Home-  Sailboat-  Submarines-  ROVs-  Metal Working-  Other Stuff -  About Us

Why Build a Sub
   Submarine 101
   Wet Sub
   Planing Wet Sub
   Dry Ambient Sub
   Submarine Yacht
Design Evolution
Cutting Aluminum
Building the Hull
Kort Nozzles
Battery & Chargers
Engine & Jet Drive
   Engine Tune-Up
   Hull Cooler
   Wet Manifold
   Engine Mounts
   Engine Box
   Jet Pump Valve
Helm Controls
Ballast Sled
Landing Gear
Trim Tanks
Wiring Harnes
Tow Truck
Rules of Thumb



Hull Cooler

So many things in life just look like simple problems.  I needed a replacement for the radiator. My "Inboard Motor Installations" book that mentioned coolers could be build into the hull. I have and aluminum hull that conducts heat like there is no tomorrow, so if build a thin channel like box into the hull and connect it into the water coolant circuit of the engine, and it's done.  Right!  I just need an idea about how many square inches of surface area need to be in contact with the water.  So I post my simple question to the forum, and now learn about BTU's, heat transfer coefficients, and laminar flow.  I just wanted to know if 220 square inches sounded about right, and now I am looking at mechanical engineering text and remembering why I chose to data processing as a career; NO MATH!

In very layman terms here is what I think I know:

A) The cross section of the channel needs to be no more restrictive than the hose.  Since the hose is 2 1/4" diameter that means 3.15 x (2.25/2)^2 using area = pi*R squared or about 4 square inches

B) According to "TSITL": Since only the wetted side of the hull and the channel are cooled the channel needs to be relatively flat in order to keep the hot water from rising to the top of the channel away from the hull and the cooling effect of the outside water.  And 1 1/2 is likely too much.  Perhaps 1/2 inch is better, which means in order to get the 4 square inches of cross section needed the channel would need to be no less than 8 inches wide.

C) "Corpus Skipper" has a 270 HP engine cooled with an external heat exchanger or "keel cooler" that is 22" long and 3 1/2" in diameter, which would make it about 225 square inches of cooled surface area.  That means my 1/2" tall by x 8" wide channel would need to be 225/8 or about 29 inches long.

D) I need to keep the area of the heat exchanger in the water.  Not a problem, it's on the ass end of a 5,500 pound boat.  And the water needs to flow even when the boat is stationary.  Also not a problem since the heat exchanger area will be next to the water intake for the jet pump, so if the engine is running, the water is moving.

E) "CaptPPan" asked "...what do you plan to do with the exhaust?"  Hmmm, good point!  I plan on building water cooled aluminum exhaust manifolds and those will be dumping a lot of heat into the cooling system too. From "Gonzo": "A too large heat exchanger won't hurt."   Sounds like good advice.  "Suede" says that the exhaust loss is 28% and cooling loss is 57% so if I doubled the square inches of cooled area it would more than compensate for cooling all of the exhaust loss, even if that were possible.  So now it's 450 sq inches and a channel 1/2" tall x 10" wide by 45" long will do the trick.   Raw water will be introduced into the exhaust but that in only the help keep the hull temperature down. 

F) The bilge area will get hot.  But that's ok because its a long way from the cabin on the bow. 

G) From "FlybabyJim": "Next you need to think about the water pump and is it suited to the task. To much velocity and the coolant will make the loop to fast to cool it. Also you may need to design in some baffles to slow done the water as well as to mix it."  Hmmm, the pump is stock so it should do fine, and the 1/2 height should help with the mixing, but adding baffles could be done as long as it did not restrict the width to less than 8 inches.  Might be good for stiffing the channel too so it would not be damaged when stood on.

H)  "FlybabyJim" also eludes to gear box cooling, and yes there is a gear box to get the diesel rpm up to where the jet pump need them, but that will get cooled by a raw water set up if it is required.

I) "Danielsan" points out the need for a reservoir and radiator cap which will accommodate expansion and air collection.

J) "FastFred" "On commercial boats I have seen the hull cooling surface was a smooth built with internal web flanges for bolts. The top plate bolted on inside and had a welded passage or maze that ran the water back & forth over the exterior surface." --Sounds perfect.

(1) Hull Coolers with simple
baffles and hose fittings.

(2) Hull coolers and oil cooler
set in place.

(1) And then you get to the part where you can actually build the thing and you get to try to implement your good intensions.  First the height was raised to 5/8 of an inch so that the 2 1/4 in water coolant hose had a little more room where it connected onto the cooler.  Yes I could have build in a raised area, but I don't think it is necessary. So the final design is 450 sq inches and a channel 5/8" tall x 10" wide by 45" long with a couple of baffles to help the flow stay mixed up.

(2) The real thing is also built is two pieces designed to fit between the engine compartment wall and the longitudinal frames for the engine bed and then turn 90 degrees and enter the forward end of the engine compartment.  The aft end extends just past the engine compartment so that a 2 1/4 inch coolant line can be used to connect the two sides together.

The coolant will flow forward from both exhaust manifolds and join at a tee before entering the front end of the right side heat exchanger.  It then flow aft through the cooler passing two simple baffles that help mix and spread the flow.  At the aft end the coolant flows into a 2 1/4 inch hose that crossed to the other side over the gear box. In the left side heat exchanger it flow forward before turning and re-entering the engine compartment where it is connects to the coolant return.

(2) An oil cooler was build on the same basis as the coolant heat exchanger.  It is separated from the hull by 3/8 inch and has two length wise partitions that will force the oil to flow in a long S pattern from one end to the other. The oil cooler is completely house in the engine compartment between the engine bed and the keel. 

(3) Kay pulls bilge duty and
clean up after the failed oil


(4) The return hose is still
connected to the oil cooler which
is bent outward and has a weld
split open along the engine
(5) Marine oil cooler with out the
water lines attached.

(3)(4) With the engine installed for a test the oil cooler failed in about 60 seconds. Once the oil psi came up a weld gave out and the engine compartment flooded with oil. After pulling the engine we could see that the oil cooler was seriously deformed by the oil pressure so it was clear that an oil pressure regulator was needed.  The original oil pressure regulator was build into the oil cooler and that was removed in order to make from from the wet manifolds.  Read more about the regulator here: Oil Pressure Regulator

So we welded the oil cooler back together and pressure tested it which exposed some pin hole leaks. Have successfully used Seal-All to fix pin holes in the exhaust manifolds it seemed like a good idea here too.  After a two treatments the cooler was pressure tested to 90 psi. and then the engine installed for a second test.  This time the engine ran for about 3 minutes with 50 psi on the oil gauge before the pressure dropped off to 46 psi and oil cooler again blew apart. A port mortem showed that Seal-All that had not firmly adhered to the inside of the oil cooler, flow to the oil filter and clogged the inlet holes causing the oil cooler to overload.  There is likely more sealer that will pull again from the cooler so we are going to scrap that cooler and add a off-the-shelf oil cooler.

(5) Our new, $130 marine oil cooler arrived is now installed up against the oil pan, just a beneath where the stock oil cooler was. The marine oil cooler will be feed raw water from the jet pump and the discharged water will be delivered to the exhaust.