Editor’s note: The following is Part 3 of a four-part series on Peter Rooke’s restoration of a 1-1/2 HP Bates & Edmonds Bull Dog.
Replacement flywheels had been cast, which needed cleaning, boring to the crankshaft diameter of 1.375 inches before cutting slots for the gib keys.
The flywheels were mounted on the milling table and it was just possible to bore them by moving the mounting arm for the vertical mill farther out beyond its normal working position. The wheels were levelled using various parallels under the rims and clamped to the table before facing one side of the hub. The center point of the flywheel was measured and centered under the boring tool and the center hole bored.
When both flywheels had been bored, they were in turn mounted on a mandrel that was fitted to the horizontal output of the mill so the flywheel could be turned, to face the other side of the boss, and turn the outer surface of the rim and its sides. This was a time-consuming exercise as only a slow speed could be used with a fine cut in view of the jury-rigged mounting of the wheels. The skin of the cast iron was so hard that this turning exercise used up several replacement carbide tips.
The next step was to cut the keyways. Having been quoted $100 for a used broach, which I would then have to set up in a press I did not have, I decided to make a slot cutting tool, and spent a couple of days cleaning up some scrap metal and machining to make a small toggle press. This filled a gap in my equipment and would be used on other occasions.
The flywheel was mounted on the side face of the mill, after removing some bolts that secured the ring around the horizontal output, to provide mounting points. Each flywheel was then set up, ensuring that the bore was square to the milling table. This enabled the press to be securely mounted on the milling table, and the adjustments it offered could be used to control the depth and position of the cut.
Measurements were taken to identify the center of the hole in the flywheel, and then the center point of the high speed tool to be used to cut the slot was adjusted to this position using the table movement screws.
As a general rule, keyways are normally tapered by 0.125 inch per foot of shaft, meaning 0.031 for the 3 inches of the flywheel hub. The slotting tool was offset from square by this amount to create the taper with measurements being taken using a dial gauge.
The cutting tool was moved to touch the bore, the adjustment dials zeroed and the lengthy, tiring process commenced, cuts being no more than 0.002 to 0.003 inch per stroke.
The fuel tank was in the base of the casting and the sealing plate to the hand hole has long disappeared. The first step was to wash out the tank and use a small wire brush to remove any loose material. The threads for the fuel pipe and drain were re-cut, and all traces of dirt and dust removed, and the area around the hand hole cleaned down to fresh metal.
A new cover plate was made from 0.125-inch-thick steel and both it and the cast iron around the hole tinned with soft solder, applying a flux of powdered tin to the cast iron rather than the normal soft solder flux. It took me three attempts to arrive at a leak free joint, testing with water.
Once the fuel tank was sealed, and after flushing out the tank, new drain plugs were fitted and I turned a piece of scrap brass to fit as the tank filler cap.
The tube was still in place to connect the drip-feed oiler to the cylinder. While it looked to be in order I decided to remove it as it was necessary to drill a hole through the new cylinder liner for the oil to feed into the bore.
Little of the oil tube projected above the top of the water hopper, and I did not have a reverse thread large enough to fit the adapter at the end of the oil tube. A 0.375-inch National Pipe Thread was cut on a piece of hex steel to fit the end of the oil tube, which was coated in thread lock before being screwed in and left to set. A stronger alternative to thread lock would be to use some spots of weld. The next day, as soon as a wrench was used to turn the tube, it broke off above the cylinder, and I then found that it was riddled with rust.
Fortunately, I had a reverse thread that fit the internal diameter of the tube, and the remains were unscrewed quite easily using a socket spanner and extension piece.
In order to drill the oil hole, a 0.125-inch hole was drilled in the end of a piece of 0.313-inch steel, cross-drilling a small air hole. A 0.125-inch drill bit was then soft soldered into the end of the steel, taking care not to heat the tip of the drill bit and destroy its temper. This extended drill was then used to drill the hole through the cylinder liner, applying a very slow speed and light pressure to avoid splintering the inside of the liner.
Replacement of the oil tube was a simple matter of threading a piece of 0.500-inch internal diameter steel pipe and making a new adapter at the top to fit a drip oiler.
