verify hole position at rfs with adjustable gage

[bxbzq]

Hi,
This may be a simple question. Let's say my part is a rectangular plate with holes being perpendicular to bottom surface. As usual the bottom surface is primary datum feature A, two edges are secondary B and tertiary datum feature C. I use position tolerance to control the pattern of holes on the plate. The position tolerance is applied at RFS and relative to A, B and C. Now instead of using CMM to verify the holes position, can I verify the holes position with a gage on which the pin diameters are adjustable? The adjustable range of the pin diameter is the inner boundary to outer boundary of the holes.

 

[pmarc]

That depends what you mean by adjustable gage.

It is doable, if the gage pins for position check - apart from being perfectly oriented and located relative to datum feature simulators A, B, C - have ability to be set (fixed) at certain size somewehere between inner and outer boundary of each hole individually depending on the size of unrelated actual mating envelope of each hole. However these cannot be gage pins which expansion and/or contraction will be stopped only when in the pins are in contact with actual surfaces of corresponding holes, becasue this would not reflect "physics" behind position at RFS.

And even if the gage pins are as described above, you will still need additional pins (unrelated to any datum) that would simulate size of actual unrelated mating envelopes of each hole. These in turn would have to expandable in the way described in the second sentence of previous paragrpaph (following "However...").

 

[Belanger]

It also depends on what you mean by "inner and outer boundary." Certainly the inner boundary would be one limit (MMC - position tol) but the pin really wouldn't need to go out to the outer boundary (LMC + geo tol) because any position error makes the hole appear smaller, not bigger. (This may be the same idea that Pmarc has given.)

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[bxbzq]

pmarc,
I'm not sure what you mean by "this would not reflect "physics" behind position at RFS."

The inner and outer boundary are the same terms in Y14.5. Unless I miss something, to me it's not just doable, the concept is quite simple. The gage is not simple though. Like pmarc mentioned, the pins are located and oriented perfectly relative to datum feature simulator, and expandable. If the pins go into the holes on the part, you measure the expandable pin diameters. I don't need additional pins to get the unrelated AME value, a micrometer does it. I subtract the measured pin diameter value from the micrometer reading, and divided by 2 I get the position and/or perpendicularity error. If pins don't go in, the part is rejected.

 

[pmarc]

bxbzq,
Yes, this can be done that way and is easier in comparison to what I proposed. However I think that micrometer itself will not allow you to find a size of unrelated actual mating envelope of each hole properly. This value is needed, because it is the size of UAME - not of any local two-point measurement - from which the size of expandable gage pin is substracted.

 

[bxbzq]

The micrometer I mentioned is not two-point measurement, but rather the type that has 3 or more jaws and can expand inside a hole.

 

[pmarc]

The micrometer would have to have ability to act as expandable cylindrical gage pin unrelated to any datums. If you think your micrometer acts like this, I have no further remarks

 

[dingy2]

Well, this is a mess. It is too bad that you could not convince the Designer that having positional tolerances at MMC since it will not hurt the integrity of the design. I have just got to see you gauge since it is not practical (nothing is impossible) to use a hard gauge in a positional application in RFS. You must contact the inner boundary of the hole and then how do you apply the diametrical tolerance zone. Also, you must come up with a value the same as using a CMM. Just get the Designer to change this before you blow your mind.

Dave D.

http://www.qmsi.ca

 

[MechNorth]

One huge issue, which I think pmarc was heading for, but nobody picked up on (or at least stated directly) ... the simulator for the feature to be inspected needs to be the largest size pin that fits within the hole and is NOT oriented wrt the datum simulators. Huge difference there. You are not looking at boundaries, you are looking at the actual axis at RFS. The proposed fixture requires MMC to be anywhere close to valid.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[bxbzq]

Jim, I think I see where the problem is. But using swing gage pins like you said does not solve problem, only brings in complexity to gage pin.
A sketch would help. Maybe tomorrow.

 

[MechNorth]

bxbzq; what you are asking may not be possible within a correct interpretation of the standard. Often what we want isn't what we are allowed. Hopefully your sketch will clarify.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[bxbzq]

Ok, the problem is from the perpendicular error. See attached. Since there is no difference between hole pattern and single hole on a plate, so I just sketched one-hole scenario. So, the calculation would be,
position error = (UAME/2)/cos(α)-pin_dia/2

Not sure if it's doable, or easier than CMM, though.
Maybe an approximate approach?

If use a swing gage pin, you still need to account for the perpendicular error. You will also need a special gage pin that has a pivot inside the pin at true position and at middle of the plate height. Even this is possible, the result will be same my calculation.

Comments?

 

[bxbzq]

By approximate approach I mean omitting the cos(α).

 

[pmarc]

I think I misunderstood your original question. I thought you asked whether it is possible to use hard gage in order to verify positional callout at RFS without getting a quantifiable data. This is possible, if you know the actual value of UAME. And this is what I was trying to say throughout the whole thread.

However, if - basing on the gage built - you want to know what is the value of actual positional error, I say it is impossible without making the assumption you want to make (the one with cosine).

 

[MechNorth]

If I ever saw data come across my desk as you propose, I would reject the lot, and probably cancel the contract. I have delisted and back-charged suppliers for invalid inspection. That is NOT what the callout means. If you elect to inspect it using an RAME pin, then your numbers are invalid and, if you know the difference between the methods, potentially fraudulent.

Knowledge of the standard guides you on how to validate controls. For your proposal, you would still have to use a RAME pin to find the angle, so what's the benefit?

If you want to hard gauge, the control tolerance needs to be modified MMC.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[bxbzq]

pmarc,
I'm trying to understand your idea. Use the sketch I created, if the as-produced hole UAME value is 20mm, and the gage pin dia can be adjusted from 19.2 to 20.8mm, then I can use the gage to verify the hole position by getting attribute data (pass not not pass). Am I correct?

Jim,
If you are saying omitting the tilt angle is not acceptable, I have no problem and will give up the idea. But from your posts I feel my fundamental understanding is being challenged.

That is NOT what the callout means.

I thought I knew the differences between MMC position and RFS position. Now I just don't see which difference makes my numbers invalid.
Sorry for being silly.

 

[pmarc]

Yes, that is correct. And when I think of it more, it is even possible to get exact value of actual positional error out of such check.

If you know that UAME is 20.0 and you put the part onto the gage pin, its size has to fall in <19.2-20.0> range.
In other words:
- if the pin is 19.2 - the hole is at maximum possible positional error = 0.8;
- if the pin is 20.0 - the hole has 0 actual positional error - it is perfectly perpendicular to A and exactly at true position relative to B and C;
- if the pin is in between 19.2 and 20.0, for example 19.5 - actual positional error is 20.0-19.5=0.5;
- if the pin is less than 19.2 or greater than 20.0 - the hole does not meet the print.