|
Flow Specialist, Cammer, Want-a-be, Mach 9 and Lasse,
My guess here is that in spite of sounding contradictory you are all right!
I like to have all my reference surfaces accurate so I know where I am. This means at some point of the game you do know what the hell is going on. But the truth of the matter is that under the heavy loads seen in a high output race engine things move - and they move a lot more than you may suppose.
Here are some examples: While working with some of my racer friends in UK some years back (about 89) we had occasion to rebuild an 850 A Series Mini Crosser race motor. This engine powered a car that was kicking butt at almost every meeting so won the regional championship and qualified to run in the final. No good being second here it was all or nothing. So my friends rebuild the engine and during the rebuild found that the center main was almost 0.003 thousandths out. The guy that built the engine had not checked for this at the time of the original build and it is doubtful that the block, which was well seasoned, could have moved that much in a six month period. Anyway the block problem was fixed and the engine rebuilt to the exact same spec. On the dyno the rebuilt power curve virtually overlaid the original. No extra power was seen.
A similar case was some tests that my friend, Paul Ivey, one of UK's most famous Mini engine builders did with the factory. They progressively bent the crank until a power drop was seen. It took just over 0.003 bend in the crank to show a drop in power on a 28 hp engine.
I also heard, but cannot confirm, that Cosworth did some experiments here on the old and highly successful DFV F1 engine. So the story goes, they measured the block distortion of an engine running WOT at near max rpm. They then machined each main bearing housing off by that amount reversed so that, in theory; all the main bearings would be inline at 10,000 rpm. It made no difference to the output!
You may ask why - well here is at least some insight to this. Let us assume that the bearing, when supported by oil, has no friction. This means that there is no torque lost due to bearing friction (load times friction coefficient). At this point let us push the main bearings out of line by say 0.004. That is enough to bind them up during the build. But an oil film is incredibly stiff and can support a very sizable load. This oil film picks the journal off the main bearings. Under these conditions the friction torque would be the load - we will say 15,000 lbs per bearing times zero answer frictional torque loss = zero. Ok that may be an over simplification but it does make the point. I have line honed blocks which were a little out (0.0015 max) and seen no measurable gain on the dyno.
Does all this mean I put together motors without due consideration to the mains alignment - certainly not. If the mains are at all suspect I have them honed at a trustworthy shop. However If the build is a low buck street deal and the crank rotates like it's on ice and all the clearances are right I don't bother. The money otherwise used can be spent on better parts elsewhere.
At the end of the day it really is a question of knowing where to draw the line (unless you work for a cup car team that is!)
DV
|