Tech – A ‘Knock Knock’ That Was No Joke

From Moss Motoring 1985

A knock in a worn engine is relatively easy to diagnose, but-what about a knock-in a freshly rebuilt engine? Especially one which has been assembled with great care in respect of bearing clearances and piston fit? We recently came across a very annoying problem which had an unusual cause…

Subject: An MG TD engine which had been rebuilt from the bare block up. Virtually all moving parts had been replaced and the engine had, of course, been bored oversize and the crank ground undersize. Re-installed in the car, the engine had good oil pressure, started easily and ran quietly. After the initial period of fast idle (to let the new cam and lifters seat), the engine proved crisp and responsive when revved. Valves were re-adjusted head nuts re-torqued, and the carbs adjusted. Smoother and quieter than before! So, test drive time.

Problem: About three minutes into the test drive, a rapping noise begins. The car pulls over and the noise goes away while its source is being sought. Everytime the car is driven, this pattern recurs. After a fifteen minute drive, the noise becomes so loud that a chase car following (in case of emergency!) can hear the knock! Again, it goes away within a minute or so of the car sitting at idle. Oil pressure and water temperature were normal throughout this exercise. Cause: We won’t bore you with details of all the attempts that were made to find the source of the knock. Suffice it to say that several partial tear-downs found nothing, so the car was driven for a couple hundred miles to see what happened. The theory was, it would either go away or something would break but, of course, neither happened. When the engine was taken out of the car and dismantled completely, the cause finally came to light.

It was discovered that the pistons were just hitting the cylinder head when the heat generated by driving under load caused the pistons to expand. Apparently, the cylinder block had been surfaced once too often; the pistons now protruded and the compressed head gasket was not thick enough to prevent them reaching the head when they expanded.

Moral: Always check whether your pistons protrude from the cylinder bores when you have had an old block surfaced. If they do, have a machine shop turn off the top of the piston on a lathe, tapering from the outside edge (as many thousandths as necessary) to the center of the piston (no metal removed). The alternative is to run two headgaskets cemented together with a good gasket compound.