This is where all MID-LIFT® geometry has evolved from since it was developed ...more than 43 years ago!




Ford's BOSS 429 crossed over into CAN AM racing, displacing 494 cubic inches. It's torque was phenomenal, prompting quite a few of these FIRST aluminum block Ford BOSS engines to be used in pursuit boats, sold to Arab nations; similar to Scarab (Miami Vice) style boats, with two and three engines each. Ford's independent Grand National contractor, Holman-Moody, was behind all these famous variations, and many other of Ford's exotic racing endeavors.

Miller BOSS 429 Series III Rockers (1983). Jim Miller was the FIRST designer who understood the detriment of side thrust harmonics and their loads upon compound geometry valve trains. Jim has used Torrington Thrust bearings since 1983, and implemented them on all PRO-SHAFT NASCAR designs whenever any side load inclines from the push-rods was evident on the head's X-axis. But it was on the extreme angled BOSS 429 where they were developed and used first.

BOSS 429 Ford



Designed in 1968, implemented in 1969, the BOSS 429 was Ford Motor Company's answer to their failed attempt to introduce the Over-Head Cam 427 Side Oiler into NASCAR's Grand National racing, and regain their competitive edge against Chrysler's dominant 426 Hemi. Ford took their "attitude" for overkill engineering originally directed into the 427 SOHC, and took NASCAR's no over-head-cam rule to the recently new block originally designed for the truck and big car market, known as the "385 Series," for its 3.85" stroke crank; but it was more familiar as the "460." It was soon de-stroked to a 3.59" crank and adapted as the 429 wedge, 429CJ, and 429SCJ. All these basic variations followed a common cast iron cylinder head canted valve array, including a few quirk aluminum head castings based on the SCJ. They were very similar to the Chevy 427 combustion chamber, except all intake ports were parallel designs (not symmetrical - left and right variations, as the BB Chevy). The BOSS 429 "head" was adapted to this engine block, but only after some oil galley and deck transfer holes were engineered to be precision machine for the novel individual gasket rings. The BOSS 429 sported many conceptual ideas of airflow from the 427 SOHC, and 427 TUNNEL PORT, mainly round intake ports, but HUGE (for their time); a hemispherical combustion chamber, that was "clipped" on each side just a tad, creating a design that was coined "semi-hemi" and "crescent" -- and later became known as the BLUE CRESCENT Hemi. Where Ford really got complex, was their initiative to introduce a flow technology that had been briefly tested on more exotic engine designs, but never an American OHV engine. It was later coined as "swirl" technology. One of the concepts to swirl technology is to purposely place the valve angles and their related port windows feeding them (as well as the port shape) in such a manner that a preset direction of flow is used to do more than just fill the chamber. It induces a cross-flow over the piston and to move across to the exhaust valve which sets up a circular pattern of chamber burn and purging. The aim is more efficient flame propagation, which wedge chambers are good at, but lack in the cross-flow efficiency to purge as well as a hemi. To accomplish this the BOSS 429 had its opposing intake and exhaust valve axis rotated 26° from perpendicularity to piston wrist pins. This required considerable creative thinking for passing the pushrods through the heads for complex mounting angles of the rocker arms. Unlike the 427 Tunnel-Port, where they simple went straight through the port with surrounding sleeves to isolate the two systems, the valve train was laid over on compound angles, which I have stated many times is the most complex rocker geometry of any American made Overhead Valve (OHV) engine, ever; even to this day, nearly 50 years later.

--jM (Revised, 2016)  ^

The concept of MID-LIFT® geometry began back in 1973, and by 1974 it was finalized into the principles of precision geometry that Jim Miller developed for the Ford's BOSS 429 Grand National NASCAR engine. You don't have to be a "Ford fan" to appreciate that in 1968, when this engine was first taken from the drawing boards of Ford's racing division to a real engine for racing in 1969, it had set a precedent for airflow, volumetric efficiency, horsepower per cubic inch, and sheer, all out "WOW" factor, that no other engine of that era had done. But it was flawed in a couple of small respects.

One, was its valve train. COMPLEX. Meaning: it was a semi-hemispherical combustion chamber, with quenched side walls to contain the cross flow intentions of a hemi chamber, but the intake to exhaust valve's opposing angles were "twisted" on a 26 degree rotation to the piston's wrist pin. The rocker arms were independently mounted on pedestals cast and machined at precise compound angles. But like another engine Ford made the same mistake with, the attachment to the block for pushrod angles was required to work with lifters that were "in-line" rather than laid over for dedicated exhaust and intake rocker locations. Plus, the pushrod's side angle array was heavily leaned away from the tappet centerlines too.  ^

Factory INTAKE Factory EXHAUST

The cylinder head's huge round ports and location, required this twisting of the valve to valve centerlines, but Ford was also chasing induced swirl technology, very new for OHV engines of that period. But to make all the parts fit between where the pushrods would exit through the heads, and the valve tips needed to be, meant that the rocker arm dimensions would be extreme in both directions. The INTAKE rocker was one of the shortest designs used on any American engine, rivaling the small block Chevrolet. The EXHAUST rocker was the longest, exceeding the Chrysler hemi. The pushrods themselves were also among the longest in use, with the exhaust reaching nearly 11" in length.  ^

In 1972, for "Grand National" racing (as Winston Cup style NASCAR was known in those days), the typical 7,800 rpm hemi engines with barely .650" valve lift could accept these crazy valve-train arrays. But drag racing's high rpm, quick shock to the drive train was entirely different. For Jim Miller's de-stroked 409ci B/MP BOSS Mustang, that launched from the line at 9,000 and hit the 1/4 mile traps at 9,300 rpm; no such designs were possible, without breaking something on each pass. This was the beginning of questions for an engine few people understood, compared to the masses of information accrued over the years for Chevrolet and Chrysler engine builders. Rocker arm questions, for instance. To begin designing his own rockers, Jim needed a foundation from how other designs were done. But even questions asked with the common Chevrolet engines, Jim Miller soon learned there was no comparable dimensions for various manufacturers on their rocker arms. Center-to-center lengths for the critical stud to valve, as well as the arcing motion across the valve, were all undecided specifics between individual manufacturers. Rocker arm "height" on the stud (or shaft, as in the Ford 390-428 FE engine) was also a question that no cam manufacturer selling rocker arms would answer, beyond "keep the rocker on the tip of the valve."  ^

The Standard By Which All Is Measured!™


"MID-LIFT" & "PRO-SHAFT" are ® Registered Trademarks of MILLER ENGINEERING INC; Copyright © MMIII--MMXVI JM Miller