Camshaft & Crankshaft Centerline Dimensions Relative to Case Deck

Hi, Paul.

Thanks for the response.  I have done what you have suggested.  This is how I know I have an issue.

If you look at the third image above, you'll see that I have a to variation of 0.013" across the deck. I do not consider this variation to be minor. 

Everything critical in the engine is referenced from the deck; when the cylinders are bored, they are bored relative to the bottom of the barrels.  Because the barrels meet the face of the case deck, the crankshaft should be dead parallel to the deck or else the barrels will be askance and you will lose a lot of power to unnecessary drag with the added disadvantage of increased wear. 

Finally, you can have similar issues if the barrels aren't correctly configured fore and aft.  A good reference to achieve this correct "ideal plane" would be from the factory blueprints which should provide the centerline dimensions I am seeking.

Any input on this front that you could offer would be greatly appreciated.

Thanks,

-Robert

Hi, Neill.

Thanks for the feedback.

I am governed to an extent by the lowest point, but what I don't know is what's been done before and if by splitting the difference I might be going in the wrong direction.  And while I could certainly cobble the whole affair back together and use any number of sealers, but that's not how I operate, and I'm not looking to be "as good as most."

My goal with my engine builds is to "blueprint" the engines, at least as a starting point or baseline condition.  By "blueprinting," I mean to bring the engine as close to the original specifications as intended by the engineers.

With the goal of blueprinting this engine, the missing piece is the answer to this questions I have posed.  Perhaps it would make more sense for me to reach out to the archives to see if they have any of the historical drawings on hand.

Thanks again!

-Robert
 

Thanks, Tony!  

You were reading my mind!  I think this should work a treat.

Many, many thanks!

-Robert

Hi, David.

YOU WROTE:  It looks as if each side is planar to within 0.001 on the right and 0.002 on the left. 

The left is off by 0.008" front to rear as is the right.

YOU WROTE:  Perhaps the two halved didn't leave the factory together? 

You are correct.  The cases are mismatched, but the primary side case is is a "matching numbers" case, hence the desire to keep these cases.

YOU WROTE:  There is a step of 0.003 between them. 

Agreed.

YOU WROTE:  So you need to establish the exact height (or heights) of the plane faces of each half relative to the crank. 

Exactly.  This is what I described in my initial post, how I arrived at the values I provided, and why I am seeking CL dimensions for the crankshaft and camshaft.

YOU WROTE:  Presumably you should insert a mandrel through the main bearings to find out what's up. 

Agreed.  The mandrel is the big shiny thing running through the bearings where the crank goes and is resting on the V blocks in the pictures attached to my post above.  To provide an extra measure of security, I also placed a mandrel where the cam goes.  Both mandrels turn freely which tells we that the alignment between the cases is spot-on.  The difference on the deck is because the cases did not leave the factory together as you had suggested. 

YOU WROTE:  If they do match - shouldn't the cases be finally tightened together with the barrel (or another flat surface) is in place?

The cases are mismatched, but the primary side case is is a "matching numbers" case, hence the desire to keep these cases.

-Robert

Hi, Barry.

The head and barrel are from different machines than the cases, but the barrels have been bored, the top and bottom contact faces are dead perpendicular to the bores, and the mating surface on the head is as close to perfect as I have ever seen and does not appear to have ever been cut or abused in any way.

Thanks for your enquiry!

-Robert
 

Hi, Ashley.

I agree entirely with your observations.

Thank you for sharing them.

-Robert

Hi, Peter.

In addition to the dowels being in place, I have installed a precision ground solid SS bar in the main bearing where the crank belongs and another precision ground solid SS bar in the bushing where the cam goes.  Both rotate freely which suggests that the cases are well aligned and not "clocked."

I should ass that the cases are a mismatched pair, but the primary side case is is a "matching numbers" case, hence the desire to keep these cases.

Regardless, Mr. Ripley was able to provide me with the information I needed. 

Kind regards,

-Robert
 

Hi, David.

Thanks for sharing your thoughts.

YOU WROTE:  Only one remaining comment...you say that both are out by the same amount (0.008") front to rear.  So if you rotate your reference plane by 0.008", they are 0.000" in that respect.  So where is the horizontal reference plane? 

The horizontal reference point, as described in my earlier posts, is an imaginary plane that is dead parallel to the "dummy" crankshaft and camshaft.  The "dummy" crankshaft and camshaft established this imaginary plane along the lateral axis.  However, the piece of information that was missing from my analysis was the factory's target dimension from the CL of the camshaft and the CL of the crankshaft.  This is the information Mr. Ripley so graciously provided. 

