Countersink Angles 2

[LadyKat13]

I'm hoping that you guys can provide a little insight.

I have a new component that is essentially the design child of an existing component. It is attached with flat head screws (82° angle) but the countersink that they sit in is 90°. I find that I have copied this mismatch in my new component, and I am wondering if there are any foreseeable problems likely to occur because of this. Both components being attached are anodized aluminum and the screws are alloy steel.

My thanks in advance!

 

[SwinnyGG]

Depending on how much loading your part sees through those screws, creep of the very small part of the countersink that is actually in contact with the fastener may be a problem.

It's also possible that the screws bottom out on the chamfered edge of the threaded hole before they seat fully against the countersink. If you 'own' the mating part as well, you can help prevent that by specifying larger chamfers on the edges of your threaded holes.

 

[JohnRBaker]

Not sure if this will help you all that much, but the 82º is the standard countersink for Imperial (inch) fasteners. The 90º is the Metric standard.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

 

[BUGGAR]

I've been testing similar, but with matching bevels. The photo shows a shear load with bearing failure commencing at the shank. I don't know if the taper has any influence here. This is 0.125-in 6061 T6 fastened with 0.164-in dia high strength screws, hand tight (to limit shear friction for my experiments).

 

[Tmoose]

I'm picturing a screw with an 82 degree head making contact somewhere at the bottom of the 90 degree cone, as jgKRI describes.

I think the correct answers ( are there any foreseeable problems) will likely depend on a bunch of things like the loads on the secured components and the actual situation at the bottom of the cone. And in one extreme case, possibly the screw threads just jamming in the tapped hole's threads, and not clamping the secured component at all.

It may make sense, if the 90 degree c'sinks are otherwise accurately done, and the fasteners are working for a living, to use threaded inserts to permit using inch fasteners so the fastener and c'sink will get along.

 

[LadyKat13]

My thanks for your input.

Additional information:
There are a total of 8 screws (alloy steel), 2 on each quadrant 45° angle around two concentric parts (anodized aluminum). The surface that the countersink sits in is not flat but rather multiple angles approximating a cylindrical surface. The head of the FHCS sits below all exposed surfaces. The countersink holes are also not (real) holes but rather small slots. The cutter travel is relatively small (only .050) but enough to allow some adjustment of spacing prior to "locking down" the position by tightening the screws. For proper assembly the instructions call for the screws to be tightened in a pattern similar to how you would tighten lug nuts. The fasteners definitely "work for a living" but the forces in play are primarily axial to the two cylindrical components (at 90° to each screw's axis). So the primary stress would be shear, except that the "slot" design essentially alleviates it.

 

[Tmoose]

Hi LadyKat,

What stage of manufacturing is this at?
You say an existing component has the same angular mismatch.

I believe there are absolutely "foreseeable problems likely to occur."

If the parts exist, I'd test fit the parts in question to see how the mis-matched angle affects things. If the angled faces touch (as tested with bluing or a Sharpie pen, etc) and the parts are secure, maybe all is well.
If on the other hand the screw threads or the cone close to the threads jams barely touching the milled slot, or without touching the milled slot at all, then an adjustment needs to be made.

I personally consider multiple c'sunk screws a mistake for any application remotely structural There almost always will be a fight due to varying locations. Rarely will the full c'sink be in contact for all screws. Some heads will have line contact and nearly all will be forced sideways or even bent while being tightened.

 

[hydtools]

Like Tmoose has noted, I find that only one screw will properly seat in a related pattern of countersink holes. All the rest have varying mismatch head to hole. That is more of a concern than c'sk angle difference. Match drilling at the time of assembly gives the best chance of position alignment.

Ted

 

[BUGGAR]

With respect to the last two posts, I have been successfully using multiple c'sks to replace rivets in aluminum plate structures. Testing indicates I am achieving above-design strengths (>1000 lb before yield vs. 700 lb calculated). And note the deformation after initiation of yield. The relative "softness" of the aluminum allows all fasteners to engage before and during failure. These screws are to resist longitudinal shear in torsion shear box structures for performance vehicles. At present, we are relying on bearing area at the shank only. We will be testing for shear friction capacity next. We are working to Aluminum Association standards.

 

[CWB1]

My only concern would be ensuring that material grades will allow the parent material to yield before the head of the fastener does. Beyond that, its a basic bolted joint analysis. Ensure you have a proper 70-90% stretch of the threads and you won't have issues with fasteners loosening. As to multiple countersunk fasteners being a mistake, if that causes issues I'd recommend the tradesman join the 19th century in terms of accuracy.

 

[Tmoose]

Hi Buggar,

".......... The relative "softness" of the aluminum allows all fasteners to engage before and during failure. .........At present, we are relying on bearing area at the shank only. ...........We are working to Aluminum Association standards."

I don't think I'm picturing the fastener detail correctly.
"bearing area at the shank" makes me wonder, are the (unthreaded) shanks (body) closely fitted into holes (Counterbores) in both panels of the box? If the holes in one panel are threaded for the fastener, I'm picturing those holes need to be a close fitting, accurately located counter bore, and the shank needs to be unthreaded for a ways to engage it. And the top panel is thick enough that the countersink leaves a close fitting hole for the fastener shank/body too.

"all fasteners .. engage before and during failure.." - Is this describing the countersunk fastener heads all making over, say, 80% circumferential contact with the c'sunk hole in the top panel .

http://www.mutualscrew.com/blog/assets/content/Countersunk bolt parts definition.jpg

=======.
Is this your source of "Aluminum Association standards ?"

https://www.aluminum.org/aluminum-design-manual-2015

thanks,

Dan T

 

[TimSchrader2]

Hello
They use alot of CSk's here. And IF they fail they do so as BUGGAR shows in shear. As I recall there is also a Moment/bending stress on the screw shank as the CSK usually contacts with the CSK surfaces first. The picture shows the head bending. Not sure if this stress is considered in above calcs. Info on this effect is not extensive

There are some charts in MIL HDBK 5J on allowable shear for CSK's.