Ottawa Engine Restoration

Author Photo
By Staff

1 / 21
The 2-1/2 HP Ottawa Engine as purchased.
2 / 21
The hopper ready to have its piston freed.
3 / 21
The hopper resting on its cylinder head bolts so the penetrating fluid can soak around the piston.
4 / 21
The hydraulic puller clamped to the hopper, pushing against the piston.
5 / 21
The puller attempting to remove the key.
6 / 21
The removed piston.
7 / 21
The unusual key that was holding the pulley in place.
8 / 21
The removed key.
9 / 21
Cutting a slit in the remains of the pulley hub, with the crack visible after splitting.
10 / 21
The puller daws resting against the clamp block.
11 / 21
The block welded to the angle piece and taper key being used.
12 / 21
The completed rocker arm along with pivot pin and adjustment bolt. 
13 / 21
The cylinder head after replacing the valves, showing the rocker arm bracket with tapered sides.
14 / 21
The lockout assembly as removed from the engine.
15 / 21
The completed assembly being trial-fitted.
16 / 21
The completed governor weight repair.
17 / 21
The broken governor weight after cleaning.
18 / 21
The broken pivot arm.
19 / 21
The brazed repair, ready to cut off surplus rod and finish file.
20 / 21
The metal block brazed to the arm.
21 / 21
The completed lockout arm repair.

My latest purchase is, according to the tag, an Ottawa 2-1/2 HP engine, serial no. C30705.

The origins of the Ottawa Mfg. Co. start with Warner Mfg. (later the Warner Fence Co.), a very successful business that manufactured woven fence wire in the 1880s. In 1903 the company relocated to Ottawa, Kan.

Warner Mfg. then produced various gasoline products, which were sold under the name Union Foundry & Machine Co.; the engines were sold by independent franchisers. These engines were also sold direct by the company under the Warner Engine name. The use of the Union name on engines disappeared around 1913, and they then became known as “The Ottawa” engine. Recognizing an opportunity, the company later designed and sold tree felling and log saw engines and is probably better known for these tools.

I was very fortunate to have obtained generous assistance from George and Helen Myers. They allowed me to tap their huge resource of knowledge on these engines. Helen was very patient, answering numerous questions and providing photographs. Their help ensured this project resulted in an extremely accurate restoration of this engine. There is no accurate dating information available for Ottawa engines, although Helen confirmed that this was a stationary engine, not a saw engine, and it was probably manufactured around 1917.

I agree with the comments by C.H. Wendel in his book American Gasoline Engines Since 1872, the Ottawa engine bears more than a passing resemblance to Associated engines manufactured in Waterloo, Iowa. The air-cooled head, hopper/main casting, muffler, governor weights and igniter are very similar to the Associated.

Removing the Ottawa engine piston

When I purchased the Ottawa engine, I was told that the piston was seized in the bore but the rust in the cylinder was thin surface rust, not deep pitting.

After the Ottawa engine was delivered, I partially stripped it, first removing the cylinder head and then the big end cap. Once I removed the four nuts holding the cylinder/hopper in place, the cylinder was taken off the main casting so that I could begin working to free the piston.

As a first step I used some fine emery cloth to clean the cylinder bore at both ends of the piston. I then stood the cylinder block upright so that I could pour some penetrating fluid in around the base of the piston and leave it to work. I rested a block of wood on the head of the piston and gave the block of wood moderate taps with a mallet, but there was no sign of movement. After a few days there was no sign that any of the penetrating fluid had seeped down around the piston, so another approach was needed.

I turned a 1-inch-thick block of cast iron to the bore diameter of 4.125 inches, then I recessed the cast iron to leave a band around the edge of the bore that would rest on the top outer edge of the piston. I rigged the hydraulic gear puller up so that it could press down on this cast iron plate, its claws catching on the other side of the cylinder block. I pumped the hydraulic cylinder up to its maximum (10-ton pressure), but there was no sign of the piston moving. I poured more penetrating fluid in and left the piston under pressure for another couple of days in hopes the fluid would work.

When I next tried to move the piston, I also shocked the metal block while it was still under pressure by hitting it with a large hammer. There was a loud crack and some movement. I had measured the distance between the top of the temporary iron block and the end of the cylinder at the outset, which enabled me to keep track of progress.

Even after the piston started to move, it was still held tight by the rust. It took a couple of days to ease the piston out using the puller, full pressure being needed right to the very end. I found that there was a little more than light rust in the bore and around the piston, more than enough to hold it tight. Fortunately, the rust left no major pitting to the bore, and it looked as though water had seeped through the oil hole to create the rust. There was some pitting in the bore but this was mainly caused by casting flaws, luckily near the skirt area of the piston so I could ignore them.

