A Bull Dog gets its bark back
Editor’s note: The following is Part 2 of a four-part series on Peter Rooke’s restoration of a 1-1/2 HP Bates & Edmonds Bull Dog.
Pouring the bearings
Before starting to pour the bearings, all bearing areas on the main casting were brushed and cleaned with emery cloth to remove all traces of rust, in order to improve adhesion. In particular, the recesses were cleaned out to keep the bearings in position. Fortunately, I possessed some formers from another bearing repair that would fit this job. They consisted of a central core (some 0.100 inch smaller than the crankshaft diameter) and two recessed end caps. These end caps had supports to keep the caps aligned in the bearing recess.
The formers were coated in oil to prevent the white metal from sticking to them then clamped in position. When white metal cools it forms a curved rather than level surface, which can drop below the required line at the edges. To prevent this, a small dam was built using babbitt putty, thus increasing the height of the white metal poured. It was also necessary to seal the gaps between the edge of the formers and the engine casting to prevent the poured metal from leaking out. (If you use fire clay rather than babbitt putty, it is essential to ensure it is fully dried when pre-heating the casting or else you will run the risk of the molten metal spitting back if it hits any moisture).
On a cautionary note, adequate safety precautions are always essential when heating white metal: safety glasses, trousers, long-sleeve shirts, preferably an apron and leather gloves. The white metal was heated in a pot, stirring from the bottom with a thin pine stick to keep the alloy fully mixed. With a low tin, white metal is at pouring temperature when the pine stick starts to singe after being held in the metal for 3 or 4 seconds. When the metal reached pouring temperature, the mold and casting were heated with a gas torch to preheat it and stop the white metal from cooling too quickly and cracking when it hits a cooler casting.
Once the metal was cool, the formers were removed and the bearing inspected to ensure it was clean and unbroken. If there are any imperfections in a poured bearing, it should be scrapped and the process started again.
The bearings in the caps were also cast using the same process although it was necessary to fill the threaded portion of the grease cup hole with babbitt putty.
Boring the bearings
Once the main bearings were cast, surplus metal was removed from the mating surfaces of the main casting and the bearing caps, so the white metal was flush with these bearing surfaces. Shims were then made, an equal number of different thicknesses of stock for each side of the cap, to a total thickness of some 0.050 inch.
The shims were then fitted and the bearing caps bolted to the casting so the bearings could be bored. The engine casting would just fit on the table of my milling machine so it was set up with the cylinder bore parallel to the table. When it was set up with a square, it became evident that the newly sleeved bore was well out of line with the rest of the engine. The crankshaft would have to be out of square with the main casting if the piston was going to slide down the bore, unless there was to be a lot of slack and tolerance in all bearings. This meant a bit of measurement and adjustment to ensure there would be an adequate thickness of bearing metal across the bearings and the engine casting would not be cut into when boring out.
Using a boring bar, the bearings were bored to diameter and faced off square. A form tool was used to round off the inside shoulders of the bearings. Then the final fitting of the crankshaft was achieved using engineer’s blue and a scraper.
The big end bearing in the connecting rod was cast in one piece. The big end cap was bolted to the connecting rod, with two temporary brass shims 0.062 inch wide. This was then set on a fire brick, with an oiled former plate underneath and oiled central core. Again, a dam was made so the white metal would extend beyond the cast iron of the connecting rod.
After pouring the white metal, the bolts and temporary shims were removed when it had cooled. The bearing was split using a hacksaw, before cleaning the mating surface with a file. Again shims were made using different thicknesses of stock, before the bearing was reassembled and then mounted on the milling table to bore out the bearing. When setting up the bearing, a piece of drill rod was clamped in the little end and was checked to be square to all axes using a run-out gauge.
When all the bearings had been bored to size, final fitting with the crankshaft started, and was achieved using a scraper and engineer’s blue, in particular the curved shoulders of the bearings. The crankshaft journals were covered in blue and the bearings put together which marked high spots for removal using a scraper.
