Feature request - user defined check add variables to tolerance fields

[[Ti...]]

There are many times we cannot use the Position checks inside GOM b/c they are not your typical features of size. 

We all know you probably should not use a Position check on a non-feature of size let me say again that not all customers know GD&T very well and they don't supply perfect drawings. 

That said, when I get a non-feature of size with a position callout at MMC I must go into the user defined checks and do some math.

The problem is I need to apply the Bonus calculation in the tolerance field. As of right now, you cannot apply a variable into the tolerance field.

So what I am doing is simply this:

(stated_toleranence.actual_value + bonus.actual_value) - (sqr(location_x.actual_value^2 + location_y.actual_value^2)

if this value is positive the check passes. I'd prefer to input the variable stated_tolerance and bonus_tolerance in the tolerance fields so the output replicates the normal Position check inside GOM.

[[Ti...]]

Should be 2*the radial deviation, missed that during my original post

[[Na...]]

Hello Tim,

maybe you can work with a scripted check here:
Scripted checks — Add-On Documentation documentation (zeissiqs.github.io)

Deviating from the gd&t standards is of course at your own risk 😉

Nanno

[[Ti...]]

Please sign in to view this username.

 

Thanks for showing me that, I'll have a look at that. I've got 2 other programmers to consider that cannot understand the python side of this software.

It isn't just deviating from GD&T...for instance we are on 2021 and it cannot apply MMR to spherical 3d position. That is where I developed my technique above except for 3d location you must add the z component to the equation. We have a lot of spherical positions that are critical dimensions. Bouncing back and forth from the cmm to the CT was getting tedious for everyone so I just played with the user defined, but the tolerance I use is not normal but gets the job done.

[[Ji...]]

Just curious - bonus tolerances accumulate as a features deviate from MMC.  If the feature(s) you are referring to are not features of size (as you mentioned), how are you calculating the bonus tolerance?  What are you using as the "size"?

[[Ti...]]

Please sign in to view this username.

 

Below is an example of a simple case where the 100mm +/- 0.1mm width. The ends are radiused so the length is to and from the extreme tangent points. Using a position tolerance on this length is not a perfect way to do things, but as I have said plenty of times on here, customer prints are not always to the standards.

 

Furthermore, you could argue that a positional tolerance on this length 'could' be the most efficient way to measure a desired outcome that both 'extreme tangent points' are a distance away from the center hole to a pretty tight tolerance. You can't reliably use each end radius as a separate position b/c those cylinders you would create may not be stable due to the small degree of arc. Surface profile would be the go to solution, BUT surface profile factors in the lows and the high points AND surface profile does not allow Bonus.  So how do you control the below example...while position is not ideal per the standard, you can easily make a hard gauge for this to control position, you can easily do this on a CMM, and on a vision system.

 

i personally don't care that this isn't a 'feature of size' per the standards definition, I know that is is a  size that can fluctuate and I can make an accurate measurement of this feature of length.  In GOM you could collapse a disc caliper using the longwise surface planes for direction, and use a disc intersection point to get this points in space, then make a mid point.  I then use the user defined checks to calculate my bonus, another user defined check to find my diametrical mid point deviation, then a final user defined check to use the formula stated earlier to check that my bonus + stated tolerance is larger than my diametrical deviation.

 

A real nice solution that you can use in pcdmis cmm is make a circle out of your mid point and one of the outer extreme points. This circle allows you to get a feature of size check and position check.  In fact this strategy actually works inside GOM except one big problem....it allows you to create a linear size of the circle and then allows you to do a MMC position check....BUT the problem is GOM is not passing the MMC data to the position check. So it appears to work, but it is not passing along the bonus to the output.

 

Edited November 23, 2023

[[Ji...]]

One option that may work for you is to use a CAD file that has square ends.  Then you can create a size dimension that can be used in the position callout of the centerplane.  The algorithm that you choose for the size dimension calculation can be modified to your liking.  This size dimension will then be associated to the positional callout and bonus tolerance will then be applied.

I created a model based on your sketch and tried it - seems to work:

[[Ti...]]

Please sign in to view this username.

'

Jim...thanks for sharing. That is an interesting approach, I'll have to give that a shot. quite intersting, thank you.

[[Ji...]]

You are welcome.  As a follow-up, I believe it can be argued that this example you provided is indeed a feature of size by definition.  Perhaps you should urge Zeiss to consider including features such as this for creating GD&T size dimensions.

[[Ti...]]

Please sign in to view this username.

I agree with you capturing a length like this should be able to be done with GDT in the software. We need the ability to push the limits here and there and not be so confined to a technical nature.

That said, I'll argue from a purist standpoint this is not a feature of size because it lacks what I call 'confinement'.

