The Right Angle Drives (RAD's) in the Queen of
Chilliwack pose the biggest issue in a fully R/C model of the vessel. This is an
area where I did not want to skimp out on. Having a model ship with this type of
drive system is very rare, almost to the point where I wonder if I am the only
hobby modeller in the world to build them to this level of operation, in this
type of model.
The real thing:
The Queen of Chilliwack has 4 RADs, they are placed in the 4 quadrants of the hull, two foreword and two aft, positioned one on either side of the keel. The two forward RADs operate in tandem with each other and the two Aft RADs operate in tandem with each other as well, but the forward and aft RADs are independent pairs from on another.
In the real ship the shaft from the engine runs horizontally along the inside of the hull, but unlike most ships where the shaft is down below the floor or quite close to it, the shaft in the Queen of Chilliwack is about 4' off the floor spinning freely and is very exposed. The internal component of the RAD is housed in an upright cylinder in the hull, the RAD Compartment; the shaft is supported by the orange bearing where it enters this compartment. The idea is that the mechanical components of the RAD sit inside the purple removable basin inside of the compartment. For maintenance the shaft to the motor is disconnected at the orange bearing and the entire basin is removed from the compartment.
Inside the tub the shaft enters the blue gearbox which powers the vertical shaft that runs down the green drive casing. The shaft continues down the casing and into the lowest part of it, where there is a gear set built into the casing. From this final gear set a third shaft runs through the red bearing / pressure seal and into the propeller.
Hydraulic lines for the CP Propeller enter the RAD casing at the grey stator pressure seal inside of the purple tub. The fluids are then pumped through the green casing, into the 3rd shaft through the pressure seal, and then into the Propeller Hub.
Rotating the RAD Casing is achieved by hydraulic pumps in the machine room adjacent to the RAD compartment. The fluid is pumped into the red hydraulics servo motors, which are geared to the RAD casing. The servo motors turn the entire RAD casing.
The entire green RAD casing can rotate inside the purple basin so the propeller can be oriented in any direction, allowing the thrust to be pointed in any direction as well. This system generates a level of manoeuvrability to the vessel that is only imagined about with twin shaft vessels, and only dreamed about in single shaft vessels. The 4 RAD's on the Queen of Chilliwack are capable of pushing her in any direction that is wanted. This allows her to come up alongside a pier and nudge herself up against it with less than 2m of clearance on either end of the vessel. Of course this isn't the easiest thing to do, and does require some assistance from deckhands as to vessel position queries, but is entirely possible.
Diagram shows forward RADs looking from Bow to the stern.
In my model I am using a very similar setup except that due to my engines being
electric motors and small I am skipping the first shaft and gearbox and
orienting my motors in a vertical fashion so they are parallel to the 2nd
vertical shaft. This eliminates the need to have a second set of gears, getting
them to line up, added noise, and it is much more compact to install the entire
drive system within the first 3" of my engine room compartment. I still have the
exact same setup as the real ship does in the drive casing part of the RAD.
I needed to design something that looked like the real RAD's, functioned just like the real ones, weren't too complex, didn't leak, and I had to be able to build them myself.
This started a lengthy struggle through a series of designs and possible solutions. The preliminary design was very complex and involved over 15 hand machined parts, 4 brass bevel gears, springs, and a large coffee can; and this was all for one RAD. Slowly I redesigned small parts of the system until I ended up with 4 hand machined brass parts, 2 brass bevel gears, and two rubber 0 rings. The brass parts and lower RAD assembly can be noted in the attached PDF.
Click Here for RAD PDF
Building the RADs was an interesting experience for me; I have never machined anything out of metal before on a lathe. My dad is a mechanics teacher, so we were able to use an old metal lathe at the high school to get the work done. The lathe is quite old, possibly more than 40 years old, and is big as well. It is in no way a hobby lathe. I had to learn quickly, and I managed to only screw up one piece. The parts were all machined out of round brass stock. We chose brass because it is easier to machine and can easily be soldered together. The heavy machining was done at the high school and then the parts brought back to the shop at home for clean up and to drill and tap the 7 screws in each RAD. The screws are #0 Stainless Hex Head Cap Screws, these things are so small that you need a special drill bit to drill out the hole to tap. I will NEVER do this again; I broke more taps and spent more time frustrated and very carefully tapping all these stupid holes, than I did laying up the hull in the first place. If I were to do it again I would have just soldered the cap on or designed it to screw on.
They were assembled using a plumbing torch, solder, some small files, and some patience to get it all lined up nicely and get the shafts turning smoothly. In the end I did get them running very smoothly, almost as nicely as a straight shaft setup. One other problem is I tried to solder the Brass gear to the shafts, this proved to be difficult as the gears were so small that the solder would wick into the teeth. I would then have to heat it up and use an old paintbrush to work the solder out of the teeth. Also it proved to be to weak of a connection and it kept on breaking under load. I ended up drilling a hole through the hub on the gear and the shaft with the drill bit used for the screw holes and putting a piece of 1/16 piano wire through with a little of epoxy to hold it in place. That seems to work much better.
For steering I thought it would be best to control the model just like the real thing. With the two Aft RADs in tandem but independent of the forward RADs, and with the two forward RADs in tandem but independent of the Aft RADs. So, the RADs are dived into two independent systems in the model, one for the Forward RADs and one for the Aft RADs, the two systems are simply mirrors of the other. I will only discuss the operation of the forward RADs as the rears ones are just mirrors of the forward ones.
I decided it would be best to use a single servo to control both RADs. Not only to save on cost, but to make sure the RADs always stayed aligned with each other. I went with Hitec high torque servos; I cut out the pots and soldered in 2.2k resistors to make them run just like continuous rotation servos. You also have to cut out the safety tabs on the output gear, the servos used had metal output gears, so I had to carefully use a Dremol tool to grind off the tab. I used more plastic timing pulleys and fibreglass reinforced belt to link all the parts up. Two 2" pulleys on the Brass RADs, a 1" pulley glued to a circular servo disk on the servo and a 2" pulley was used as an idler. I installed an MA3 continuous rotation encoder into the idler so that it would tell me the angle the RADs were pointing. The system now consisted of two RADs, a servo as a motor and an encoder that gave the position of the RAD. A friend and I are currently working on building a PIC microcontroller board to control the RADs.
The motors are quite powerful but after assembling the first RAD I found that I would get better results by gearing the motors down. I elected to use plastic timing pulleys and fibreglass reinforced belts to create a 3:1 gear ratio. Not only did this allow for better use of the motors, but it allowed for a better mounting position for the motors.
At the moment the entire system is running manually, in that the RADs aren't proportional to the transmitter, yet. Currently pushing the stick to one side tells the RAD to rotate either CW or CCW, the farther you push the stick the faster it turns, but there is no way to tell where the RAD is pointing, unless you can physically see it underwater. This will eventually be replaced by the Microchip system, but that could be a ways away.