Gas Engine Magazine

John Smyth 4-1/2 hp Restoration – Part 3 of 5

By Staff

Early examination of the cylinder bore revealed severe rust pitting, and the only way to rectify it was to install a sleeve. It could have been left as was to see if there was any compression, but as the piston required machining of the worn ring grooves it made sense to sort out the bore first, if possible leaving it a few thousandths of an inch undersize so that the rust marks and unevenness of the piston could be removed.

Sleeving the Smyth first involved setting up the engine block for machining, no easy task in view of its shape and size. The old cylinder would need to be bored out, removing enough metal to insert a sleeve and leaving a shoulder at the crankshaft end as a stop for the sleeve. Once the new sleeve had been fitted it would also need boring to size, being supplied undersized. Finally, it would have to be honed ready for the piston.

The first step was to locate a suitable cast iron sleeve, and a search was made on the Internet for a supplier here in the U.K. A sleeve 4.75 inches in diameter with a 4.5-inch bore and 14 inches long would be the ideal fit, leaving 1 inch of the original bore at the crankshaft end.

Nothing approaching this size was found in the online catalogues; however, the nearest being only 10.5 inches long. That length could be used, buying two and cutting one down, but that would leave a “join” that the piston rings would have to cross. Not ideal. Enquiries were made about buying a custom-made sleeve, but the longest that could be made was just 13 inches and the price, with delivery, was quoted at nearly $450.

Enquiries started further afield and revealed a standard Melling sleeve the right outside and inside dimensions, but only 12.25 inches long. However, this was at a far more realistic price, approximately $60. A 12.25-inch long sleeve meant the piston rings could operate within the length of the sleeve. This meant fitting a 1.4-inch collar at the cylinder head end, but it would be beyond top-dead-center and therefore clear of the piston. A sleeve was ordered and shipped to our son in Texas, and with the help of a returning friend the new sleeve found its way to the U.K.

There was no way I thought I could machine an engine of this size with my modest workshop equipment, so the next issue was to find someone who could bore out the block and install the new sleeve. Enquiries revealed a suitable workshop, well recommended and not too far away, so the engine was craned into the back of my SUV and taken there for examination. This shop was quite busy, and the foreman said he could do it in three months — but the ballpark cost was around $650. Another machine shop a friend knew was tried, and while they had the machinery they could only think of problems with the project, so I decided to re-think my options.

Portable borer

Searching the Internet, I had noticed commercial boring equipment manufactured to repair large items in the field. There was also a thought at the back of my mind that this sort of portable tool was used in years gone by. This prompted further research, and the idea began to form to make a boring tool.

Thoughts were sketched out and then discussed with an engineer friend, Alan, who is a “proper” engineer, highly skilled and adept at using basic equipment to produce first-rate jobs. The basic concept was to fit cones at each end of the bore, with a bearing in each and a boring bar holding the cutter to run between these bearings. This would make setting up easy and facilitate accuracy, crucially important with the bore having to be machined to a size 0.002 inch smaller than the sleeve to give an interference fit.

After several days of thought a design emerged. A steel disk would be turned to be a tight fit in the crankshaft end of the bore. A steel plate would be bolted to the cylinder head end using the existing studs. This plate would hold supports for two parallel rods, on which would slide the second bearing bracket and the fixture to hold the power source and gears for running the borer. This steel plate would be aligned using another steel disk, turned to be a precise fit in the cylinder. Once the steel plate was in place it would not have to be moved until all operations had been completed, ensuring accuracy could be maintained. Precision ground steel bar would be used for the boring bar and the two guide rails to make the device as accurate as possible, so an order was placed for 30mm and 20mm bar.


Cutting cast iron on a 4.5-inch radius meant the cutter would have to be turning at around 150rpm. An old fixed-speed drill operating at 1,000rpm would be used as the power source, so this had to be geared down. Initially, I though about hand-feeding the cutter into the cylinder, but that would be painstaking work so further gearing was necessary to give a feed of around 0.002 inch per revolution.

