|The MGA With An Attitude
MISALIGNED COMBUSTION CHAMBERS - TC-326
in the Twin Cam Cylinder Head
This is going to be a long introduction, followed by a longer response, so hang on.
On 11/23/2011, Mick Anders in Australia wrote:
"Below is an email that I sent to Mark Hester, Roy Crosier, and Dirk Van Ussel. All these people have reported a misalignment of the Twin Cam combustion chambers, i.e. they are not in a straight line, parallel to the engine axis. Note all three are using an MGB five bearing block with a Twin Cam cylinder head. Do you agree/disagree"?
Mick's message to the others was:
"It seems a coincidence that all the people reporting a combustion chamber misalignment are using the MGB five bearing block. I do not believe that the company AMAC that made the cylinder heads would make such a simple error as to have the combustion chambers not in a straight line. This company built millions of alloy cylinder heads, some for aircraft and some for very high performance engines. It is not known who machined the holes in for the cylinder head studs. AMAC could have drilled the holes in the head, but more likely they were done at the Morris Engines Branch in Coventry, or where the stud holes in the blocks were done, at Austin in Longbridge. All these companies would have done it correctly.
"What needs to be done is to take measurements which will clearly show what causes the problem. On an unmodified head place a straight edge along the stud holes and measure the distance to the edge of each combustion chamber. (The actual distance will vary if the head has been skimmed, but what you are looking for is any variation for the four chambers). Measure the distance from the studs (or stud holes) to the edge of each cylinder bore on a Twin Cam block. (Again this can be affected by a re-bore but variation is what is sought). Do the above measurement on the MGB five bearing block.
"My suspicion is that the error has arisen during the setting up of the stud changes required for the MGB block, both in enlarging the diameter of re-used stud positions and in the placement of the new stud positions. The above measurements will confirm where the problem arises. If the valves clear the pistons I do not think it matters much anyway as long as there are no exposed sharp edges with the head, block, or gasket".
Then Mick asked me if I agree or disagree. Oh, this is going to be fun. Take a seat right over there.
The short answer is, I disagree. But I have some strong hunches on the source of the problem. Being a machine design engineer I know how castings are machined in mass production. First thing to remember is that all MG engine blocks and cylinder heads are sand castings with an internal mold core to create the voids that will ultimately be air space in the finished parts. You likely know most of this but I will run though it all just to be sure we have the details.
For the engine block the mold core occupies the inside of the crankcase, most of the volume of the cylinder bores, all of the space that will be water jacket, the tappet galleries, the pushrod holes, most of the tappet bores and cam bearing journals. The mold core pieces have fingers that stick out the sides to engage part of the outer mold cavity to hold the cores in alignment. The function of "core plugs" (not freeze plugs) is to close some of the necessary but bothersome core holes in the side of the castings where the water jacket core was being supported on a few side fingers.
For the cylinder head the casting core occupies the space that will be water jacket and intake/exhaust ports, and most of the space for tappet bores and bearing journals. The combustion chambers are formed (held void) by extensions of the main mold cavity. The support fingers for the port core extend into the combustion chamber area through the rough valve seat holes to find support on the main casting cavity, and are also supported on the sides at the intake and exhaust ports. The water jacket core is supported on fingers through the water pump and thermostat openings, and through a few more peripheral holes that will ultimately need to be closed by core plugs.
The casting cavity is made in two pieces, left and right halves, or top and bottom halves, with a split line in the middle. These cavity parts have notches formed in them to accept the supporting fingers on the core pieces. The core pieces are carefully placed it one cavity half, and then then the other cavity half is placed on top to close the cavity. Molten metal is poured into the mold and allowed to cool. The mold is then opened to release the raw cast part, and the part is passed over a shaker table to knock all of the sand out of the core to leave the desired voids inside. At that point you have a raw casting that needs to be machined.
We all should know that placement of the cores in the casting cavity may be as much art as science, and the cores are not always perfectly aligned. In fact there is always some misalignment, the question being how much? For this reason you get variations in thickness of cylinder wall and certain dislocation of intake/exhaust ports. Any misalignment of surfaces that will be machined is of little consequence, as machining will produce the surfaces in the correct position. Casting surfaces that are not machined will sort of lie where they may. In some cases, like inside and outside walls of the crankcase for instance, the misalignment doesn't matter much, as long as the walls are thick enough to support the function. For intake and exhaust ports that are not machined, the raw casting core alignment is more important.
In any case, any casting surface may be misaligned relative to other surfaces. Such is the nature of sand castings. After you do a little machining on the part you may have a few flat spots or tooling holes to use for fixturing the part to keep it in constant alignment while doing the rest of the machining. The big question is, how to you hold the irregularly shaped raw casting before the first cutting tool gets to it?
