Distance Dimension Options

[brianparker]

Would like to better understand how the options work here. Looked in the "Help" file, did not find it helped me much (I typically need things dumbed down a bit when it comes to PCDMIS).
If I am looking at distance from Corner-Corner (point-point-same Z height) on the pictured tub, would this always be "3D-shortest? Makes a sizable difference if I leave it in "2D-to feature", but not so vast a difference that it couldn't be possibly correct.
Thanks in advance for any help
Distance Options.JPG

[Matthew D. Hoedeman]

That would be 2D as you only want the distance between 'as seen from the Z workplane'. No 'to feature', nothing, simply 2D, pick the 2 points. MAKE SURE you are in the correct workplane

[brianparker]

That would be 2D as you only want the distance between 'as seen from the Z workplane'. No 'to feature', nothing, simply 2D, pick the 2 points. MAKE SURE you are in the correct workplane

Thanks,
It's in Z plus workplane, appreciate the clarification.

[AndersI]

How do you manage to put the points exactly on the 'corners'? Seems to me you will mostly be dimensioning CMM or manual accuracy in placing the points?

2D distance: the points are projected to the current workplane, then the point-to-point distance is calculated
2D To feature: According to the Help, this choice needs a third feature indicating the direction for the distance (in combination with Perpen to/Paralell to).
3D distance: point-to-point distance is calculated - will be the same as the 2D distance unless the points are taken at different heights - if so, the 3D result will be longer.

[UKCMM]

My preferred method for his type of feature/dimension combination would be to measure the corner radius and call out a 2D distance of the radius with Add radius.

[DAN_M]

This is unrelated to your original question..though I feel it must be said. Regarding your dimension name...

PC DMIS allows you in part name fields, dimension name fields, and other data entry fields, to use math. If you use special characters (like the "-"s you used that PC DMIS has reserved to mean minus), spaces, OR ANYTHING OTHER THAN letters, numbers, and underscores... you'll eventually begin having issues where your programs will become corrupted. That being said, just some friendly advise, I would highly recommend you get away from doing stuff like that. I would have named that dimension something to the effect of "C_C__U_R"

If you're writing lower complexity programs with no variables/funky stuff going on, you'll most likely be OK. But the more advanced you get, the more issues you'll have.

Goodluck my friend, hope you get your issue figured out!!

[KIRBSTER269]

I don't know about you, I get tired of repeating this over and over and over. Just wonder how many, of these people that come on here asking questions, because they're lacking the KNOWLEDGE, say to themselves, "That's irrelevant to my question, ignore" or "That guy don't know what he's talking about"

[Douglas]

I'm finding exactly this if I add any assignments to an older existing program... just spaces in feature names are a problem and I need to replace them with underscores. With a feature name like above any assignment of measured values to variables would be a total crapshoot

[brianparker]

This is unrelated to your original question..though I feel it must be said. Regarding your dimension name...

PC DMIS allows you in part name fields, dimension name fields, and other data entry fields, to use math. If you use special characters (like the "-"s you used that PC DMIS has reserved to mean minus), spaces, OR ANYTHING OTHER THAN letters, numbers, and underscores... you'll eventually begin having issues where your programs will become corrupted. That being said, just some friendly advise, I would highly recommend you get away from doing stuff like that. I would have named that dimension something to the effect of "C_C__U_R"

If you're writing lower complexity programs with no variables/funky stuff going on, you'll most likely be OK. But the more advanced you get, the more issues you'll have.

Goodluck my friend, hope you get your issue figured out!!

Thanks for that,,,had some trouble getting variables to work awhile ago, and was informed of that very issue. appreciate the input!

[brianparker]

How do you manage to put the points exactly on the 'corners'? Seems to me you will mostly be dimensioning CMM or manual accuracy in placing the points?

2D distance: the points are projected to the current workplane, then the point-to-point distance is calculated
2D To feature: According to the Help, this choice needs a third feature indicating the direction for the distance (in combination with Perpen to/Paralell to).
3D distance: point-to-point distance is calculated - will be the same as the 2D distance unless the points are taken at different heights - if so, the 3D result will be longer.

The points are not randomly placed, built utilizing trig function of cosign/ radius, see attached.

[louisd]

This radius/cosine point you are making is dependent upon the shape of the corners being a perfect square (vector of 0.707168).
The picture of your part shows a rectangle.

You are better off measuring each radius. Then construct a rotation alignment to a line between opposing radii centroids. Rotate about X or Y
From there you can construct a generic point, if you rotated about X to the line,
set the generic theo's to x = CIR2.TX+CIR2.TR, Y = 0, and Z = 0
set the measured values to x= CIR2.X+CIR2.RAD, Y = 0 and Z = 0

do the same for opposing point. just subtract the radius on the other one.
measure a distance between both generic points

[brianparker]

This radius/cosine point you are making is dependent upon the shape of the corners being a perfect square (vector of 0.707168).
The picture of your part shows a rectangle.

You are better off measuring each radius. Then construct a rotation alignment to a line between opposing radii centroids. Rotate about X or Y
From there you can construct a generic point, if you rotated about X to the line,
set the generic theo's to x = CIR2.TX+CIR2.TR, Y = 0, and Z = 0
set the measured values to x= CIR2.X+CIR2.RAD, Y = 0 and Z = 0

do the same for opposing point. just subtract the radius on the other one.
measure a distance between both generic points

Hey Louis, thanks for the help, lord knows I can use all I can get on PCDMIS! Seems like we've tried something similar to what you've listed, and we get something more aligned with the opposing tangencies, like the red line shown below, kid of the same as going circle-circle, when looking at a rectangular container..., am I wrong on this? Going to try the above as soon as I get time, just "thinkin out loud" here.....
Again, sincerely appreciate any and all input,

[louisd]

No. You would be creating a line from measured radius centroids using my method. With this align function then generic point added, you are simply extending that line the distance of each of the measured radii, and constructing a point precisely where the line would pierce the radius.

[louisd]

come to think of it, yes, it would be opposing outer tangents of the circles. the longest 2-point measurement you could extract if you used a caliper would match the method i described. Is this not what you want? Why would anyone want the distance between non-tangental points (45° corners) as a feature of size?