The lathe has a set of change gears for screw cutting, and after undertaking some calculations it was found there were nearly enough gears in the set to make up the combinations required to give the required speed and feed rate. A used gear found on eBay completed the gearbox. The speed of the drill was reduced to 150rpm to drive the cutter. The post for the intermediate gear was fitted in a slot so that it could be moved to allow different gear sizes to give a range of final drive speeds from nearly 100rpm up to 200rpm. The feed screw was driven from the final drive, gears being assembled to give a step down of 20:1 to a 20mm by 1mm pitch threaded rod. In the first gearbox some eight gears were used.


Trying to work to an accuracy of thousandths of an inch on the tired lathe and mill took some time, some measurements being taken several times with any cutting being of small amounts well before getting to the required measurement.

The plate to fix on the cylinder head was made first, first drilling the four holes for the cylinder head studs and two 0.5 inch holes to mark the position of the guide rails. A digital readout was used to ensure complete accuracy, with the holes for the mounting studs drilled 0.125 inch oversize so the plate could be moved to accurately align it to the bore. Two lengths of 1.5-inch-diameter steel were cut 2 inches long then faced true on the lathe before being drilled with 0.5-inch holes. These holes would later be bored out to hold the two parallel guide rods. A length of 0.5-inch steel rod was then used to hold these sleeves of steel precisely in place while clamped and then welded to the plate.

Once the guide rail supports were welded to the plate it was again mounted on the milling table and the digital readout zeroed to it. The supports were first drilled undersize, then for accuracy bored out so the 20mm guide rods would be a precise slide fit. To aid accurate measurement when boring out, a test piece was machined to the diameter of the guide rods, with the first 0.25 inch being 0.02 inch smaller. This would be was used to check progress.

The center of the plate then had to be cut out to allow the boring tool to pass through. Again this would be accurately set to the center of the guide rods and a hole precisely 4.751 inches in diameter cut. This was 0.002 inch smaller than the sleeve and the diameter the cylinder would be bored to. This would act as a check when machining to ensure the hole was not bored oversize. To cut the center hole, a series of holes were drilled on 4.25 inch diameter to give an undersized hole. The resulting hole was then enlarged to 4.751 inches using the boring head.

A 1 inch length of 5 inch steel had already been turned to this diameter, to be used as a check piece to check the bore had been machined to the required size along its full length. Furthermore, this “gauge” had a step machined for the first 0.25 inch to a reduced diameter of 4.502 inches so that it would fit the present bore of the cylinder and could be used as a setting gauge to align the cylinder head plate. Additionally, this piece could be used to fit against the sleeve if it needed to be pressed into place.

A small piece of plate steel with a round steel plug was welded to the bottom of the main plate to act as the support piece for the threaded rod acting as the feed. This was drilled so the threaded rod would slide in it then a cross hole drilled and threaded for a clamping screw. The clamping screw was slackened to stop the feed and allow the cutter to be withdrawn from the bore. Finally, holes were drilled for bolts to secure tie bars to hold the crankshaft end bearing in place.

Using the same concept, the bearing support was made. Pieces of steel rod were welded to a 0.5-inch steel plate before being drilled then bored out to provide the support at the cylinder head end. This support had leaded bronze in the center as the bearing, carefully bored to size.

The gearbox would be made by fitting two pieces of 0.25-inch steel plate to a pair of tubes that would be a sliding fit on the guide rails. It was necessary to calculate the distance between the centers of the gears that would be used so holes could be drilled in the steel plate to mount custom made pivot pins. The electric drill would be connected to the gearbox by clamping at its front, the chuck being removed and with the first gear being on a shaft screwed directly into the drill.

Some 20mm bar was then threaded with a 1mm pitch and a similar thread cut in the center of the final drive gear to provide the forward feed. Turning this gear wheel moved it down the threaded rod when its end was clamped at the cylinder head plate. This pulled the gear assembly and boring bar into the cylinder.

The boring head was made to accept a radial tipped tool. A carbide tipped tool was selected as it would be easier to replace if worn or broken, with the cutting edge remaining exactly in the same angle position.

