Repairing the Raymarine ST1000+ Tiller Pilot

Dismantled tiller pilot.

I have been using the ST1000 since June 2013 (around 7 years) and have logged around 3,700 nautical miles in that time, most of it on autopilot. I have tried to be easy on it by not expecting it to deal with heavy helm or crazy running conditions. The Tiki is also a very light boat, contributing to my expectation that the tiller pilot has had a relatively easy life.

Broken ram seal housing (Xyzzy: Sailing Anarchy).

Recently it became clear that there was a water ingress issue as the LED screen was cloudy, or cutting out, and was not clearing up after a drying period between sails.

I disassembled the tiller pilot (top picture) and found two issues: the ram seal housing was broken, and the rubber ring and gasket that cushion the ram at maximum and minimum travel had disintegrated. I repaired the housing before taking pics, so the pic above is from another on Sailing Anarchy. My housing was broken on both sides, not just one side as shown here.

Arrows show G/flex glued fillets.

To repair the housing, I assembled it on its steel carrier and glued it with thickened West Gflex epoxy. I don't consider this a permanent repair and expect that it will fail sooner or later. I think that the broken housing allowed water to get past the seal, explaining the fogging of the LED screen.

Reassembling with the repaired seal housing in place.

I brushed down the electrical control board with alcohol and a toothbrush and removed some obvious spots of corrosion. I then let it dry for several days under a lamp, and also to let the Gflex set good and hard.

Electrical board refitted.

The rest of the tiller pilot did not show any obvious signs of wear and tear. The gearbox cogs and belts seemed to be operating OK. I reassembled the pilot taking extra care to replace the case seal without pinching it. The case seal had aged and was too long (stretched?) to fit neatly in the grooves, so I put it into the freezer for an hour to shrink it, and that did the trick. See videos below for testing after reassembly. The pilot is fully functional again

Testing the switch functions.

Testing the fluxgate compass in "auto" mode.

Close up of drive belts in operation.

So here is the thing: the reason the seal housing was broken and that the rubber gasket stops were mincemeat is that Raymarine does not install ram travel cut-out switches in the ST1000 and 2000. So when the ram reaches the extent of it's travel it just mashes into each end of the tiller pilot assembly. The motor and gearing are stronger than the construction of the seal carrier housing and it bends the mounting plate and breaks the housing. To me this is incredible, but is common knowledge to longterm users of these products, and clearly Raymarine. I share the amazement of many folk online who have shared their experiences about the demise of their tiller pilots! It is even more surprising in that the engineering standard of the pilots seems high overall - I guess that the cost of adding ram travel limits to the design was just too much for them to remain profitable. Anyway, if you have an ST 1000 or 2000, DON'T LET IT RUN OUT TO MAXIMUM TRAVEL in either direction if you want it to last. However, it has worked flawlessly for 7 years and nearly 4 thousand miles, so maybe that is a return on the original investment?

Voila! Ready to go again.

I don't know how long my repair will last, but fingers crossed for 12 months or so to build up the boat budget for a new pilot. I have done lots of online research and found that the Simrad 22 and 32 pilots have a ram position sensor that stops the motor at full travel. This is probably where I'm headed for a replacement, but it is too pricey at the moment.

 

Update 10/2020

The pilot only worked for a couple more hours and then the fluxgate compass failed. I was on my way to the Farallons and had to hand steer for the remaining  50 NM on that trip. A new fluxgate was more than $100 on Ebay, so I bit the bullet and bought a new ST2000 (around $450). I wanted to try the Simrad 32, but the covid prices had climbed to the $700 - $800 mark, which was just too much.