Because the factory specified that the CL from the crankshaft to the camshaft was to be 3.125", I removed the radius of the crank journals and added the radius of the camshaft to the factory's target value to determine the vertical displacement from the top of the "dummy" crankshaft to the top of the "dummy" camshaft.  With this in hand, I was able to determine how to set the cases fore-to-aft by pivoting the cases around the "dummy" crankshaft and carefully measuring the relationship between the two shafts until I achieved the desired displacement.

The values I originally used were somewhat arbitrary as I originally fixtured the cases such that the trailing edge of the deck on the timing and primary side was zero, and then I swept the other dimensions.  This was not done to determine where the plane was, or should be; it was done to determine the overall relationship between the datum points.  I only chose them as a reference point because the trailing edge of the deck on the timing and primary side was at zero.  Because I knew these points to be in A correct, but not necessarily THE correct relationship to the plane, I chose them.  I could have chosen any of the datum points to be the "zero" reference point, but the overall relationship between the datum points would not have changed.

To prove this point, as you can see from the chart below, I adjusted the values in each row such that the zero datum changes from one column to the next. 

You can see that, despite the arbitrary "zero" reference point, the maximum difference is always between the front "step" of the case on the timing side and the rear "step" on the primary side. 

And while I understand that this difference is likely because these are not matched cases from the factory, regardless of the cause, what does not change is the absolute value of the difference between those two data points.  It is always 0.013" in this case and will generally remain in that range.  (The only exception to this would be if I were to radically rotate the cases around the axis of the crankshaft, for example, if I were to rotate the cases, for example, 90 degrees.)

YOU WROTE:  The only one I can think of is the base of the crankcase around the drain plug.  I doubt if that is reliable, nor if it is important that it be parallel to the top surface(s).
Nonetheless it seems you have the job in hand. 

I did consider this and concluded that the base of the crankcase around the drain plug is a reference plane, but only in theory, so its practical use as such is approximately zero. 

As you can see from the photos I have provided throughout this thread, I went to considerable lengths to secure the cases such that the "dummy" crankshaft was dead parallel with the mill table and therefore dead perpendicular to the quill.  Even the piece of paper you see to the left in the picture below was intentional, as I didn't have any cigarette papers on hand to dial in the shims.

I suspect that once the deck was milled, its surface was then used as a fixturing face to bore and mill the flat for the drain plug before threading the hole.  But, because it has such a small footprint, its use as a canonical reference plane would be questionable at best.

YOU WROTE:  Although I don't know how the design dimension will help unless your cases are over height from the factory and you can reduce all surfaces to the correct plane.

I don't know if the cases are over height from the factory or not, but it's also not relevant, at least not at this level of analysis.  Now that I know where the target plane lies, I will use the lowest point on that plane as my zero reference and go from there.  This approach ensures that I remove the minimum amount of material needed to achieve my goal.  

YOU WROTE:  Or go lower and restore height with a compression plate and increase the chance of oil leaks.

While oil leaks are often a concern, I am not overly concerned about them in this application.  Instead of having a planar surface, namely the base of the cylinders, mating with a non-planar surface, namely the case deck as it exists at present, I will soon have two truly planar surfaces mating.  I prefer the latter of these two options by far.

Once I go up with the engine, I can check to see if the 0.013" I have skimmed from the decks has any appreciable effect on the compression ratio, pushrod length, valve geometry, etc.  Spoiler alert:  It won't!  

But, assuming it did, I could use a base gasket along with Wellseal, Hylomar, Gorilla Snot, or whatever sticky stuff you have on hand, to adjust the height of the barrels.  If I don't need to adjust the height, I am still not concerned about the potential for leaks, as I will install a PROPER breather to the crankcase that will place its pressure at a value slightly lower than the ambient atmospheric pressure.  My experience has been that reducing the crankcase pressure gets one 90% of the way to eliminating the all-too-common leaks on our beloved parallel twins.

I appreciate your considered thoughts and collaboration.

Kind regards,

-Robert

Hi, Ashley.

Your comment regarding the CR is wise counsel.  

I added the comment about using the base gasket as a variable for the record, more than anything.  If someone is having issues in the future and comes across this thread, I wanted to put this out there.

I plan to run an SS cam in this engine, so my target CR is something around the 10:1.  Once I get the bottom end together, I can turn my attention to dialing in the CR.

Kind regards,

-Robert