To make the boring head setup as rigid as possible, tool overhang was kept to a minimum by using 3-inch-diameter steel for the body. To cut the square hole for the cutting tool, I cut two 1-inch long pieces of this 3-inch steel, with one piece milled to form a slot for the cutting tool before joining the two pieces together using four cap head socket screws, with a locking screw also fitted for the cutting tool. The position of the slot was carefully measured to ensure that the edge of the cutting tool was aligned with the iddle of the boring bars axis.

The block was then centered on the lathe so the hole for the bar itself could be cut and the tool was finished off by drilling and tapping two more cap head screws to lock the head on the boring bar. Finally, a 1-inch thick piece of steel was turned to size to be a tight fit at the crankshaft end of the cylinder. It was then center drilled to receive a press fitted piece of leaded bronze as a bearing. The bronze bearing was bored out 0.002 inches larger than the boring bar. A crossbar was fitted to hold this in place, with tie rods passing down the side of the cylinder into holes drilled in the plate fitted at the cylinder head end. A small half-moon was cut away at the bottom of the bearing plate to allow access to clean out swarf during machining.

It was originally planned to thread the end of the boring bar so that it directly connected to the final drive of the gearbox. However, to provide some flexibility and allow clearance if there were small errors in measurements an alternative was used. A 0.5-inch-diameter hole was drilled through the bar so that a steel peg could be inserted, held in place with a lock screw. As a simple dog clutch, a sleeve was made to fit over the boring bar, threaded at one end to fit the gear box and two slots cut to fit around the peg.

Starting to bore the cylinder

Unsurprisingly, there was a small amount of bend at the end of the guide rails when the gear box and drill were in position, so a roller used for support when planning wood was placed under it to take some of the weight.

To set the cutter accurately, I welded together a few scraps of steel then machined this to fit around the boring head. A 90-degree V was cut, with a short bolt on the opposite side so when tightened the round head of the boring bar was pushed into the V. An old spindle from a micrometer was then fitted and zeroed so the amount of cutter protrusion could be measured. It was not pretty, but it worked.

The cutter was initially set to the bore diameter of 4.5 inches. The boring bar was moved down the cylinder, turning it by hand, to check for free movement. It was then set to take a skim cut of 0.005 inch to check everything worked satisfactorily. Everything was connected up, the drive was engaged and the motor started.

Unfortunately, the old fixed speed drill I chose as a motor decided to give up the ghost some 20 minutes into the cut. It was, to be fair, more than 30 years old. Looking for alternatives available in the workshop, I selected a 1hp motor I had previously ignored in view of its weight. This also meant changes to the mounting would have to be made, and as it spun at a faster 1,400rpm more gears would be needed to slow it down.

Additional tube was welded to a base plate to fit the new motor, allowing it to slide on the guide rails and follow the gearbox being held in position by a G clamp. This enabled the motor to be easily moved out of the way to enable the boring cutter to be fully pulled out of the cylinder.

It took around an hour for each cut, with the cutting depth being 0.010 inch per pass as the whole arrangement was a little Rube Goldberg in construction and care was necessary to ensure it was not over stretched. For the majority of the time the rig was left running while other jobs in the workshop were completed.

Once the required diameter of 4.751 inches was met the depth of cut was reduced to 0.005 inch and careful measurements were taken, being double and triple checked. The final cut was only 0.002 inch and then the dimensions verified by sliding the specially made gauge, (0.002 inch less than the diameter of the sleeve), down the bore. Finally, a knife tool was put in the boring head and set 0.010 inch smaller than the bore and used to clean up the step in the bore at the crankshaft end so that the sleeve butted fully against it. After cleaning out the bore all was ready for the next stage, fitting the sleeve.

Contact Peter Rooke at Hardigate House, Hardigate Rd., Cropwell Butler, Nottingham, NG12 3AH, England –

  • Published on May 10, 2016
© Copyright 2022. All Rights Reserved - Ogden Publications, Inc.