Tolerance Stack-up with O-ring Fit

[Raddy13]

If you are doing a stack-up to determine how far off-center a plug inserted into a hole can be, how do you handle it if the plug has O-rings? I have dimensional tolerances on the O-rings, but if the ID is shifted 1mm to the right relative to the OD, will the plug then be 1mm off-center as well, or will the compression even out the O-ring depending on the material?

Is there a good rule of thumb for how to treat O-rings in stackups?

 

[3DDave]

The guiding surface will be the bottom of the o-ring groove unless it's so far off center to make it a fight between the o-ring on one side and metal-to-metal on the other. Typically the runout of o-ring grooves is small enough that the item it is in will be properly centered.

IOW assume the diameter of the bore and diameter of the o-ring bottom surface are perfectly concentric.

I don't know of any o-rings that are shifted that much. If it was, compression would have the highest spring rate where the o-ring was thickest.

However I see no reference where o-ring concentricity is a problem and, given the way they are made, I see little means for it to happen in a measurable way. There is usually some inner diameter tolerance and section diameter tolerance, but never anything about section diameter variation within an individual single o-ring.

 

[greenimi]

I suspect that you already know that a catalog, such as Parker O-ring handbook, could be used as reference to calculate extruding, maximum stretch at installation and may many other conditions.

https://www.parker.com/Literature/O-Ring Division Literature/ORD 5700.pdf

 

[dgallup]

It depends on the forces involved. If the side loading forces are small compared the the force to compress the o-ring it will act as a self centering mechanism. If the side loading is large enough to compress the o-ring, the it's dependent on the other tolerances. In the Parker handbook linked above, they assume worst case, that the o-ring is not self centering. There are standards for o-ring quality, we use MIL-STD-413C. It controls mold mismatch, etc.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[3DDave]

Mold mismatch won't force the o-ring off center. The section area is the same and the displacement will be the same as the section deforms to fill the groove. However, if elastic deformation due to side force is a consideration that will likely require an FEA to resolve and is not part of a typical tolerance analysis.

Also - elastomers aren't compressible, they are elastic; their volume remains essentially constant and if completely contained are more rigid than steel, a factor that has ruined parts when 100% fill is combined with a temperature rise.

MIL-STD-413C is still available, was cancelled for use by the DoD in 2011.

 

[3DDave]

Notice 2, replacement info, for MIL-STD-413C:

MIL-STD-413C, dated 8 December 1980, is hereby cancelled. The acceptance criteria for O-rings as defined in ISO 3601-3, ―Fluid power systems — O-ring — Part 3: Quality acceptance criteria,supersedes MIL-STD-413.

It is recommended purchasers specify the ISO 3601-3 Grade to be used with a sampling plan on the purchase order.
SAE AS5752 is recommended for use as the acceptance criteria for other (non-O-ring) aerospace elastomeric seals.
CAUTION: The supersession information is valid as of the date of this notice and may be superseded by subsequent revisions of the superseding document.

(Copies of ISO publications may be purchased from the International Organization for Standardization’s Website at

http://www.iso.org

and from the American National Standards Institute’s Website at

http://www.ansi.org.)

 

[dgallup]

When the "powers that be" decide to replace a widely used industry standard with an arbitrary new standard that you have to spend $$$ to obtain and are then not allowed to duplicate or distribute, we simply ignore it. MIL-STD-413C is every bit as usable today as the day it was written.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[3DDave]

Even so, '413 has no information that can be used to analyze any aspect of o-ring performance. It's a workmanship standard.

 

[dgallup]

That's what I said 5 posts ago.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[CheckerHater]

When the "powers that be" decide to replace a widely used industry standard with an arbitrary new standard that you have to spend $$$ to obtain and are then not allowed to duplicate or distribute, we simply ignore it.

in keeping with the DoD practice of adopting non-Government standards whenever practicable, Chapters 600 and 700, as contained in previous versions of MIL-STD-100, have been entirely replaced by ASME Y14.35M and ASME Y14.34M respectively, and Chapter 200 is largely based on ASME Y14.24M. An accurate perception of DoD Engineering Drawing Practices therefore necessitates user recognition of MIL-STD-100G, ASME Y14.24M, ASME Y14.34M, ASME Y14.35M, and ASME Y14.100M as being a composite set.

In 1998 MIL-STD-100 was replaced by ASME Y14.100-1998. But I guess you are "not buying it" (pun intended)

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[3DDave]

Y14.5 was a couple of DoD standards: The US Army Ordnance Corps. Standard for Dimensioning and Tolerancing, ORD 30-1-7, and then US MIL-STD-8. I have seen neither of these, but there are references that suggest they supplied the initial symbols and rules before ANSI got their mitts on them.