15-20 HP Carl Anderson Co. engine gets a tune-up

Big Waldo's 21st century modification

carl anderson1

Curt Holland used some 21st century technology to get this rare 15 to 20 HP Carl Anderson Co. engine running properly.

Photo by Curt Holland

Content Tools

The religious conservative Waldensians trace their roots back to 46 A.D. to the valleys of Northern Italy. Modern-language Bibles owe a debt of gratitude to the Waldensians as they were the first to translate the Bible from Hebrew and Greek to French. Population growth and persistent persecution led many of the Waldensians to seek a new life in the Americas, and they settled in one area known today as Valdese, N.C. They were phenomenal stonemasons and many of the houses they built as contract laborers can still be seen dotting the Carolinas in a 100-mile radius around Valdese. Duke University, as well as many of the stone bridges through Virginia and into Washington, D.C., illustrate some of their handiwork.

The original Waldensian settlers operated a sawmill, clearing land and producing lumber to build their homes and sell to pay for their land. The old engine they were given to pull the sawmill soon got to be known as “the damned old engine” because it was so difficult to operate.

North Carolina State Senator Jim Jacumin is a third generation descendent of the Waldensians and has worked tirelessly to document the life of the Waldensians in Italy and in Valdese.
Over the last 15 years he led the construction of a beautiful tribute to them in the form of a village that represents the life in Italy and in North Carolina, the Waldensian Trail of Faith in Valdese, N.C. Included in this village is a sawmill belted to the engine operated by the Waldensians.

Looking for help
Last summer I was asked to look at the engine and see if anything could be done to make it work properly. It had to be helped with an electric motor to operate the sawmill for show.
There were no identifying marks on the engine and it looked to be between 15 and 20 HP. There was no cam gear on the engine and it had what appeared to be ported exhaust.

My first impression was that it was some sort of unusual 2-cycle engine. I took several photos and then began the long task of going through all the pages of C.H. Wendel’s American Gasoline Engines Since 1872 to try to identify it. Not too far into the book I found a Carl Anderson Co. engine that looked very close. Luckily, I found mention of a patent number, and a search on the number turned up a lot of information about the operating principle of the engine.

The essence of the patent was the exhaust pushrod mechanism and the linkage that accomplished the “skip” feature. Actually, it should be called an “engagement” feature instead of a skip feature as the engine is designed to omit exhaust cycles all the time except when a power stroke occurs.

On the crankshaft is a simple eccentric, which operates the exhaust pushrod. This pushrod moves at the same frequency as the piston, or twice as fast as needed for a 4-cycle engine. At the end of the pushrod is a slip mechanism. In this slip mechanism is a small pin that is extended to link the pushrod with the exhaust valve stem. By default the small pin is retracted and the exhaust valve is never activated.

In the side of the valve chest is a small cylinder and piston. When the engine fires, this small piston moves out as a result of the high cylinder pressure. This small piston bears on a linkage that moves the small pin into the mechanical pushrod linkage.

On the next cycle of the exhaust pushrod, the exhaust valve is opened and the burned contents in the engine are exhausted. As the exhaust pushrod closes the valve, springs retract the small pin, and this pin stays retracted awaiting the next power pulse.

Not quite a hit-and-miss
The other feature covered by the patent is the governing system. This is not a hit-and- miss engine, but a rather unusual throttle-governed engine. It has a mixer like you would find on a hit-and-miss engine, but there is no throttle plate. The engine speed is regulated by a flywheel governor weight, spring and rod system that acts on the stiff intake valve spring. As the engine comes up to speed, the governor increases the spring force on the intake valve, limiting how far it can be sucked open on the intake stroke. Rather clever, I thought, although I must admit to safety concerns with this arrangement.

Pondering an age-old question
After studying these papers a long while I began to question how the engine could run. I could see how the engine might work well with the skip/engage feature if used in conjunction with a hit-and-miss system, where there was a powerful power stoke every time to assure the small pin was inserted in the exhaust linkage. This mechanism, in conjunction with throttle governing, just didn’t make sense. If there was ever a soft fire or a misfire, the small piston on the side of the valve chest would never extend and thus the burned cylinder contents would never be exhausted and the engine would simply coast to a stop.

I consulted with some well-respected engine men (Terry Tanner and Ed Rowlands) and they too were puzzled at how this combination could actually work. All of this echoed the 100-year-old Waldesian journals in which Jim had read about how difficult the old engine was to start and keep running. Many a confession and subsequent penance were probably served as a result of this temperamental engine!

A 21st century solution
After pondering what to do for the engine over the winter I made a set of recommendations to Jim. The purists may scream at these recommendations, but rest assured – no original engine parts were altered or destroyed to make the modifications!

