Cirrus Construction Log: Valve Gear

Created: June 2006


The valves and associated parts are astoundingly numerous. In all, there are twelve different parts for a grand total of 88 items (excluding another 28 commercial fasteners, not to mention the camshaft and gears). When faced with serious repetition jobs, I generally prefer to make a "test item" at the same time as any jigs and fixtures that my cunning plan says will be required. This process frequently uncovers flaws in said plan, leading to (mostly) a simpler and better approach, frequently with fewer jigs! An experienced machinist would probably view this as time wasted, but then experienced machinists are probably not reading this!

   Valve Caps and Retainers
   Rocker Posts and Pivots
   Cam Followers

Valve Caps and Retainers

The valve springs are retained by a turned cap that centers the spring and is held in place by a "C" washer the sits in a recess in the top of the cap, thus preventing the washer from sliding out. These are made by drilling a length of drill rod, then locking the headstock and using the lathe saddle as a shaper with a parting tool on its side as the cutter. The washers can then be parted off to width. It's best to have these on hand before making the valves so the grooves in the valve stems can be machined to suit.


The valves ride in valve guides that will be pressed into the head. Not relishing the prospect of reaming 1/8" holes in 1/2" long sections of bronze, I've decided to use brass tube for the guides. My test item showed that the valve stem diameter for the required close, sliding fit in the nominally 1/8" ID tube was 0.1275". This ruled out fabricating the valves by silver brazing heads onto drill rod stems as described in the first issue of SIC and used by Les Stone with great success in all his four-strokes. Effort wise, I don't think machining valves is much more than fabricating them. The first step was to rough down the 5/16" stainless steel rod to near the final stem diameter by step-turning to about 0.132".

The Cirrus valves feature a 0.100" neck above the head to assist gas flow. In this shot, a cutter with different nose radii is being used to form the neck and radius its transition to the valve head at one end and valve stem at the other. The stem is supported by the travelling steady during this operation to prevent deflection under the cutting forces.

Forming the neck before finishing the stem confers a side benefit. The neck forms a "run out" area for the travelling steady that allows the steady to lead and lag the cutter. This ensures the cutter is fully backed up by the steady fingers over the length of the cut. The stem was turned to 0.128", the polished with 600 grit paper backed up by a steel rule to obtain the required fit in the guide tube.

The valve seat can now be formed by setting over the compound slide 45°. Although doing this, and setting it back for each valve is a lot of effort, it does ensure that the seat will be concentric with the stem. The groove for the valve cap retainer is also cut at this stage. My first plan was to drop valves into the inside of a 1/8" Myford collet to cut the valve clip groove as I had done for the Morton M5 valves. Good plan, but my trial item showed the 5/16" head would not fit into the collet as the 1/4" diameter Morton valves had. So the grooves had to be cut before the valves were parted off. Be sure to touch the groove cutter to each side of the groove after cutting to remove the burr the operation will have raised.

With all the stem work complete, the valve can be parted off (sawn in my case—less waste and less wear on thin parting tools). The valve is then gripped in the chuck for facing. Shim is used to protect the stem and the rear face of the head pushed back against the chuck jaws to provide support during facing. After a cleaning cut, the valve is removed to measure the head thickness and determine how much more needs to come off. Laborious and inefficient, but the foreman does not seem to mind.

Here, at the front, we see two good, finished valves—one complete with a retaining "C" washer resting in the groove. At the rear are a pair of rejects. Of the rejects, the one furthest away has no groove (and no way to cut it now) while the other is perfect except I somehow forgot to face the top of the stem before parting it off. As this surface needs to be smooth for the rocker to ride over, it's not a lot of use except as an ornament.


This must be the most labour intensive item on the entire engine, and there are eight of them to do! Eric Whittle's instructions in SIC said to make them from 3/16" steel plate. The 2-56 tapped hole for the valve clearance adjusting screw is offset 0.015" from the rocker centerline (meaning there are four each, right and left handed). Something did not sound right, so I drew the rocker out in CAD and sure enough, it is not possible to form the 0.175" diameter eye specified without having a flat on the offset side. So I decided to make them from slices of 1" diameter 12L14 rod. Lots and lots of waste and swarf, but 12L14 machines better than the 3/16" barstock I have, so it's sort of justified.

