Quarter Scale Merlin V-12

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
My next step was to match-machine the heads to the blocks. Earlier, when I surfaced the heads I left them tall so I could finish their surfaces in the same set-up where all the critical drilling and boring will be done. I also thought it best to calculate the static compression ratio before surfacing the heads. The c.r. came out to 7.8 compared with 6 for the full-size engine, which may be a bit high if the supercharger turns out to be functional. Since most of the foundational machining will be finished by the time I get to the supercharger, I should be able to lower the compression by shaving the pistons.
I'll also model a complete operational cylinder soon since I'm planning to increase the size of the water jackets as well as the wall thicknesses of the liners by reducing the diameter of their bores. Although this seems like an inconsequential change, it provides opportunities for valve and connecting rod interferences.
The faces of the heads were mapped to determine the best centerlines for the bores before doing any of the head machining. The cast features in the heads must be machined to match those features already completed on the blocks. For appearance sake I also wanted the outer perimeters of the assembled heads and blocks to align as closely as possible.
The mapping showed I could easily achieve near perfect alignment between the left-side head and block, but there would be a compromise misalignment of .007" on the right head because the cast central stud tube hole in the right block was just too far out of place on its top surface. This small misalignment won't be noticeable because the heads don't actually mount against the blocks. Even though all the cast holes in the heads had to be relocated as mentioned earlier, I was not able to align the already cast central stud hole in the right head to the one in the block without shifting the head .007".
The heads were surfaced to produce an average combustion chamber depth equal to that called out in the drawings before the sealing shoulders on the combustion chambers were bored. It was important to simultaneously keep the intake mounting flanges parallel to these surfaces and also to match their heights on both heads to reduce difficulties later when fitting the intakes. The rest of the interiors of the combustion chambers were left as cast since the valve guide machining will be done later. The Merlin heads are designed for separate valve seats and guides, but I hope to come up with an integral valve cage design. It looks to me like the multi-angle and multi-level contours cast into the heads behind the valves will make it difficult to end up with separate seats and guides which are concentric.
Since the stud tubes penetrate the coolant jackets in both the head and block they must be sealed. I ended up with a slight mismatch, after all, on the troublesome center stud tube hole in the right head; but I believe it can be sealed with gap-filling Loctite.
In order to relocate the pre-cast stud tube holes in the head I fabricated a flat-end drill from a slightly undersize 4-flute carbide end mill. Starting a quarter inch behind the end of the cutter I ground the flutes down slightly in order to clear the hole left behind by the cutting portion of the tool. In the past I've plunged deep holes using unmodified 4-flute end mills, and the results were usually inconsistent. I don't know for sure if my modification helped, but 23 of the 24 relocated holes came out on size and where they were supposed to be. The reamer I used to bring the holes to their finished diameters was dulled rather quickly by investment packed in the coolant passages under the surface of the head. After some research I learned the investment typically used in aluminum casting can be dissolved in water. I soaked the heads in warm water for several minutes, but it didn't seem to have much effect. After the next machining operations on the head I'll try boiling them in water.
I turned partial dummy liners out of Delrin for use as fixtures to assemble the head/block pairs for match drilling and tapping the 24 auxiliary head bolt holes in each head. These liners were turned for press fits in the blocks, and the sealing spigot diameters were turned undersize by .002" just as will be done on the actual liners. The blocks, with their press-in plastic liners, assembled onto the heads perfectly with no alignment issues; but, of course, I'm not yet dealing with the interconnecting tubes.
These plastic liners will protect the sharp sealing edges on the combustion chambers in the heads later while checking the alignments of the 28 fluid tubes running between them. Note how the heads do not sit down against the blocks when assembled but are, instead, clamped against the elevated liner spigots. This gap is what creates the need for all these crazy fluid tubes. My guess is that with the level of horsepower generated by the full-size engine, Rolls-Royce engineers didn't feel that the head gaskets available at the time would be reliable. The high clamping pressures in the sealing corners of the narrow liner spigots may also be the reason that an alloy steel rather than brittle cast iron was specified for the liners. - Terry

091.jpg


092.jpg


093.jpg


094.jpg


095.jpg


096.jpg


097.jpg


098.jpg


099.jpg


100.jpg


101.jpg


102.jpg


103.jpg


104.jpg


105.jpg
 
The sound of a full size Merlin in a Spitfire at an air show is something else. And to build a working model of one is just as exciting I think.

This is an absolutely superb project, thanks for sharing progress with us. I would like to try this at some time but I'll watch and learn first.

Joe.
 
Awesome stuff Terry

Only 2 valves per cylinder on the scale model then? Well! It'll save you machining an extra 24 valves and associated components!
 
