Ignition Timing Advance

All engines have a manufacturer-specified ignition timing. An engineer considers the role of the typical delay of several systems in ignition timing advance.

Initially all engines have a manufacturer-specified static ignition timing. Combustion, and therefore, cylinder pressure build-up is a rather slow process. In order to have the cylinder pressure peak at the proper time in the power stroke, combustion must be started before the piston reaches top dead center (TDC) on the compression stroke. The higher the engine rpm, the earlier the required spark. The spark coming in advance of TDC is measured in degrees of crankshaft rotation. Some engines like John Deere, Economy, Sandwich, R&V and others have the spark time mark stamped or cast on the flywheel. Others may have relied on measuring the crankshaft throw angle.

When the original engine manufacturer specifications are not available and an obvious flywheel mark is not visible, deductive methods must be relied on. “David’s Old Engines” recommends 8 degrees of advance per 100rpm, others recommend 5 degrees per 100rpm. The mark on a 1-1/2hp 600rpm John Deere is 25 degrees before TDC, making it about 4 degrees per 100rpm. To set the advance, the flywheel is slowly rolled forward until the trip mechanism trips, or the buzz coil switch closes, and adjustments are made until the desired advance is achieved. This is a static procedure.

It is unknown if those original specifications or criterion optimize the engine for power, fuel economy, smooth running or some other performance feature. Independent of the optimized performance feature, the static advance is unusually large. Although compression is much higher, a modern engine would run at most 10 degrees of advance at 500 or 600rpm, not the 25 to 40 degrees these old engines seem to have.

The large advance is the result of static timing. The trip, or closure, of the switch does not immediately result in a spark, rather it begins a process that will create a spark at a later time. In a typical igniter-fired engine, a mechanical assembly that is tied to the crankshaft spring loads the igniter hammer. At trip, the hammer begins rotating toward the anvil. When the hammer strikes the anvil, it transfers momentum to the anvil, which in turn causes the anvil to leave its sitting position and initiates the spark. This process takes time and results in a retardation of the static advance.

Evaluating Typical System Delay

Figure 1: The typical delay of several systems measured by the author.

Regarding the 1-1/2hp John Deere mentioned above, that process takes about 5ms (0.005 second). At the specified 600rpm, the flywheel rotates an additional 18 degrees after the igniter trip before the actual spark is initiated. Rather than 25 degrees of advance, the actual advance is 7 degrees (25 – 18 = 7).

The Webster M1 system behaves in a similar manner. The large springs begin rotating the armature with its attached hammer at trip. The hammer strikes the anvil and initiates the spark approximately 7ms (0.007 second) later.

The trip of a Wico EK starts the armature moving away from its static location. After it travels the specified 1/16 inch, or 3/32 inch, the points open and spark occurs. This process takes approximately 7ms.

Closure of the timer switch on a buzz coil engine causes current to begin building up in the primary winding of the coil. When the current reaches a high enough value, the electro magnet formed has enough strength to pull the contact flapper in and the spark occurs. That process takes about 2.5ms at the original 6v.

Figure 1 is the delay of a few systems recently measured.

Most high-tension systems don’t have the same timing discrepancy. Those systems use adjustable points opening to set the termination of the spark process rather the initiation. Regarding the early Clinton engine that was featured in the February/March 2021 Gas Engine Magazine, when when the engine is running a full three times the speed of the John Deere, it has only 18 degrees of actual advance. This was measured with an automotive timing light.

Figure 2 is a convenient graph for estimating the actual engine ignition advance. Based on Figure 1, or other inputs, estimate the engine spark system delay and choose the appropriate delay line (or midpoint between lines) in Figure 2. Go vertically from the intended rpm to the delay line, then horizontally to the degrees of retard. The engine running advance will be the static advance minus the value in Figure 2’s “Degrees of Retard.”

For example, the earlier mentioned John Deere has a delay of 5ms, halfway between the 4ms (orange) and 6ms (gray) lines. The engine is specified at 600rpm. Draw a line straight up from 600rpm to midway between the 4ms and 6ms lines, and from there straight over to 18ms of retard. The engine is statically timed at 25 degrees, therefore the actual running advance is 25 minus 18 degrees, approximately 7 degrees. It is a common practice at shows to run a small 500 or 600rpm engine at 200rpm. If that engine is set up at 10 degrees advance (5 degrees per 100rpm) and has a 5ms delayed igniter, the actual advance is around 4 degrees.

Conclusions about ignition timing advance

This article is not intended to change the way ignition timing advance is set but rather to satisfy some intellectual curiosity. However, it does lead to some interesting facts. A gummy/worn igniter, Wico sticky armature or incorrectly set points could easily add 2ms to the time to spark and significantly change the actual spark advance, in addition to reducing spark energy.

John Deere used the same flywheel, thus the same static advance, on the igniter and buzz-coil fired engines resulting in a different actual ignition advance on those engines.

Dr. David Cave is a regular contributor to Gas Engine Magazine and can be emailed at jdengines@cox.net

Originally published as “Spark Timing Advance” in the August/September 2023 issue of Gas Engine Magazine.