From a machinist's standpoint, the cam followers will compensate for the machining errors that have been accumulating in the valve train. From an engine builder's perspective they provide a method for setting the valve clearance. After installation they won't be adjustable and can't be used to compensate wear, but in a model engine wear is typically less of a problem than a lash adjuster that won't hold its setting. The full-size engine didn't use lash adjusters either - probably for reliability reasons. Hydraulic lifters were still fairly new at the time and probably not yet race proven.
I used .372" for the diameters of the followers so they'd be sliding fits in the .375" cam box bores. The first step was to reduce the diameter of a length of 3/8" drill rod using abrasive paper and a shop-made ring gage. While still in the lathe, the rod was parted into some thirty .290" long blanks, each with a shallow hole drilled in one end as a start on their interior machining. The followers' lengths were chosen to keep them entirely within the cam boxes to avoid rubbing or binding in the head. I wanted each valve stem to push against a truly flat surface inside its follower, and it turned out the most efficient way to accomplish this was to pocket the interiors of the followers on the mill. I initially tried to bore them on the lathe, but tool deflection became a problem around a nub that tended to form at the center of the bore.
A custom holding fixture was used to cycle parts quickly through the mill where a pocketing routine machined their interiors. Based upon measurements of the valve stem heights, a variety of depths was used with allowances for the tops of the followers that would be finished later. The parts were then heat treated and tempered at 300F. After heat treatment, their outsides were polished on the lathe with the help of a quick change mandrel fabricated from a wood dowel and a couple o-rings. The parts were easily slid on and off the mandrel while the lathe was running. Their interior polishing was done with Scotch-Brite but took longer since they had to be gripped in a collet with the lathe continually started and stopped.
After polishing, the top of each follower was ground on a surface grinder as needed to realize a .007" lash clearance (including the .005" cam box gasket) for each valve. The spread in the final follower thicknesses was .004" with a couple outliers at .008". Step feeler gages made up from brass shim stock were used to measure the clearances.
Final assembly of the head required drilling out the ten holes in the cam boxes which up to this point had been threaded for the screws securing the camshaft bearing caps. The cam boxes, through their bearing caps, were then bolted to the head using longer screws in holes previously threaded into the head. The delicate teflon gaskets made up earlier for use between the boxes and head were a bit tricky to keep in place during assembly. After assembly, compressed air forced into the oil feed port at the front of the head could be felt leaving the oil holes drilled through the lobe heels of both camshafts indicating the top-end's oil distribution path was likely clear. The camshafts smoothly opened and closed the valves in the completed assembly as hoped (whew!), and the lash was verified one last time.
Fabrication of the water outlet pipe should finish up the remaining work related to the head. - Terry
The wall thickness is .015". I too was concerned about distortion, but a few test pieces moved maybe a tenth or so. Just in case, I added the extra thousandth clearance which turned out to not be needed. Perhaps the parts are small enough and being essentially hollow they can't help but be uniformly quenched. They were heat treated in three batches. Each batch was sealed in a stainless steel package filled with argon and heated to 1475F for an hour. The package was held over a can of old transmission fluid and its bottom cut off so the parts could randomly drop into the oil. I could see them still glowing red as they slid out of the package. One part from each batch was file checked to make sure it had indeed hardened. All three batches were tempered at the same time a day later. - Terry
You wouldn't think something as mundane sounding as a water pipe would be much of a challenge, but this one turned into one of the most difficult parts so far in this build. The water outlet pipe is actually a three-input manifold that will return coolant to the radiator from three internal passages in the head. Coolant flow in this engine is a known issue, and the walls of its scaled-down manifold are necessarily thin. The water pipe is a prominent part that will sit at the very top of the engine, and so in addition to providing important functionality it needs to look good.
The main body started out as a 7" length of 5/16" brass rod whose interior was drilled out in three contiguous passages using 1/4", 3/16", and 1/8" drills. The entire o.d. was then taper turned starting with the .312" diameter on one end and ending with .187" on the other.
