Complex Datum Schemes for non-rigid parts

[designmgr]

Hello,

I am trying to dimension a print for an assembly made of multiple formed plastic lines, joined together by "T" and "Y" connectors, and terminating in multiple quick-connectors.

In use, this assembly is installed by first connecting these QC's to their mating parts, then the plastic lines are snapped into retaining clips at various locations along their length to keep them from moving around. This assembly is typically 1-2 feet long with multiple bends.

Typically for a single quick -connector, the diameter of the mating feature (cylindrical in shape) would be a primary datum, and the feature that limits the insertion of this mating feature would be secondary. This particular assembly has 3 QC's, so I now have 6 datums.

When this part is gaged, I want it to be fixtured as it is used. My main concern is that when installed per intent, the profile of the tube falls within a certain tolerance from the basic dimensions. This means that in addition to the 3 QC's, I also want it snapped into gage features representing the clips that hold the lines in place. There are two of these clips, so here are 2 more datums.

A voice in my head tells me no more than 3 datum references, or on can combine datum features within a single box in the frame by joining them with hyphens. Not sure how to proceed:

[profile of surface|x.x|A|B|C|D|E|F|G|H] with some note detailing Gaging procedure?
|A-B-C|D-E-F|G-H] or some other combination to honor the 3-datum "rule"?

Thanks!

 

[drawoh]

designmgr,

There is no ASME Y14.5M-1994 rule that limits you to three datums. Even on a rigid part, you have the option of using datums A, B and C to locate the part, and applying datum D to locate one feature to another. There is no reason why you cannot constrain each connector or clip with three datums each. Hou do you want the thing inspected?

If you have the ASME standard, read section 6.8 on Free State Variation. This covers the datums and fixturing procedure on flexible parts.

JHG

 

[ringman]

drawoh,

Did you mean to say datum features rather than datums when you stated there is not a limit to the datums. It seems that in our three dimensional world 3 datums provide all we need for a basis of inspection in a single callout. We can use more for the 'datum features' which in turn determine the datums. Or, more if multiple set-ups are required.

 

[ctopher]

I do cable assy's that go on engines similar to your plastic lines. We do not use datums for the assy dwgs, just dimensions. Too complicated to check. Fixtures do all of the alignment and check for fit and tol and it is accepted by our aerospace customers.

Chris
Sr. Mechanical Designer, CAD
SolidWorks 05 SP3.1 / PDMWorks 05
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

 

[drawoh]

ringman,

Each datum is attached to a feature.

I can immobilize a rigid part using three datum features. When I apply tolerances like position, profile, parallelism and such, each tolerances specifies one or more datums. Obviously, it is desirable to use one or more of the three datums used to locate the part.

Take the case that we have a manifold that performs three or four separate functions. DFMA is fun! Your datums are the bottom surface and two sides. You have two holes for gears that must be very concentric to each other. Location from your standard datums is not accurate enough, even though the datums are ideal for everything else.

One of your holes becomes your fourth datum feature. The other hole is called up concentric to the first hole.

It appears that designmgr's part cannot be immobilized by three datum features.

JHG

 

[designmgr]

My real-world concerns for gauging this part are two-fold:
1) Is the part of a form such that it is possible to simultaneously make all of the required connections and snap the tubes into their retaining clips?
2) Assuming I can meet the above requirements simultaneously, is the form of the tube that is not restrained falling within a given profile tolerance for clearance reasons.

Similar to ctopher's situation, this is typically measured with a painted wooden gage where these connections / constraints are applied and then a visual inspection is done to see if the tube falls within the bounds of the green-painted areas of the form gage.

These tubes are more rigid than cabling, so they have a definite form, but they are not as rigid as steel tubing. They could also be measured by CMM if constrained in a fixture that uses all of the mating part connections as datum features as well as datum surface targets simulated by the clips.

I do not place GD&T tolerances on the locations of the connectors because if the gage demonstrates that I pass tests #1 and 2 above, the part is fine. Hence a surface profile on the tubing when the part is fully constrained as in it's installation is enough.

I will go read section 6.8 per drawoh's suggestion and see what I can find.

Thanks,

WJS

 

[ringman]

drawoh,

I think my point was missed completely. There is a clear and definite distinction between 'datums' and 'datum features'. ASME Y14.5 is a precise language and we need to make sure we are using the correct terms. I think you referred to datums when you intended to state datum features. If I am mistaken, my apologies.

ringman

 

[drawoh]

ringman,

Practically, datums are defined by datum features. For the purposes of fabrication and inspection, I do not see a difference.

Perhaps you are thinking of the datum reference frame. This is a feasible concept on around 98% of the drawings I have worked on, but it does not work very well in the example above.

JHG

 

[ringman]

Drawoh,

A quick look at the Standard and I found figure 4-38, which might come close to approximating this problem we are discussing. It appears that there are 7 'Datum Targets' used in this example. (I believe this is what you call datums.)

In this example they define A, a datum plane,B a datum point, and C the mid plane. These are the three datums that are used for inspection.

Does not this come close to the problem being discussed?.

 

[TheTick]

I think you need to consider datum zones or targets: define a datum by a set of features.

Consider a planar datum on a nearly-planar surface. One acceptable practice is to use datum targets to show where the part is fixtured to get the planar datum.

Also consider stepped datums, where targets on non-coplanar surfaces define a single datum.

It is reasonable to expect you could come up with a "legal" scheme of datum targets defining a single datum frame of reference across multiple features.

I could be the world's greatest underachiever, if I could just learn to apply myself.

http://www.EsoxRepublic.com

-SolidWorks API VB programming help

 

[drawoh]

ringman,

Datum targets are an entirely separate issue from what we are discussing. For a normal rigid object, you need three targets for the primary datum, two for the secondary datum, and one for the tertiary datum. You wind up with three datums, and a datum reference frame.

Datum targets solve the problem of irregular surfaces where you want to define exactly where everything gets jigged.

designmgr's problem here is that the part is not rigid. Three datums control the position of one connector, leaving everything else dangling ambiguously.

JHG

 

[designmgr]

Drawoh gets the jist of the problem, pretty much. One connector can serve as 2 datums (I have the axis of the connector as a datum as well as a feature that limits my motion along that axis. At this point, I have a gangly assembly that is free to rotate about the axis of that first connector. I then need to make at least one other connection to stop rotation, but that still leaves other arms of this assembly free to dangle.

Due to the nature of this part, there really are no requirements for its form in the free state. The form is only important when it is forced into a particular orientation, and there are many, many physical constraints required to do so.

Section 6.8 is just brief and generic enough not to provide much insight on this issue. My thoughts at this point are that I should call out each datum in my profile tolerance, even if there are 7 or 8 of them, and keep them in the order of the assembly sequence. I will accompany this with a lengthy narrative describing the gaging sequence. One of my frustrations with all of the GD&T standards, classes, etc. is that all of the examples seem to be rectangular blocks with flanges and bolt-hole patterns, where 1/2 of my bill of materials is made of non-rigid parts.

Thanks all for your help