Making the "trial item" and the jigs took five days of evenings. Machining the rockers themselves was done in a single afternoon, despite the number of separate operations required (20, I think). The photos below show most of these steps. I will describe only highlights worthy of mention that may be of use in similar circumstances.


Photo 1

Photo 2

Photo 3

Photos 1 through 3

A pot chuck is used to face band sawed slices of 1" diameter 12L14 to the required width (0.196"), rebate reliefs in either side, and drill centrally for the rocker bush. The depth of these is asymmetric (0.62" and 0.72"). The marking out of each was a waste of time as the limits for the cross-slide were established on the first item and all the rest made to the same readings, ignoring the scribes marks.


Photo 4

Photo 5

Photo 6

Photo 7

Photo 8

Photo 9

Photos 4 and 5

A 0.350" slice is milled off each blank so it can rest in the jig used to drill two holes located precisely in relation to the pivot axis using the mill table. This avoids suffering through an "interrupted cut" as would happen if the holes were drilled before rebating the side countors. The location of the holes is asymmetric with respect to the rocker pivot location, as are the rebate depths, so care is needed to end up with four each for left and right hand use.

Photos 6 and 7

Another slice is removed parallel to the first so the blanks can be aligned horizontal to drill the adjusting screw hole #50 for later tapping. The cross bar of the jig has been drilled to act as a guide that positions these holes from the center line and jig base by the correct amounts to achieve the offset mentioned earlier. Two holes are required to make left and right handed rockers. The hole is drilled blind, only 1/8" deep, so the jaw of the mill vice is in no danger.

Photos 8 and 9

The final fore and aft sizing cuts are made by aligning the blank with the edges of the mill vice jaws, touching the cutter to the blank, then milling off a predetermined depth of slice. Then placing the blank on a parallel, the adjusting screw side is milled to final thickness of 0.087" and the other side (the "nose") is rough milled for later radiusing.


Photo 10

Photo 11

Photo 12

Photo 10

The securing post in the jig is used to align the jig in the GHT rotary table to form the lower radius cut that blends the two holes together. The CAD drawing had been used to determine the limits of travel for a 1/4" cutter and the table stops set to the required positions. This table was shop made as one of my early "training exercises". It does not get a lot of use, but when it is called on, I'm glad to have it.

Photos 11 and 12

The aligning pin from the previous step is moved to another hole in the base of the jig and a different rocker positioning pin that does not protrude inserted. This re-positions the jig relative to the axis of the GHT rotary table to radius the rocker nose face. The finish is good, but will be stoned to a fine finish after case-hardening.


Photo 13

Photo 14

Photo 15

Photo 13

A simple jig with a 2-56 clearance size hole drilled normal to one face is used to align the tap as it is started into the hole drilled earlier (in photo 7).

Photo 14

The mandatory arty, posed shot. Still five operations to go.

Photo 15

Each rocker is clamped at about 4.5° to taper the nose section. A jig might have been nice, but this sufficed.


Photo 16

Photo 17

Photo 18

Photos 16 and 17

The jig is tilted 13° with the lower edge set on parallels to mill away excess width at the nose of the rocker. The quill downstop is set after the first to make all to the same size. After doing one side, the cutter is moved to the other side of the pivot to form the taper on the other side of the nose, resetting the down-stop as required.

Photo 18

The last operations are to form the radius around the tapped adjusting screw hole, and to radius the tip of the rocker nose on the top side. I decided to use "filing buttons" for the first task rather than risk having a run-away milling cutter chew up a rocker beyond usability after all this work. The buttons are 0.175" diameter, tapped 2-56, fully hardened to resist the file.


Photo 19

Photo 20

Photo 21

Photo 19

The nose section of the rocker is now case hardened using "Casinit" powder. This is done by bringing the steel to red heat, then rolling it in the powder. A blob sort of melts and sticks to it. The mess is then heated to red heat again, well an truly melting the blob, then quenched in cold water. The result is a change in the steel composition to a high carbon content for a depth of a few thousandths of an inch. This, combined with the heating and quenching results a thin but very hard layer, hence the term "case hardened". As the core is still comparatively soft, tempering is not needed. The process can be repeated to increase the thickness of the "case".

Photo 20

The rockers were then given a quick sand blast to give them a uniform finish (and rather attractive too, if I say so myself). After this, the radiused contact area of the nose is stoned to a smooth finish.

