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




The Alignment Problem

Needless to say the Berkeley Jet Pump was not intended to be connected to a Ford 7.3 Diesel.  But if they can make gay marriages work then this should be easy.  The distance from the bottom of the oil pan to the center of the drive shaft is about 12 1/2 inches and the Berkeley's shaft is about 8 1/4 inches from the bottom with about a 3 degree down angle on through the transom.   I am tempted to elaborate on the gay analogy but given the new FCC rules I better just let it lay. 

One option is to build a new oil pan with 4 1/4 inches removed from the sump portion of the pan and lower the engine.  (1) Another option is to build an extension for the jet pump that fits between the intake plate that mounts to the hull and the bottom of the pump and thereby raise the pump's shaft.


Lowering the Engine

Lowering the engine would reduce the oil sump area and cause the crank shaft to slap more oil as it rotates.  That could be remedied with a dry sump oil system but that would cost more than the engine.  I would also have to modify any replacement engine that also needed 13 inches.  As a plus it would lower the center of gravity by 5 inches.  However, I am going to rely on ambient pressure to keep water from rising up the exhaust and entering the engine through the exhaust manifold and that's 5 less inches between the water and the exhaust ports.  Also lowering the engine would place the starter very close to the hull and more susceptible to bilge water.




(1) The drive shaft on the jet
pump is about 4 1/2" lower than
the diesel engine so considered
casting an aluminum extension
using the "lost foam" process to
raise the pump's drive shaft
up lever with the engine.

(2) Hot wire used to cut the hole.

(3) Making a template for the
outside of the extension.

(4) Hot wire cutter is used to cut
away the foam between two

(5) The extension before flanges
are added.

(6) The extension as it would be
mounted between the intake and
the pump.

Raising the Pump

(1)-(6) Raising the Pump is a matter of casting about 15 pounds of aluminum and a big pour sounds like fun.  I can also incorporate a valve into the extension that will allow the hull to flood or be pumped dry. 

The other choice is to modify the cleanout port on the top of the pump, but locating the valve next to the hull will eliminate some of the complexity.   That being the facts as I see them it looks like the best solution is raising the pump.

The pump extension is approximately 10" wide x 23" long x 5" high with an estimated 450 sq inches of surface.  If it is 1/2 inch thick that would be 225 cubic inches.  The crucible a 6 inch diameter pipe. The area of the bottom is then area = pi * radius squared or 3.15 x 3^2 or about 28.5 square inches.  So 225 cubic inches / 28.5 cubic inches per inch equals 8 inches deep in the crucible.  Not a problem, because the crucible can hold about 12 inches.

Speed-up Gearbox

Sitting an looking at the extension it occurred to me that there is another option.  I could put a gearbox, or perhaps a v-drive box, or a belt drive between the engine and jet pump that would align the drive shafts. 

The down side is that any gearbox or belt arrangement will rob some of the horse power.  However in this case it also presents a solution for another problem.  The Berkeley Jet Pump is designed for 3000 rpm and up, and the 7.3 diesel runs at 2600 rpm at WOT (wide open throttle) and its' best tork is generated about 2000 rpm. So a gearbox can convert the diesels tork into more rpm for the pump. I had considered switching to a larger impeller, but even with the larger A2 impeller needs 3800 rpm in order to completely absorb the available hp.

(1) Berkeley's Impeller to
Engine Matching chart. The red
line marks were we want to be.

(2) Spine belt reduction drive
used on air boats.

(3) V-belt drive

(4) Overdrive

(5) Jeep Hyvo type chain drive
transfer case

(6) Sprint car "Open Tube"
rear end.

(7) Quick change gears in
a sprint car differential.

(1) Berkeley's impeller to engine matching chart shows the horsepower absorption characteristics for a Berkeley 12J size jet drive with various impeller sizes.  My 7.3 diesel will hopefully output 150+ hp. I may be too hopeful here but I am counting on the frequent between gas engine hp and diesel engine hp. Knowing the Berkeley used the inflated gas engine hp for their calculations I am using 200 hp for my engine. The 7.3 can run at 2600 rpm WOT, but is best operated around 2000 rpm, which is closer to the peek of it's tork curve which is around 1500 rpm.  Using the chart; for 200 hp and the "A" impeller, I need the pump to run at 4000 rpm.

