Positional tolerancing of "pure" countersinks 1

[Tunalover]

Peers-
I do a lot of sheetmetal design and often run into the situation where a countersink has to be so big that it removes the clearance hole altogether (and the bottom of the screwhead can actually go through to the other side in the assembly). How does one apply positional tolerancing to a pure countersink? If you make the swept angle (e.g. 82, 90, or 100 degrees) and diameter basic, what's the best way to control the position?
TIA!



Tunalover

 

[ctopher]

Have one callout for the hole and csink together with true position to the callout.

 

[ewh]

ctopher is correct. Unless otherwise specified, the positional tolerance of the hole also applies to the countersink.

 

[Tunalover]

Guys-
But there is no more hole. The countersink took it away entirely. How can you inspect the position of a hole that isn't there any more?



Tunalover

 

[ctopher]

You asked how to control the position. You need to either change the size of the hole or the C'sink.

 

[Tunalover]

ctopher-
You don't understand. There are times when a flathead screw is used and it's not possible to provide a clearance hole; it is only possible to provide a countersink. This is because the required countersink to insure a below flush condition is so large that a realistic clearance hole is removed. Basically what I'm trying to do is to control the position of a steep, tapered hole.


Tunalover

 

[ewh]

It does not matter if the hole is no longer present. The positional tolerance would still apply to the countersink, the same as it would the hole if it were still present. The only problem I see is if the countersink was toleranced to the hole itself.

 

[btrueblood]

TL,

What does your drawing callout for the c'sink look like? If it looks something like:

V (diam. symbol)0.56 X 90(degrees symbol)

then yes, a positional tolerance can be applied to the countersink (since the c'sink has a theoretical centerline). The callout shown above controls the outside diameter of the countersink and the total included angle (side to side). A positional tolerance could control both the x-y location of the hole as well as the perpendicularity of the centerline, depending on the choice of datums.

The machinist seeing your print shouldn't even blink - a c'sink is just a fancy drill to him, and he indexes the table in x and y to locate the holes where your drawing shows 'em.

Inspecting the callout might be a little tricky if the tolerance is very small, but for typical positional tolerances, either templates, jigs, or jig + tapered pins would work. Templates would have holes at the LMC. A jig would have pins at the minor diameter, and inspects the position by inference, a more correct approach with a jig would have pins (with tapered heads, like a flathead screw with no threads) dropping into holes in the jig plate.

That said, I've worked for people who thought putting fat flathead screws in sheet metal was perfectly fine. I've always tried to avoid it, personally, since in practice such holes are easily "punched out" by the screw heads. A better approach is to use a larger number of smaller diameter screws (whose heads are a closer match to the particular sheet metal gauge), or use pan-head or similar fasteners.

 

[Tunalover]

btrueblood-
Since the fit-critical diameter is the one at the BOTTOM of the countersink, that's the diameter that needs to be controlled the most. Your method controls the TOP diameter which is less critical. Actually, since I first posted the question I have come up with a solution. Bruce Wilson's "Dimensioning and Tolerancing Handbook" (Genium Publishing) describes a way to control tapered holes. A size (diameter) callout and feature control frame is provided for each diameter. Optionally, the BOTTOM diameter can form a local datum from which the TOP diameter can be toleranced.

As for good vs. bad design practice, I agree that it's not good practice. My situation, however, is:
A. I can't thicken the sheet metal due to space constraints,
B. The size and positional tolerance of the tapped hole
are of a part for which I have no control (it's purchased),
C. I must have a below-flush condition (e.g. pan head screw is not a possibility), and
D. Because I am engaging with a tapped hole I can't use a flush captive stud.

Thanks to all and Happy Holidays!

Tunalover

 

[btrueblood]

Hey, I didn't say that was my method, just "a method" or "your method" . Yes, controlling the bottom diamter is better, but also more difficult to "manufacture that way" (due to warping of metal during drilling/c'boring).

 

[MadMango]

Wouldn't the top diameter be more critcal due to the need for flush or below flush installation?

[green]"But what... is it good for?"[/green]
Engineer at the Advanced Computing Systems Division of IBM, 1968, commenting on the microchip.
Have you read faq731-376 to make the best use of Eng-Tips Forums?

 

[rmetzger]

as long as the angle is defined and toleranced it shouldn't matter if you call out the top or bottom. Regardless of the edge dimensioned (top or bottom) its where the screw will seat thats important for a below flush mounting and thats easily accomplished using either diameter and the angle to define the cone. If the drill point angle is less than the head angle of the screw then the top hole would be more important to measure because the top of the fastener is where the point of contact will be and the inverse is true if the angles are reversed. If you are going to match the head angle of the screw with that of the hole then it shouldn't matter as you won't know what side the deviation will take place and will have to rely on the clearance and tolerance to accommodate the difference.

 

[btrueblood]

Mango -
"Wouldn't the top diameter be more critcal due to the need for flush or below flush installation? "

That's what the 90 grit sandpaper is for

I still think the bottom diameter is more critical to control, to prevent punch-thru of the head. I've seen so-called machinists "deburr" the sharp edges of the hole (small diameter) to the extent that the fastener would nearly drop through. Agree with rmetzger though, that which end doesn't matter if the tapers of head & hole are close, and tolerances tight enough.