Ignition Timing on Corvairs

Builders,

One of the things that sets our work apart from other engine options is the amount of direct interaction we have with builders. A salesman can spend all day talking to a guy about buying something and still have no real measure of what that guy’s understanding of the product is. Conversely, I can spend 10 minutes showing a single step to a builder, and then ask him a few questions. I will then have a good gage on his understanding, and tailor the delivery of information to his level. Although we have several hundred builders, I am an ‘idiot savant’ when it comes to perceiving and remembering our builders development of mechanical skills and understanding.

In conversation, one of the things I have consistently seen is that many builders don’t have a good picture of how ignition timing works on an engine. There are three levels of understanding on this: 1) being able to follow the directions on our website and use a timing light. 2) Knowing a little more about the why of timing in our engines, and 3) having a grasp on the factors affecting the behavior of combustion chamber reactions.  With the help of the graph below, I want to move a number of builders from 1) to level 2).

.

.

Above is an ignition graph I ‘borrowed’ off the MSD ignition website.  Although this timing ‘curve’ is not an exact match for a Corvair flight ignition, it is close enough to support a good explanation and aid builder understanding.

.

First, some language translations: What they are calling ‘initial’ timing is what I call ‘static’ timing. On a Corvair we usually start with 8 degrees to get the motor to fire up on the stand. But this number isn’t important, ‘total’ timing, or what I call total advance is. This is the number that your engine will fly around at, and it is the one you must check with a timing light. Also, people use the term “ignition curve” to describe the advance section of the graph, because theoretically it would graph as a curve, but in practical reality it graphs as a straight line.

When an individual cylinder is coming up to fire at idle, the spark takes place about 10 crankshaft degrees before the piston gets to the top of the bore. This allows the peak pressure and the effort of the cylinder to take place after the piston passes over top dead center onto the power stroke. As the engine rpm comes up, the spark needs to happen sooner because there is less physical time for the combustion to take place. This change in timing is called ignition advance, and in the Corvair this is done with mechanical flyweights inside the distributor body. Cars have a very complex set of running parameters, but airplanes do not. With fixed pitch props the power required by any rpm is fairly consistent. Thus the mechanical weights can cover the task that cars must have complex systems like vacuum advance (then) and computers (now) for. Airplanes like ours can use a very simple timing setting where the engine stays at one setting most of it’s airborne operation.

.

Total advance on a Corvair flying on 100LL is 32 degrees. On 93 auto fuel it is 30 degrees.  Total advance is Static + the Mechanical (centrifugal).  Cars have both mechanical and vacuum advance, but in planes we just use mechanical advance. The engine makes it best power at these settings, with a good margin protecting it against detonation. Why not leave it at these all the time? It would be very hard to start.  While you can say we ‘advance the timing for full power’, you can use different language to say ‘timing is retarded for starting.’  It is saying the same thing.

Here is the critical point to understand: Having the timing set too far advanced, to say 38 or 40 degrees, is the easiest and fastest way to destroy the engine. This isn’t just true for Corvairs, it is true for all gasoline engines. The specific number has a lot of variables, but on any engine where you advance it too far, it will detonate. On a Corvair, the first thing to go are the head gaskets. (This is actually a good thing. On most car engines, it is the cast piston that breaks up first, leading to a stopped engine. A Corvair will run with several blown head gaskets and make about 60% power. the forged pistons we use will suffer detonation but stay in one piece. )

Why do some people get away without setting the timing for a while? How fast the engine is damaged is a variable. Lets say a guy never checks the timing and it is accidentally set to 42 degrees. On his first flights he uses 100LL and it is 50F outside. 180 seconds after starting the take off roll he is at pattern altitude and throttling back. the CHT is only 285F. He will get away with this. Change one variable, and bets are all off. Make it 80F outside, add a passenger and lengthen the climb to 300 seconds at full power, or worse, load the tank with auto fuel, and it will start detonating. Now, it isn’t likely to do it in the first 60 seconds, it is going to wait until the CHT climbs through some threshold. This is how people can fly the first 20 or 30 hours on a plane with the timing never checked, and then one day scare the shit out of themselves (and their passenger) when the engine suddenly starts losing power when they are at 300 feet climbing out. This is entirely preventable by just checking the timing with a light. If the same guy just took 10 minutes to set the timing back to 30 degrees total, he could run the engine at full power, fly at gross weight, at any OAT, any time he liked with no damage to the engine. Seems simple, but 1/5 of builders will not do this. I would like to tell you why they don’t, but I can’t. My expertise is aircraft maintenance, not mental illness.

.

Above a 2008 picture from my shop. on the left is my 1950 Allen distributor machine. I actually first used this machine to build drag race distributors at Speed World of Union, N.J.,  from 1982 to 1986 when Englishtown was my Oshkosh. After I got into aviation, I went back to my original ‘alma mater’ and bought the machine  from Speed World’s owner Ron Murphy. I have been working with distributors for a long time.

.

Second critical point to understand: The engine must reach total advance BEFORE it reaches the static rpm of the propeller.  If you tie the tail down and run the engine to wide open throttle (WOT), you will be at ‘static rpm.’ All the distributors we build are internally modified and test run to make sure they have all of their advance in effect before 2,600 rpm. (The exact number is written on the bottom of each distributor after it is tested.)  Go back and look at the graph above. If a KR-2 builder used a car “curve” like the one pictured and he set the timing to 32 degrees at the plane’s static rpm of 2,700, he would then take off and have the engine wind up in flight, but the actual timing would advance to 40 degrees and the engine would be in the detonation zone. Although I warn people against doing this exact thing in my manual, a KR-2 builder from out west, used a regular car distributor destroyed his plane on the very first flight doing this. I later confirmed this by testing the distributor. Believe it or not, the guy wanted it back unmodified, because he was building a second plane and was going to use it again. As I said before, I would like to tell you why they thought it would work better on round two, but it isn’t in the scope of my expertise.

