T Camshafts Testing, Part I

During the past several years I have heard many complaints from Model T owners about the lack of performance from their cars. Most of these cars had newly rebuilt engines. The complaints centered on a lack of low speed pulling power and acceleration, not top speed performance. I have worked on several cars that suffered from poor performance and could never find anything wrong with them. I have also noticed that cars with a stock camshaft in good condition seemed to perform better than cars equipped with a modern reground cam. This was just a seat of the pants feeling, since I had never done any type of testing. I do not have a large camshaft collection or access to an engine dynamometer. Therefore, there seemed to be no way of ever testing different camshafts with different engine configurations. A good answer to this question seemed impossible.

Early last year I mentioned to Larry that I felt there was a definite lack of performance from modern reground Model T camshafts. Larry said that he had software for his computer that would simulate torque and horsepower curves for an internal combustion engine. I really was excited because suddenly the possibility of testing camshafts looked feasible. About a week later Larry told me that he had tried to simulate a Model T engine without much success. The program would not simulate an engine running at less than 2000 rpm. Thus it could not simulate a stock Model T engine that develops its maximum horsepower at 1,600 rpm. I was disappointed but felt that testing cams on a computer was possible, if we could find better software. I felt it should be possible to simulate a Model T engine with a stock cam, then rerun the simulation with different cams and get different results.

Several weeks later I found out that a friend named Terry Shaffer had bought a software package called Engine Analyzer by Performance Trends Inc. This software simulates an engine dynamometer test. Terry and I experimented with it and decided that this software might do the job. I put on my salesman hat and persuaded Larry to buy a copy and help me with the testing. When Larry started using the software, he discovered that extensive engine airflow data was required to create an accurate simulation. This is something that Terry and I had noticed in our experiments but did not think would be a problem. However Larry felt that getting this type of data for the Model T engine would be impossible. It looked as if we were back to square one! Then Larry found an article written by Wayne Atkinson and published in The New Secrets of Speed magazine. In the article Wayne published airflow bench data he collected on Model T intake manifolds, carburetors, and valve ports. This was just the data we needed!

Using Wayne’s data, we began to build the simulation. First, we collected timing and lift specifications for twenty-six camshafts that could be installed in a Model T engine. Included in this collection are stock Ford factory cams, antique T racing cams, and modern reground cams. Also included are stock Ford Model A, B and C cams, and modern reground Model cams. Model A based cams are included in the collection because they can be modified to fit a Model T engine. Table 1 provides the specifications for the cams tested. You will notice that we decided not to include the brand name of the Model T and Model A cams that are currently available today.

Next we input the specifications for the stock Model T cam that was used from 1913 until the end of production in 1927. Larry ran the simulation and compared the results with horsepower and torque curves published in 1919 from an engine dynamometer test on a stock engine. Various parameters in the simulation were modified until the simulator results matched the 1919 curves.

Each cam was tested in four different engine configurations. We did this to determine if the cams performed differently as engine performance increased. The first configuration is the stock engine produced from 1913 to 1927. The next two configurations are modified engines using a flat cylinder head. The last configuration is a modified engine using a Frontenac SR OHV cylinder head. The Table 2 below outlines the specifications of each configuration.

Table 2, ENGINE SPECIFICATIONS

The air flow data for the modified engines were extrapolated from Wayne’s data for the stock engine. Other data such as valve port friction coefficients were developed from guidance provided with the software.

Since all cams produce their maximum torque and horsepower at a particular RPM, it is necessary to be able to relate engine RPM to road speed. Table 3 shows the relationship between engine RPM and road speed. For example if you find 40 MPH in the Road Speed column, you will see that the next column shows that the engine is turning 1627 RPM in high gear if the car is equipped with 30 X 3 1/2 Inch tires and the standard 3.63 to 1 rearend ratio.

