A sample problem about choosing datum for machining and its effect on assembly accuracy

[Vah1D]

Hi all,

The attached pictures was given to us by my employer, I asked them if it's ok to share and they were ok with it since they are not the original producer of the information.

I had a little bit of debate about this with my co-workers, a few said in this particular example it doesn't not matter if all your datum surfaces (blue colored) used to machine the dowels are on same side as shown in the picture, but others said it matters, that it can affect the final position of the final unit on the top.

I'm with the first group. I understand generally what the article is trying to say, but let's say in the example provided, for the green block, if I had the A' surface exactly on the opposite side in the 2D print, and ask the machinist to drill the (b) dowel holes from the top surface, I still get my (c) dowel holes where I intended to be, even though the datum now is not on the same side as datum of main base. Am I missing something? Any ideas?

https://imgur.com/a/JgFLQfv

Thanks,
Vahid

 

[drawoh]

Vah1D,

Do you actually care how your blue datum faces line up? If you don't, then it does not matter which face you use as a datum.

In most applications like this, I actually do not care (much) about the component outlines. It is good practise to use attachment and positioning features as datums.

My primary datums would be the horizontal attachment faces. My secondary datum features would be the dowels. You would need to specify datum targets on the appropriate quadrants of your dowels, as opposed to their diameters. This is weird, but it conforms to your assembly and positioning scheme. You need some sort of tertiary datum features.

Your plate outlines would be controlled by sloppy profile tolerances.

--
JHG