[chiphead]

In theory, I agree with the "breakaway torque" discussion on a properly torqued fastener without a gasket, like a main journal bolt or something. A clamped joint is held together by the bolt, which is stretched by being torqued to a given value. The amount of rotational torque applied to the bolt gives a predictable amount of stretch, and thus a given clamping force for thread pitch, bolt diameter and bolt material. When a bolt is torqued to X value, it will require more than X value to further turn the bolt against friction in the threads and under the head.

This is why the type of lubricant on the threads is critical when torquing bolts. ARP suggests the use of their lubricant, since the amount of torque with lubed threads will more reliably produce the expected amount of stretch in the fastener.

However, when re-torquing head gaskets, you are RE-tightening fasteners that could have less than the correct torque value when you re-torque them. Torquing to the correct torque (without loosening) should properly stretch the bolts that are at less than rated torque. The loose bolts will be readily apparent by rotating but the tight fasteners will simply "click" when you torque with a clicker style torque wrench.

When retorquing, I watch for bolts that never seem to stop turning, they just keep going and going as you try to reach rated torque. Those bolts are likely stretched beyond yield and may break in half. I remove and replace those.

If you back the bolts off first, you will have no idea if the bolt was near rated torque or not. I do back off head bolts if I suspect that the original installer did not jump around and tighten the head bolts in the correct pattern, which does not evenly clamp the gasket.

If re-torquing does not shut the leaks up, it's time to pull the heads and see what's up. Could be too long bolts, improper deck surface (pits, etc), uneven deck, cheap head gaskets, fatigued bolts, bad install procedure, dowel pin issues, etc.

my .02, anyways......