Reduced Allowable shear for Countersunk screws

[TimSchrader2]

Hello ... Does anyone know of a chart or link that gives the above for countersunk screws in single shear (as compared to hex heads)? When a CSK screw is in single shear the loads on the CSK angled faces induces moments that can rotate the head out of the countersunk hole. I have seen this happen. The angled faces contact the material imediately upon torquing, before the hole diameter would create bearing loads on the body of the screw. Especially if a large hole clearance was used. So, it makes sense that a CSK screw would not handle as much shear load as a hex head where most all shear is thru the body (neglecting clamping friction).

The exact solution is more complex as it would have to deal with the preload values and the K stiffness constant of the surrounding metal, which can be about 8 times the screw stiffeness constant. But I would think there would be a rule of thumb. It should be a certain % of the shear capacity of a hex head. I have seen a few charts that do not list any reduction. Which I believe are not accurate.

MIL STD HDBK-5H does address CSK rivets.


This is a 'remotely' similiar situation with pin bending (seperate topic). Which relates to bolt bending to some degree. Standard methods of assumed load distribution yield very conservative moments. I have one book that says too assume the loads as a point load at the centerline of the thickness of the 2 or 3 plates the pin goes thru. This gives results that are way too conservative, unless maybe for a bridge where that safety factor is justified. Using a FEA to better define the load distribution between the bearing surfaces gives more accurate results on pin bending. I have done some tests to verfiy this.


Thanks for any insights
Tim

 

[Tmoose]

I'd use the ~ 50% less recommended torque and resulting 50% less clamping for the flat head SHCS as a start

 

[3DDave]

I think the rule of thumb is to not use countersunk fasteners in shear critical applications.

 

[Tmoose]

What are you making?

How many bolts in the pattern? Assumptions about load sharing being "good" due to each countersunk head contacting the countersunk hole are flawed. I'd suspect many of the bolts would simply be forced to lean this way and that. The bolts leaning >>that<< way are not going to participate in load sharing until the component has slipped >>this<< way far enough to make them resist, by which time the bolts that started off already leaning >>this<< way will REALLY be heavily loaded.

5 bolt patterns using nuts.screws with Tapered seats, torqued to provide clamping force, and some tricks to provide compliance for insurance, have been around a long time for torsionally slip critical joints .

http://chevy.oldcarmanualproject.com/shop/1955/images/55csm0302.jpg

 

[TimSchrader2]

Hello..... Thanks for the input.

I agree with 3DDave, but the aircraft industry does it all the time. There must be some shear on those panels. I also saw a chart on line form some manufacturer giving shear values for CSK screws at about the same level as hex heads. Although, I do not believe it. The above MIL STD lists allowable shear for CSK rivets, which probally should be conservative values if applied to screws. But they look higher then I would think they should be. The chart is at about page 9-142

Of course all rules of thumb are not 100% accurate.

So, TMoose you only count on the shear provided by clamping force and not the shear capacity of the CSK screw. Which should normaly or even always be a conservative assumption. If I read the above correctly. Although many do this I usually am a bit leary of depending on clamping force as we mostlly use stainless screws which do not develop the clamp pressure of Gr8 Bolts/screw. But the 50% rule is conservative enough to avoid any issues. If slipping did occur the shear at the body of the screw shear plane and the moment on the tapered screw head should be considered to calculate the reduced shear load. But have not found a chart to back my numbers up.

We use a fair amount of CSK screws in brackets and flanges that attach to the decks of Yatchs. So for cosmetic and clearence reasons I am sometimes pressured to use them at times. So I have some numbers of what shear loads these screws have been subjected to, but have not fully defined how much the safety factor has been compromised. They have done so for years, before I got there.

 

[Tmoose]

"We use a fair amount of CSK screws in brackets and flanges that attach to the decks of Yachts. "

Is this the application you were asking about ?
If so, are these essentially wood screws, screwed directly into fiberglass or mahogany?

 

[tmschrader]

Hello Tmoose.

My question was in general as we do use alot of CSK screws. But in this particular case I was working on a bracket on a crane that is screwing into 3/16 aluminum.

The screws are self tapping #14 (1/4) screwed into a fiberglass sandwhich in most applications. Simliar or alamost the same as wood screws. Not very often into wood, but sometimes. That is the worst case scenerio cause the screws can pull out of wood.

 

[Tmoose]

#14 self tapping screws into 3/16' thick aluminum?
I wouldn't be focusing that hard on the fastener.

http://www.itwbuildex.com/pdf/UPDATED BX Catalog.pdf

https://www.hilti.com/medias/sys_ma...chnical_information_ASSET_DOC_LOC_1577675.pdf

 

[TimSchrader2]

Hello
#14 screw in 3/16 alum is not something I like to do or do very often. I have seen it here a few times in many years. Which is why my concern about the CSK heads was amplified enough to do this post. Thanks for the links. I realize it is not up to the 'rule of thumb' of 1 diamter, some say 1 1/2 Diamter min thickness. I have some formulas and charts from the aircraft industry that give reduced allowables for thinner materials. In this case most of the load is shear. 0.220 Min thickness is not too far off from 3/16, as mentioned in the one of the above links.

It is hard to get them to do extra work when something has worked in the past.

There was a 1" pin on a crane here that if analyzed by the the distributed load method for moment bending stress would bend at below the operating load. The Standard methods of calculations are often conservative. A FEA will usually give stresses that are about 60% to even as low as 40% of the bending stresses from the distributed load method. I have confirmed this with some testing.

 

[hydtools]

Here is a link to fastener handbook discussion of csk screw loading.

https://books.google.com/books?id=N...KHZkSC_w4ChDoATABegQIBRAB#v=onepage&q&f=false

Also refers to MIL-HDBK-5.

Ted