Third important point: Every distributor we sell has three numbers on the bottom that you need to know. They typical look like 700-20-2350. The instructions explain this also, but many people don’t read them. Maybe 10% of they builders I visit with a running engine or a flying plane have these numbers written down in their log book or notes. A great number of builders actually paint over them before reading the instructions, which requires sending the unit back to me so I can run it through the distributor machine to reestablish the numbers. Here is a hint: if a builder calls me up with a running engine and wants to know how to set the distributor, all I have to do is ask him what his three numbers are to know if he read the 9 pages of installation instructions or just threw them out. (It also gives a lot of insight to how closely he followed other directions.)

If we made 3 numbers for the graph above, they would be 1250- 18- 3750. Our first number is the highest rpm that you can have with no advance. The second number is the degrees of advance, and the third is the rpm where all the advance is in by.

Fourth important point: The static advance and the high rpm that all the advance is in by are not critical, only the total advance is.   OK, lets look at some samples; A 750-16-2200 distributor would yield 16 degrees of static when it was idling below 750 rpm, if the total was set with a light at 32 degrees with the engine running above 2200 rpm.  What about the 8 degrees static? That was just to start the engine the first time and get the break in run done. If the guy never when back and set it with a light, he would only have 24 degrees total, and he would make his 100hp Corvair into an 85 HP one, maybe shaving 300 FPM off the rate of climb. I have only seen that done 40 times or so. At least the engine is in no danger of detonating. I have actually had people decide their didn’t make enough power and remove it from the plane without setting the timing. I have also had people cut down their wood props until the static rpm came up, and only then set the timing with a light and were now greeted with a prop that over sped. Timing lights cost money, but not reading instructions can cost a lot more.

.

Above, a closer look at an E/P distributor in the machine ( circa 2008). The machine has a large electric motor inside that spins the distributor. The silver degree ring is fixed, but the spark is projected onto the inner black rotating disc. The motor is variable speed from 0-6,000 rpm. I can easily see the exact advance at any rpm. The machine is equipped with coils and plugs, so the ignition have a load on them. The machine is fully instrumented for dwell, and the points in every distributor come pre-set. Using this machine I can make sure that the two systems fire at exactly the same time.

.

Next example: 650-24-2400. This is a good hand prop distributor for a retrofit on a Pietenpol with a blower fan. The engine will be easier to prop because if the total timing on this distributor is set to 30 degrees, it will only have 6 degrees static when the owner is propping it.  Once the plane starts, the idle will typically be 750-800, and some advance will come in, and the timing will be 8 or 10 degrees and smooth out. A Corvair will idle good at 8 degrees better at 12-16. The number is not important, only the total advance is critical. Likewise, it doesn’t matter if the full advance comes in at 2000 or 2650 rpm, just as long as you set the timing with a light to the total advance ABOVE that rpm.

Why not make every distributor the same? Inside the distributor their are 6 different pin holes in the plate, 9 different advance weights, 12 different advance cams, and 3 different springs that can be used in several combinations. If you didn’t get an “A” in statistics, let me share:  it is over a million permutations. With 24 years of experience on the distributor machine I have it down to 60 effective and useful combinations we use, but this is why every singe one of the is verified on the machine by me personally.

If a builder reads and follows the directions, he has mastered level 1). If he reads, considers and understand this story, he has moved his understanding up to level 2). Does he need to know more than this to effectively use the engine? No, but if he would like to know far more, it is one of the things I have a good understanding of in engines. This did not come from years of being a mechanic. The further understanding came from a number of years in Engineering classes at Embry-Riddle, Particularly the Chemistry classes. While the subjects we studied were academic examples for almost all of my younger classmates, I was 26-28 years old then, and the information was enlightening when I had a sudden understanding of combustion dynamics that I had observed for years in automotive racing, but didn’t have a detailed view of how the factors worked together, far less that you could make calculated and predictable changes.

Many factors are at work in the cylinder as the piston comes up to the ignition point. RPM, Charge temp, Density, surface area, compression ratio, the rate that the mixture is being compressed at, the octane of the fuel, how well it is atomized, how homogeneous the mixture is, hot spots in the chamber, angle of the plug, and the big one, how turbulent the flow in the chamber is. All this comes into play before the spark ignites the mixture. As the flame front propagates, a complex new set of variables happens as the pressure in the cylinder and the heat goes up, but the mixture becomes diluted by byproducts of combustion. These are interesting to the design of the engine, but you need not be conversant in them to use the engine and be its Master.

Over the years, modifications we have done and recommendations we have made have come from this category 3) understanding. These tests verified on our own dynamometer, confirmed many of the things we do today as common practice. For example, the tight quench area of 2,850 and 3000 cc pistons was solely to improve the turbulence in the cylinder, which in turn raised the rate of reaction, and gave us full power with far less ignition advance. The dished pocket on these same pistons was to lower the static compression to suppress detonation. This type of work is a very good example of why I like to call the Corvair an “Automotive Conversion engine”  It is far more than a car’s engine plucked from a modern wreck or a roll over. Yes, the origins came from the labs of General Motors, the worlds largest corporation, but we have gone to great lengths to tailor this engine to make it an outstanding aircraft power plant. It makes todays aircraft builder the beneficiary of a tremendous amount of work. All he has to do to utilize this is read the directions and use a timing light. -ww