Position without Datum Reference

[[Te...]]

So I have a print that is similar to the sketch attached.

I know that this is a good callout I am just not sure the correct way in Calypso to check it.

It is a pattern so the print I am using has the position of the pattern (2 holes) to ABC and then the position with no datum reference

What is the correct method, I do not want to make the position of one diameter to the other because of the gap between the holes which we all should know does not work well and the error gets worse the farther apart they are.

The tolerance for my position is 0.00 with MMC

I think I should do the position of the pattern to the best fit alignment created from the pattern position to ABC but with only 2 holes won't it be perfect?

Any help will be greatly appreciated

Thanks

TP NDR.JPG

[[Ro...]]

Hi Ted,
For that callout I would measure 2 cylinders, create a 3d line through both of them, and then measure them again using the 3d line as your origin in your TP. You're essentially checking coaxiality to a common centerline.

[[Te...]]

Can I just do the position of the pattern to the 3d Line?

I really don't want to or think I need to measure them a second time and I really just want one position on the report but if I need 2 that is fine.

The BIGGEST problem is that it is for a demo and I have no part or CAD so I am writing the program before hand and then running it for the first time at the demo and I really need to make sure I am using the best practice.

Thanks again

[[Ro...]]

The problem you're going to have is that the cylinders create the 3d line, so you'll have a circular reference if you're trying to use the original cylinders and the 3d line for position. What you can do is recall the original cylinders into new ones and use those for your evaluation.

The reason I normally would remeasure them is that it will reduce any probing error. It may be small/insignificant/nothing, but depending on the form/location of the cylinders to each other, it could make a difference. It's really just a refinement. It may or may not help. Also, I would put them in a new alignment that uses the 3d line as an origin. If you don't there's no point in remeasuring them, as it will use the same alignment as the first time.

Since it's a pattern I think that 2 results would be the proper way to go.

[[Da...]]

If I'm understanding correctly your print has one callout for 2 holes to ABC meaning each hole has to be within this tolerance.
Then you have another position callout without datums that is the relationship of one hole to the other, Why would you have to check them twice? You check the cylinders as separate features, then report 2 positions to ABC and 1 coaxial using those cylinders.

[[Ro...]]

There is no coaxiality callout from what I understand. Coaxiality checks one feature to a datum structure. If you check coaxiality of one hole using the other hole as the datum, you'll get projection error on the axis across the span between, as Ted pointed out. This is a pattern, so they both need to be checked as a best fit to themselves. That's what the 3d line does.

[[Da...]]

Please sign in to view this quote.

Thank You, Robert.

[[Te...]]

Thanks for all the replies.

The holes are Ø 31.27 but only 5.7mm deep and just over 14mm apart

I am going to measure them twice for this project because of the short depths.

I am going to do circles at the 2mm immersion depth and create a 3d line and use that for the planner rotation and then I will measure them as cylinder it the new alignment (I will probably measure the circle and then the cylinder one right after the other to speed the program up, because of the need to use a Star Probe or the RDS, and then Recall One Feature for the Cylinders to the new alignment).

I will then report one cylinder to the other to meet the position to no datum features. I expect a good result because the holes are aligned which is allowed because there a no datum features.

[[Br...]]

There's probably a few ways that would work. You could create a "geometry best fit" alignment for the two cylinders, then use that alignment for your datum reference. Alternatively, you could do use the bore pattern best fit option. I like the idea however that was mentioned above, creating a 3D line and making an alignment from that. Then your nominal XY locations would be zero.

[[Ch...]]

A couple notes to add to this - as I've gotten deeper into the "coaxial" position requirements.

There is a Y14.5 section (7.5.1.8) for this with a similar looking part ..

Where you measure the cylinder paths will have an effect I believe. (e.g. they both should be equal percent from each thru wall) - due to how Calypso creates 3d line through each cylinder axis centroid point.

Another crafty option might be to pick each side of each thru wall, make the THEORETICAL features, project each thru cyl axis to each thru wall, create a new thru axis from these 2 points, this axis will be correct length, etc.

I sure wish there were an easy way. If it were thru three walls for example, you would have 3 position to the 'common axis'.

Good luck, happy measuring, and happy holidays !

[[Cl...]]

Please sign in to view this quote.

Brett, with that GBF would you allow trans/rot in all axes?

[[Br...]]

Please sign in to view this quote.

Yes, check every box