Engine Kickback - Causes and Correction
Taming the Threat to Your Starter's Drive Train
Kickback has several meanings: graft in politics, feet up with shirt collar open, and sudden reversed engine rotation at starting.
Each is a "condition" caused by some defect. The first is greed, the second indolence, and the third "defect" is a set of inter-related factors, possibly damaging to the engine starter train.
There are three simultaneous and necessary conditions for engine kickback at starting:
- gas vapor in the cylinder that "kicks,"
- spark with sufficient voltage to jump the plug gap and enough current to ignite that vapor,
- the piston in the ignited cylinder to be some crankshaft degrees before top-dead-center (BTDC).
Engine kickback, if severe, can cause damage to the starter mechanincal train, usually the Bendix Drive spring/gear or shaft, so the condition should be avoided.
As noted, there must be three simultaneous and necessary conditions at start for kickback. The only one that can cause the effect is ignition and sufficient gas expansion while a piston is moving up in some cylinder. This gas expansion force must be great enough to overcome the starter motor torque and momentum of the engine rotating masses.
It has been well established by both theory and measurement that the gas vapor is completely burned in approximately 0.002 seconds of time. At an engine cranking speed of 150 RPM, the crankshaft has moved only 1.8 degrees, not a significant change in piston position or gas compression.
150/60 X 360 degrees X 0.002 = 1.8 degrees
That small rotation determines that kickback is caused by spark, gas ignition, and burning well before top-dead-center (BTDC) piston position. If it were after TDC (ATDC), the engine would start, idle or run normally.
To experimentally investigate that simple analysis, tests were recently conducted on two representative "A's." Both were stock 1931's, 6 volt positive ground.
Tests were made both with point contact and condenser ignition and with a transistor system. The gas vapor and burning should not know which spark started ignition, but the tests would confirm if any difference existed.
The test programs follow:
- Initial timing set to 5 degrees ATDC, full retard at steering column.
- Run at set timing for 15 seconds.
- 6 consecutive starts - normal start sequence (Ignition "ON" before starter is actuated).
- 30 to 60 seconds between starts.
- After the 6th run, reset timing by advancing spark control rod and check setting with timing light.
- Repeat at 0 degrees, 5 degrees, 10 degrees, 15 degrees, 20 degrees, 25 degrees, and then at 30 and 32-1/2 degrees, all BTDC or heavy bucking occurs.
All timings were measured with timing light and crankshaft indicator as noted in reference.
- Repeat all above with starter actuated, next followed by ignition switch turn "ON" for ignition.
Smooth start and run?
Any evidence of hard start?
Test Data Results and Comments
There was no "kickback" using the normal start procedure (ignition "ON" and then engaging the starter) through timing set to 25 degrees BTDC.
Repeat tests with the starter engaged and then turning the ignition switch "ON" were substanitally the same as the first tests. When advance was set to 30 degrees BTDC, sporadic and slight "kickback" started to occur. At 32-1/2 degrees BTDC, HEAVY "KICKBACK" was experienced. Further advance could result in starter damage so the tests were stopped at this point.
To investigate if cold or hot engine temperatures were a factor, the tests were repeated after a 12 hour interval. They repeated as above within the limits of observation.
The tests (well over 100 trials) showed NO KICKBACKS and start and run with timing set at less than approximately 25 degrees BTDC. At greater setting, approximately 30 degrees plus or minus 2-1/2 degrees BTDC, "kickbacks" may well occur on start, with damage to the starter or other components. So, engine start should be made with the column spark control set at FULL RETARD (UP) and timing set to 5 degrees to 10 degrees ATDC for ease of starting.