In the stock engine horsepower varied from 22.5 to 16.6, a range of 5.9 horsepower. Torque varied from a high of 84 foot pounds to 53.6, a range of 30.4 foot pounds. The maximum horsepower was produced at 1600 to 1700 RPM while maximum torque was produced at 900 to 1600 RPM. The test results revealed that the antique racing cams produced more horsepower and more torque than the stock Model T cams. The antique cams also produced their maximum torque at higher RPM than the stock cams. Most of the modern Model T regrinds produced about the same or slightly more horsepower than the stock T cams. However, all the modern regrinds made less torque than the stock T cams. Some of the modern cams made 30 foot pounds less torque than the stock cams. A car equipped with any of these cams would have noticeably less acceleration and hill climbing ability, but top speed should be about the same as a car equipped with a stock cam

Both the stock Model A and B cams produced slightly more horsepower but a little less torque than the stock Model T cams. Performance on the road with an A or B cam would be about the same as performance with a stock cam. The modern Model A or B regrinds produced about the same or slightly more horsepower but significantly less torque than the stock Model T cams.

We felt that we should determine an overall ranking for each cam. Table 4 is an attempt to do this. We decided that the overall ranking should be a composite of the torque and horsepower rankings. However, since the variance in the torque range (53.6) is greater than the horsepower range variance (5.9), we decided to favor in the ratings' cams that made more torque. We did this by multiplying the horsepower ranking by 20% and then adding the rankings together. The larger the overall ranking number the lower the ranking. As you can see the Muskegon antique racing ranked number one with 1.8 points and the .290 lift T Racing cam rated last with 31.2 points.

Table 4, CAM RANKING FOR A STOCK ENGINE

In the Stage 1 Modified Engine horsepower varied from 28.1 to 21.7, a range of 6.4 horsepower. Torque varied from a high of 101 foot pounds to 66.5, a range of 34.5 foot pounds. The maximum horsepower was produced between 1,600 and 2,000 RPM while maximum torque was produced at 1,000 to 1,700 RPM. The rankings for the Stage 1 configuration engines are about the same as the stock engine configuration. The antique racing cams produced more horsepower and torque than the stock Model T cams. All the other cams produced about the same horsepower but less torque than the stock Model T cams. From Table 5 you can see the Muskegon antique racing was again ranked number one with 1.8 points and the .290 lift Racing cam rated last with 31.2 points.

In the Stage 2 Modified Engine horsepower varied from 41.6 to 34.9, a range of 6.7 horsepower. Torque varied from a high of 122.4 foot pounds to 85.6, a range of 36.8 foot pounds. The maximum horsepower was produced between 1,800 and 2,300 RPM while maximum torque was produced between 1,500 and 2,100 RPM. In the Stage 2 engine configuration, the antique racing still performed the best. The stock Model A and B cams also performed well. The .270 Lift T Touring modern regrind cam also did well. Several of the other modern grinds moved up in the ratings but most continued to perform poorly. The stock Model T cams produced good torque but the horsepower falls off as the rpm’s increase. Table 6 shows that the Laurel (Roof) antique racing cam and the Muskegon tied for first place with 2.4 points, and the .290 lift Racing cam again ranked last with 31.2 points.

In the Fronty OHV engine horsepower varied from 96.3 to 84.1, a range of 12. Torque varied from a high of 122.4 foot pounds to 85.6, a range of 36.8. The maximum horsepower was produced between 3,600 and 4,250 RPM, while maximum torque was produced between 3,000 and 3,500 RPM. The surprising result from these tests was the torque produced by the two stock Model T cams. The early cam ranked number 1 while the later cam ranked number 3 in torque produced. This may explain why many Fronty Fords used the stock cam. These cams probably performed well on short tracks with sharp corners where a lot of low-end torque was required to exit the corner quickly. In the Fronty OHV configuration, the antique racing cams ranked 1, 2 and 3, while stock Model A and B cams ranked 4 and 5. The stock Model T cams ranked 6 and 7. The rest of the cams were "also ran’s". Table 7 provides the detailed rankings. The Muskegon antique racing cam again ranked number 1 with 5.2 points and the .290 Lift Racing cam again ranked last with 36 points. Since the range between torque and horsepower was not as great for these tests, the weighting factor was changed to .4. This change reduced the advantage of cams that produced the greatest torque.

