It requires time, and an ability to express myself clearly without getting myself in a muddle and without confusing you or being ambiguous - skills I don't have!!!
I'll start again, if I may. 'Cheap' rods use an H section. I suggest that I section rods are what you want. This is with the neutral axis horizonatally through the middle of the section, because the rod is free to rotate around the journals, and hence has no bending on it, whereas in the other direction there can be bending loads (and of course there is the compressive/tensile loading). Therefore the axis of the journals is vertical, and this applies to both the H and I examples.
So, in an ideal (simplified
) world we want enough material to manage the tensile loadings and material distribution to best manage to bending/compressive loads.
I can't be bothered to draw stuff, or scan something, so bear with me with the explainations.
I section - The horizontal width of the top/bottom 'webs' is B. The length of the web to the middle is b. The length of the middle is a and the total height of the I is A. Does that make sense? I shall say that A and B are 10 and the material thickness at any point is 1, so that makes b = 4.5 and a = 8
I = 1/12 * (BA^3 -2ba^3) = 450
H section - B is the height of the H, T is the thickness of the materal and a is the length of the inner horizontal (which would be A-2T). Hope that makes sense. We'll stick with the same material dimensions for simplicities sake - thus 10 tall, 1 thick and (10-2*1) = 8 wide across the middle of the H.
I = 1/12 * (aT^3 + 2TB^3) = 167
For a given weight (it just so happens that my dimensions give a cross sectional area of 28 each, so would weigh the same) the I section is vastly stronger in bending in the direction that matters.
Obviously extremes of use might change what you're after. Very low revving, high BMEP engines might actually benefit from a different cross section (many ships use round section for instance), and that will vary with load, stroke, journal sizes, rpm etc, but for 'car engines' which operate over a broadly similar rpm range and power range then the I section conrod is much, much better.
It's not explained very clearly, and I think I might have actually contradicted what I said earlier - not on purpose, but because I found the I/H section thing very confusing, trying to visualise the bending axis, the journal axis, the Ixx diagrams and the name. Apologies for that; I've literally had to sit here for 5 minutes sketching things to try and get my head around it. But the jist is that the one on the right (the I beam by my terminology) below is the one that resists bending better in the direction bending is likely to occur.
To make is easier, the overall width of the bits you can see are (on the left) is dimension A and (the one on the right) is dimension B. The depth (into the picture) is B and A respectively. Maybe that doesn't clear it up... Maybe there will be no hope but to draw a picture.