Stock vs Regrind

In order to check out the results from the engine simulator (see simulation), we replaced the cam in a 1926 Tudor sedan. The engine was a recent and typical rebuild by a reputable engine rebuilder in this region. It had modern valves and aluminum pistons, but was otherwise stock. We estimate the engine had about 2,000 miles on it, so it was fully broken in. On the advice of the builder, a 290 lift cam was installed during the rebuild. He called this a touring grind - the cam installed in the majority of his rebuilds. The engine builder knew the lift, but none of the other cam specifications, e.g. duration.

The car was first dyno tested by two methods: (1) a local commercial chassis dynamometer and (2) our Cheapo Dyno. The Cheapo Dyno performs calculations on data collected while accelerating the car (this program may be downloaded for free). The results of the tests are shown on the following graph. Keep in mind that a chassis dyno measures horsepower at the rear wheel, so drive train friction loses are reflected in the measurements. If the engine produced 20 horsepower and the drive train friction loss was 25%, then a chassis dyno would measure 15 horsepower, i.e. the peak horsepower measured here. The power we measured on this car is somewhat higher than that reported by Milt Webb and Carl Amundson in Vintage Ford Vol. 12, No. 3, p.43.

The chassis dyno results are the average of three runs, and there was no measurable difference in power between the use of high gear (Ruckstell high) or Ruckstell low. Since the cheapo dyno takes measurements while driving the car, we must account for wind resistance or drag in order to compare the results with the chassis dyno. A drag coefficient-frontal area product of 28 square feet produced the results shown in the plot. The good comparison between the high gear and Ruckstell performance suggests that this value is reasonable. There is agreement between the chassis dyno and cheapo dyno in peak horsepower, but some discrepancy on the low end. The low RPM performance from the chassis dyno looks suspicious, since there is no peak in the torque curve down to 600 rpm. A stock Model T produces peak torque at 900 rpm. For the typical regrind, the simulator predicts the peak torque to occur at somewhat higher RPM. The cheapo dyno predicts a peak torque at about 1100 rpm, in agreement with the simulator. Despite this discrepancy, both dynos gave reproducible results on multiple runs. There were typical fluctuations on the order of a half horsepower, which we attribute to variations in manual advance of the spark lever. Either dyno should be adequate for evaluating camshafts, provided that the differences are at least one half horsepower.

After dyno testing with the 290 cam, we installed a degree wheel on the crankshaft and pulled the head so we could use a dial indicator on the valves. We also measured all the valve clearances. This equipment produced the measurements summarized in the table below. The 290 cam had a seat-to-seat duration of about 264 degrees, 46 degrees more than a stock Model T duration of 218 degrees. The duration at 0.050 lift was 232 degrees compared to 196 degrees for a stock cam.

Intake

Cam	Open (BTC)	Close (ABC)	Duration	Valve Clearance
290	8	76	264	.022,.020,.020,.017
Stock	-12	50	218	.021,.026,.025,.023

Exhaust

Cam	Open (BBC)	Close (ATC)	Duration	Valve Clearance
290	61	23	264	.020,.020,.015,.022
Stock	46	8	234	.013,.011,.012,.010

Next, we removed the 290 cam and replaced it with a NOS Model T cam, which was contributed to the project by Steve Coniff. Measurements with the degree wheel and dial indicator gave excellent agreement with the lift curves calculated from the specifications for a 1912+ stock Model T cam (see Cam Design). We installed the NOS cam as shown in the table. The intake valves were set to give the stock duration of 218 degrees, whereas, the exhaust valves were set to give greater duration. This method of installation was suggested by Steve Coniff.

After installing the NOS cam, the car was again dyno tested. The results for the two cams are shown in the graphs below. The results with both dynamometers agree that the NOS cam produces 10 to 15% greater power up to about 1300 rpm, and essentially no difference in power above 1500 rpm. This change improved the car's acceleration and made it into quite a good hill climber. We would expect this result from the discussion of power and torque.

These dynamometer tests confirm the conclusion reached using the simulator, i.e. 264 degrees duration is far too large for a Model T engine. However, the simulator predicted a much greater difference in power than we observed in this test.