I had new castings for the muffler, which was a closed cylinder rather than the 3-part pancake style seen in the later Bull Dog catalogs. I am told this was also a correct style, having a photograph of a similar one on an engine that was described as all original. The drawback with this style of muffler is that the exhaust gasses exit around the perimeter of the muffler, close to the long spars of the skid. With the pancake style, the gasses exit through the center, and there is a thread provided for an elbow or pipe to be fitted to further deflect the gasses.
The first task was to clean up the casting and the mating surfaces on the lugs before drilling holes for the securing bolts. The main part was set up on the lathe and then a hole was bored to a diameter of 1.14 inches before screw cutting a 1-inch-by-11.5-inch thread for the pipe nipple. The pipe nipple was threaded for nearly 2.50 inches on one end so that it extended nearly to the bottom of the muffler, forcing gasses to bounce back over the baffle to exit.
To finish the muffler, it was given a couple of coats of black engine paint that is able to withstand high temperatures.
All that remained of the original igniter was part of the block and the holes through which it had been sealed with weld. I had a new casting for the main body and was able to take some measurements from a reproduction igniter.
The first step was to clean the outside face of the casting and rough-clean the inside. Measurements were taken of the distance between the two mounting studs on the cylinder head, and two holes were drilled and tapped with a 16 TPI thread in the new casting for 0.375-inch studs the same distance apart.
The center point of the igniter casting for turning the inside was measured next and marked with a punch. Then two studs of 0.375- inch steel with lock nuts were used to clamp the igniter to the lathe face plate, the length of the studs being adjusted to align the face of the casting square, and also center it. A center was used in the tailstock to help steady the casting when turning it.
Light cuts were then made to clean up the inside of the casting, and produce the raised center to locate in the opening in the cylinder head. Once this stage had been completed, the stud holes were drilled out to 0.438 inch, to fit the studs on the cylinder head.
The next step was to mark out the position of the igniter pivot pin and fixed terminal. The casting was mounted on the drill table to drill the first hole 0.375 inch through the support arm and block, then I drilled the 0.4375-inch hole for the mica tube around the fixed contact.
The fixed contact and pivot were then fabricated, using the drill rod for the main part of the pivot.
The trip arm was milled and then filed from pieces of mild steel to match the photographs I had taken. The other items were simple turning jobs apart from the spring, with a number of efforts being rejected before I was happy with the one I had made from 0.063-inch piano wire. A series of holes were drilled in the casting for the spring anchor point so it could be adjusted.
The fitting of the various components took a while, but eventually I managed to adjust the length of the igniter contacts (made from soft iron) and the spring position/tension until I was happy with the gap between the contacts and the “snap.” Once satisfied with the working of the igniter, the final step was to drill a hole for the fixed post to limit the opening movement of the moveable contact giving a gap between the contacts of 0.035 inch to 0.070 inch.
Unlike most engines, the igniter on the Bull Dog is tripped on the return stroke, the trip having a hook that engages with the igniter lip and pulls it. Most of the photographs I had acquired never showed a clear shot of the whole fitting, but I was able to make a passable copy although some dimensions were estimated.
The main body was no more than a clamp to the pushrod to support for the trip hook with adjustment. In some photographs this appeared to be made out of brass or bronze, rather than steel, so I decided to use brass, which would be easier to work.
After locating a block of brass the first step was to drill a 0.375-inch hole drilled lengthways through the block for the push rod. Next, two cross holes were drilled for the clamping bolt and for the hook pivot screw. Both of these holes would be threaded later, but drilling them at the outset provided reference points for shaping the block.
The shape and profile of the part was then marked and it was roughly formed by milling off the majority of scrap material before slitting the bottom half lengthways for clamping and then tapping the threads. Next was the time-consuming part: Using new, sharp files the block was shaped to mirror the photographs I had obtained. (I have three sets of files: old/scrap for first cutting cast iron, less old for steel, and new for brass and other soft materials.)
The trip hook was an easy task, milling the hook and flat section from some mild steel. For the securing screw to the body of the block, a piece of 0.50-inch round mild steel was drilled through to the diameter of the pivot pin, before being brazed to the hook. The slot for the adjusting bolt was not cut at this stage, but later after trial assembly when the relationship between the parts could be more accurately assessed.
