Position tolerance for non-piloted weld nut

[hygear]

I am trying to determine the best way to dimension/tolerance a set of drawings where non-piloted weld nuts are involved (why we insist on using non-piloted nuts I have no idea). I have one level of drawings which details a formed plate with all the proper GD&T for placing the holes used for the weld nuts. I then have a level of drawings where the weld nuts and a few other bits and pieces are welded onto the formed plate. There are three schools of though in our office on the correct way to place the nuts for welding:

1. On the welded level drawing, the same datums used on the formed plate drawing are utilized and the nut is placed with a position tolerance similar to the way the holes are placed on the formed plate drawing.
2. On the welded level drawing, a set of datums are created and the weld nuts are positioned using +/- dimensions and a note that says something like "BOLT MUST PASS THROUGH WITHOUT INTERFERENCE".
3. On the welded level drawing, simply show that the axis aligns between the hole and the nut and do not add any dimensions (or use reference dimensions). Then add the "BOLT MUST PASS THROUGH WITHOUT INTERFERENCE" note.

I believe all three options could be correct, but I wanted to reach out and see what other companies are doing and also see what others believe is correct according to Y14.5-2009.

 

[CheckerHater]

1. is a correct way.

2. using +/- dimensions is not explicitly forbiden, but strongly not recommended. Also, "BOLT MUST PASS THROUGH WITHOUT INTERFERENCE" statement must be described in terms of GD&T.

3. may be acceptable, if you care more about every nut being aligned to its hole rather then nuts being aligned to each other, and once again, "BOLT MUST PASS THROUGH WITHOUT INTERFERENCE" statement must be described in terms of GD&T.

GD&T does not describe method, so you specify how much misalignement is allowed, piloted or not piloted.

 

[ewh]

... or you can take a detail view of one hole/nut location, make the hole a datum and control the nut to that datum. Include the weld symbol in the detail. Put the number of occurances below the detail view label.

“Know the rules well, so you can break them effectively.”
-Dalai Lama XIV

 

[drawoh]

hygear,

I would locate your weld nut using positional tolerances, from the same datums you use to locate everything else.

Your problem is that weldments are not accurate. This makes welding drawings very much more difficult to prepare than machining drawings. Your nuts are being positioned to welding tolerances. Either your mating components must account for the positional error, or you are going to have to do some engineering of this thing.

Open up your mating holes to account for your sloppy positional tolerances.

Consider composite feature control frames. Do the nuts have to be accurate with respect to your datums, or do you only require internal accuracy? Maybe you can tap the holes in a plate, then weld that on.

Consider nuts and bolts, with large washers. Nuts permit more slop than tapped holes.

--
JHG

 

[KENAT]

Yep I'd look to do something like ewh says, it's really the thread of the nuts you want positioned relative to the holes as I understand it.

Vaguely analogous is Fig 5-39 of ASME Y14.5M-1994 & section 5.7C. Essentially show one instance as ewh says, make the clearance hole a datum & next to the datum symbol say 8X INDIVIDUALLY (replace '8' with your number of instances). Then put an FCF referencing the thread of your nuts and again put 8X INDIVIDUALLY under the fcf.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[CheckerHater]

Here is some graphics:

Case 1. Position of nut in asembly is controlled the way similar to position of the hole in part.

Case 3. Position of every nut is aligned with posion of respective hole (the way ewh and KENAT suggested) this is GD&T way of sayin "bolt must go thru".

Combo case. Both controls must be satisfied at the same time.

 

[hygear]

Thanks everyone. I believe the the detail view with the hole as the datum is the way to go.

 

[TheTick]

Meh. If tolerance was of any real importance, they'd use pilots.

 

[CheckerHater]

If tolerance was of any real importance, they'd use pilots

Once they found nuts without pilots for a penny less, common sense went out of the window.

Have seen it before: regular nuts are cheaper than PEM nuts, let's use them! Who cares that it takes twice as much time to assemble.

But to be serious, it may be possible to devise simple jig to align nut with the hole as they are welded, so not all hope is lost

 

[MintJulep]

What you really want to achieve is control of the resultant threaded hole. Right?

And I'm guessing that you tolerance the machined holes to meet that goal - as if they were the threaded hole.

In that case you cannot tolerance the nuts with the holes as a datum. That would result in adding the tolerance zones.

 

[ewh]

you cannot tolerance the nuts with the holes as a datum

A tolerance zone is being added to the nut relative to the hole. Yes, depending on the mating part, tolerance zones may need to be tightened, and this will affect cost. I don't know if I would go as far as forbidding it though.
A very legitimate consideration. Thanks MJ.

“Know the rules well, so you can break them effectively.”
-Dalai Lama XIV