Putting up with intolerance

by Simon Hargreaves

Sometimes the worse something fits, the better it works...

It’s long been a biking myth that in Japanese engine assembly plants, female workers are preferred to males for some jobs on the production line because they have smaller hands and greater dexterity. Having spent time at Suzuki’s vast Takatsuka Engine Plant in Hamamatsu, I can safely put that one to rest; it’s not true. They have plenty of both on the production lines.

But while I was there I saw something else intriguing: as a chap put together piston and and rod assemblies for GSX-R750 engines, he selected the parts from three colour-coded batches, according to weight. So each row of four pistons, rings, gudgeon pins, circlips and rods (which took him around two minutes to assemble) would be matched; effectively blueprinted.

Blueprinting used to be chapter one in the engine tuner’s handbook. It is stripping an engine to its components, then refining them to more closely match the original engine designer’s intent; ie removing the inevitable tolerances and imbalances introduced by variations in manufacturing and production. Blueprinting consists of three main areas: 1) balancing: if pistons, rods, crank, cams and any other reciprocating parts are accurately balanced, less energy is wasted through vibration and bearings get an easier time. Even with a factory assembly check, pistons can vary in weight by a few grammes, rods can be out of true by a similar amount and a crank can be anything up to four or five grammes out of balance. 2) Gasflowing: if all ports and areas of gas flow are smooth and even, there’s less disruption and impedance to getting air and fuel into, through and out of the combustion chamber. Factory production techniques can create thing like burrs from valve seats being pressed in, height differences in the intakes where inlet rubbers meet the manifolds, or overhanging exhaust gaskets – even valve cutaways in the piston crown aren’t always even. 3) Tolerances: if the clearance between moving parts is optimised, less energy is lost through friction. This includes piston/bore clearances (which vary when hot), freeing up any tightness in the cams and crank to make sure they spin freely, tidying up sharp-edged gears, and optimising everything from valve seats to deck heights to valve timing.

Doing all this all means your standard engine will make its designed power – it runs the way it was meant to. But blueprinting is a painstaking, time-consuming and expensive process. And the efficiency of modern manufacturing methods means these gains aren’t always worth the expense, unless you’re serious about racing (the rumour is some World Superbike teams, forbidden by the rules from machining pistons, would buy them by the container-load to ensure they’d get four of identical weight).

In fact these days it works the other way round: some modern engines are put together too tightly, and will make more power if they’re looser inside. Tuners will actually try to build wear into an engine (by polishing crank journals, going up a main bearing shell size, putting more oil clearance around the cam journals etc).

It’s effectively like putting 20,000 miles on a brand new motor. I’d like to see them try that on the production line at Hamamatsu.

shuggiemac
shuggiemac's picture
Offline
Joined: 23/11/2008

malauder wrote:

ie removing the inevitable tolerances and imbalances introduced by variations in manufacturing and production.

Removing them totally is impossible as there will always be some deviation. It ultimately comes down to the accuracy of the equipment you measure it on. To quote a rather deep physics/metrology notion "“Any measurement that you make, without any knowledge of the uncertainty, is MEANINGLESS!”
-Walter Lewin

Navy Boy
Offline
Joined: 12/02/2009

That's deep Shuggie - Real Deep man...

I feel a running-in article coming on!

shuggiemac
shuggiemac's picture
Offline
Joined: 23/11/2008

Sadly Navy Boy its not something deep or earth shattering that I came up with. I have worked most of my professional life in precision metrology (still do) and it is the kind of thing that is every day. I am sure we can become real pains when we challenge people who say that they have made something exactly x mm or made a perfectly round hole. These things exist only in theory though of course in practical terns, who really gives a monkeys?

pittsy
pittsy's picture
Offline
Joined: 06/08/2011

".........who really gives a monkeys?"

Oi! I heard that.

;D

pittsy
pittsy's picture
Offline
Joined: 06/08/2011

I agree with shuggie. I mean look at the circle itself, the definition of. The centre on which the whole thing is based isn't actually there. You'll chase yourself to infinity trying to find it (and measure it). The deep thing is that the perfection of the circle depends on the, in effect, non existence of the centre point. Eh?

shuggiemac
shuggiemac's picture
Offline
Joined: 23/11/2008

oooh we could take this to the n th degree here ! On a measuring machine the only way to return a perfect roundness value is to measure just three points but as soon as you measure a fourth then it is no longer perfectly round - brilliant !!