Difference between revisions of "Best-Fit Alignment Versus Hard-Point Alignment"

From ATTWiki
Jump to: navigation, search
 
(Why Use Hard Point Alignment Methods At All?)
 
(28 intermediate revisions by one user not shown)
Line 23: Line 23:
  
 
== Best-Fit Alignment versus Hard-Point Alignment ==
 
== Best-Fit Alignment versus Hard-Point Alignment ==
 +
 +
<table cellpadding=20>
 +
<tr>
 +
<td width=400>
 
The goal in designing a qualification system for a tube shape is to designate the appropriate tolerance in each portion of the tube along the straights, then determine if the tube shape fabricated fits within that system of tolerances.  In the tube fabrication world, tolerances generally imply an allowance for deviation in all parts of the tube simultaneously.  In most cases, the issue is not if there is a tolerance - but it is how much of a tolerance is reasonable in each part of the tube.
 
The goal in designing a qualification system for a tube shape is to designate the appropriate tolerance in each portion of the tube along the straights, then determine if the tube shape fabricated fits within that system of tolerances.  In the tube fabrication world, tolerances generally imply an allowance for deviation in all parts of the tube simultaneously.  In most cases, the issue is not if there is a tolerance - but it is how much of a tolerance is reasonable in each part of the tube.
 
<br><br>
 
<br><br>
Line 28: Line 32:
 
<br><br>
 
<br><br>
 
Alignment methods that work best with this understanding of tube tolerances are those methods that take into account the entire tube shape during alignment.
 
Alignment methods that work best with this understanding of tube tolerances are those methods that take into account the entire tube shape during alignment.
 +
</td>
 +
<td>
 +
[[image:alignment_tube_image.jpg|300px]]
 +
</td>
 +
</tr>
 +
</table>
 +
 +
== About Best-Fit Alignment ==
 +
<table cellpadding=20>
 +
<tr>
 +
<td width=400>
 +
This is why best-fit methods are superior to hard-point methods in tube shape qualification.  Both methods are iterative and rotate the tube in space to find the smallest deviation possible.  However, unlike hard-point methods, best-fit methods use more sophisticated averages throughout the entire tube shape – and can even “weight” sections of the tube (like A End and B End) if appropriate.
 +
<br><br>
 +
Best-fit alignment can be thought of as a simulation of a real gauge in which the tube is dropped for fit check.  However, the similarity ends if a tube must be forced-fit into the physical gauge.  When a tube is force-fit into a gauge, it may appear to qualify - but only through changing the shape.  Changing the form of a tube is never allowed in VTube-LASER alignments.  Therefore, qualifying tube shapes using VTube-LASER ensures that deflection of the tube can’t be a factor in simulations of the gauge.
 +
</td>
 +
<td>
 +
[[image:best-fit-even-weight.jpg|300px]]<br><br>
 +
[[image:best-fit-endA-weight.jpg|300px]]
 +
</td>
 +
</tr>
 +
</table>
 +
 +
== About Hard-Point Alignment ==
 +
<table cellpadding=20>
 +
<tr>
 +
<td width=400>
 +
As the term “hard-point” implies, this method is locked out of the flexibility given to best-fit methods.  It must translate to a single point.  It must orient the tube based on a given plane formed by three points.  The final orientation may not be the best one available for qualification.
 +
</td>
 +
<td>
 +
 +
</td>
 +
</tr>
 +
</table>
 +
 +
== The Problem of Exaggerating Deviations ==
 +
<table cellpadding=20>
 +
<tr>
 +
<td width=400>
 +
Some users are tempted to believe that hard-point methods of alignment are “more truthful” than best-fit alignments.  One implication with this perception is that best-fit alignment methods are biased toward making the result appear better than it is in reality (or it “cheats.”)
 +
<br><br>
 +
It is true that best-fit reduces deviations (or it least it should).  But, for properly operating measuring centers, any reduction in deviation simply from better alignments can in no way be used to exaggerate system (or a tube shape’s) accuracy.  The opposite is true though:  A reduction in deviation brings us closer to the truth about qualification (and further from “cheating”) than alignment methods that may exaggerate the deviation.
 +
<br><br>
 +
The important point is this: Exaggerating deviation using less sophisticated methods of alignment (like hard-point) can make a part more difficult to qualify - but with no real gain to anyone.  In that case the deviation values are not the truth for that inspection regarding how close the measured part is to the master part.
 +
</td>
 +
</tr>
 +
</table>
 +
 +
== Why Use Hard Point Alignment Methods At All? ==
 +
<table cellpadding=20>
 +
<tr>
 +
<td width=400>
 +
Hard-point methods allow quality-control to lock a tube shape’s dimensions in order to emphasize a feature.  The problem is that this method comes at a cost.  As shown above, it tends to exaggerate deviation in the shape of the tube.
 +
<br><br>
 +
Another perceived benefit to hard-point alignment is that the alignment is easier to replicate downstream in the inspection process.  The problem here is that there is not a real need to replicate an alignment once a qualification has been achieved.  If VTube-LASER can fit all parts of the straight along the center of the MASTER tube within tolerance, then the goal of qualification has been achieved.  If qualification can be achieved based on comparison of tangent points along aligned straights (measured and master), then the method used to arrive at the qualification is of little consequence.
 +
<br><br>
 +
If validation is required, then as long as it can be proved that the aligned and non-aligned tubes have the same shape, then the qualification is validated.  The best way to compare tube shapes is to compare the bender data calculated from both sets of data.
 +
</td>
 +
</tr>
 +
</table>

