LOCKDOWN PROJECT – PETROL ENGINE
Stuart beats lockdown by resurrecting a petrol engine idle for more than half a century.
In 1963 my father and I were members of a model boat club in Southport based on the marine lake. Although my father’s main interest was steam locomotives he had seen a Sea Queen model boat and decided to build one and power it with an Edgar Westbury water-cooled Kiwi 15cc overhead -valve petrol engine.
He constructed the engine but never finished the model boat and so the engine has lain idle in his and subsequently my workshop ever since. Although I cannot remember seeing the Kiwi running I presume it had, albeit for only a short time since it was not fitted with petrol tank, float chamber, oiler or any cooling system.
I was looking round for my next project having just completed a large steam-driven tug boat (EIM, October 2020) and so decided to refurbish and finish the Kiwi. If the project was successful, and for something different, I planned to fit it into a small 3½-inch gauge early shunting engine.
Before I start recounting my experience during the process, I must explain that my knowledge of IC (internal combustion) engines is a little faded, since the last time I worked on them was also in the 1960s and they were full-size car and motorbike engines. Some of the issues I encountered might have been recognised and dealt with more easily by an experienced IC modeller, but my dialogue may be of some help to the first-timers.
Ring cycle
The engine, having been left untouched for nearly 60 years, would not turn but once it was stripped down it soon became clear that the issue was the piston rings – the oil had dried and stuck the rings to the bore and slots in the piston. A gentle tap removed the piston from the bore, but while removing the rings from the piston one of them broke.
I ordered a new set from Hemmingway (www.hemingwaykits. com )as well as a float chamber casting. Butr the casting was out of stock so it took a month before it was delivered. I used the time to inspect, clean and paint the rest of the engine and did not find anything wrong so it all went together well, or so I thought...
I made a mechanism employing an electric drill to turn the engine over for starting. I later replaced this
ABOVE RIGHT: Stuart got his Kiwi petrol engine to run, but it took a lot of effort...
BELOW: Jig made up to turn flank radii on cams. with a drive and housing utilising a roller clutch which proved a far better solution. However when the engine was spun it made no attempt to run.
I had previously checked that the original magneto produced a spark, so believing this to be okay I set about checking the valve timing against the timing diagrams in Edgar Westbury’s articles. I found interpreting the diagrams awkward because my father had intended the engine to run anti-clockwise as viewed from the flywheel end – this was because it was common in the 1960s to drive model boats with left-hand propellers and the diagrams are drawn for clockwise rotation.
I eventually concluded that both valves were opening far too late and closing far too early – this had to be the problem, new cams were required. This was a new venture for me as I had never made cams before or cut internal keyways.
Studying the original articles it was clear that the engine will run in either direction by simply turning the valves round and that it would be easier for me to make the valves separately, not in one piece as drawn. With this design of cams the keyway is set at top-dead centre and was already in the engine camshaft so my keyway had to match both the cams and the jig used in the machining process,
I found that a 1.5 mm slot drill cut the size exactly, a 1⁄16-inch one as suggested by the drawing would with my equipment cut over-size, so the cutter that produces the internal keyway was ground to this size before starting on the blanks.
My advice when starting from scratch is to mill the keyway on the shaft first and make the internal cutter to match. Turning the blanks is now straightforward but it is important that the hole is reamed to exact size, not drilled.
Cutting the keyways is also straightforward, first making sure that the cutter is set accurately to centre height, and using the lathe as a horizontal shaping machine with plenty of cutting oil and tiny cuts.
To turn the flank radii I produced a jig that fitted into a four-jaw chuck – it is simply a piece of bar machined square and accurately marked out with the spigot position centre, a top-dead centre line, the other four radii centres for the flank angles and then spotted with a centre punch
“The last
time I worked on IC engines was in the 1960s and they were full-size car and motorbike engines....”
exactly on the cross marks, making sure that the centres are marked for which cam they are to produce.
The jig was mounted in the four-jaw with the spigot centre running dead true. I reamed the hole for the spigot and with the top-dead centre line locked in the horizontal position cut a key way. The spigot is then keyed onto the jig so that the cam blanks are always in the correct and identical position.
I then set the jig to run true on the flank angle centres. Westbury tells you to measure the first flank across from the external diameter of the blank and when turning the second flank to go down to the cross-slide index. This did not produce the desired result, the thickness around the hole was different and so the first one was rejected as scrap.
If the radii is to produce a tangent intersect at the correct angle, then the edge distance across from the centre hole has to be exactly the same, so the flank angle dimensions were produced by measuring the thickness of material across to the hole. This has to be as the same as that produced by the base diameter when machined and was only made possible by being able to remove them from the jig and to refit them exactly.
The jig was then transferred to a rotary table set up on the milling machine so that the spigot was perfectly true. It is so easy to make a mistake and over run the angles so sticky tape was used to mark the start and end positions of the radii to be nibbled away. Again the cams could be taken off the jig so that the edge thickness could be measured and produced the same as for the flank angle curves – this way the tangents must be in the right place.
The cams were then polished and case hardened, the engine rebuilt, valve timing checked and all was well. This time she would surely start but no, she never even tried...
Ignition issues
My experience from years gone by told me that quite often when an engine did not even try to start the problem was ignition. I knew the coil, plug and points were at least 60 years old and that even though I was getting a spark outside the cylinder was I getting a strong enough one under compression? I ordered and fitted a new plug and condenser but still the little engine made no attempt to start, so I concluded it must be the coil.
I replaced the ignition system with a Minimag electronic system using a Hall sensor instead of conventional points mounted on a disc driven from the camshaft, but guess what, still no joy, this engine was not going to run.
By now I was starting to run out of ideas so I removed the carburettor for inspection and found that the primary air intake was missing. This is a small hole which bypasses the main intake and enters the body of the carb just prior to the jet. I made the modification and refitted it.
Shy plug
I also noticed at this point that the hole into which the spark plug screwed was not as deep as the drawings showed, which meant that the electrodes did not protrude into the compressed mixture space, so I set the head up in the lathe and corrected it.
Furthermore, the gap on the new spark plug had mysteriously disappeared. On investigation the earth electrode was a different shape to that on the original plug and when screwed in fully it fouled the couterbore in the hole and closed up. Everything was refitted but yet again the engine refused to make any attempt to run.
By now I was clutching at straws and after talking to some members of my club I decided to check that the piston conformed to the design standards, but all seemed to be okay. However I realised at this point that Westbury never used gaskets – all his joints were metal to metal. Yet on my engine a fibre gasket of some 30 thou thick was fitted between the crankshaft casing and the cylinder, which with this size of engine must have reduced the compression ratio considerably.
So I again reassembled the engine minus this gasket and when spun with the starter it fired up perfectly on every stroke!
This engine however still had a trick up its sleeve. When the starter was switched off the engine stopped – it would only run whilst being turned from an external source.
Whilst scratching my head I turned the engine over by hand with the flywheel and it seemed to slip when turning through the compression stroke. Westbury used a split taper collet to hold the flywheel in position and retightening this prevented the slip and bingo it fired up straight away and ran perfectly. It was time to put the kettle on or to have a stiff whisky...
I could only run the engine for a few seconds at a time because there was no cooling or oiler fitted, so the next stage will be to fit it to a test rig with the ancillary equipment and do the necessary adjustments before designing a model to fit it into.
The whole episode had taken me three months but the engine now runs and although some of the work done may not have been essential, I have to believe that the modifications I made will result in far more efficient running when fitted into a model.
“The hole into which the spark plug screwed was not as deep as the drawings showed, which meant
that the electrodes did not protrude into the compressed mixture space...”