Is a countersunk flathead screw in and of itself a locking fastener? 1

[mecengtec]

A discussion has come up in our engineering department concerning countersunk flathead screws being a locking fastener. I disagree with this statement. I believe the additional surface of the conical head may reduce loosening, but don't believe it should be refered to as a locking fastener. What are my fellow engineer's thoughts?

 

[JohnRBaker]

I agree with you. I would never trust something like that to be 'locking' without some additional 'intervention'.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA

http://www.siemens.com/plm

UG/NX Museum:

http://www.plmworld.org/p/cm/ld/fid=209

To an Engineer, the glass is twice as big as it needs to be.

 

[hydtools]

I would not consider it a locking fastener, the angle of the countersink is too wide. Also, other than the first screw in a pattern, the odds are against the remaining screws having 100% underhead contact.

Ted

 

[rb1957]

third vote for not locking ... sure there's higher friction in the CSK, but ...

 

[TheTick]

Maybe if the c'sink and the screw angles didn't match, and the part deformed to act like a lockwasher.

 

[CoryPad]

Flat head (countersunk) fasteners do not provide locking nor extra loosening resistance. They do provide alignment and shear resistance in the absence of preload.

 

[dhengr]

Drive those mothers home, and then put a puddle weld over/around each head. That’ll lock em down. Of course you gotta do the same thing at the nut, if one exists. And, you may want to weld the piece on the other side to the first piece unless there are several welded screws btwn. the two pieces. Or, you could use locktite or some such.

 

[tph216]

I thought they were locating, rather than locking. We tend to avoid them in general, preferring counter-bored, if we need to hide the head.

I can't see how they are locking though. I don't think the area under the head is even detrimental to the friction / locking property. Rather, I think that prestraining the shank to a value at which rotation cannot occur, is the mechanism by which locking occurs (don't quote me on that though, I'm not 100%).

Also, F = u*R - Surface area doesn't come into it, and pondering further, wouldn't the reaction be grounded over a smaller radius anyway, for a conical head of the same surface area as a flat head?

 

[mcgyvr]

Our UL inspector at one time said they accept flat head screws to fulfill any "no loosening" requirements on certain products/certain situations.

 

[MintJulep]

Fundamentally, no difference from a bolt or screw were there under head surface is perpendicular to the thread axis.

The only forces acting are frictional, developed from the preload in the fastener.

 

[Tmoose]

Flat head SHCS accept far less torque than regular SHCSs, so already are competing with one hand tied behind their backs.

Like in McGyvr's example, there may be some "special" circumstances that could lead to locking behaviour, or the appearance of locking behaviour.

Post Hoc, Ergo Propter Hoc

Accepting something as standard practice or physical law based on a single event, or worse yet, extrapolating data to all designs can be looking for trouble. When I was 21 I had unprotected sex one time with Nancy, but I held my breath at "the moment", and she did not become pregnant. I would not suggest any useful technical conclusion can be drawn from that single data point, factual as it is.

 

[ornerynorsk]

In actual practice and in absence of a seconday locking component, FHCS's require considerably more torque to loosen than do the typcial square shoulder configured bolts due to the conical geometry (wedge).

In addition, a square shoulder bolt is constraining movement in one plane only, given the fact that the clearance hole for the bolt IS going to be slightly oversize. The FHCS effectively constrains in 3 planes, and, properly pre-loaded, will therefore reduce or eliminate the possibilities of minute movements, which I believe is the root cause of loosening in square shouldered (clamping perpendicular to the axis) hardware.

No, I still don't think you can rely on them to be a self-locking fastener in a critical application.

My 2 cents.

It is better to have enough ideas for some of them to be wrong, than to be always right by having no ideas at all.

 

[MintJulep]

The FHCS effectively constrains in 3 planes, and, properly pre-loaded, will therefore reduce or eliminate the possibilities of minute movements

Interesting thought.

True for a two-ply joint where ply one has the countersink and ply two is tapped.

Not true if there are more than two plies with the last ply tapped, or if used with a nut rather than tapped ply.

 

[IRstuff]

All the above said, there have been a few threads here concluding that most accepted "locking" approaches are not, as well.

faq404-1257
thread404-230741
thread404-295664
thread725-85323
thread725-188550
thread725-89146

Mostly talking about lock washers in particular, but some of the comments might be useful/interesting. Somewhere in the vast reaches of the forum threads was a link to a video showing a fastener undoing itself under vibration; quite fascinating.

TTFN
faq731-376

 

[DesignerGuy16]

This is why loctite and nylon patches in the threads were devised.

Now to throw the whole thing for a loop, check out

http://www.spiralock.com/

You'd be ahead of the game with a flat head screw with that type of thread profile, not that they even exist or anything. It's just a cool setup.

James Spisich
Design Engineer, CSWP

 

[hydtools]

I would suggest more torque is lost in friction due to the conical seat, therefore less torque is available for initial clamp load and the joint is loose from the start. The screw may be bound but the joint is not.

Ted

 

[dvd]

I have been involved in several instances where fhcs' loosened. An additional issue seems to be the surface finish on the c'sink. With fluctuating fastener loads, high spots in the c'sink will burnish over time and tension is lost on the fastener.

 

[rmw]

I've got a mental image of taking a flat headed countersunk screw, a punch and a hammer and staking the edges of the screw into the countersink in a couple or three spots.

Did I dream that up or has anyone else seen it?

PS: after tightening of course.

rmw

 

[MikeHalloran]

>>>Our UL inspector at one time said they accept flat head screws to fulfill any "no loosening" requirements on certain products/certain situations. <<<
Now, that's scary.


WRT staking the screw head, no, I haven't seen that, but I swear that I've seen the substrate extruded into the slot of a flat head screw by means of deep punch marks. Don't lose track of which screw goes in which hole...



Mike Halloran
Pembroke Pines, FL, USA

 

[ajack1]

I could be completely misunderstanding the question here, but isn’t the reason car wheel nuts are designed with a taper to increase alignment and reduce the likelihood of loosening from vibration?