Stackup - paralelism

[roberto1brazil]

Please, could someone help me with the following issue?
My objective is to calculate the maximum and minimum gap possible (RSS case) when assembling two identical washers into an envelope (part) of 10,10 +/- 0,02mm of length (refer to the figure showing the assembly. The two washers have the dimension shown in the figure attached and a geometric tolerance as required in the same figure. In my mind besides the dimension and its washer tolerance, we would have to add the tolerance of +/- 0,01mm (twice) between the washers regarding the parallelism contribution. The flatness, in this case, has no influence in the stack as I have researched (see figure about flatness). Isn’t it? Inputting all those data in the table stack up – I have find a nominal gap of 0,10 mm with a tolerance of +/- 0,06 mm –RSS case. So a minimum of 0,04 mm and a maximum of 0,16mm gap would be the limits (RSS case). Could you please say if my assumptions are correct?
Second question – The thickness of the washer 5, 00 +/- 0, 04 mm can coexist with the parallelism zone of 0, 02? In my mind, I would say yes. The thickness of washer could be between the maximum and minimum dimension and at same time keeping the parallelism required. . Please refer to the sketches attached.
*FLATNESS CONCEPT in stackup
Considering flatness tolerance within dimension tolerance (in the case described above is 5,00 +/- 0,04mm). Showed in the sketch.

Thanks and regards

 

[pmarc]

Before anything else, which GD&T standard are you working to? ASME or ISO? This is very important to know as the stack-up interpretation will be different depending on the standard.

 

[roberto1brazil]

Despite of making use of milimeters in the dimension of the parts, I am having as reference the ASME.
Thanks

 

[pmarc]

OK, so if ASME is in charge here then:
- both specified geometric tolerances for washer (parallelism & flatness) have no influence on the tolerance stack you are considering;

- based on assumption that you meant worst-case scenario not RSS (root sum square), you are right that minimum gap possible is 0.04 and maximum is 0.16. This is a result of simple addition and substraction of nominal sizes and their tolerances.

- both geometrical tolerances for washer are applied correctly. As long as tolerance values of flatness & parallelism are not greater than tolerance for width of the washer 0.08 everything is absolutely fine.

 

[dingy2]

First of all, I can see a bit of a conflict where the parts might not fit and everyone pointing fingers at each other.

Lets go theoretical first.

The large the washer can be is 5.04. So we have 2 washers at 5.04 and the minimum opening is 10.08. Looks like we might need a shoe horn and some grease to get those together since, in theory, we have no gap. We have size on size.

Here is the problem as I see it. In theory, parallelism must be inside the size tolerance but people on the shop floor check the size or thickness in a couple of spots and then check the parallelism separately. We could have an oversize product this way. The overall width should be shown with a profile of a surface of 0.08 with the 5 mm dimension as basic. If one wanted the refinement of the profile tolerance with parallelism, then go for it. No conflict now and the shop floor would confirm the profile first and then the parallelism. I believe that this is a better way but I still would not want size for size. We still need a bit of wiggle room when the washers are at their maximum size and the opening at its smallest.

The maximum gap as it stands now is (10.12 - 4.96 - 4.96) 0.2 mm.

Dave D.

http://www.qmsi.ca

 

[pmarc]

roberto,
Dave is right that minimum worst-case gap is 0 and maximum is 0.2.
Sorry for the confusion.

 

[roberto1brazil]

Hi Pmarc and Dave
Thanks for the answers and explanations. According to I have understood the geometrical feature parallelism (in this case in particular?), does not affect the stackup result. When I had calculated the minimum value of 0,04 and maximum of 0,16mm I had introduced the dimension +/- 0,01 (total of 0,02 multiplied by two because I had two washers) regarding the parallelism contribution and I had the result at RSS and not worst case. I agree that the worst case would be zero and 0,2 max with no parallelism contribution. Just another question – Are there cases that parallelism is took into account to calculate a stackup gap or not?
Regards.
Roberto

 

[pmarc]

In your example parallelism or flatness would be taken into account if you were searching for allowable gap (amount of air) between two washers for instance.

Imagine that the space for placing two washers is smallest possible 10.08 and the washers are at maximum width 5.04. So theoretically there is no space between parts within the assembly but in reality due to flatness or parallelism error there might be a space between nominally flat surfaces of elements. This is especially important when considering applications where leakage control is really important.