  • Like
Reactions: gus
Mattsta,
Yeah, I've been studying the top end, and it is pretty complicated with only two valves. I'm still trying to wrap my head around the dramatic differences between the left and right camshafts. I'm going to have to enter the top-end into SolidWorks in order to be able to even begin to visualize it. While between planes during the past few days I've been reading up on the history of the Merlin's development. It's a remarkable story with its mechanical complexity, wartime schedule pressures, and company politics. It's mind-blowing that over 160,000 of these engines with their 11,000 parts were built on two continents in such a short period of time with such great reliability and no modern high tech tools. It makes my machining challenges on this model seem whiny and insignificant. It's a shame that when the war ended nearly all of them were destroyed and its story allowed to fade into obscurity for most of us. - Terry
 
  • Like
Reactions: gus
Hi, my name is Morten and i'm from Norway. I have followed your 18 cyl. Build and now this one. I just want to say that I enjoy reading about your work very much. 17 years ago I visited Norwegian aviation museum in Bodø and they had a fullsize Merlin cutaway model. I was very impressed of that engine. It had technology that was not used in automotive engines before in the 80th ( 4 valves/cylinder). A complex engine constructed with pencil and paper, made with manual machines and handwork.

Morten
 
  • Like
Reactions: gus
I met Terry at NAMES this weekend. He said that this build is "frustrating" since he's spending so much time tapping holes and not making any parts. :)
 
Mattsta,
Yeah, I've been studying the top end, and it is pretty complicated with only two valves. I'm still trying to wrap my head around the dramatic differences between the left and right camshafts. I'm going to have to enter the top-end into SolidWorks in order to be able to even begin to visualize it. While between planes during the past few days I've been reading up on the history of the Merlin's development. It's a remarkable story with its mechanical complexity, wartime schedule pressures, and company politics. It's mind-blowing that over 160,000 of these engines with their 11,000 parts were built on two continents in such a short period of time with such great reliability and no modern high tech tools. It makes my machining challenges on this model seem whiny and insignificant. It's a shame that when the war ended nearly all of them were destroyed and its story allowed to fade into obscurity for most of us. - Terry

I recently found a thread on an engine building forum posted by a someone in the US who rebuilds Merlin engines and he posted some very interesting photos and accounts of how exacting and precise the tolerances are in these engines. He detailed the procedures for reconditioning the fork and blade connecting rod assemblies and their respective bearings and I was amazed at how precise the tolerances specified for correct manufacture and assembly are. As luck would have it, my hard drive self destructed and I lost the link to this forum. I'll try to find it for you.

Indeed! How on earth did the British and Americans mass produce these engines under wartime conditions with slide rules and pencils? This applies equally to Bristol, Napier and the American manufacturers, Curtis and P&W and Packard.

It's mindblowing

I've visited the Science Museum in London where they have a cutaway Bristol Centaurus and a Napier Sabre engine in addition to a Merlin and a Griffon engine and wondered in awe..............how the hell did they design and mass produce these things in the mid 1940s???

Makes you realise how ingenious human beings are when their backs are against a wall and their creativity and enterprise is not strangled by government.............but encouraged.


PS. If you need any CAD skills in this enterprise, I'll gladly help you out
 
It's my understanding that the Merlins bulit by Rolls-Royce were actually built up completely by small teams using craftsman techniques common to coach building at the time. The parts from one engine were 'filed to fit' and didn't necessarily fit into another engine. When Packard started their production in the U.S., their engineers spent the first year converting Rolls drawings over to their drawing and production standards which included producing the engines using assembly line techniques and, for the first time, interchangeable parts. Considering these engines were typically rebuilt after 100 hours service, their repair and overhaul was another unbelievable feat. - Terry
 
