings which meant disabling and disconnecting the Auto arm from the rudder shaft , Since then my Auto Pilot does not work and sends me into a spin soon as it is engaged.
I'm just picking up on this problem again as had a chap at the boat yard tried to help me but never got anywhere.
Im going through my ST 60 owners manual , but losing my a bit , as I'm unsure if I need to re-set it to dealers settings or try and re-calibrate it . Can anyone help at all please.

Cheers Gary

Further update : Chap just emailed me and suggested I look at the 'Rudder position indicator ' which he explained is a metal bar attached to the rudder shaft below decks and we may have put this back on the wrong way when putting the auto arm assembly back on the rudder, I didn't help with this bit when we put  the rudder back in after changing the bearings in the rudder , so I will have to get up to speed on the RPI indicator .

[Lazy Pelican]

I’ve had a similar problem, which turned out to be the AP compass. Once it failed completely and another time it came off its gimbals- I took the cover off and placed it back on the gimbals.
Though in your case I’d look at the RPI, in the first instance.

[Yngmar]

If you've removed and reassembled things, the calibration between RPS, drive and rudder has likely shifted. You must do the calibration procedure, which is outlined in the manual. I'd also strongly recommend familiarizing yourself with the autopilot a little more, as the rudder position indicator on the screen is indeed the first step to troubleshooting and also quite helpful sometimes! 

Also check the fluxgate compass bearings on the display as you do a circle and see if they at least coarsely match up with your steering compass. If not, there's another procedure for calibrating the compass which involves driving in slow circles in a calm piece of sea until the fluxgate has recorded a new deviation table.

[MagicalArmchair]

I had my rudder off last year and replaced the bearings and whilst I did recalibrate the autopilot after (...I think?) I had no significant problems with the autopilot following that. I have had this issue before after taking off the autopilot drive unit. If the two (identical, black) wires that go to the drive unit are reversed, the autopilot will go round in hard circles. Has the drive unit been unwired at any past of the process?

As the wise and sage Jedi master Ynymar has said, the rudder position sensor is a good place to start. Engage the autopilot in dock, what does it do? It should just twitch about. If it flings hard to port or stbd I would wonder at those autopilot drive unit wires....

[PEA-JAY]

I think I had a similar problem some years back. I was told locally I needed to replace something quite expensive! 
 I phoned Raymarine and explained what was happening. It resulted that the red and black wire connected to the ram, that I had recently re-connected, red/red and black/black had to be black/red and red/black!

[Jeffatoms]

We had a similar issue a few years back having to do with rudder quadrant becoming loose and losing its grip on the autopilot ram arm.  The autopilot arm would activate and send the  rudder hard over because the sensor at the quadrant and rudder post position were out of synch due to the loose connection.  After remanufacturing the quadrant, I also deepened the keyway to insure a better and more reliable alignment.

[marioxp]

I have a problem, the autopilot stopped working, the rudder position does not give me the rudder position. I noticed that when I touch the cable from the rudder position arm, I lose the signal. I noticed that one wire is not connected, I disassembled the rudder position arm but I can't see anywhere where the fourth wire is connected.

[SYJetzt]

The missing wire you are looking for is the shield of the cable, which is NOT connected to the sensor arm (only on the computer side!).
The rudder sensor is a potentiometer with 3 terminals. Use an ohmmeter to measure between red and green terminal (disconnect the sensor from the computer first). There should be a resistance measurable. Then check between red and blue, note the measured value, then measure between blue and green (without changing the sensor position!). The sum of ohm-values between (blue and red) and (blue and green) should be the same as between red and green.