The Berkeley web site says that my Berkeley 12JC Pump is ideally suited for vee and modified vee hull boats, measuring 14 to 24 feet in overall length, with a gross weight of not over 5,000 pounds per unit, including passengers and gear.   So at 5,500 pounds I am a little over weight; nothing new there.  They also recommend at least 1 hp for every 27 pounds.  For 5,500 pounds that's 203 hp.  That puts the my 7.3 diesel powered sub just below their minimum recommendation which is fine.  I want the sub to plane, but I am not interested in winning any races.

Convinced that a speed converter was worth the extra work I started looking at the various options.

Spline Belt Drive

(2) Spline belt drives are used on air boat to decrease the rpm but it would work equal well for increasing the rpm. It will require moving the engine about 8 inches from the pump but that is no further than what the pump extension would have done.  The converter will need to be a box, fixed to the hull and constructing of two wide timing pulleys.  Both pulleys will be mounted on short shafts with bearings to absorb the lateral force. The top pulley will be slightly larger and connect to the dual u-joint drive shaft that came with the pump.  The bottom pulley mesh with the 1 3/8 inch 10 spline shaft from the pump.  The pumps 1 1/8 inch stainless steel shaft is designed for applications up to 500 HP.

V-Belt Drive

(3) I might get by with casting and turning v-belt pulleys and using a set of 10 v-belts, however it will take up a lot of room, and will likely burn up belts quickly.  Also since v-belts rely on friction the belts must be tensioned.  If a tension pulley is used it would allow a quick way of disconnecting the engine from the jet pump, which is beneficial because the jet pump can not be operated while not immersed in water.  Another benefit is that the v-belt drive would be the failure point should an obstacle become lodged in the pumps impeller.  However because  v-belts rely on friction they are not an efficient method for transferring power, and I would need more than 11 inches between the centers.  Gates has a program named "Design Flex" that you can download from their site, that helps you select find the proper belt and pulley combinations.  It is very helpful when trying to understand how belts work without doing the math.


(4) I took a quick look at automotive overdrives too.  They would need some refitting to make them work, but possibly better than building from scratch.  A used unit should only a few hundred dollars, but the speed ratio is only about .71 meaning the 2000 rpm becomes 2816 rpm.  Also the input and output shafts are in-line so I would need to built the pump extension in order to raise the pump.

Transfer Case

(5) I was actually looking at chain drive options when I came across transfer cases.  They are ready made and sitting in the local junk yard. They have a 2 to 1 speed ratio; the shafts are often offset, everything seemed perfect except that Mr Young of Young's 4x4 shop swears they will blow apart if I try to run one backwards at 2000 rpm.  He is probably right, but it might be fun to try it. (6) However Mr Young suggested a gear arrangement such as the rear end of sprint cars and stock cars.

Stock Car, Quick Change, Rear End Gears

(6) (7) Using the actual rear end differential will not be possible but building my own box so I can use quick change gears is very appealing.  Pairs of gears can be purchased new for around $60 to $90 and ratios range from 1 to 0.60.  The ability to change the gear ratio without replacing the entire box will allow this or any other engine to be tuned to the jet pump.


(1) Gearbox with 2 sets  of
quick change spur gears in cast
aluminum box.

(2) Drive shaft offsets.

(3) Drive shaft details.

(4) Completed speed converter.

(5) Completed box mounted on
the jet pump and with the
engine drive shaft cover in place.

(6) Double U-joint drive shaft
that goes between engine and
the gear box.

(6) Using an angle grinder to
round the new mounting plate in
the lathe.

(7) Joint bolted to the fly wheel.