The project intent was to make the engine a reliable and safe runner, allowing the consistent demonstration of the engine and sawmill in the Trail of Faith public forum. By profession, I am a mechanical engineer, but also do a fair amount of electrical engineering too, using programmable logic controllers (PLCs) with the machines I design. Since the skip/engage mechanism on the old engine was the primary source of trouble, I recommended making a new mechanical skip mechanism that would incorporate a small electric solenoid to extend and retract the small pin. This solenoid would be accurately energized by a PLC.

Belted to the crankshaft would be an encoder turning at half-crankshaft speed, accomplishing what a traditional cam gear does. There would also be a “flag” on the encoder timing pulley that would reset the high-speed encoder every time the piston passed through bottom dead center (BDC). By utilizing the A/B channel inputs and the flag reset (proximity switch) the PLC would know the exact location of the piston at all times and would also know the rotation direction.

Programming the PLC
Jim liked the recommendation and so it was time to start machining new parts and programming the PLC.

A graph of the engine motions was made, showing the sinusoidal motions of the piston and the exhaust pushrod. This helped with the programming and at what encoder pulse counts to turn on and off the solenoid. The control panel was built and set up on the workbench, long before ever installing it on the engine.

Once the mechanical parts were machined and the solenoid programming was worked out, it seemed only natural to go ahead and control the ignition timing with the same PLC too. The existing crude homemade wiper system that operated a buzz coil was unreliable and this had long ago replaced the original hot-tube system in the top of the valve chest. A few lines of PLC code later and the ignition system programming was complete.

The PLC gives a simple relay closure output, energizing the buzz coil. And we thought that while we are playing with the ignition system, why not add something fancy! Modern automobile engines utilize spark advance, so we decided to do the same. Some more code was written, advancing the spark linearly from 15 degrees to 30 degrees as engine speed goes from 50 RPM to the 300 RPM maximum speed.

The next focus was a scheme to allow easy starting. We all know how difficult a 15 to 20 HP engine can be to start, so the electronics were configured to aid this as well. A few mechanical changes were made to accomplish this.

The original hot tube in the top of the valve chest was long ago replaced with a spark plug. I had a hunch this was not an ideal location, so the plug was removed and a priming cup added there. The spark plug was relocated to the middle of the valve chest in the location previously occupied by the small cylinder and piston that were part of the skip/engage mechanism. A new part was made, so again, no original parts were destroyed.

We are all familiar with how the flywheels are walked backward on an oil field engine to start them, so I thought this would be a simple, easy and safe way to start this engine. Since the PLC knows rotation direction, code was written to activate the buzz coil at 30 degrees before TDC when walking the flywheels backward against compression. Starting procedure would be simple: With the piston at TDC, open the priming cup and prime as you rotate the flywheels clockwise. At BDC close the priming cup, walk the flywheels hard backward against compression and wait for the buzz coil.

The last bit of programming was to address my safety concerns about the governing system. Given that the engine is to operate in a public forum, a run away engine could have disastrous consequences. The encoder being driven by the crankshaft has two other A/B channel outputs. These are fed into another high-speed input on the PLC and converted to a frequency. More code was written comparing this input frequency to a maximum safe frequency. From old literature it was determined the maximum speed for the flywheels of this diameter was 300 RPM. It was planned to run the engine at approximately 225 RPM. If any part of the mechanical governing system fails and the engine runs up to 300 RPM, both the ignition and the electric solenoid are de-energized to shut the engine down and serve as a double safety.

But will it work?
All the programming was worked out on the workbench at home, and in early May, it was time to install the PLC cabinet and the mechanical parts that had been made.

It took us about 3 hours to install all the parts and the rebuilt fuel pump, and then it was time to try it out. We primed the engine, rolled the flywheels to BDC, and walked the flywheels backward against compression, and the buzz coil fired. BANG! The engine fired, reversed and was off and running on the first pull of the flywheels! Those Waldensians of yesteryear wouldn’t believe it!

We spent the next couple of weekends working out the engine speeds and altering the gearing of the sawmill carriage to slow it down.

It’s showtime!
The last weekend of May was Jim’s big show at the Trail of Faith and we were in the log sawing business, sawing some 20-inch diameter poplar logs all weekend. This big engine with its huge stack sounds so good thumping out a steady deep exhaust rhythm as the blade digs into a log to the delight of spectators. It was a great weekend!

An opportunity appreciated
The opportunities to do projects like this in one’s life are rare. I would like to thank Senator Jim Jacumin for affording me this opportunity, and giving me the leeway to design and build this project in my vision, incorporating the features I had in mind. In essence, a modern-day electronic engine management system was designed, built and fitted to a 100-plus-year-old engine. I’m a purist and adamant about maintaining the originality of rare one-of-a-kind engines, but given the forum the engine is being asked to perform in, this seemed the only workable solution. All the original parts have been preserved and are safely stored. This is the best of both worlds, a reliable running engine that can easily be returned to its original, as built condition.

Contact Curt Holland at (704) 853-2992 •  curt@rustyiron.com

For more information on the Trail of Faith, visit  www.trailoffaith.com