water in the keel

  • June 02, 2012 7:34 PM
    Message # 941390

    the following is a report regarding a recent bout of "water in the keel" suffered by my W28, "narwhal." i am sharing it here in the event that it may helpful to someone else who discovers a similar problem. bud has indicated that he has not come across this problem before.

    during a haul-out for bottom painting, i came across a few largish (3" dia.) hull blisters about 2' up from the bottom of the keel. i accidently drilled a hole into one blister a little too far into the fiberglass. the bit struck the lead ballast with a dull thump. when i withdrew the bit it was followed by a slow dribble of brown fluid and a few shiny twists of lead.

    by thumping on the surface of the hull nearby, i could get the fluid to spurt out a little with each thump. the fluid didn’t smell strongly of blister juice, but maybe more like diluted blister juice. it didn't taste particularly salty, but there was a hint of salt.

    now, "where did this fluid come from?" perhaps i had a leak in the keel somewhere, but it wasn't making its way to the bilge, which is characteristically dust dry.

    i consulted a shipwright in the yard. he felt that the fluid had to be coming from inside the hull somewhere. i disagreed, because as far as i could tell, the bilge is sealed from the keel space below. he advise that i drill 3/8" dia. holes at about 4" spacing over the entire surface of the keel to allow for drainage and drying out over time, probably months.

    i drilled about 30 such holes at a much larger spacing and collected about 2 gallons of the brown fluid in a bucket. i probably lost about a third of the drainage. some of the holes were dry, but most drained whatever fluid was above them. tapping a coin along the length of the keel indicated that some areas are solid and probably adhered to the lead and some areas are hollow or drummy, so not in contact with the lead.

    a careful search for possible water entry points was carried out, but revealed nothing. after about 3 months the bottom-most holes were still dripping very sowly, maybe less than a quarter of a teaspoonful a day. i borrowed a large manometer from a friend and approached the problem with the idea of pressurizing the keel space and then applying soapy water wherever i could reach and watch for bubbles in much the same manner as one would search for air leaks in a punctured innertube. i plugged off all the ventilation holes except for the lowest one and then attached an air hose to that one. with the aid of the manometer, i then tried to pressurize the hull to about 80" of water head. (that's a little less than 3 pounds per square inch and is the standard used here for pressure testing fuel tanks and the like.) at first it wouldn’t hold any head at all. i then got somebody outside continuously on the hand tire pump, while i went inside along the bilge with a large squirt gun full of soapy water. a huge pile of bubbles appeared around the short stump of a compression post that supports the inner floor liner just forward of the icebox.

    further investigation and removal of the post revealed that the post had sheared off the bilge bonding at the forward edge of the lead ballast. this allowed any water coming from the chain locker or the forward hatch to drain into the keel space rather than flow aft to the bilge pump. the plywood bilge support forward of the ballast stopped at the ballast without an overlap, so there was no structural support there. (refer to the w28 construction manual, page 2-a-3.) i repaired the breech, by chiseling out a good deal of the bonding and some rotten plywood and fiberglassed it over with epoxy. i fabricated a stainless shoe to support and spread the post load across the bonding to bear on the hull and the lead ballast.

    pressure testing was resumed, but would hold only about 40" of water head for a short time. more soapy water and a series of photographs of the bilge under the engine (there is no way to visually observe this area) revealed a clump of bubbles about a long foot forward of the stern tube. i lifted the engine and moved it forward far enough to drill a 4 3/4" hole in the liner under the engine. this revealed a rotten plywood plug in the bilge bonding that would have leaked into the keel space anytime water came from the stern tube. i chiseled out the plug. i then drilled a hole through the foam in the keel space there, to the bottom of the keel. the upper foam was white and springy, but the bottom foam was saturated, kind of gooey and black. i left a long, 1" dia. plastic conduit tube in the hole and then fiberglassed over to seal the leak.

    more pressure testing would hold only about 60 inches of water head for a short time. more soapy squirts revealed a few bubbles at the aft end of the forward floor hatch. i fiberglassed over another rotten plywood plug in the bilge there. i also found 2 more plugs, but these did not leak. i assume the holes that the plugs were intended to seal, were used to inject the foam into the keel spaces forward and aft of the ballast during construction.

    further pressure testing held about 80" of water head for 3 days. and so i declared the keel space sealed.

    why did i leave a conduit tube in the hole to the bottom of the keel? since i do not seem to be able to vent the space bone dry, perhaps some sort of internal ventilation may assist with ongoing drying. my intention is to bond a small air tube through the forward end of the bilge bonding and into the keel space. and also to insert a 1/4" dia. air tube through the new conduit tube to the bottom of the keel and seal this to the bilge bonding as well. by pumping air into the forward tube any fluid in the bottom of the keel should be pushed out the aft air tube and thus some ongoing ventilation may continue. i have warned myself that in the event of a leak from outside the keel space, this arrangement could sink the boat, although perhaps more slowly then before i sealed the leaks. with that in mind, i intend to be able to seal the forward air tube and also lead the aft tube to a point above the waterline.

    boating is fun!

  • February 05, 2016 12:22 AM
    Reply # 3801882 on 941390
    Deleted user

    I goin a spot on the Keel where the fiberglass was deteriorated and drilled a hole deep into it today with the die grinder. water begun dripping out but not a great deal. I'll tap along the keel and see if I can find any dead spots like you and drill them as well. What did you do to fill the void here?

  • February 05, 2016 4:23 AM
    Reply # 3802089 on 941390

    i measured the thickness of the fiberglass hull in several of the drilled holes.  it is a fairly consistent 15 mm thick, so you need to go in at least that far before hitting the lead.  in most places there seams to be about 3 to 5 mm of void between the lead and the inside of the hull.  this should have been filled with foam to prevent the lead from knocking around in there, but i suspect the foam job was a little patchy.  if you have the construction manual, page 2A3 shows the location of the lead ballast. the ballast extends from the bottom of the keel upwards, a little less than 400 mm. it extends from the lowest point in the bilge forward nearly 2000 mm and aft about 600 mm.  when i tap a 20 cent coin over this area on the outside, some places sounds like tapping on solid concrete and others like a hollow drum.  i assume the hollow areas are where there is no foam in the void.  i plan to fill the void area with warm neat epoxy. the void needs to be dry or ordinary epoxy wont cure properly.

  • February 07, 2016 12:26 PM
    Reply # 3805523 on 941390
    Anonymous member (Administrator)


    " i assume the hollow areas are where there is no foam in the void.  i plan to fill the void area with warm neat epoxy. the void needs to be dry or ordinary epoxy wont cure properly." 

    How are you thinking of getting the epoxy into the voids?  

    If there are foam voids would filling the voids with foam also work? 


  • February 07, 2016 1:23 PM
    Reply # 3805600 on 941390

    i have drilled a lot of holes through the hull into the void  for ventillation, so i can inject the epoxy into the holes with a syringe.  it will have ten minutes or so to penetrate before it gels.  

    my experience with foam is that it is difficult to get it to flow into thin crevices.

  • February 08, 2016 8:02 AM
    Reply # 3806870 on 941390
    Deleted user

    A note, that can work to draw much more into cracks  (vacuum can draw epoxy into areas that would not be meet by injection alone)

    Please do not use vacuum with foam; as it will give a very very low density i.e weak result.

    Also heating the (surface or area) before will allow the epoxy to penetrate better. (if possible)

    Lastly; they do make low viscosity epoxy which will flow much much better than west system. like http://www.uscomposites.com/epoxy.html#epoxhard

Powered by Wild Apricot Membership Software