Building a Speed-up Gearbox

(1) I designed my own aluminum housing for the quick change gears.  Power from the engine would is input on the top shaft, the delivered to the 2nd shaft through the "quick change gears" and then on to the bottom shaft through a second set of quick change gears.  The 3rd shaft would also return the rotation to the proper direction for the pump.  (2) It's a lot more work than I'd really like to do, but It will be the most flexible solution and the 7 inches between centers of the input and output shaft is perfect to align the pump and engine drive shafts.

Spur gears are suppose to be the most economical gears in the power transmission industry and the most efficient way to transfer rotational force.  Spur gears with 20 degree pressure angles are considered best for load carrying capacity and the lower 14 1/2 degree angle is best for backlash and noise reduction which is not a concern for this drive gear.

So after looking at lots of spur gears I've decided to just use common sprint car quick change gears. They are relatively inexpensive at about $80 a pair and they are sold in standard, numbered sets, with each set having a slightly different ratio between the gears. Since I will use two set of gears the sets can be further combined to create several possible gear ratios.  Purchasing 3 sets, the #6, #18, and #39 allows for 2000 rpm from the diesel engine to be converted into output rpm ranging from 2600 to 4100 as shown in the tables below.

Quick Change Gear Sets
Gear Set Teeth Teeth Ratio
#6 23 25 0.92
#18 19 24 0.79
#39 18 29 0.62




RPM Output for  2000 RPM Input
#6 Reversed & #18 2600
#6 & #18 2700
#6 Reversed & #39 3000
#6 & #39 3500
#18 & #39 4100

Getting back to the original problem of the Jet Pump and Engine Alignment; the distance from the bottom of the oil pan to the center of the drive shaft is about 12 1/2 inches and the Berkeley's shaft is about 8 1/4 inches from the bottom with about a 3 degree down angle on through the transom. The gearbox provides exactly 7 inches of offset which will leave about 2 3/4 inches between the bottom of the oil pan and the bottom of the boat.  I could reduce the offset to less than 7 inches by offsetting the center shaft, but 2 3/4 inches will allow for a larger replacement engine should one be needed. 

(3) After a  trip to the local Motion Industries  ( store for some advice regarding bearings I had enough information to finish the gearbox and more importantly the drive shaft designs. Below is a list of parts for the bearings and seals used in the gearbox. The 10 Spline coupler came through Motion Industries and is manufactured by Hub City ( for farm implements.  The coupler is only 2 1/2 inches long so it required cutting 3/4 of an inch off of the Berkeley's drive shaft.

GearBox Parts List
Qty Part# Description
2 CR Seal # 534951 Oil Seal
2 6307 Open Ball Bearing 35mm ID
4 6305 Open Ball Bearing 25mm ID
1 Hub City 0332-00365 1 3/8" x 3" 10-Spline Coupler

The 3 custom drive shafts cost about $650 from Riverside Spline and Gear. Their web site is  It took about 4 weeks to get the shafts. Note, that is shafts not shaft.  If I wanted the shaft I would have taken the quote for $3600 that I got from via This may not be a bad source to check but finding the people who routinely cut splines was the key to getting a reasonable quote.

(4) The aluminum casting for the box came out better than I expected on the first attempt. You can see details for casting and milling the gear box on the "Lost Foam Casting" page under "Casting a Gearbox".

(5) The rear thrust bearing plate and seal were removed from the Jet Pump and replaced by the gearbox.  Four additional bolts were tapped into the Jet Pump to help secure the gear box.  Attaching the gearbox to the pump instead of the traditional approach of attaching it to the engine, allows for the engine to be replaced without the need to adapt the gearbox to a new engine.  (6) The short double U-joint drive shaft that normally goes between the engine and Jet Pump will instead go between the engine and the gearbox.

(6) The bolt pattern for the drive shaft was wrong for mounting to the engine so it was cut away and a new plate was welded on using a wire designed for welding cast iron. Without a torch at this time I rounded the new bolt plate by welling a temporary stud onto the back and then put it in the lathe and ground the corners off with an angle grinder as it spun in the lathe.

(7) A new bolt pattern was drilled to match the engine bolts that also hold the fly wheel in place, and the new bit bolted down.