Once the block was shaped, the smaller items of clamping bolt, pivot screw and adjustment bolt were made. The trip was then assembled and fitted to the pushrod in order to test the action of the trip against the trip finger. It took some time to make adjustments, both to the operation of the trip finger and position of the trip to get the spark at the right time and a strong snap from the igniter.
Gears, rock arm, pushrod
Before starting on the various castings for the rock arm and governor, new pins were made to replace all that were missing from the main casting. It was also necessary to drill out the locking bolts for two of them and re-tap the threads.
A replacement small gear wheel for the crank had been cast from an original with no build up and there was little surplus metal to remove from the teeth. The wheel was centered and the center bored to a diameter of 1.375 inches to match the crank, broaching the slot for the key.
Fortunately, the original main gear was present and after trial fitting both, I found that they meshed fairly well so all that was necessary was to clean off any imperfections with a grinding wheel and a file. The roller for the rock arm was missing, so a new one was made on the lathe using some 0.750-inch diameter drill rod, which was then hardened by heating and quenching in brine. To retain full hardness it was not tempered as it should not be subject to any shocks likely to break it.
The rock arm casting was cleaned up, the 0.6875-inch hole for the main pivot pin was drilled, and the mounting points for the governor catch and the spring plate to rest on the rocker roller were cleaned up. Holes were also drilled for the spring securing screw and the pivot bolt for the pushrod.
The latch block was made from drill rod and again was hardened after the grooved slots were machined and thread-tapped for the securing bolt. While the plate to rest on the roller is called a spring, there was little spring in the plate I inspected, which consisted of a piece of 0.125-inch-thick hard steel.
Always looking to recycle material, an old file was retrieved from the scrap drawer and heated bright red for a few minutes before being allowed to cool slowly to soften it. The serrations of the softened metal were then machined off to leave a smooth surface, which was trimmed to size, drilled and countersunk for the mounting rivets.
To get the relationship right between the latch block and the latch, it was necessary to bend the spring before hardening. The plate was then heated cherry red and quenched in brine to harden, before being polished and then left in a warm oven (put in while my wife was not looking) to temper. The following day the tempering was tested by a scientific method: laying the metal on a flat surface and hitting it with a large hammer to see if it would crack.
After cleaning the casting and drilling the hole for the pivot pin, the governor shoe was placed in its position with the rock arm and flywheel – complete with governor weight – to check clearances and operation. The casting was adjusted by grinding where necessary to get the correct fit before taking measurements for the length of the latch. Again, an old file was used for the latch, following the same procedure as for the rocker spring. The latch was made 0.125 inch oversize so that it could be ground to fit on final fitting.
To make the spring that keeps the governor shoe against the flywheel hub, a piece of metal strapping was ground to shape and the mounting hole drilled with a carbide drill.
The original pushrod came with the engine, completed with the pushrod ends. While the rod ends were usable, the rod had to be replaced using new 0.375- inch steel rod as it was so badly rusted the trip arm would not fit properly.
The cylinder head was bolted onto the main casting and the gear wheel fitted to the crank, so all elements could be assembled in order to check the fit and action of the rocker arm and related parts. First, the fit of the parts was adjusted, and it was necessary to fit spacing washers on some pins so that the pushrod moved smoothly and parallel to the bore. Secondly, the action of the upper rock arm was checked so the exhaust valve opened and then closed on the out stroke of the piston, and the pushrod started its back stroke to trip the igniter just before top dead center. This was achieved by changing the mesh point of the gears and adjusting the length of the pushrod.
The correct timing for this engine is for the ignition to snap at 15 degrees before inner dead center, and for the exhaust to open at approximately 30 degrees before outer dead center. Once happy with the settings, the mesh points of the two gear wheels were marked with a center punch.
The governor weight is fixed to the flywheel by a bolt into one of the spokes and pivots on this bolt, tension being applied by an adjustable spring. When the weight moves out it pushes the shoe toward the front of the engine, catching the lower rock arm and holding it.
Fitting was a case of drilling holes in the spoke and the weight, tapping a 0.438-inch-by-14-inch thread in the spoke and making a step bolt to fit. The end of some threaded rod was heated and flattened to form the spring adjuster.