Rotation Control

[Scott Brown]

Hello Everyone

I have a question about controlling rotation in an alignment. Simular situation as posted earlier by tnguyen, Datum A is a plane and datum B and C are holes on datum A but are not in a straight line to print dimensions or any features of the part. I see 2 methods I could use, #1 Create an offset line from B to C and use that for rotation or #2 Create a 3D line from B to C and rotate to that line and then rotate by a basic angle to square the coordinate system. Which method is right/ wrong / prefered. If I rotate to the offset line the basic angle would not match and if I use method #2 the basic dim from hole B to C would not match. Kind of like the tail wagging the dog....

Thanks
Scott

[#2#]

I don't see how you can do #1. It seems to me that creating an offset line is going to be dependent on your current alignment. If your current alignment is way out of whack then your alignment using the offset line will be out of whack. Maybe I'm wrong - anyone, anyone?

[Maqboul]

For Me :
#1 Iterative alignment.
#2 Use B_C circles + Use Theo. Rotation Angle.
#3 Construct 2D lice from B_C circles + Use Theo. Rotation Angle

my 0.02tl
Maq

[JoeRZ]

I've always used 2d when involving a plane as primary level, diameters would lock in to that plane and less hassle afterwards.

1) Level plane (origin if needed)
2) Measure dia.s to working plane rotate to these 2, assign 1 of them as orig.
3) Offset to trigged out angle.

3d works a little different and not as stable from part to part.

my input - lol

JRZ

A man who says it cannot be done should not interrupt a man doing it!

[Maqboul]

I don't see how you can do #1. ?

I'm using 3 vector point for the Level and 2 AutoFeat-Circle for the Level.

I'm writing the nominal values manually and the vectors.Then i'm following the procedure for Iterative align.

Maq

[Maqboul]

It seems to me that creating an offset line is going to be dependent on your current alignment. If your current alignment is way out of whack then your alignment using the offset line will be out of whack. Maybe I'm wrong - anyone, anyone?

According my experince , you are right.

[tking]

I've always used 2d when involving a plane as primary level, diameters would lock in to that plane and less hassle afterwards.

1) Level plane (origin if needed)
2) Measure dia.s to working plane rotate to these 2, assign 1 of them as orig.
3) Offset to trigged out angle.

3d works a little different and not as stable from part to part.

my input - lol

JRZ

A man who says it cannot be done should not interrupt a man doing it!

This would be my exact method, always.

.02

TK

[Paul Sarrach]

If you have CAD use iteritive. then you can make new alignment by Just zeroing out at Datum -B-

Or do like Tking said.

[cmmguy]

Hello Everyone

I have a question about controlling rotation in an alignment. Simular situation as posted earlier by tnguyen, Datum A is a plane and datum B and C are holes on datum A but are not in a straight line to print dimensions or any features of the part. I see 2 methods I could use, #1 Create an offset line from B to C and use that for rotation or #2 Create a 3D line from B to C and rotate to that line and then rotate by a basic angle to square the coordinate system. Which method is right/ wrong / prefered. If I rotate to the offset line the basic angle would not match and if I use method #2 the basic dim from hole B to C would not match. Kind of like the tail wagging the dog....

Thanks
Scott

You dont say if the basic dim between B and C are both X and Y or just Y(Y Basic). IF they are both basic then trig the angle based on the basic values. If ONLY the Y is basic then you will have to create a flexible rotation angle that calculates what the angle should be to satisfy the basic Y dimension. You will have to trig the angle based on the straight line (2D) distance between the two holes as the Hypotenuse and the opposide side as the Y basic. This has to be calculated after measuring the two holes. It is dynamic for each part.

Do the alignment between the two holes and then rotate to this new calculated angle.

[VONDY]

Construct an off-set point from the second circle so that it is the same basic distance in one direction, use the smallest of the two, if the hole is 10" over and up 1"up, construct the offset point of the the circle 1" down. Then construct a line from the first circle and this constructed point.

[#2#]

Construct an off-set point from the second circle so that it is the same basic distance in one direction, use the smallest of the two, if the hole is 10" over and up 1"up, construct the offset point of the the circle 1" down. Then construct a line from the first circle and this constructed point.

How will this work? In order for this to be anywhere near accurate the alignment you are in when constructing the point will have to be pretty close to the final alignment.

In your example the calculated "basic" angle would be 5.71º (inversetan 1/10). If your alignment when you create the point is off by 1º your 10" basic will measure 12.137" (1/tan 4.71º)!!!!!

If you calculate the "basic" angle using trig and your basic dims. then rotate to that you will be more accurate.

Somebody check my math please. If I'm wrong I want to know.

[cmmguy]

your right GoodLuck, that wont work.

[Scott Brown]

Rotation Control

Thanks everyone for your input. I agree an offset point would not work properly without an alignment before hand. However the offset line works fine provided that you create a level alignment prior to the constuction. Looks like the line rotation and basic angle rotation is the prefered method. Both methods will work, I guess it's just a matter of how you want to lock out Datum C. (which alignment reads zero on one axis)

Thanks again

Scott

[Matthew D. Hoedeman]

Thanks everyone for your input. I agree an offset point would not work properly without an alignment before hand. However the offset line works fine provided that you create a level alignment prior to the constuction. Looks like the line rotation and basic angle rotation is the prefered method. Both methods will work, I guess it's just a matter of how you want to lock out Datum C. (which alignment reads zero on one axis)

Thanks again

Scott

It will also depend on what is required for the C-Datum. I have not been keeping up on this post, but for an example:

B-Datum is a ROUND HOLE and the C-DATUM is a round slot, with the slot on the same rotation as the 2 holes would be. In this case, the alignment needs to be rotated by the DESIGN angle between the holes. If the C-DATUM is a round slot that is square to an axis, then you need to rotate it so the the C-DATUM is correct (in the width) to the body value, which means it may be a larger or smaller angle of rotation than design. At least, this is how I would do it.

[tking]

Matt,

I would use the design angle only, just as a gage does.

My .02,

TK