Once I had removed the piston I released the wrist pin lock bolt. I easily removed the wrist pin along with the connecting rod, and then soaked the piston and rings in kerosene and cleaned them up.

Removing flywheels

Over the course of a few days, I gave the governor pins, nuts and both the keys squirts of penetrating fluid. I then cleaned the rust and dirt off the crankshaft and nearby areas using a wire brush. I cleaned the visible parts of the keyway slots and crankshaft with emery cloth after first using a file to flatten any nicks in the metal.

My first task was to remove the pulley, the key for which appeared to be an old coach bolt of unknown length, which was firmly rusted in place.

The pulley itself did not look to be in good shape and appeared to have welded repairs on its face. I decided to split the pulley because if I needed strong-arm measures to remove it I would risk damaging the crankshaft.

First, I cut a slot in some 0.375-inch thick steel plate so I could slide it under the head of the “key.” Then I drilled a 0.500-inch hole through the plate to align with the center of the crankshaft. To strengthen the head of the rusty key I welded it to the plate before I tried to remove it.

I put a steel rod through the hole in the Ottawa engine and fitted the hydraulic puller to press against the steel rod to pull both the plate and key out. Unsurprisingly, the key sheared so it was then time for plan B. I squared the end of the key with a grinding wheel before drilling for a 0.3125-inch thread to be cut, this size eliminating the risk of damage to the crankshaft. After tapping a thread in this hole I bolted the sliding puller to it and eventually pulled the key out.

The only place that the hub puller could work was on the rim of the pulley; there was no way to get any support around the hub as it was tight against the flywheel. I cranked up the hub puller and then there was a ominous “crack” as the pulley rim shattered. Examination of the pieces revealed old cracks in the rim and that two of the three spokes had broken long ago.

This left the hub on the crankshaft, so I used a cutting wheel to cut a groove along it, taking care not to get close to the crankshaft. Then I split the hub with a cold chisel and removed it.

Looking at the key for this flywheel, there was no sharp edge to the head, so I used a file to clean it up, but I soon realized that the bearing metal on the key had been split. It was then an easy task to use a file to restore the key’s shape. I cleaned out the keyway with a file and oiled it before clamping the curved key puller in place and hitting the key puller with a hammer. The key came out fairly easily, complete with a thin metal shim at the bottom that had been used to tighten it.

Before removing the governor-side flywheel I attempted to take off the governor weights because one of them was broken. First, I applied more penetrating fluid because they were covered in rust. I took care to not apply excessive force when removing the pivot pins to avoid cracking the mounting bracket. Gradually the pins moved and I was eventually able to punch them out.

The key on the governor-side flywheel had no head, so I welded an angled fitting to the key so I could couple it to the slide hammer to free it. This had no effect, so I welded a lump of steel onto the bracket so the tapered key remover could be tried. This only resulted in the weld breaking, so the next step was to drill through the key.

I drilled a 0.125-inch hole through the center of some 0.375-inch square steel that I would clamp in the keyway as a drill guide. I then used a 0.125-inch extended drill bit to drill through the key and then opened up this hole with larger drills. When doing this remember that the keys are tapered and drilling too far through with too large of a drill might cut into the flywheel hub.

During the drilling process the extended drill snagged and the tip broke off. This small piece of drill remained deep in the hole. To remove it, I shaped a piece of drill rod to fit the flutes of the drill, then ground flats for the wrench before hardening the drill rod. I twisted the extractor counter-clockwise to free the remains of the drill that then fell out.

Once the hole in the key had been opened as big as was deemed safe to avoid damaging the crankshaft, I attempted to collapse the key and remove the remains. Bumping the flywheel and crankshaft on a block of wood did nothing so I used a steel pick to remove as much of the key as possible.

Using the puller on the flywheel presented problems: There was nothing for the claws of the puller to grab against — the thin spokes of the flywheel were definitely not an option. I was able to remove the crankshaft and pull against the gear wheel (and, through this, the flywheel) but the puller would rest against and stress the governor ring, potentially damaging it.

I found that when unscrewing the governor weight bracket and then pushing the governor ring against the bearing there was a 1-inch gap around the crankshaft. I made some scrap steel 1 inch thick into a clamp block by boring it out to fit around the crankshaft, then cross-drilled for two clamping bolts before cutting it in two. I bolted and clamped the clamp block around the crankshaft to press against the hub of the flywheel to provide a solid platform for the puller claws to rest against.

