Partial cyl (radius W/ target lines) as Primary Datum

[dtmbiz]

Often we have parts that are partial diameters that fit over cylindrical surfaces.
I believe that the DRF defined by a partial cylindrical surface with target lines
is a way to define a primary datum plane related to a radius. With a hole as
the secondary datum I believe that a DRF exists which constrains all DOFs.

The attached pdf with datum A as the primary datum and datum B the secondary datum
illustrates what I am talking about.

(No FCF’s using this DRF; just showing what they are)

Comments appreciated.

 

[pmarc]

How flexible is the part?
In my opinion third datum target line in the center would be useful:
- for better stabilizing the part,
- for preventing potential deformation due to its flexibility
- for better reflecting mating conditions since (if I am imagining correctly) the part fits over solid cylinder along its whole inner circumference.

I am also not so sure if the current datum targets A1 and A2 should be where they are now - rather closer to the ends of the arc, if not exactly on the bottom inner edge of the part. I am just thinking about the datum feature B hole. If for some reasons you need to have it centered on the part, current datum target location may not assure it, because the part is not constrained in rotation.

BTW, I know it is not the key point here, but the arrows terminating datum target leader lines in the bottom view are illegal.

 

[MechNorth]

For starters, '94 or '09? Makes a difference for my suggestions at the end.
The issue with your datum target line method is that the radius can be any of a huge range of sizes, and you would make contact with the two lines. If you want to use datum target lines, you would be better off doing it across the part rather than along its length. It will be easier to quicker assess whether or not you are close to nominal dimension on the radial feature.

Next, I understand what you mean from your poste, but that's not what you've shown on the drawing. You can't put the Datum Feature Symbol on a non-existent line (the horizontal of your centerline) in '09. You are also implying that you only want the vertical (in the bottom view) dof constrained, but this is not absolutely clear, as the geometries in conjunction with the datum targets would constrain vertical and horizontal dof. Using '09, you can specify which dof you want constrained by each datum reference.

As an alternative, you can use the entire surface as a datum feature by putting the callout on a leader which points to the surface, which is probably the best in this situation. In '94, you would have had a mathematically defined surface, but there was little guidance on how to do this or what it meant; '09 would allow you to specify the datum feature at MMB, LMB, BSC, etc. Also, the radius should be basic, not reference.

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[dtmbiz]

This drawing is contractually to the 94 standard. My objective is to define a DRF that a hole chart can refer to. There are maybe 50 holes that are arranged around a contoured gasket.

MechNorth it did escape me that datum A could also control horizontal which is good from the point of the objective because the holes are oriented thru the case axis.

Next I did not realize you can use target lines as a radius (across the part) if I am reading you correctly.

The reason I want the actual datum offset from the contact points on the surface is because this coincides with the case axis. This is where the chart values are functionally related; to the case axis. I was thinking this might be achieved with the stepped datum concept, however I do recognize in that senario that there is physical contact with a feature.

If you are talking about using a radial line element established with a radial target line, then that sounds better. This would be like a radius knife edge, correct?

pmarc,

This is high strength steel that is approximately .100" thick.

If you are referring to the "arrowheads" pointing to the target points and that it should be a plain leader without arrowheads, then I am on board with that.

It is true I didnt put in a basic angle and agree that it is necessary in the configuration of target lines shown.
(Just wanted to get the basic concept out there quickly for folks like yourselves to see - missed it; it is important)

I really appreciate both of your inputs!

 

[dtmbiz]

More like this?

(see attached)

 

[pmarc]

Hmm, interesting...

1. I do not think you need 2 phantom lines in the top view. I would simply use 6 datum target points for clarity. Just like in fig. 4-50 in Y14.5-2009.

2. I would still like to know how are you going to control position of hole B relative to datum axis A derived from the set of 6 datum target points A1, ...., A6. Or are you going to leave it as it is? If so, then I am inclined to say that the specification of this hole is incomplete.

 

[dtmbiz]

pmarc,

I was thinking of putting a positional control to Datum A only while showing that the axis of datum B sharing the axis of Datum A.

If I use target points in the top view then it would be 6 spherical tipped gage pins. I actually want a radial knife edge at each target line location. I am not sure if this DRF is being shown correctly to do that though. Maybe I just need target line A1 and A2 in the top view. and remove the target points in the front view.

Comment?

 

[dtmbiz]

hopefully getting there....

pls see attached

 

[pmarc]

Yes, if you plan to use a radial knife edge, I would recommend removing datum target points from the front view. I think it should be even acceptable to show nothing instead. I know you said '94 edition governs the print, but 4.24.11 of '09 edition says that phantom line arcs could be omitted in your case.

As for hole being datum feature B... This was actually the reason why I mentioned this in my very first post. If you want to control its position relative to A only, I am afraid you may have some problems with getting repeatable and reliable positional measurement results. Why? Because the part will be free to rotate around datum axis A when installed on 2 knife-type datum targets A. So one inspector may get the part "symmetrical" on A having it oriented exactly as shown on the print, but the other may get the part rotated a little bit and this rotation may significantly affect hole's position results. Therefore my bet would probably be to use additional datum to limit this rotation during inspection.

Do you see my point?

 

[pmarc]

Just before others will correct me, I'd like to withdraw my assertions about ambiguity of positional tolerance wrt to A on hole B. As I look at it now, there is nothing wrong with this. There should be no issues during inspection with interpretation of positional callout due to potential rotation of the part around datum axis A. My apologies for that.

 

[dtmbiz]

thanks pmarc
your comments were very helpful

MechNorth
thank you for the direction

 

[MechNorth]

The last image is pretty much what I was envisioning. Given that the Datum-A callout is on the basic dimension, I agree that the phantom lines for A1 & A2 in the bottom view wouldn't be necessary. I don't know if I would use +/- degrees to control the limits of the arc segment; I'd probably use profile instead. Otherwise, coming along nicely. As for the datum simulator for the arcs, a roll-formed arc of wire/rod would work better than a knife-edge (wear and error perspectives).

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

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[dtmbiz]

Good information Jim...

Yes, I just put up some quick angle dimensions.
Agreed. Profile is the way I was headed for the final dwg.

Appreciate your time...

dtm