Latest revision as of 00:17, 10 April 2012

Vtube-laser logo.jpg



Vtube-laser screen 1 54.jpg




Best-Fit Alignment versus Hard-Point Alignment

The goal in designing a qualification system for a tube shape is to designate the appropriate tolerance in each portion of the tube along the straights, then determine if the tube shape fabricated fits within that system of tolerances. In the tube fabrication world, tolerances generally imply an allowance for deviation in all parts of the tube simultaneously. In most cases, the issue is not if there is a tolerance - but it is how much of a tolerance is reasonable in each part of the tube.

The tolerances are assigned per straight section. They can be reduced at certain sections to ensure that the tube passes through or into tight sections properly. But it’s rare, and usually not helpful, to attempt to fabricate a tube that has zero tolerance.

Alignment methods that work best with this understanding of tube tolerances are those methods that take into account the entire tube shape during alignment.

Alignment tube image.jpg

About Best-Fit Alignment

This is why best-fit methods are superior to hard-point methods in tube shape qualification. Both methods are iterative and rotate the tube in space to find the smallest deviation possible. However, unlike hard-point methods, best-fit methods use more sophisticated averages throughout the entire tube shape – and can even “weight” sections of the tube (like A End and B End) if appropriate.

Best-fit alignment can be thought of as a simulation of a real gauge in which the tube is dropped for fit check. However, the similarity ends if a tube must be forced-fit into the physical gauge. When a tube is force-fit into a gauge, it may appear to qualify - but only through changing the shape. Changing the form of a tube is never allowed in VTube-LASER alignments. Therefore, qualifying tube shapes using VTube-LASER ensures that deflection of the tube can’t be a factor in simulations of the gauge.

Best-fit-even-weight.jpg

Best-fit-endA-weight.jpg

About Hard-Point Alignment

As the term “hard-point” implies, this method is locked out of the flexibility given to best-fit methods. It must translate to a single point. It must orient the tube based on a given plane formed by three points. The final orientation may not be the best one available for qualification.

The Problem of Exaggerating Deviations

Some users are tempted to believe that hard-point methods of alignment are “more truthful” than best-fit alignments. One implication with this perception is that best-fit alignment methods are biased toward making the result appear better than it is in reality (or it “cheats.”)

It is true that best-fit reduces deviations (or it least it should). But, for properly operating measuring centers, any reduction in deviation simply from better alignments can in no way be used to exaggerate system (or a tube shape’s) accuracy. The opposite is true though: A reduction in deviation brings us closer to the truth about qualification (and further from “cheating”) than alignment methods that may exaggerate the deviation.

The important point is this: Exaggerating deviation using less sophisticated methods of alignment (like hard-point) can make a part more difficult to qualify - but with no real gain to anyone. In that case the deviation values are not the truth for that inspection regarding how close the measured part is to the master part.

Why Use Hard Point Alignment Methods At All?

Hard-point methods allow quality-control to lock a tube shape’s dimensions in order to emphasize a feature. The problem is that this method comes at a cost. As shown above, it tends to exaggerate deviation in the shape of the tube.

Another perceived benefit to hard-point alignment is that the alignment is easier to replicate downstream in the inspection process. The problem here is that there is not a real need to replicate an alignment once a qualification has been achieved. If VTube-LASER can fit all parts of the straight along the center of the MASTER tube within tolerance, then the goal of qualification has been achieved. If qualification can be achieved based on comparison of tangent points along aligned straights (measured and master), then the method used to arrive at the qualification is of little consequence.

If validation is required, then as long as it can be proved that the aligned and non-aligned tubes have the same shape, then the qualification is validated. The best way to compare tube shapes is to compare the bender data calculated from both sets of data.