When I first started this build I thought it would be wise to work top-down and verify the assembly of the castings before starting on any of the bar-stock components. Thanks to the large number of castings and an engine that seems to have been hermetically sealed with bolts, I've been drilling and tapping holes for nearly three months. It's been a tedious process because the precise location of each fastener usually has to be individually found, nearly all the holes require match-drilling, and I seem to be able to only do a half dozen or so per hour. In addition, 3-48's have become the 'large' bolts in my shop.
The next hundred fasteners, though, will complete an important milestone for this project which is the fitting of the intake manifold and a trial assembly of the major castings forward of the wheel case.
I first had to make some Delrin liner collars for insertion between the blocks and the crankcase. These collars are temporary replacements for the metal ones which will be machined later. Their purpose is to precisely locate the blocks on the crankcase and set the proper heights of the heads so the intake manifold can be properly fitted. Notice in the photos that the blocks are separated from the crankcase by the thickness of these liners. The gap they introduce creates the need for oil seals between the blocks and crankcase to contain the top-end drain-back oil.
The crankcase and blocks in the early Merlins were actually a single casting, but performance testing uncovered a tendency for the blocks to crack during full-power runs. The engineers felt that a redesign of such a complex casting would have seriously delayed production, and so it was decided to separate the blocks from the crankcase. This decision complicated the engine's assembly and added a number of additional parts; but it solved the reliability problem and, with war looming, minimized the impact on production.
The Merlin's intake manifold is made up of three separate castings: a center 'log' section and two side sections. These three pieces will be combined into a single assembly which fits in the 60 degree valley between the heads. Measurements on the temporarily assembled heads showed the included angle between their flanges was exactly 60 degrees and, within the limits of my measuring capability, they were parallel.
Machining the center log section was fairly straightforward and required only a simple facing operation and 30 transfer-drilled tapped holes. The critical side sections, though, were another story. Both were badly warped, and each had a spiral twist along its major axis. I couldn't come up with a coherent plan for straightening them, and so I just did the annealing and then spent hours randomly pushing, pulling, and twisting them until they looked acceptable. The process wasn't at all satisfying, but it seemed to work.
The side castings, even after straightening, were extremely difficult to fixture for machining. Each had a pair of mounting flanges that needed to be machined at 30 degrees to each other; and, upon assembly, the flanges facing the heads must end up parallel. The head mounting flanges on both side sections were hand-lapped after straightening to provide a starting point for the machining. The only fixture I could come up with to support them while machining their second surfaces, though, was a kluge; and I held my breath during every .002" pass.
I tried to adjust the height of the assembled manifold in the valley between the blocks by only milling flange material from the heads since it was a straight forward fixturing and machining operation. After removing as much as I possibly could, the manifold still set high by about .030". Some simple trig showed that .013" still needed to come off each side of the manifold. Since this was too much material to lap away, another fixture was created to support each manifold side section so its head flange could be milled.
Another important requirement in the fitting process is that the intake must also end up exactly centered between the blocks. A 1-1/8" diameter rigid fuel tube will eventually connect the manifold's center section to the supercharger at the rear of the engine, and there will be no way to later adjust for an off-center error.
Eventually, I had the assembled manifold resting between the heads. The angles matched perfectly, the log section was centered, and there were no measurable gaps which showed the flanges were indeed parallel. Twelve of the bolts that hold the three castings together actually pass through the thin wall manifold, and they will be joined by several more after being secured to the heads. The issue is that the internal supports for these bolts were left out of the designs of these castings even though they were a part of the full-size castings. Since they're very fragile, the slightest over-tightening causes the side sections to flex, and as a result a slight gap is opened between them and the heads.
I was hoping that the 70 bolts that I will be installing to secure the manifold to the heads would also seal them so I could avoid intake gaskets. However, it now appears that nearly a third of them will be counterproductive during final assembly, and so I will have to use intake gaskets after all.
I measured the thickness of some 1/64" gasket material samples I had on hand because the exact gasket thickness will have to be removed from each side of the manifold before the mounting holes are match-drilled. I measured as much as .005" thickness variation between my samples. In practically any other application this would have little effect, but for an angled manifold installation it raises a red flag. The .004" hole clearances I've been using for the recommended 3-48 manifold screws is limited by the small diameters of the cast bosses. Because of the manifold angle, a .005" change in gasket thickness will move the mounting hole locations about .011" along the manifold, and this is much greater than the hole clearance. This means that the holes must be drilled for the exact batch of gasket material that will eventually be used, and I need to plan for spares. I decided to delay the mounting hole drilling until I have time to think about this some more. I'm dealing with the flu right now, and I don't want to make any irrevocable decisions with my currently unclear head. - Terry

106.jpg


107.jpg


108.jpg


109.jpg


110.jpg


111.jpg


112.jpg


113.jpg


114.jpg


115.jpg


116.jpg


117.jpg


118.jpg


119.jpg


120.jpg
 
Astounding work!!

As far as the gasket material, how about using a curing gasket material like high temp silicon or something along that line?

Ron
 
not a project for the faint of heart thats for sure....
 
Terry
Can you not find some uniformly thick gasket material.
Get enough to make a few gaskets and a piece to go between mating faces for match drilling.
Robbie
 
Robbie,
So far, that's my plan. Right now, I'm trying to decide if it would be best to do the match-drilling and then mill away the flanges for the gasket or, as you say, mill the flanges and then insert the gasket while match-drilling. It would be easier to do the drilling without involving the gaskets, but then I would get only one chance to mill away the correct amount for the gasket. Terry
 
Terry
I would mill flanges first. Any miss calculation can be rectified with thicker gasket, and the match drilling will still be perfect. I wouldn't make gaskets till all drilling is complete. Only require a piece of gasket material between mating faces. Maybe a spot of oil to hold in place.
Robbie
 
Wow!

Can I say "Master"

To get to your level for me is... un-achievable. ;D

Fantastic work.
 
Impressed with the workmanship and the huge intake manifold on that engine.
 
I would have loved to shake the mans hand who made the patterns for these amazing castings.
Love the patience for straightening those casings. Great work.
 
Back
Top