total runout vs flatness

[gnichols73]

is there a difference between using total runout of a plane and flatness of a plane?

[KIRBSTER269]

Flatness is Form to itself, Runout is in relation to a Datum, it is controlled by another feature, Flatness is not

Untitled.jpg

[gnichols73]

forgive my crude hand drawing, I don't think they want me posting prints but this is sort of what I am looking for.

[gnichols73]

drawing.PNG

[KIRBSTER269]

This doesn't help me, help you, not sure what I'm looking at.

[dph51]

It’s a breast, sheesh, Kirb!

[KIRBSTER269]

dph51 you should know me by now, that was definitely my first thought, but I figured this thread would end up in a ditch real fast.

[KIRBSTER269]

You forgot the eyelashes, I helped you with that, give me more detail.

Capture.JPG

[gnichols73]

first of all thanks for adding the eye lashes. second of all the part is lets call it a saucer. the center is flat and is datum A. then it tapers around the outside at around lets say a 16 degree angle. (I cannot neither confirm or deny any of these dimensions are accurate as to protect the identity of the innocent). they are wanting a runout (one arrow runout) lets call it around 1 inch polar radii from center. in times past have successfully created a plane at these hits and used flatness to represent the total (2 arrow) runout. however pcdmis will not let me select a plane for circular runout of a plane.

[KIRBSTER269]

https://www.pcdmisforum.com/forum/pc-dmis-enterprise-metrology-software/pc-dmis-for-cmms/461001-thoughts-on-circular-runout/page2

check this thread out. I wouldn't call a runout to a surface, the opposite yes, but I would say calling a cylinder back to a surface would be more along the line of perpendicularity.

[Rich P]

to OP
YES
Very different
flatness of a plane is independent of datum's and is only interested plane surface by itself.
Run out as drawn is looking to constrain geometry revolved around the datum axis.

I have only seen run out used on a planar surface that is perpendicular to datum axis but if you have equally disposed flats around and axis such as a hex nut outside then I could see the practical but maybe not legal callout to this.I think profile of the surface would better application GD&T if control of the planar surface is desired.
I have seen a position callout to the theoretical intersection points that constructed makes a circular feature.

[KIRBSTER269]

beat me by 2 minutes

[gnichols73]

thanks for all the help. I found the link you provided informative. as well as all other responses. thanks gang

[louisd]

Just want to clarify that Kirb's wobbly datum in his linked post is absolutely incorrect, EXCEPT in the instance of rotating the part with datum in a V-Block.

Per ASME y14.5 2009:

1.3.13 Datum
datum: a theoretically exact point, axis, line, plane, or
combination thereof derived from the theoretical datum
feature simulator.

To successfully establish the datum, you need to include all form error and derive the true center axis of the cylinder as attained by encapsulating the AME actual mating envelope of the datum surface.

-On a CMM, this is relatively easy with a scanning probe. With a touch-probe, you need to take many circular hits, and at least 3 circular cuts about the axis. Try to use prime number of points to ensure lobing is captured.

-On a lathe, this is also relatively easy, as long as the lathe can be confirmed to not contribute to runout. To confirm: stick a deltronic pin in the chuck, and measure pin for TIR total indicator readout with one indicator. See if runout is over 50 millionths. If runout of your deltronic is abnormal, then you should use two indicators to isolate whatever rotational error your lathe is inducing.

-On a V-Block, you are not establishing the datum axis effectively, as you are not capturing the actual mating envelope of the datum.
This is where you can include form error/wobble of the datum into your runout measurement. If this is your only means of measuring runout, you need to use the 2-indicator setup. I personally would use 2-indicator setup with both indicators at top "12 o'clock" of v-block, then repeat measurement with indicator on datum at bottom/6oclock (or 3/9oclock if not possible at 6), just to confirm how the datum and feature runout values correlate when being spun on the v-block.

[KIRBSTER269]

https://www.pcdmisforum.com/forum/pc-dmis-enterprise-metrology-software/pc-dmis-for-cmms/394863-cad-coordinate-system-disappears-once-i-make-an-alignment-how-to-fix

here delete that attachment, then try uploading a pic

[KIRBSTER269]

Technically it is not incorrect, I just left out basically circularity of that Datum, I blew the picture out of proportion, so others get a better picture of the difference of concentricity and runout. I always True up my Datum as best as the material is given, I check circularity on it, before I use the 2 indicators. I also added info on the other thread. Give me a break louisd I'm getting old, I hate typing, I expect people like you to fill in the blanks, It's enough I have to display all the pictures, since you lazy turds won't do it.

[louisd]

Still scratching my head on how you upload all those pics. I tried following your post about how you upload pics, it and still gives me the middle finger saying i'm over my enormous 1mb allotment.

https://www.pcdmisforum.com/forum/off-topic/off-topic-aa/400936-here-s-a-load-for-you

[KIRBSTER269]

You upload Attachment? louisd I believe it gave me a popup like that a long, long time ago. I believe I deleted an old picture on a post, then I have been posting pictures ever since? I think it was something to do when they changed the site. try deleting some old picture from a past thread.

[[email protected]]

I have query regarding mating GD & T. Face has flatness of 0.05 & run out also of 0.03.Which value should be bigger? run out or flatness?

[neil.challinor]

Generally I would always expect total run-out (double arrow symbol) to give a larger value than flatness. As other people have already stated, flatness is only reporting form, run-out reports form and orientation in relation to a datum and is therefor capturing more information. If it is circular run-out (single arrow symbol) it will depend on how much of the surface you sample. Strictly speaking, you are supposed to sample the entire surface. For circular runout you should extract individual, circular sections and report the worst single value. For total run-out you would report the total range of values across all circular sections combined. To use the dial indicator analogy, it is the difference between taking a series of discrete readings at different radial distances from the centre, reporting the worst single value and of sweeping across the entire surface, reporting the total range.

[[email protected]]

Hi Neil..plz find attached snap for more infi

[neil.challinor]

I'd need to see a bit more of the drawing to be certain - at least what datum A & B are and what shape the surface being measured is. For now, I'm going to assume that the surface you are measuring is something like the bottom of a counterbore or a face that a bearing would but up to and your datum A is a cylinder that is concentric to that counterbore / bearing location. Datum B would have to be another face parallel to the one being measured. For the flatness you would need to measure a plane across the whole face. For the run-out - which if I'm correct in my assumptions, would be singular axial run-out - you would need to measure a circular pattern of points at a fixed diameter of Ø133. The best way to do this would be to use an auto-plane set to circular mode rather than square and set the spacer value to 66.5 (spacer is a radial value - 66.5*2=133). If you have a scanning probe it would be better to use the adaptive plane circle scan strategy.

[[email protected]]

Yes..you are right.its a single run out.run out should bigger than flatness?.I am assuming that for every datum it should have flatness of 0.05

[neil.challinor]

In this case the run-out would probably give a smaller value than flatness since the run-out is only looking a single Ø133 ring of hits whereas the flatness if for the whole face.

[[email protected]]

Still not convinced.what happened if flatness is less than run out?

[UKCMM]

What are you not convinced about.

[[email protected]]

Also I am not able to put snap due to size constraint