The asme standard states in the 2009 standard that a regular feature of size(fos) is:

'one cylindrical or spherical surface, a cicrular element, and a set of two opposed parallel elements or opposed parallel surfaces, each of which are directly toleranced'

Then it states there are irregular fos which are:

directly toleranced features that "may contain or be contained by an actual mating envelope"

 

This feature I provided is not regular, and then to be considered irregular would need to be able to be 'contained' ie. locked down by a hard gauge. So my purist opinion is it is not a fos.  But a quick hard gauge would be to make a pin for the center hole(primary datum) so you can spin the part around with the flick of a finger. Then you collapse a plane perpedicular to the longwise body surface(secondary datum) until it cannot rotate around anymore. Then rotate the part to the other side and push a feeler gauge until you get a solid contact. Now you know your position ;)

 

[[Ji...]]

This is certainly an interesting topic of discussion.  My view is that this condition falls under the “Irregular Feature of Size” category. From ASME Y14.5-2018:

3.35.1 Irregular Feature of Size
irregular feature of size: there are two types of irregular
features of size, as follows:
(a) a directly toleranced feature or collection of
features that may contain or be contained by an unrelated
AME that is a sphere, cylinder, or pair of parallel planes.
See Figure 7-41.
(b) a directly toleranced feature or collection of
features that may contain or be contained by an unrelated
AME other than a sphere, cylinder, or pair of parallel
planes. See Figures 7-40 and 11-29.

Consider that the features under consideration can be contained by two parallel planes:

 

It would be interesting to get some other expert opinions.

[[Ji...]]

Some more information for consideration:

Here is figure 7-40 from ASME Y14.5-2018:

 

The title of the figure is "Irregular and Regular features of Size as Datum Features".  Since datums B and C are considered regular features of size (a set of opposed parallel surfaces), it can be concluded that datum A is an irregular feature of size.  Datum A is similar to your example, so I believe your example can indeed be classified as an irregular feature of size.

 

[[Di...]]

Hi,

regarding the provided CAD element I created my own one using the data from your sketch. And I do not see the problem. What I'm doing different compared to your approach?

Regarding more complex features (ASME freeform shaped irregular features): This is indeed not yet supported.

So if you like to have this please send a software wish to our support department.

Thank you very much for sharing your wishes with us.

Best regards

Dinko

Edited November 29, 2023

[[Ji...]]

Dinko,

I see the "tolerated element" you used for the size is a cylinder:

 

 

If you look at his sketch, there is no such cylinder that exists.  There are two separate arcs (one on each end) that don't share a common center point.  The feature of size that is referenced is the outermost distance of these arcs.

[[Ti...]]

Please sign in to view this quote.

 

You cannot 'contain the radii at the end of this long part with two plane. You can collapse two planes and 'touch' each end but the slightest flick of a finger and the part would dislodge and spin freely. Containment means you have immobilized the features with the UAME, unrelated actual mating envelope. Two planes collapsing upon these radii end points is not a mating envelope.

 

 

It is real simple to add this to the software....We can already create two opposing points in the software and then a midpoint. Then we can create a normal circle using the midpoint and one of the outer points to parametrically define the cirlce.....Now there is a feature of size built 'generically' same as we do in PCDMIS.

 

You can even go into the linear dimension and generate a GDT size for this generic circle. You can also then apply this MMR feature to a bonus position callout.

 

BUT....as I have said before....all this works EXCEPT....the bonus does not get pushed over into the positional check. Everything works, except the bonus doesn't carry over.  if you all can add the carryover into the position check, i can create all the position checks I want that are not purist feature of sizes.

Edited December 1, 2023

[[Ji...]]

Hello Tim - 

This is certainly an interesting conversation.

Based on your last comment, I see where we differ in our viewpoints is in the definition of the word "contained".  I believe you have interpreted  it to mean "restrained" or "constrained" as are typically used when a part is to be fully immobilized.  In this case, the word "contained" means that the feature is question must be "enclosed" or "bound" by two parallel planes.  Of course, the vector direction of measurement has to be defined and would be the same direction as the two  bounding parallel planes.

So, from my experience, the example you have provided is indeed considered an "irregular" feature of size. Also, as Dinko Sabo has commented above, "Regarding more complex features (ASME freeform shaped irregular features): This is indeed not yet supported."  

From this standpoint, I believe the ball is in Zeiss' court add functionality to the software to support irregular features of size.

[[Ti...]]

Please sign in to view this username.

 

To have a feature of size regardless if it is irregular, it must have a 'mating envelop'.  You quoted the standard above referencing the abbreviation 'AME'...actual mating envelope.

A feature to have a size must mate up with another feature. Two collapsing planes can only touch two extreme points in this situation...that does not mate, it only touches. Try that mating strategy in Solidworks, or Siemens NX.  If it did everything under the sun would be a feature of size. So two planes collapsing upon a part can only find two identifiable extreme points is now considered a 'mating envelop'? Those parts are now mated together by those two points?

 

 

Edited December 10, 2023

[[Ji...]]

The mating envelope would be the extreme boundaries of the surface in consideration (in this case the small radii on each end).  The size dimension would be the extreme distance in the direction of two parallel planes.