ISO 1101 concentricity (concentricity is still evil)

[joebk]

Since our parent company is German we have to deal with the ISO system of GD&T regularly. They use tend to use concentricity to control coaxiality of features. Since they have design control of these drawings and indicate that ISO 1101 must be used to interpret them we are struggling to adapt. Unfortunately they are not forthcoming with explanations for call outs or consistent with the definitions in the standards so we are somewhat up the creek. So the best we can currently do is refer to the standard and hope for the best.

We have the ISO 1101 standard to refer to and it would appear that there are two definitions for concentricity in the standard (as if concentricity wasn't bad enough already). One that mirrors the ASME Y14.5 definition and one that defines concentricity of a point.

I am not sure when to apply each definition. It is my current assumption that if concentricity is called out to a single datum (the reference diameter/cylinder) concentricity of a point applies. If concentricity is called out to two datums (primary plane and reference diameter/cylinder) concentricity of an axis similar to ASME Y14.5 definition applies. This doesn't seem right to me but the definitions and figures in the ISO standard are poor. There is no statement indicating when each condition applies.

In the example within the standard, the modifier ACS is used (any cross section) above the FCF. This figure is contained within the definition for concentricity of a point. Is this the note that indicates that concentricity of a point is indicated? This modifier is never used (to the best of my knowledge) on any of the European devision drawings.

Concentricity is covered on pages 38-39 of the standard.

Thanks

JoeBK PE

PS not to impressed with the ISO standard. The definitions don't seem to be as well through out as the ASME Y14.5 definitions. I am not to happy about having to purchased a whole bunch of separate standards either!!

 

[PaulJackson]

My copy of ISO 1101 is old (1983), the latest revision I think is (2004), I don’t have it. The (1983) revision has two tolerance descriptions as well, (concentricity to a datum point) and (concentricity to a datum axis). The distinction between the point and the axis seems to depend upon whether the datum reference (not to be confused with datum feature) is considered as a circle (2D if you will) or a cylinder (3D). I don’t believe that the distinction rests in whether there is one datum feature as opposed to two as depicted in the illustration. I would assume from your comments that some refinements (specifically ACS) may have been added to the (2004) revision so I cannot comment on that.

I think that the ISO definition of concentricity more closely resembles the (1982 ANSI 14.5) description of concentricity where the “individual cross sectional centers” of the considered feature form must reside within a (cylindrical) diameter tolerance zone.
I have not read the (2004) revision of ISO 1101 but I don’t think that the TC 213 (ISO’s equivalent to our Y14 standards committee) has ever entertained the “median point of diametrically opposed surface elements” that our 1994 ASME standard revisers “unfortunately” saddled us with.

When the ASME definition changed in 1994 I was disappointed because as powerhound stated in CONCENTRICITY VS. THE CMM (CONCENTRICITY IS EVIL) “there is no substitute for concentricity when balance …is a factor” Although the new ASME definition can reasonably control the average center of uniform continuous surfaces with even lobbing i.e. ellipses, etc. it fails to control odd lobbed or interrupted surfaces i.e. odd tooth gears, vane pump rotors, etc.

The 1982 standard cautioned that “determining the average center may be difficult to measure” but that is precisely CMMs do with circular regression equations figuring “best –fit” centers from measured surface points. I know that the ASME committees sought to refine the analysis distinctions between Runout, Position, and Concentricity in concert with Y14.5.1 Mathematical Definition but I think that the effort fell short with concentricity by reducing the interpretation to “median point of diametrically opposed surface elements” never-the-less that is what we have with the ASME standard.

I have consulted on numerous DIN, ISO, BS, and JIS conversions and/or interpretations to American National Standards and I have seen concentricity specified used interchangeably with position where the toleranced features were not coaxial with the datum features, so don’t be surprised if you see that. In fact, the ISO 2692 first edition (1988-12-15) “Geometrical Tolerancing – Maximum material principle” standard that I have shows concentricity specified with MMC modifiers on both the datum and toleranced features.

Paul F. Jackson

 

[dingy2]

If a Designer wants to control coaxial conditions on two cylincrical features, I would suggest positional in RFS.

If a Designer wants to control the surfaces on two SMALL coaxial features for balance, I would suggest circular runout but indicate the checking locations using phantom or chain lines where the readings should be taken. Total runout could also be used but may not affect the balance as much as circular runout.

Circular and total runout should be performed on a divider head (chuck) with a indicator and stand. There would be a 360 degree contact on each segment.

Just some thoughts.

Dave D.

http://www.qmsi.ca

 

[joebk]

Based on what you indicated I think the definition is basically the same in the newer ISO standard Paul. The definition and figures seem vague to me and confuse the issue by using modifiers such as ACS, multiple or single datums, etc without ever saying exactly when each situation applies.

I don't think we can convince our European counterparts to change, we have to figure out exactly what is intended on a case by case basis. Yuck. So the hunt is on. We will probably have to figure out what the engineering department means, how the machinists interpret it, and how quality confirms it. Double yuck.

I would guess at the end of the day it will boil down to position RFS.

In our corporate addendum (only applies to USA division) to the ASME Y14.5-1994 standard we reiterate the ASME definition for concentricity (and symmetry for that matter) and strongly recommend against the use of the spec unless absolutely necessary. We have had way to many problems with these geometric specifications. There seems to be to much historical baggage attached to these callouts. I.e prior practice in the shop and QA as opposed to the standard definition.

If it is of interest, when balance is necessary we specify a balancing requirement.