My first two attempts collapsed while turning the taper and, after some study of the drawing, I realized that a portion of the 1/4" i.d. section had been designed with a wall thickness of only .008". I eventually reduced the diameter of the drill for that section to 7/32" but ran into similar problems elsewhere. The drills for the deep 1/8" and 3/16" sections tended to wonder off course in the brass and leave a paper-thin wall at their intersection. These problems were invisible until after the unwieldy taper (with its own challenges) was turned. Then, they'd typically appear as innocent looking blemishes in the surface finish that might not show up until hours later. Some remained invisible until they wrinkled under the soldering heat. Reducing the diameters of the smaller passages wasn't a good option since it would have restricted an already limited coolant flow. Several tries using a combination of parabolic drills and reamers was required before I got my first of several usable main bodies.
The documentation contains a helpful template of the manifold. A full-size photocopy was glued down to a block of wood and used as a forming guide. A pair of grooved Delrin rollers was then machined and attached to the block and used to form the tubing.
Before bending the main body, its small end was annealed and temporarily soldered closed. Its interior was then coated with cooking oil - a release agent for Cerrobend. The Cerrobend was melted in a beaker sitting in a pan of boiling water to prevent overheating it. Immediately after its pour, the manifold was placed in a glass of ice water to crystalize and strengthen the Cerrrobend against cracking under the bending stresses. When the bend was completed, the tube was gently warmed with a torch, and the Cerrobend drained out cleanly.
The first of several difficult silver-soldered joints attached a mounting flange to the end of the main tube. I made a (low thermally conductive) stainless steel soldering fixture to hold the manifold parts in alignment during soldering. My preferred method for soldering is to include the solder in the fixturing so I only have to hand feed heat rather than the solder. The flange mounting holes are so close to the tube though that even after turning down the heads of the hold-down screws, the flux and solder tended to flow around them and attach them to the flanges. This was eventually solved by switching from stainless to plain steel screws, turning down their heads, and then blackening them using a cold bluing solution. For good measure, I also added temporary mica washers, cut from a TO-3 transistor insulator, under the heads.
The remaining two inlet tubes were also drilled out from solid rod, taper-turned, and bent according to the template. Ron recommends using thin wall brass tubing for these, but I was concerned about the tiny contact area that would be presented to the main body. One of my weak-walled main bodies failed under the heat one of these soldering steps. I also had problems with the hidden entrance holes in the main body being inadvertently soldered closed. The curved inlet tubes made it impossible to drill these out, and the only solution was to start over. I went through almost two six-piece batches of machined flanges before the manifold was finally completed.
Soft soldering would have eliminated a number of problems that took a lot of time to work out. It did, however, weed out a defective manifold that might have otherwise failed on a running engine. Strength-wise, soft soldering would have been fine for the flanges which had machined recesses for the tubes. I'm not so sure it would have been reliable for the butted intersections between the two auxiliary tubes and the main body.
In any event, I stuck with Silvaloy 355 for all five joints since I'm confident it can be nickel plated. I'm not so sure about plating the tin/lead solder that I have on hand. I plan to test this as well as a non-lead alloy later.
Although the manifold is currently bright and shiny, its finish will eventually dull and look like something more appropriate on a steam engine. The next step will be to try my hand at DIY nickel plating so I can maintain the authenticity of the part with white metal. - Terry
I think the purpose of the taper inside the main body was to equalize the coolant flow in each the three lines coming from the head. In the full-size engine, that main tube was probably a rolled and seam welded affair whose outside naturally took on that same taper.
I was originally going to fabricate the manifold from stainless steel so I'd have it in white metal. In hindsight, that would have been madness. Ron's version in brass got me thinking about nickel plating - something I've been wanting to try for some time. There are a lot Youtube videos on DIY plating, and the one by Doc1955 who occasionally posts on this forum is as good as any.