Once the bearings were a good fit on the crankshaft the grease grooves were cut in the white metal. The shape was first outlined using a marker pen before they were cut with a small rotary file in the Dremel.
Time for the piston
I already noticed that there was a small slit in the bottom of the piston skirt, where it looked as though someone had tried to cut it with a hacksaw. The piston ring grooves had already been trued up, to nearly 0.030 inch oversize, although the width varied by up to 0.003 inch. After contacting Dave Reed at Otto Gas Engine Works, he suggested using standard 0.250-inch rings and spring steel spacers (0.030 inch) rather than having custom rings made.
At this point I examined the bore and the fit of the piston. With no rings, the piston was a good fit in the bore, and when I measured the bore I found it was 0.050 inch under size. I then realized that the bore had been sleeved, and the liner had been left unfinished and had not been bored and honed. I have already mentioned that I found the bore was not square to the crankshaft when boring the bearings, and it had not been faced off at the cylinder head, leaving a step evident on one side.
The slit in the skirt of the piston was repaired by filling it with a fillet of braze, taking care not to overheat the cast iron and gradually reducing the heat when finished. The next step was setting the piston up on the lathe to cut the piston ring grooves to a constant width, 0.2815 inch, including 0.0015 clearance. To accurately measure the width of the piston ring groove, a piece of square High Speed Steel was used together with feeler gauges. The width of the HSS was constant and known after measurement with a micrometer, so adding the size of feeler gauges used identified the groove width.
At the same time the piston was also skimmed, as when setting up with a dial gauge, it appeared oval in shape. The piston diameter was reduced by a further 0.015 inch above the top ring and 0.010 inch above the second ring to allow for the greater heat and expansion at the top of the piston.
Removing the cylinder head
Once happy with the piston, it was measured and the bore was then honed 0.005 inch bigger than the diameter of the piston. While sorting out these problems I also noticed that no hole had been drilled to allow oil from the oiler into the bore and this would need to be fixed.
I made an attempt to remove the cylinder head studs after soaking them in penetrant, using two counter-tightened nuts, but despite applying a lot of force, they showed no sign of moving. The cylinder head studs were in acceptable condition, so no further attempt was made to remove them in case the main casting was damaged through using excessive force. In any event, the safest way to remove them would be to saw them off, and then drill out and re-cut the threads before making and fitting new studs.
The presence of the studs made the removal of the step at the head of the cylinder more difficult. The excess liner was first filed away as far as possible without touching the main casting. Once nearly level with the casting a piece of plate glass was covered with engineer’s blue and pressed against the cylinder to identify the high spots that were then scraped and the process repeated until the top of the cylinder was flat. In hindsight, it probably would have been easier to have drilled out the studs, had the cylinder machined flat, and made and fitted new studs.
Inside the cylinder head
The first step was to remove the remains of the valves, one of which did not look original. The inlet valve was not only bent, it was stuck solid, so it was sawn off near the top of the casting before popping with a center punch and drilling the stem out of the valve guide. The remains of the mixer were removed as well as the exhaust rocker. The studs for the igniter were already missing so replacements would have to be made.
The first challenge was to repair the hole that had been cut in the wall of the inlet chamber for a spark plug. This hole was drilled fractionally larger to uncover clean metal and it was then tapped with a taper pipe thread and a plug made to match out of cast iron. The plug was made longer than necessary so it could be screwed in tight then trimmed to size after it had been fitted. Both items were then degreased and covered with brazing flux before the plug was screwed into the hole.
When heating cast iron, particularly old cast iron, it is important that heat is gradually applied and reduced to allow the cast iron to normalize and not become brittle. Particular care had to be taken with this cylinder head, which was irreplaceable.
The whole cylinder head was then put in the oven for two hours to warm up before being moved to the brazing hearth where the plug and surrounding area was heated to sweat braze into the gaps in the thread to lock it firmly in position. As soon as this was completed the head was returned to the hot oven, the temperature of which was gradually reduced over more than an hour. Once the head had cooled, the plug was ground down on both sides to provide a smooth surface.
