Dear TWL, We have had some not-so-good experience with our 24VDC B&G autopilot hydraulic gear pumps in 2002. We have two completely parallel pilots. Pump #1 failed at about 1400 hours, almost causing some serious damage. This is one of our very few adventures - described in the new website report "Passage Hobart to Sydney". My question to TWL is - why would DC motor failure occur so early in life? Why so much brush wear on both pumps? For background, here is the msg I sent to Alan Wilson at Hypro UK, who supplies the pump assemblies to B&G. <<<<< We selected the Hydra pilot with 24VDC T3 pumps thinking we were acquiring the most reliable pilot system suitable for our yacht and its useage. Then we completedly duplicated the pilot to purchase insurance that we would never have to deal with a pilot failure in a remote area (we are sailing for Patagonia next year). We thought the probability of ever actually needing the redundant pilot was extremely small. That hasn't turned out to be true so far. So, before we install the replacement T3 I thought it prudent to investigate further why we are experiencing such short life on the T3 pumps. I say pumps plural because we have concerns that the other pump in our dual pilot system may shortly fail as well - since both pumps have approximately equal service hours. History: at first failure a bit less than 12,000 nm over 24 months, or about 1400 hours on both pumps/pilots. We estimate the usage of each pump to be roughly the same, based simply on measuring the remaining brush lengths. These are 11mm remaining on both pumps (18.5mm new). So the first failure was at about 700 hours life - which seems incredibly short. Very few hours of that have been in DWIND/downwind steering mode - almost all have been NORM/normal mode. Installation parameters: 52-ft Morrelli & Melvin sail cat, 34,000 lb displacement true loaded displacement. Hynautic 17.6 cubic Inch K-2 cylinders, series driving two fluid coupled, balanced spade rudders. The yacht is very easily steered, and very directionally stable. We measure 4.5A at 24VDC when the pilot is driving rudders hard over/hard over at dock side. And 7A when release valve opens at 900 psi when rudders driven against mechanical stops. More information on Adagio can be found at www.adagiomarine.com. Pilot duty cycle is apparently normal - just based on listening to pilot plusing and discussing with Stan Montgomery whilst trying to imitate pilot pulsing. We don't have first-hand experience to judge normal duty-cycle. >>>>> Hypro wrote back: > Your k2 ram has 288cc volume = no load hard over time 7 sec. ( to fast ) > and is back loaded by the second ram & rudder. Double the normal load. > The fast hard over time will cause over correction and motor reversing, this + > the extra load is causing the short brush life. > A 1.5 lpm pump usng the type 3 24 volt motor will give a 12 sec. hard over. I replied: > It is double the load of one of Adagio's rudders. But it is hard to believe it > is double the load of an equivalent 34,000 lb. displacement monohull, > assumming a modern hull form and rudder design. Or that it is double the load > that Adagio would require if she were designed with a single rudder of > equivalent performance. > > I would expect our cat to require the pilot pump to perform less work > (power-time integral) than the typical boats on which T3 pumps are fitted(?) > > Another way to look at is to consider the work required to steer Adagio if the > fluid rudder link was replaced with the common mechanical linkage driven by a > single K2 cylinder. The results should be similar. Further comments... First, our measured rudder rate is actually faster than he calculated, about 11 degrees/second for +/- 30 degrees (stops are +/- 40deg). HO-HO time is 4.8 sec. Second, I think he is dead wrong to say "too fast". B&G pilots aren't like the old dumb pilots that will overshoot or go into servo-instability. They are pulse-width modulated within 3 speed ranges (fast, med, slow). This only happens in BIG seas, but if the pilot commands a large correction, the computer phases down thru the 3 speeds, finally pulsing into the exact heading. It never overshoots. Normal operation the pilot just pulses tiny corrections. Summary: the B&G computer knows how fast the rudder responds, therefore it only commands what it needs (not always asking for Full Speed that the pump is capable of). Third point, I'm 99% certain that the pilot has never encountered a steering system fault - such as rudder doesn't move when commanded - which would cause high current draw. And even if that had occurred, the whole system is protected by the Hynautic pressure-relief valve set at 900psi (if you manually drive the pilot against stops, the for-sure working relief valve operates, the pump draws only 7 amps, not the 25A that B&G defines overcurrent fault). Sum up: Mainstay in Sydney examined the failed pump motor, commenting that it sure looked like it had been through more than one set of brushes. I.e., more carbon buildup than they expect. My worry is that we still don't understand why this happened. Further that both pumps have consumed the same amount of brush material. So when will pump #2 fail? What are we overlooking here? Many thanks, Steve