Measuring Piston-to-Valve clearance

I've got a set of soft test springs, and a dial indicator for checking the clearance.

Question is: Should the indicator be setup to point parallel to the piston travel? Or parallel to the valve travel? Which is the proper standard to go by?

I've found mixed information on this when I search online. Assuming the valve travel is angled, then measuring these two ways will get a slightly different result.

It seems to me that measuring parallel to the piston is the better way. But all the examples I find of anyone doing this measurement, they almost always set it up parallel to the valve.

Since you can't move the piston without moving the valve, but you can move the valve without moving the piston -- I would say that dictates that you have to measure in the direction of the valve. Also in practice that is where your contact is going to occur -- when the valve is not keeping up with the cam.

Use the valve travel, as that's the thing that is affected by the valve timing and is the variable.

The piston, or rather the crankshaft, position is the fixed value you're checking against and, unless you change the rod-to-stroke ratio*, or deck height, it is fixed.

*Something that some overlook when changing that aspect, is the piston position to crank' relationship will change, and has to be checked, if changed. Not so common nowadays, but it was possible to run 2, or 3, common commercially available, different 'rod lengths with the SBC, with suitable pistons.

I think I didn't ask the question clear enough. Both answers didn't seem to quite get it.

The core of this question is Pythagoras.

If you correctly measure valve travel, then you're not correctly measuring piston-to-valve clearance. Unless the valve travel is exactly inline with the piston.

No, I get it - go back to what I said about the reference you are checking against, and what it is that you are checking. Least ways, that's the way I look at it, others may have a different idea, but I'd remind you that with the 'plasticine' method the material is cut and the cross section is measured in the plane of the valve face's impression and not in line with the bore which would require the measurement to be inclined to correct the valve angle.

You could do it the other way, I guess, setting the camshaft angle and checking the piston against it, but it would be much more confusing and a lot more work.

The clearance angle affect I think you are thinking of is "cosine error", it shouldn't be a problem here.

Perhaps if you consider what happens at, say, 5 degrees BTDC, if you advance or retard the camshaft slightly. The reference crankshaft angle is the same, but the variable which is valve lift changes, as does the clearance.

As you're using the light spring method, I'd do three initial checks, the first at the nominal camshaft timing and then with them advanced/retarded 5 and 10 degrees (bearing in mind the affect variable valve timing is going to have on overall valve lift), checking the available valve travel from 30 crank degrees before to after TDC in 5 degree steps - that should establish the basic clearances you have to work on. I'm 'old school' so would plot the clearances for each camshaft's clearance on the y-axis of a graph, with the x-axis being the crankshaft position. With some variable camshaft set-ups you may even need to go more than 30 degrees, but that will show up as the clearances being less there.

It is confusing, though, I agree.