Case Study, 1959 Porsche

By Scott R. Tucker

 

Design of ignitions system for MBT (mean best torque) in a four cylinder air-cooled engine.

 

Vehicle:

1959 Porsche 356 Outlaw

Engine:

2.5 liter flat-four

                Naturally aspirated

                2 valves per cylinder

                2 spark plugs per cylinder

                Points type ignition system with mechanical advance only and dual coils

                Dual Weber carburetors

Symptoms:

Engine exhibited strong misfire and poor acceleration in 1500 – 4000 rpm range.  Above 4000 rpm, it pulled smoothly although labored.  Engine idled normally, torque curve was not smooth, and throttle response was poor.

Initial observations:

Misfire did not feel to be fuel related.  I checked the engine on ignition oscilloscope and gas analyzer.  Ignition patterns at idle were unremarkable except for minor glitches at the points-closed position.  Increasing revs increased severity of glitches.

Initial gas analysis was as follows:

At idle:  7% CO (carbon monoxide) and 2000 ppm (parts per million) HC (raw hydrocarbons)

At 3000 rpm:      6.5% CO and 1300 ppm HC

Mixture at both idle and 3000 rpm are both determined to be excessively rich, but lean enough that it should provide sufficient drivability.

Initial repairs:

Because of the glitches seen on the scope at the points-close position, it was suspected there was an issue in the primary ignition system.  Inspection of the points found them mildly pitted.  Dual condensers were fitted.  First, I replaced the points and condenser.  The engine now ran well, but it had a poor torque curve, and had incipient detonation between 2000 and 3000 rpm.  The detonation was detrimental because driving in-town would favor the lower RPM range and thus would be prevalent.  Oscilloscope patterns still showed glitches at the points-closed position.

After further test driving, it was found that the misfire was slowly returning.  The points were again found to be mildly pitted.  I measured current at the points and found that they were conducting 9 Amps @ 6 volts (due to ballast resistors).  This is approximately double what a normal set of points would see and was contributing to the burning of the points.  The torque curve of the engine was poor due to the incorporation of mechanical advance.

Solution:

There were two problems which needed to be solved here.  First, the ignition system needed to be reliable, which it was not.  Second, the engine needed to have good drivability around town as well as ample power on the freeway.

The MSD 6AL-2 Programmable ignition system was chosen because it would provide low current flow through the points for reliability, as well as a programmable ignition curve, which is superior to any mechanical advance.

In the time before computers and electronically controlled ignition systems, the breaker point ignition system with mechanical advance was the standard in the performance aftermarket.  For its day, it was reliable, simple, and powerful.  However, due to its mechanical design, it could not provide an optimal ignition curve.  Mechanical advance incorporates weights that are thrown out by the centripetal force of the spinning distributor shaft.   This caused the breaker plate to advance the timing.  Timing increases at a linear pace in direct proportion to the engine RPM until a mechanical stop is reached, which ceases any further timing advance.  This produces a curve similar to the one seen in figure 1.

 

1

 

 

 

 

 

Figure 1 – Typical mechanical advance timing curve

 

Unfortunately, this advance curve is far from optimal.  There is a lot of power which has been left unclaimed due to this imperfect timing curve. Every single time an engine runs for 1 revolution, its requirements for fuel and spark change, even if just an extremely small amount.  In the mechanical advance curve, timing does not follow that which is dictated by rpm and volumetric efficiency.  A more correct curve can be seen in figure 2.

 

2

 

 

 

 

 

Figure 2 – Optimal ignition timing curve

 

You can see that the curve looks more like an ‘S’ now.  This is what the timing requirements of the engine are for maximum power.  Due to the mechanical nature of the distributor, a mechanical advance cannot recreate this curve.  A programmable ignition can.  By adding a programmable ignition, we were able to gain both low end torque, and high end horsepower.

If you overlay a mechanical advance curve and the programmable advance curve (Figure 3) you will see the areas on which improvements can be made.  All of the shaded areas represent spark advance on top of what a mechanical advance would have provided.  This translates into increased efficiency and increased torque.

 

3

 

 

 

 

 

Figure 3 – Mechanical advance/MSD timing curve overlay

 

Conclusion:

The redesigned ignition system solved both requirements of increased reliability and increased drivability and torque.  The breaker points now only conduct a very small amount of current and the current draw by the coils is well within the limits of the MSD unit.  The mechanical advance has been locked out, therefore it cannot wear out.  This should provide good service life.

The programmability of the MSD box has allowed us to create an optimal spark curve.  I estimate that torque was increased 5 – 10 ft/lbs in the 1500 – 3500 RPM range and 5 – 10 ft/lbs in the 5000 – 7000 rpm range.  This would correlate to an increase of 10 – 15 horsepower.

In the world of street cars, drivability will always trump horsepower and torque and I think this was the biggest gain on this car.  The car now has great drivability.  It is responsive around town and very easy to drive.  Yet when you wring it out on the highway, it pulls hard up to its maximum rpm.  The optimized spark curve should provide increased efficiency too, which translates into better gas mileage.

I have not always been a fan of MSD’s products, but this ignition has helped that.  The software interface for this system is terrible, but once I was able to figure it out it was rather easy to program.  My past judgement of MSD always had to do with their reliability as I’ve had to change quite of few burned out MSD boxes throughout the years.  However, in addition to being more reliable than points ignition by default, I have seen fewer and fewer failures in the last 10 years.  I trust this can be attributed to the reduction in price of high temperature integrated circuits and refinements in thermal control of modern microprocessors.

 

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>