Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations SDETERS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Profile measurement question

Burunduk

Mechanical
May 2, 2019
2,470
What would be your preferred way of manually inspecting this profile tolerance?

1000019119.jpg
 
Replies continue below

Recommended for you

Burunduk
I would use a sine plate with a magnetic v block
On a surface plate. Sin plate kicked to 105 degrees. Dial indicater holder on a surface plate.
Slide dial indicater over the face while locating on datum-A- and -B-

An other would be same set up except CMM
 
Burunduk
precisely my friend.
the gage block for zero is dead on
 
Burunduk,
In this setup, is the height of the stack of gage blocks used to establish the zero reference calculated based on the distance between the intersection point of datum axis A and datum plane B and the surface plate? Is the point-to-plate distance going to be maintained for the measurement of all piece parts (assuming there is more of them to inspect)?
 
A quick calculation of the stack up would determine gage block setup.
Also a quick cad drawing would make short work.
Once this setup is completed. It is a very fast method to inspect a lot of parts.
I done it many times.
I can ad one two three blocks can aid in the location of datum B
 
pmarc,
I merely illustrated mfgenggear's suggestion as I understood it. I think yes, with the dimensions of the sine bar and the exact and constant placement of the pair of V blocks taken into consideration, it will be necessary to calculate trigonometrically the vertical height from the surface plate to the intersection point of datum axis A (center of the pair of V blocks) and datum plane B (the inclined top face of the sine bar). Can't tell the exact math right away, I'd probably check that with CAD. To that, I would add 35cos15° to know the height of the zero reference stack of gage blocks.
 
Last edited:
pmarc,
I merely illustrated mfgenggear's suggestion as I understood it. I think yes, with the dimensions of the sine bar and the exact and constant placement of the pair of V blocks taken into consideration, it will be necessary to calculate trigonometrically the vertical height from the surface plate to the intersection point of datum axis A (center of the pair of V blocks) and datum plane B (the inclined top face of the sine bar). Can't tell the exact math right away, I'd probably check that with CAD. To that, I would add 35cos15° to know the height of the zero reference stack of gage blocks.
Got it.

One more question: Assuming the flat surface was manufactured exactly at the true profile (for this condition we know the measurement reading should be zero), how does this setup ensure that the flat surface is parallel to the surface plate?
 
pmarc, unless I'm missing something critical, this nominally parallel orientation is obtained by setting the sine bar's angle to 15° (can be done by piling the small stack of gage blocks shown on the right side in my illustration to the height corresponding to the opposite cathetus of 15° in a right angle triangle with the hypotenuse being the distance between the rollers). Obviously the accuracy of all the gage components in this set up will contribute to the measurement uncertainty.
 
Burunduk, I was not thinking about the accuracy of the gage components, but rather about something as shown in the picture below (it illustrates an exagerrated situation of the part rotated 180 degrees about datum "axis" A).

Does this make sense?

d.JPG
 
pmarc, that's why I said "unless I'm missing something critical" :)
But speaking seriously, I think that the unconstrained degree of freedom makes it OK to optimize the rotation of the part around the datum axis for having the "best-fit" to a horizontal condition. Although on second thought, it would be an optimization not just for horizontality but also for the location relative to the zero - a bit of a compromise of each for minimizing the measured value. Anyhow, I think it should be OK. Do you agree?
 
pmarc
if the correct angle is used on the sin plate.
and sliding the dial indicator to cause the
dial indicator over the profile of the part.
the dial indicator should not move more than
.050 wide. if it is ground it will be .0005 inch..
to grind the parts the same setup would be used to grind the surface after A and B are
trued up. all degrees of freedom would be restraint.
 
pmarc, that's why I said "unless I'm missing something critical" :)
But speaking seriously, I think that the unconstrained degree of freedom makes it OK to optimize the rotation of the part around the datum axis for having the "best-fit" to a horizontal condition. Although on second thought, it would be an optimization not just for horizontality but also for the location relative to the zero - a bit of a compromise of each for minimizing the measured value. Anyhow, I think it should be OK. Do you agree?
All I wanted to accomplish with my question was to hear that an optimization would be needed as an additional step in the inspection process to ensure satisfactory level of confidence of readings.
 
pmarc
if the correct angle is used on the sin plate.
and sliding the dial indicator to cause the
dial indicator over the profile of the part.
the dial indicator should not move more than
.050 wide. if it is ground it will be .0005 inch..
to grind the parts the same setup would be used to grind the surface after A and B are
trued up. all degrees of freedom would be restraint.
My additional question was not about the manufacturing precision and the accuracy of gage components used in the proposed setup. Notice that in my example I made the flat surface perfect, yet the dial indicator reading could get quite significant due to a non-optimal installation of the part before taking the readings.
 
"All I wanted to accomplish with my question was to hear that an optimization would be needed..."
And it's good that you did 🙂
It's a good and important point.
 
That's my point if the indicator verifies in tolerance across face. You know it is good.
 
mfgenggear,
One more question if possible:
How would you make sure that the location of datum axis A is repeatable for all measured parts, and independent on the variation allowed in the diameter of datum feature A (Dia. 20+/-0.05)? Note that in my illustration the axis location is approximately in the middle along the sine plate/bar, but obviously it can be set up differently. I ask because that longitudinal location also impacts the required vertical height of the stack of the zero reference gage blocks, and it better be kept constant.
 
Burunduk and pmarc
with part in vblock is slid and concurrently is leveled out by indicating the face.
the height of the indicator is not moved , but the entire base of indicator is slid on the surface plate.
"as an experience mechanical inspector and machinist it is easier to instruct than to verbally explain."

the vblock with the part is slid length wise on the sin bar. it sounds cumbersome., but here is the deal.
i would inspect and record each diameter , and segregate all parts with the same diameter.
a note: during machining, I would have tighten the diameter and face dimensions for tooling.
 
mfgenggear,
Sorry, I'm not sure I follow. Possibly I wasn't clear enough with the question too.
I'd like the inspected profile to be based on both dimensions 35 basic and 105° basic, not just 105° basic which is leveled to horizontal condition by the sine bar set at 15° (or that would only be an angularity check). So I'd want the v-block not to slide, up/down the sine bar, because if it can slide and get that way to the zero reference height then dimension 35 basic is not part of the control. On the other hand if the axis is fixed along the sine bar and the gage blocks height for zero reference is set up right, the location of the profiled face in the A,B datum reference frame is under control. I'm basically asking how would you fix the X dimension added in the below image. If you fix just the left side or right side V-block then the diameter variation of datum feature A will make an unwanted impact, so it's needed to somehow fix the center between both.
 

Attachments

  • 1000019125.jpg
    1000019125.jpg
    155.8 KB · Views: 10

Part and Inventory Search

Sponsor