Eventually, after a lot of effort and shocking the clamp block with a lump hammer at the same time that it was under pressure, the flywheel started to move. I have never experienced as difficult a job removing a flywheel and its key. Looking at the remains of both keys later, I wondered if at some stage new keys were made, one larger than the other and then swapped over, hence the need for a shim in one and the other being over-tight. Fortunately, close examination on the flywheel hub revealed no cracks from use of the lump hammer.

The flywheels would need new keys for reassembly. I made the keys by milling a head and also a 1:100 taper in some 0.375-inch by 0.700-inch steel. I then fitted keys using engineers blue, filing to ensure a tight fit with even contact. I made the keys so that when they were knocked home, the gap between the hub of the flywheel and the head would be 0.375-inch.

Rocker arm

The rocker arm was missing, but Helen Myers provided details of the correct style along with photos and a couple of measurements. It would have been possible to purchase a casting from Helen, but I decided that it would be fairly simple to fabricate one.

As a first step I drilled a piece of steel 1.750 inches round with a 0.375-inch center hole, and two 0.750-inch diameter bosses 0.125-inch high turned on each side, the total thickness being 1 inch.

I sawed a length of steel 5 inches by 0.625-inch by 0.875-inch to the rough profile of the arm and drilled a pilot hole to be eventually threaded for the adjustment bolt. I milled a 0.375-inch-wide slot in what would be the central hub of the rocker, thinning the center section of the arm to slide in it.

After brazing the two parts together, I shaped the arm using the milling machine and then filed it to match the photographs.

Before finishing off the rocker arm, I turned my attention to its pivot point on the bracket attached to the cylinder head. The top and bottom faces of the bracket tapered toward the hole for the rocker pivot pin, which made it impossible to get a good fit where the arm operated without excessive unwanted movement.

I set up the bracket on the mill and both sides trued up parallel. At most it was only necessary to remove 0.0625-inch of metal from each side.

I then had to build up the bracket; I used two pieces of 0.125-inch thick steel. This was thicker than the 0.0625-inch needed, but if I used thinner sheet it would have curled when heated during the brazing process. Once the two pieces were brazed and cooled, I again fixed the arm on the milling machine so the surplus metal could be removed and the arm reshaped. I also restored the hole for the pivot pin.

Once I fixed the rocker arm bracket, I could mill the slot in the rocker arm to fit the parallel arm.

I center-drilled the pivot pin to the midpoint and then cross-hole drilled a 0.109-inch hole so that any oil squirted in the countersunk hole would lubricate the middle of the pivot. I made the adjustment bolt and lock nut and tapped its 0.375-inch threaded hole in the rocker arm.


One of the governor weights was broken and the governor lockout assembly had suffered damage in the past, with a temporary repair made.

My first step was to clean the governor weight back to bare metal. Then I glued the two pieces together with superglue before clamping the governor weight in the milling vice to drill down the length of the arm in preparation to insert a reinforcing rod.

I then broke the bond of the glue and cleaned off all traces of old glue, grinding back to bare metal as well as providing a “V” at the edges for the braze to fill when joining the two pieces.

I brazed the two parts together and filled the hole that had been drilled through the pivot pin support. I cleaned off the surplus braze and reamed the hole for the pivot pin to size.

The lockout assembly was in a very poor state with what looked like a “blacksmith’s repair” — a piece of rough steel had been formed around and bolted to it in order to hold it together. I stripped this down so that the damage to the separate parts could be identified.

The first item I repaired was the pivot arm, which had fractured near the pivot point. Again, I made a repair following the usual routine of cleaning, gluing together, drilling and fitting a reinforcing pin, cleaning and then brazing. In this case, I used threaded rod to match a thread cut in both pieces as there was only a short distance to hold the reinforcing pin.

One of the two arms was broken on the lockout arm. I milled away the area around the break to get to a solid area of flat metal. I brazed a block of steel to this, along with a short reinforcing rod.

I sawed this block, and then filed it to shape before drilling the hole for the pivot pin.

Then all that remained was making a new roller for the follower to fit in the collar on the flywheel, stoning the edges on the hardened lockout plate and making a new spring for the adjustment screw.

In part 2 Peter makes headway on the Ottawa, focusing on the piston, cylinder head and connecting rod.

Contact Peter Rooke at Hardigate House, Hardigate Rd., Cropwell Butler, Nottingham, NG12 3AH, England • peter@enginepeter.co.uk

Gas Engine Magazine

Preserving the History of Internal Combustion Engines