Therefore, what should you do if you want to build a good performing Model T engine? First, avoid most of the currently available reground Model T cams. For all the engine configurations tested it would be better to use a stock unground Model T cam in good condition. A good alternative is to install a stock Model A or B cam. If you must use a modern reground cam select one of the .270 lift T Performance, Driver or Touring cams. The .270 Driver works best in the Stock and Fronty engines, while the .270 lift Performance works best in the Stage I engine. The .270 lift Touring cam works best in the Stage II engine.

The get the best performance find a shop that will grind a cam to your specifications. For modified engines grind the cam with lift and timing specifications like the antique Muskegon or Laurel (Roof) racing cams. If you are building a stock engine, get the cam reground to the stock specifications with .270 to .300 lift. The added lift will produce about two extra horsepower at 1600 rpm and additional torque above 1000 rpm. For street driving the cam ramp angles should be moderate to reduce cam and lifter wear. The cam lobe should not be ground to a point. The top of the lobe should have at least a 1/8-inch radius. Anything less than 1/8 inch will wear down rapidly and reduce the lift of the cam.

Why does the stock Model T cams and the old racing cams perform so much better than the modern regrinds? Look closely at Table 1. The table has been sorted in descending order by the opening of the intake valve. Notice that the best performing cams are near the top of the table and the worst performers are near the bottom of the table. In the best performing cams, the intake valve opens after the piston has passed top dead center. It is very apparent that there is a direct relationship between the opening of the intake valve and the amount of torque produced by a Model T engine. The best performing cams also have little or no overlap between the closing of the intake valve and the opening of the exhaust valve. In other words, the valves are not open at the same time. This type of valve timing is completely contrary to the timing of modern camshafts. Modern cams open the intake valve before the piston reaches top dead center. Modern cams also have many degrees of overlap. The number of degrees that the valves are open (duration) is also much greater on a modern cam than on the antique cams.

I am not a cam or engine designer, so I cannot give an expert answer why intake valve timing appears to be critical to good torque performance. However, I believe the main reason is the extremely low amount of airflow through the Model T engine. The restricted air flow conditions require very high manifold vacuum to move air into the engine. Opening the intake valve early reduces the vacuum pressure. This in turn reduces the amount of air sucked into the engine and performance suffers.

Cams are usually installed in an engine "straight up". This means that the cam is installed without any advance or retard built in. The timing marks on the crankshaft and cam gear line up. However, it is possible to improve the performance of some cams by installing the cam several degrees advanced of retarded. Do this by slotting the guide pinholes in the cam gear. The slotted holes will allow several degrees of adjustment. Advancing or retarding the cam gear one tooth will move the cam 7.5 degrees. On of the features of the Engine Analyzer software is a provision for advancing or retarding the cam. The top performing cams listed below were retested with 7.5 degrees of advance and then 7.5 degrees retard.

Early Stock T, Late Stock T, Model A, Model B, .270 Touring, .270 Driver, .270 Performance, .275 Lift T, .300 Lift T

All the cams retested produced more low-end torque with the cam advanced 7.5 degrees (one tooth). However, peak torque and peak horsepower was reduced slightly. In addition, peak torque was produced 100 rpm lower for most cams. The .270 lift Touring cam, the .270 Driver, and the .270 Performance cam produced even more low end torque when advanced 15 degrees (two teeth). Peak torque and peak were also reduced more. Retarding the cams produced less torque and fewer horsepower.

In Part 2 of this article Larry presents the design details of the simulation. He points out the problems solved to make the simulation work. In addition, he admits that solving these problems may have effected the accuracy of the results. Please remember that each cam was tested under the same conditions. This means that differences in performance between cams are a result of the timing and lift specification of the cams. The numbers may not be absolutely accurate but the difference between cams is real. The detailed horsepower and torque results for each cam in the four engine configurations are also provided in Part 2.