The answer is, tooling pins and clamps. Imagine you put the part in a rectangular box, but the box has only a floor, one side and one end, to be open on the other two sides and the top. On the floor place three vertical pins with flat or slightly rounded nose, and set the part on top of these three tooling points. On one side there will be two horizontal pins in the wall of the fixture, and you push the part up against these two pins. On the end wall there will be one horizontal tooling pin, and you push the part up against this pin. The part is then touching six tooling pins and is resting in a stable position, and you can clamp it down in that location. Then you can start machining the part relative to the six-pin fixture.
Fairly early in the machining process you can cut some selected tooling features. For instance on the engine block you might first machine the sump flange flat and bore a couple of holes (far apart) that will later be used for fixturing. Of course you can drill and tap the sump bolt holes in the same set up if you like. When you get that far, you can remove the part from the initial 6-pin fixture and place it in a second fixture consisting of two long flat rails with a couple of alignment pins to support the part on the sump flange. In this state there will be access to all sides of the part for machining in all directions as it moves down the automated machining line. Subsequently, if the machine works as intended all of the machined features will be in the desired true position relative to the initial fixturing features.
However, any surface that is not machined is subject to positional misalignment from two sources. The first misalignment is a function of the roughness and inaccuracies of the molding cavity, misalignment in positioning the cores in the mold, possible shifting of position of the cores during handling before and during pouring of the molten metal. The second misalignment comes when the rough casting is initially placed in the 6-pin fixture. Alignment here is entirely dependent on location of the six points on the casting where the tooling pins will touch. All of the rest of the casting may be pushed out of position some depending on lie of the six tooling points on the raw casting.
End result of all this is that all machined surfaces should be accurately positioned relative to all other machined surfaces. Cylinder bores and head stud holes will always be properly aligned with crankshaft journals. Similar happens to the cylinder head where all machined surfaces will be properly aligned to all other machined surfaces. So when you assemble the engine the head will drop right onto the studs, camshafts and tappets and valve gear will all be properly aligned, and timing chain and sprockets and tensioners will also line up as intended. All of the cam, tappets, valves and valve seats will be exactly where intended relative to each other. Bearing journals, cylinder bores, top of block stud holes, crankshaft, con-rods, pistons are all machined. Bottom of head, cam journals, valve guides, valve seats and valves are all machined. And the whole machine will move in harmony.
So then where is the alignment problem? You will find it in, on, and around the casting surfaces that were not machined. In this case it will be the contour of the as-cast combustion chambers. Misalignment here comes from all of the above sources, beginning with casting tolerances and 6-pin fixturing. Combustion chambers may start out only a little misaligned relative to the outer surfaces of the casting, but the whole raw casting may be more misaligned when placed in the 6-pin fixture. So the combustion chambers will be out of position before all of the machining is done.
If some smart tool design engineer was on his toes 50 years ago there could have been an up front fix for this problem. Some of those six initial fixturing pins should be located to touch critical clearance surfaces in the combustion chambers. Initial machining of the later fixturing features would then be properly positioned relative to the combustion chambers. Then all of the final machining which is located from the fixturing features would be in proper position relative to the combustion chambers, so when you get it all assembled all of the chambers would be properly aligned with the piston crowns.
Assuming this was not done properly in production back in the 1950's, today you can have the misaligned combustion chambers that have been that way for 50 years. In the original engines it may not have been a problem, as the designers take such misalignment into consideration when building production tooling, simply allow enough clearance between parts so pistons will not hit the head or valves in cases of worst misalignment (maintaining minimal required clearances). But when you rebuild the engine many years later you might shave the top of the block or bottom of the head to restore flat surfaces, in which case the combustion chambers get closer to the pistons. Now the clearances are smaller, and various misalignments appear to be more bothersome, perhaps to the point of noticing slight differences of compression between cylinders or some parts getting too close together for good running conditions.
This is why engine rebuilding is more of an art than original engine manufacturing. When you shave a block or a head, increase valve lift or change cam timing, you have to inspect and measure all of the parts to assure that you are not creating some interference that was not originally present. If I was rebuilding a Twin Cam engine I might take the opportunity to shave the head a little to reduce chamber volume, followed by machining of the entire combustion chamber for all cylinders to assure that they are all identical and properly positioned. That sort of combustion chamber machining won't be cheap, but if you want a nice engine you might be willing to spend the money.
As to your hypotheses, I will bet this has nothing to do with any misalignment in 5-main bearing blocks being converted to accept Twin Cam cylinder heads. When they plug some stud holes and drill and tap others to mate the Twin Cam head, some of the original studs are not moved, and all of the relocated ones have to be in the right position to allow assembly of the head to the block. Therefore the head is NOT misaligned on the block. If the combustion chambers end up being misaligned relative to the cylinder bores, it is because the head was originally manufactured that way. On close inspection you will likely find that the only thing out of place is the raw cast surface of the combustion chamber in the head while all of the valve gear is in the correct location relative to cylinder bores and pistons.
Bottom line is, it all started with larger tolerances in position of the combustion chambers 50 years ago. This chamber misalignment is not a result of mating the head to a 5-main bearing MGB block. You will likely find the same chamber misalignment on any MGA Twin Cam engine.