Some thoughts on precision, tolerancing, and quality needed for racing engines.........I just finished up the post-mortem measurements on the cylinder head, and have quickly discussed what I found with Chris, by phone. I'm going to post up the measurements & some photos, but before I do that, I have some thoughts. I'm just going to tilt my head and let them spill out.
I've been building performance and racing engines a long time, longer than I'd care to admit. A person can't have a long career in this field, without a track record of success. And no one can be successful, for any amount of time, without the requisite performance AND reliability from their product. Think about that for a minute, and then ask yourself, "What is required to achieve this?"
Inherent in the quest for this performance and reliability is the need to understand precision. The dictionary definition of precision is as follows:
pre·ci·sion
n.
1. The state or quality of being precise; exactness.
2.
a. The ability of a measurement to be consistently reproduced.
b. The number of significant digits to which a value has been reliably measured.
adj.
1. Used or intended for accurate or exact measurement: a precision tool.
2. Made so as to vary minimally from a set standard: precision components.
3. Of or characterized by accurate action: precision bombing.
Precision means different things to a framing carpenter vs. a finish (trim) carpenter. They use different tolerancing values. The same is true for a general machinist vs. the operator of a crankshaft grinder. What does this have to do with racing engines? Only to lead into the idea of "tolerancing". It's pointless to ask carpenters to work to +/-.001", their work doesn't need this kind of precision. But, while +/-.001" may be OK for some CAD/CAM machined parts, this sort of tolerancing is too tight for say, gaskets. For crankshafts, however, (even stock cranks) +/-.001" is too wide a tolerance to be considered precise.
The point of all this is that: Racing engines require differing levels of precision, and tolerancing, in different areas. Quality then, as far as racing engines are concerned, is the elevated level at which precision & tolerancing can be sustained or repeated. For it does no good to fit 3 pistons correctly and the 4th incorrectly.
Some parts need to be sized (or fitted) in thousanths (.001") of an inch, others need to be sized/fitted in ten-thousanths (.0001") of an inch. For other parts, say pushrod overall length, +/-.010" would be acceptable. Parts can be made to almost ANY tolerance, IF, cost is no object. But what is the point? The trick is to know what the acceptable level of precision and tolerancing is for a given part or assembly.
That is what Chris is dealing with here. Parts fitted & assembled outside of the TOLERANCES needed to be workable & reliable in/at the rpm range the engine will operate. The cause here is: valves fitted too tightly into their silicon bronze guides. NOT, lack of valve to piston clearance, or how that valve to piston clearance was measured. Turns out that the #1 inlet valve has the largest stem diameter AND the smallest valve guide inside diameter, giving the least amount of clearance, less than what was required. No big surprise then, that this was the valve that got tagged by its' piston. At high & sustained rpm, even the higher pressure valve springs could not close the valve as it was seizing in the guide. The piston closed it, and the resulting shockwave was transferred throughout the remainder of the valve train, loosening the rocker adjuster. And things went downhill fast from there....... I'm just glad that more damage didn't occur.
A few pages back, (reply #1506, page 101) Chris has posted up the valve to piston clearances, as measured with a degree wheel and a dial indicator with .001" graduations. Determining these measurements, (at the various positions of the piston, both BTDC & ATDC) with clay, would be difficult, time consuming and subjective, rather than precise.( +/-.001") All clay dynamically reveals is the closest a valve gets to a piston. It does not reveal where this occurs, as the dial indicator & degree wheel do. And while we are on the subject, NEITHER method reveals the actual dynamic clearance at 8500rpm (or whatever rpm) under load or backed off the throttle. The static "clearances" used to set-up dynamic valve to piston "margins of safety" are based on several factors. Most notable are valve train mass; v/t moment of inertia; camshaft 'negative' accelerations, rpm range used, etc, etc. Only a spintron can reveal the dynamic conditions likely to be encountered at operating speed in the valve train. Unfortunately, spintron time is a little outside of the current budget. Like I've stated many times before: "It's complicated." AND, additional complications are added when you are trying to determine which came first, the chicken or the egg.
Fordboy
Topic: Milwaukee Midget (Read 3295473 times)