There was bad rusting on the mating surfaces between the head and the cylinder, so I decided to mount it on the lathe and skim it. To facilitate mounting on the lathe the center point of the cylinder head was found using dog leg calipers and center punched. The only mounting points that could be used were the holes for the valve stems, so two holes were drilled and 0.311-inch threads tapped in a piece of 0.500-inch steel plate. Securing bolts to match were then inserted through the valve guides into this plate, which was then clamped on to the lathe face plate.
To support the cylinder head, four bolts were mounted in the face plate to line up with the holes for the cylinder mounting studs. Nuts were screwed onto these studs to rest under the cylinder head, which enabled adjustment to be made to keep the surface of the cylinder head square. It took some time to line up the cylinder head so it was both centered and the mating surface was square, readout being checked and checked again using a dial gauge.
Given the length of the head, its mounting on the face plate was not ideal, even with the use of a tailstock center to support the head. As a result, any cuts would have to be no more than 0.002 inch at very slow turning and feed speeds. Eventually it was necessary to remove 0.020 inch to get a surface with no large rust channels. To maintain the compression ratio the raised center section of the head was also skimmed by a similar amount.
Once completed, the fit against the cylinder was checked again using engineer’s blue. A number of high spots on the end of the cylinder were scraped until the blue showed a good even fit.
The hole in the cylinder head for the mixer was not perfectly round so it was set up on the milling table with the port vertical and centered under the high- speed borer, which was then used to remove 0.020 inch and true up the hole. This was only possible as a new mixer would be fabricated that could be custom made to fit this bigger diameter. If it was necessary to match an existing mixer, then an alternative course would be to bore out the port to a larger diameter and make a sleeve with an external 0.002-inch taper which could be pressed into the port.
Valves and valve seats
The original valves appeared to be late replacements, one of which was bent and rusted in place. A bit of pressure after applying release agent removed the exhaust valve. As mentioned earlier, the bent stem of the inlet valve was sawn off then the center of the remainder was drilled out.
The guides for the valves were badly worn and rust pitted. Rather than drill out to fit sleeves, the holes were drilled out 0.015 inches oversize and then reamed 0.328. This gave clear unmarked guides, so making new valves with oversize stems was the simpler solution rather than fitting sleeves.
To make the new valves some drill rod was turned to 0.327 inch taking a fine slow cut to finish with a smooth surface. A 1.375-inch diameter steel disc was then brazed on to one end of each rod, ready to be turned to make the valve heads. The original taper on the valve heads appeared to be 45 degrees so the valve heads were finished with this, profiling the area back to the stem.
Once the valves had been completed, the seats could be renewed. As I do have access to a valve seat cutting tool, a half round file was used to remove the rust pitting in the seat. Each valve was then marked inlet or exhaust, before it was used as a pattern to renew the seat. Engineer’s blue was applied to the valve, which was inserted in the head and twisted to mark the high spots, which were then removed with a scraper and the process repeated until there was a good even contact. Grinding paste was then applied, and using a stick with a suction pad on the end, the valve was turned by rolling the stick between the hands, lifting and reseating the valve to redistribute the paste. This was repeated until there was a clear band of contact around the valve and seat. The 0.125-inch holes were drilled in the valve stems for the spring retention caps, which were made based on images in the Bull Dog catalog of parts.
The remaining task was to restore the rocker arm, the pivot being rusted, and the head that presses the exhaust valve was also in a poor condition. A new pivot pin was cut from some 0.500-inch steel and 0.125-inch holes drilled through at each end for the pivot arm and rocker head securing pins. The bearing surface of the rocker head was ground flat and the head was fixed to the pivot pin with a steel pin, which was peened over to lock it in position. A casting was used to make the new upper rock arm, cleaning the mating surfaces and drilling holes for the shaft and pushrod end securing bolt. To get the upper rocker arm in alignment with the lower rocker arm, and the igniter trip hook to engage properly, it was later necessary to make a thick spacing washer.
In Part 3 Peter tackles the flywheels, governor, igniter and much more.