With all due respect to Mr. Messina, the "rule of thumb" which advises to use a power valve rated at 1/2 of the idle manifold vacuum may or may not be adequate. A more accurate method is to mount a vacuum gauge to the dashboard and run a vacuum line from the intake manifold to the gauge. Then, drive the car on level ground and have a passenger record stabilized manifold vacuum at 10 mph intervals in each gear up to about 60 mph or so. From this data, you can determine exactly the lowest stabilized manifold vacuum under normal ("unloaded") operation. In all probability, the lowest vacuum will occur at idle, but this way you will KNOW what it is rather than guessing. Based on the lowest stabilized manifold vacuum, select a power valve which is 1.5 to 2.0 inches lower than your lowest recorded stable vacuum. See this link:
http://www.holley.com/HiOctn/TechSer...o/TI-222A.html
Not to beat a dead horse, but power valves can cause a multitude of problems if not tuned to the application. A power valve tester is almost a necessity for really knowing what's going on inside the carb:
http://www.jegs.com/cgi-bin/ncommerc...61&prmenbr=361
FWIW, my money is still on a faulty needle valve.