Photo 21

The last operation for the rockers is to machine up and fit bronze pivot bushes. Lacking any bronze smaller than 3/8" diameter, I've made these from brass. They are slip fits in the rockers to allow Lockitie grade 630 ("super wicking") to be applied with a toothpick for a permanent bond.

Rocker Posts and Pivots

These fiddly little things are turned and milled from 1/2" diameter steel. To start, the stem is turned slightly over length, with a generous radius where it transitions to stock diameter. They are then parted (sawn) off leaving a bit over 1/4" of stock material.

This excess is then faced back to be exactly .250" long. I decided to drill for the mounting screw from this end rather then the previous setup, reasoning that the collet would hold the parts concentrically and that it's easier to drill a through-hole than a blind one. Also, if the drill wanders, at least the head will be centered.

All blanks now go to the rotary table on the mill to have three sides milled away so as to reach within 0.005" of the stem diameter. The corners are then milled away (no need to measure the 45°, eye-ball is close enough).

The radius is formed by lowering the last 0.005" for a finish facing cut, then rotating the work around to the opposite side. The cutter is a 5/16" slot-drill which can span the club shaped protrusion of the head. This is an adaptation of the way I now form the radius for con-rod ends.

Here we see some cunning reuse of the rocker arm jig to position the rocker posts for drilling the hole that will take the rocker pivot pin. One post is measured and center popped, then the mill aligned over the mark with the post pushed against the stops of the jig. A pair of shop-made machinists' clamps (to a design from Model Engineers' Workshop) hold it in place for drilling. The others can then be drilled in the same way quickly and accurately.

If you visited the con-rod ends page mentioned above, you will recognize this jig. This time, a 1/4" slot-drill is used to radius the protrusion on the rocker posts concentric with the rocker pivot pin axis. A length of piano wire was inserted in the post during this operation to provide some extra leverage and get the fingers further away from the cutter.

The posts, which were made over-size, are now milled to final length by slipping them onto a length of 1/8" drill rod. The stems are gripped in the mill vice with the rod set horizontal on parallels allowing a light cut cleaning cut to be taken across the top of the posts.

The tops can now be set on a parallel so the irregular bases poke up. A piece of scrap aluminium packing allows the posts to be clamped up to mill them down to final size (note the slices of business card used to compensate for any miniscule differences in post diameters).

Almost done now. The last step is to slit one side of the post so that the screw that fixes the post to the head will also clamp the rocker pivot in place. The saw in 1/32" thich as specified. I now wish I'd used a thinner saw, the slits look a bit out of scale.

The slit post allows the posts to be clamped to a piece of drill rod and chucked (colleted?) so the dog-leg bit can be thinned down to 0.156" wide. This brings the rocker centerlines to the valve centerlines and the adjuster screw centerlines to the tappet/push-rod centerlines... hopefully. I've set the rebating tool at a slight angle so contact with the rocker is only in the bush area.

The rocker pivot pins are simple, but take quite a while to make. The plans show two designs, one threaded on the end for 2-56 nuts, the other grooved for "E" clips. As I happen to have a stock of suitable "E" clips for the 1/8" diameter pins, I've chosen the latter option. Easier, smaller, and neater. After facing all pins to the same length, the grooves are cut by referencing the side of the tool from the end face. Grooving raises a burr that would prevent the rockers sliding on, so the tip of the parting tool is touched to the pin surface and run back and forth to de-burr.

The end result of all the effort: four posts and four pins. Measurement of an assembled unit suggests the rockers are correctly spaced to align with the valves and push rods—some sort of minor miracle.

Cam Followers

More fiddly work. The tappets are a straight turning job from 1/4" drill rod. I find this to be ugly stuff to achieve a fine finish on by turning alone. A sharp, well honed tool and fine finish cuts (in-feed of 2 to 3 thou) produce a reasonable result, but polishing is always needed. The lifters need to be a highly polished and precise fit in the 1/8" reamed holes in the upper crankcase. The stems were produced by step-turning to 0.125" diameter in two 1/4" steps. Fine glasspaper and a steel rule were used to polish off a few tenths to achieve the close sliding fit. The mushroom (ie, tangential cam contact) head probably should be hardened. Maybe it will be.

This completes all the valve gear with the exception on the push rods. These will be made from 1/16" piano wire after the engine is assembled and the correct length—as opposed to the theoretical length—can be determined. Call me a pessimist.




Back to Cirrus Main Index


This page designed to look best when using anything but IE!
Please submit all questions and comments to