| 
			
				| 
					
						| Project Log:  Saturday, January 28, 2012 
 A month or two ago, I purchased the components for the 
						new hydraulic steering system.  After various 
						research and correspondence, I selected a Teleflex 
						Seastar system, upgraded with a brass cylinder (vs. the 
						stock aluminum) and Kevlar hoses (vs. the stock nylon 
						tubing).  At that time, I'd performed measurements 
						in the after steering room (the cockpit lazarette), and 
						knew from the onset that providing adequate room for the 
						hydraulic cylinder would probably require certain 
						modifications.  All of this would also eventually 
						tie in with my Simrad AP24 autopilot and RPU80 hydraulic 
						autopilot pump.
 |  
						| Photo from October 2011 showing the 
						lazarette opening
 
  |  
						| The basic components from which the new system would 
						grow consisted of a two-piece bronze rudder post; bronze 
						rudder tube and stuffing box; bronze coupler; and 
						stainless steel steering arm.  The original 
						cable-driven push/pull steering system had connected to 
						the steering arm, which was oriented in a transverse 
						direction facing the port side of the boat.
 |  
						| 
   |  
						| Earlier, I'd determined that there was not going to be 
						room for the hydraulic cylinder with the steering arm in 
						this position:  the fore and aft clearance in the 
						steering room was simply too short.  With more room 
						available across the space's width, my first test would 
						be to move the steering arm's position on the rudderpost 
						to a different position.  At this point, however, I 
						didn't know if the rudderpost (and therefore steering 
						arm connection) was round, or squared off.
 
 To begin the process, and to ensure that the rudderpost 
						and all related components would be in their correct and 
						intended positions, I temporarily reinstalled the 
						rudder.  I love this rudder system, where the 
						rudder is attached with mating flanges and four bolts:  
						easy and effective.  With the rudder back in place, 
						I knew the rudderpost components above would be at the 
						correct height and position, and could also ensure the 
						rudder's position easily by eye when it became 
						necessary.
 |  
						| 
  |  
						| Next, I loosened the setscrews (a series of hex-head 
						bolts) and pulled the upper section of the rudderpost 
						(which extended through the coaming for tiller 
						attachment) up and out of the way for now, holding it in 
						place with some locking pliers. I didn't pull it all the 
						way out because there was a groove with bronze key set 
						in plate above the coupling, and though this didn't seem 
						to serve a current purpose I saw no reason to remove the 
						key either.  The key prevented the rudderpost from 
						sliding through the round hole in the coaming.
 
 |  
						| 
    |  
						| Then, I could remove the coupler, and finally the 
						steering arm.  The rudderpost was indeed squared 
						off, as was the hole in the steering arm, so whether or 
						not I could use this arm in the new system would depend 
						on whether I could simply rotate it 90° and have the 
						cylinder fit.  However, the cylinder, which 
						required 28" of clearance for full operation, was 
						clearly too long to fit in this way.  This was not 
						unexpected, though I'd hoped for the easy solution.
 
 I thought rotating the steering arm 45° would work.  
						So using the old one as a rough guide, but changing the 
						length and shape slightly, I made a plywood mockup of a 
						new arm, with the square hole rotated 45°.  
						According to the instructions, the steering arm should 
						be 6" between the centers of the rudderpost and bolt 
						securing the cylinder to the arm, so I made the mockup 
						in accordance with these instructions.
 |  
						| 
    |  
						| The cylinder needed to be installed parallel to and in 
						line with a line drawn through two points defined by the 
						maximum angle of the steering arm, or about 30° to each 
						side.  The instructions called for a 6" rudder arm 
						(6" from the center of the rudderpost to the center of 
						the pin securing the cylinder to the arm), with the 
						maximum angles defined by a 3-1/2" offset to each side 
						of the center position of the arm.  Using the 
						plywood mockup as a guide, I made various marks and laid 
						out these key points.
 |  
						| 
  |  
						| Eventually, the steering cylinder would align with these 
						marks, perpendicular to the steering arm and running 
						forward towards the bulkhead.  Even now, the length 
						of the cylinder and its maximum travel would require 
						some modification to the bulkhead, but the opening would 
						end up inside the starboard cockpit locker (which I'd 
						yet to build).  These modifications and final 
						installation would come sometime later.  For now, 
						I'd determined what I needed and how to proceed, and my 
						first task from here would be to have a machine shop 
						make up the new steering arm based on the plywood 
						mockup, after which I could proceed with final layout 
						and installation.
 
 There was an old ring terminal left over from a 
						now-defunct bonding system secured to one of the bolts 
						holding the rudder stuffing box to its plywood support 
						platform.  To remove this, I needed to remove an 
						extra nut that was securing it, but this simple task 
						turned into something larger when I found that I 
						couldn't get the nut off the top of the stud; even 
						holding the fixing nut beneath didn't prevent the screw 
						from turning, since the action was pulling the screw 
						head (a panhead, slotted screw that was inaccessible 
						beneath the platform and visible only thanks to my 
						camera) into the plywood.
 |  
						| 
  |  
						| Eventually, with some needle-nose locking pliers that 
						just fit into the space between the two nuts, I got the 
						nut and old ring terminal off, but I couldn't leave this 
						fastener as is.  I removed the fixing nut and old 
						bolt (using the locking pliers to hold the stud) and 
						installed a new bronze hex bolt with a large fender 
						washer to protect the bearing surface.
 |  
						| 
  |  
						| The after two (of four) bolt heads were completely 
						inaccessible for all practical purposes, given the 
						confines of the space and the impedance of the rudder 
						tube; I was able to tighten the nuts slightly and felt 
						they were secure as is, so I decided against any heroic 
						efforts to do anything with them.  However, I 
						thought I might also replace the other forward bolt with 
						a new one and washers like its counterpart.  I made 
						this decision after reviewing the photos on my computer 
						vs. the small camera screen, so as of this writing I'd 
						not yet done the chore.
 
 Earlier in the week, I'd painted a small section of the 
						new engine foundations to test the paint's curing 
						ability over the relatively new epoxy:  Bilgekote, 
						my paint of choice, will not cure properly if applied 
						too quickly over new epoxy.  Allowing the epoxy to 
						reach its full state of cure, which can take anywhere 
						from a couple days to a couple weeks, allows the paint 
						to work properly when applied.  I hoped it'd been 
						long enough, as I was anxious to paint the engine room 
						and get ready to install the engine soon.
 
 The test patches seemed to have cured completely, so I 
						went ahead and painted the forward part of the engine 
						room, as far aft as the ends of the foundations; I left 
						a section unpainted as I still needed to install some 
						new through hulls there (cockpit scuppers and engine 
						intake).  The unpainted area wouldn't stand in the 
						way of other important progress, and in any event I 
						planned to do the through hulls as soon as I could get 
						the parts needed.
 |  
						| 
  
 
      |  
						| I spent the last part of the day making up some orders 
						for things I'd need for through hull and engine 
						installation and some other tasks in the coming week or 
						two.
 
 
 |  
						| Total Time Today:  5 hours
 |  
						| < 
						Previous | Next >
 |  |  |