Plating requires an electrolyte, a low voltage dc source, and a pure nickel anode (all available from Amazon). The most commonly used electrolyte can be made by dissolving a couple online-available nickel salt chemicals in water (a.k.a. Watt's solution), but another (nickel acetate) can be made from ordinary kitchen ingredients i.e. plain white vinegar and salt.
Also available is a Watt's-based kit from Caswell that contains nearly everything needed. An advantage of this fairly expensive option is that it will also include some proprietary chemicals for a bright finish right out of plating solution.
Unfortunately, I wasn't able to find any end result comparisons. Nickel acetate doesn't seem to be commercially popular - maybe because of an incompatibility with the industry's 'secret sauce' brightening agents. I initially experimented with nickel acetate since I already had everything needed on hand. My test parts turned out so well that I continued on with it.
The first step was to make the lead acetate. This involved adding table salt to 1-1/2 quarts of white vinegar in a wide mouth canning jar. The electrolyte can be used over and over (it even gets better with use), and so I used a jar with a spring lock top that I'll use for both plating and electrolyte storage. A pair of pure nickel anodes (purchased from Amazon) was then lowered into the vinegar, and 5 VDC from a wall wart power supply was applied across them.
I found little guidance on how much salt to use, and my three tablespoons turned out to be more than was evidently used in any of the videos. The salt will control the conductivity of the electrolyte which in turn will affect the speed and (maybe) the quality of the plating. In this first step, a solution of nickel ions is being created for later use in the actual plating process. Over time, the color of the solution will turn greenish-blue indicating that it's ready for use. How fast this happens will also depend upon the salinity of the solution. In the videos that included current measurements, tens of milliamps through the solution was common, and the electrolyte creation process took overnight. My own measurement indicated 2.8 amps, and my electrolyte appeared ready for use in about an hour.
Although my DC supply was capable of even more, the relatively high current draw was concerning. Everything I'd read indicated best plating results will be obtained while using low currents. These sources were referring to Watt's electrolyte, though, and so I wasn't sure if the same applied to nickel acetate. I decided to continue on.
The part to be plated is connected to the negative terminal of the supply, and one or more nickel anodes are connected to the positive terminal. Hydrogen bubbles immediately begin forming around the part, and in my case they temporarily fogged the solution. The bubbles tend to agitate the electrolyte making additional stirring unnecessary. Plating occurs quickly - in just a few minutes. Other than pulling the part up for inspection, it's difficult to judge when you're done.
Thicker plating will likely result from longer plating times, but quality nickel plating is measured in tens of microns even in plating shops. I suspect, with no evidence to back it up, that if very long times were attempted, a rough uneven finish might result from second order effects related to the shape of the part with respect to the shape of the electrode. The wire hanger used to hang my parts built up an ugly rough finish rather quickly.
The importance of proper surface preparation is stressed in any reference on plating. Even when done commercially, nickel plating isn't thick enough to fill in machining marks or surface scratches. Except for its color, a part's surface will look the same after plating as it did before plating. The part also needs to be absolutely free of any oil or grease including finger prints.
I finished all my parts with 1000g paper and a white Scotch-Brite pad while supported on a fixture to avoid finger prints. Most of the manifold's metal finishing was done while the tubes were on the lathe and still straight. Its soldering fixture was used to hold the completed manifold during final polishing. I found I got better results if the final polishing was done immediately before plating rather than over a tarnished surface. All parts were dip cleaned in acetone, air-blown dry, and then hung in the plating solution without being touched by hand.
The 2-3 minute plating time on all my test parts looked great. A couple soldered flanges using both 60/40 and nearly pure tin solder took nickel plating well. When removed from the electrolyte, all parts had a smooth gray satin finish which would be suitable for some applications. They might have had a brighter finish if the Caswell kit with its secret brighteners had been used.
