ANSI Screw specs not correct?
ANSI Screw specs not correct?
(OP)
Hi.
I have a question about the ANSI spec, B18.6.3 for a Slotted Flat Countersunk Head screw. In the specs they give a range for the allowed countersink angle of 80-82 deg, and another range of 99-101 deg. But when I do the math, many of the screws do not fall within those ranges. For instance, a nom. size screw of "oooo" has a head sharp diameter of .043" a shaft major diameter of .021", and a head height of .011". Using some simple math,
Delta Width = .043 - .021 = .022
SideDiff = DeltaWidth / 2 = .011
Half countersink angle = Atan(SideDiff / Head Height) = Atan(.011 / .011) = 45 deg.
That means a full countersink angle of 90 deg! WAAAY outside the specified 80-82 deg.
I know I'm missing something here, but the Machinery's Handbook does not specify any allowed +- error. They do for a Hexagon Socket Flat Head Cap Screw (ANSI B18.3), but not ANSI B18.6.3. That having been said, just the range alone of 80-82 deg, implies 81 deg +- 1 deg.
Can anyone shed some light on this?
Many Thanks! :)
I have a question about the ANSI spec, B18.6.3 for a Slotted Flat Countersunk Head screw. In the specs they give a range for the allowed countersink angle of 80-82 deg, and another range of 99-101 deg. But when I do the math, many of the screws do not fall within those ranges. For instance, a nom. size screw of "oooo" has a head sharp diameter of .043" a shaft major diameter of .021", and a head height of .011". Using some simple math,
Delta Width = .043 - .021 = .022
SideDiff = DeltaWidth / 2 = .011
Half countersink angle = Atan(SideDiff / Head Height) = Atan(.011 / .011) = 45 deg.
That means a full countersink angle of 90 deg! WAAAY outside the specified 80-82 deg.
I know I'm missing something here, but the Machinery's Handbook does not specify any allowed +- error. They do for a Hexagon Socket Flat Head Cap Screw (ANSI B18.3), but not ANSI B18.6.3. That having been said, just the range alone of 80-82 deg, implies 81 deg +- 1 deg.
Can anyone shed some light on this?
Many Thanks! :)





RE: ANSI Screw specs not correct?
At the very, very small screws in question (0000, etc.) the very small distances that are the discrepancy between theoretical and real-world differences between a 90 degree angle x a sine delta of only a 2/100 of an inch are, really, meaningless. The screw will deflect some, the thin sheet metal it is being squished between (the attached sheetmetal and the baseplate sheetmetal parts) are both flexible, and both will be deformed by the pressure caused by the screw being trightened.
Yes, theoretically, the difference matters. But, on a very small screw between two surfaces - both having a recess for the countersink? Not really.
Now, when I buy countersink machining heads, those DO have various angles, the right machining head (tool) needs to be matched to the screw that will be inserted. But those angles are more important for the the larger screws.
RE: ANSI Screw specs not correct?
Likewise the shaft dia is clearly stated as nominal, if memory serves it's probably at the max.
So of the 3 pieces of data you use for your calculation at least 2 are only nominal or reference.
The head diameter and angle with their assigned tolerances are what defines the form of the head.
So it is not the angle that varies to comply with the stated height, rather the height varies dependent on head & shaft dia and angle.
What is Engineering anyway: FAQ1088-1484: In layman terms, what is "engineering"?
RE: ANSI Screw specs not correct?
Thank you very much for the clarifications. I see now what is going on. I did not really examine the difference between "ref" and "nom". Makes perfect sense now.
And the real-world tolerance makes perfect sense too. I always wondered about tiny screw specs when the screw is torqued down...There has to be a lot of deformation going on that basically changes the screw forever (takes it past its elastic limit?).
RE: ANSI Screw specs not correct?
Trouble is the threaded portion is trying to center on the threaded hole, while the flat head is trying to center on the countersink - so they end up fighting each other and can end up requiring quite a bit of deformation and hence stress when there are multiple such fasteners.
However, if the screws get loaded in service then the heads can 'pop off' - the stress from the assembly combined with the service load combined with the nice stress raiser where the head goes into the shaft can easily overload the the limits of the screw.
Large patterns of countersunk screws are problematic for this reason.
What is Engineering anyway: FAQ1088-1484: In layman terms, what is "engineering"?