For the manifold, I wanted something that looked like chrome. Even the brightest nickel finish, though, has a slight yellow tinge compared with chrome. Using red rouge buffing compound on a strip of micro-fiber cloth, I was able to buff its surface to a bright reflective finish that was close enough. - Terry
I've done a tiny bit of chrome plating with a very similar setup to what you're using, but running the charge directly through the applicator 'brush' to apply the chrome solution. I chromed directly over my nickel plating attempt because the nickel looked slightly yellow and I wanted the chrome effect. Here's a picture of my exhausts to show the difference in nickel and chrome LINK - note the nickel is nowhere as yellow in person as it looks in the picture, but against the true chrome is is very evident.
Edit to add - I just noticed how yellow my picture looks. It seems the white balance was off target on my camera - the light pieces of the marble table are actually white, so the yellow exhaust is really just slightly yellow and the chrome one is true chrome colour.
Well done. You are correct in your assumption that plating time has an affect on the surface quality. Too long or to much current will yield a rough surface. My power supply lets the voltage float and you set the amperage to what you need. To figure that, multiply the surface area (sq in) by .07. Fifteen minuets will build up around .00030" of plating. If you want more buildup repeat the plating. If you need to fill in scratches or pitting plate first with copper and sand smooth. Repeat if necessary.
I looked into chrome plating but felt it wasn't something that I could easily dip my toes into like I did with nickel plating. An expensive Caswell kit is probably the most practical way to get into it, and the chemicals involved are a hassle to keep around a limited space shop for occasional use. Maybe someday though ... -Terry
Get the answer to every anodizing, plating, powder coating & finishing question or need at The Home Page of the Finishing Industry®, the world's longest running, most popular, most authoritative source
The exhaust system includes a complex set of pipes that are virtually identical to those on the full-size engine. The model's exhaust is made up of several pieces of 5/8" and 1" diameter seamless 304 stainless tubing. There isn't a straight piece among them, and so a lot of metal forming/joining is involved. Basically, it's a much larger version of the just completed water manifold.
The first step was to fabricate the megaphone style collector. Its conical shape was created by first cutting a v-shaped wedge out of a six inch length of 1" diameter tubing. This was a pretty messy operation that consumed nearly a dozen Dremel-style abrasive cut-off disks and one Covid mask. The collector's shape was formed around a custom tapered mandrel. With the slotted tube resting in a v-block, gentle persuasion from the ram of an arbor press around the periphery of the tube soon had the seam closed up tight. A piece of leather wrapped around the tube prevented it from being marred.
I chose to tig weld the seam rather than solder it so the heat created later while silver-soldering the exhaust pipes won't be an issue. However, the poor thermal conductivity of thin stainless steel can create an unbelievable amount of weld distortion. The gap was small enough that filler rod wasn't needed, but I used it anyway to avoid a slight depression along the length of the collector.
The mandrel was used to stabilize the shape of the collector during welding. A notch was milled along its length to provide a gap behind the seam to keep it out of the weld puddle. The mandrel kept the collector circular and wrinkle free, but shrinkage along the seam created a subtle bow along the collector's length that actually improved its appearance. The tube wound up shrunk so tightly around the mandrel that it had to be pressed off.
The siamese'd ports in the head will require the circular ends of the exhaust tubes to be expanded. A swaging tool was machined and case-hardened in order to prepare the tube ends for soldering to the exhaust flange. A few test pieces were swaged so measurements could be taken for the flange's machining. The exhaust tubes will eventually be soldered in shallow recesses machined into the outer side of the flange. These recesses will help stabilize the assembly during soldering and provide some additional surface area for the solder. A back-up plate was also made to support the thin exhaust flange during soldering. Finally, a couple Teflon flange gaskets were cut using a vinyl cutter on my Tormach.
The next step will be to form the exhaust tubes which must be bent on a 2" bend radius. My 5/8" tube bender works on a 3" radius, and so I have some thinking to do. - Terry
edit: the caption on the first photo should read '... 270 Exhaust'.
Brilliant! I am impressed. Carry-on, as I am enjoying your work. Excelent story and explanation. Well supported with pictures. You'll be able to sell this story to a magazine when finished. Let us know the magazine so we can read the full story as well.