Shear load tranfered by nutplate fastening system 1

[Benbarca7]

Hello everyone,

I have a design problem, and I was wondering if someone on Eng-tip could help me.

I have a design change which consists in fastening a seal attached to other plates with bolt + nut plates instead of classic rivet system.
this change will permit to change the seal without any drilling operation which could oversize the hole diameter.

Basically I have no appreciable load from the certification report, but there is probably thermal loads due to the difference of material between the seal (aluminum) and the plate (titanium) .

My question is, with this load, how would I perform a calculation to transfer it from the plate to the seal ?

For me, as there is a gap between the bolt shank and its hole, there is no bearing load or other efect due to the contact of the bolt on the skin.

I was thinking that the method to calculate a shear load transfer in a nutplate configuration is friction.

has someone an idea ? what calculation would you perform ?

I reformulate my question in a more general point of view :
What hand calculation would you perform to transfer a load from a skin to another if you have a bolt + nutplate fastening system (no contact between the bolt shank and the hole).

Thanks for reading and hope this question will help the community.

Regards,

Benjamin

 

[desertfox]

Hi Benbarca7

For shearing of the joint then yes the friction generated by the bolt preload resists the transverse force, however I am wondering how you might assemble all your bolts without any of them touching the side of the bolt hole.
For the thermal changes the stress created would also increase/decrease in the bolts and clamped materials parrallel and transverse to the bolts.
A little bit of info on thermal loading of bolts can be found at:-

http://www.roymech.co.uk/Useful_Tables/Screws/Preloading.html

desertfox

 

[RichChwa]

Hi Benbarca7,

It depends on your design application. Is this an airtight sealing? Do you use any rubber gasket and what type of rubber gasket do you use? How much is your hole diameter larger than your bolt shank diameter? If your holes are a lot larger than your bolt shank diameter you may not have any bolts contact with the material in bearing. If you have multiple bolts in parallel and rows design, some of the bolts may be lucky to contact depends on the installer how he installed it. There is no way you can predict which bolts will contact the panels and load can transfer. Using some sealant will help you transfer the shear load, if you call out the minimum strip width and length area but you need to get some reduction factors.

You probably rely on friction and bolt tension for your design. FAA/DER will not allow you to use friction for analysis. If it is going to use on airplane external air-passage and no cabin pressurization involves, you make sure the external air-pressure or suction pressure loads the panel can take and the panel will not be failed and flying off and hit the control surfaces or engine. Without knowing your geometry design it will be hard to advice.

Rich.

 

[Benbarca7]

Hello gentlemen,

Thank you for the answers Rich & desertfox.

1/I agree with you that during installation, there might be contact between bolts and holes.
But you don't know how many bolts will be in contact.

2/But you may not bear until the friction force is defeated, am I correct ? or you mean that a possible contact + thermal distortion will induce bearing effect ? so doing a bearing check is very conservative because the load is conservative. Plus in the bearing calculation, the bearing surface is smaller as you have a bolt diameter smaller that the hole. This will lead to very low margins.

That is why I wanted to do friction, to show that no shear load is transferred in the bolts.

3/ Rich, Can you explain to me why the FAA would refuse friction calculation ? Do you refer to a FAR article ?
I will answer your questions later when I'll have all the info.

4/ Desertfox, thanks for the link, I will have a look at it.

Best Regards,

Ben

 

[desertfox]

Hi Benbarca7

What I am saying is that you cannot assemble all the bolts and guarantee, that the bolt as clearence alround its diameter (ie cannot place each bolt perfectly central in each clearence hole),but you are also correct in that the number of bolts touching the side of the clamped parts is unknown. I don't know what the rules are for aircraft but I would normally do the friction force to resist the transverse force and then check the bearing, shear and crushing stress on a single bolt, if the stresses are within the deemed safe stress then I consider the joint to be safe, sometimes you need to assume two bolts in contact otherwise you end up with bolts being larger than you really want.

desertfox

 

[Dan320]

If your shear force is less than the nutplate attatchment rivets can handle you would be ok. Nutplates are thin, so check bearing on rivet. Use DD-rivets perhaps?
The friction would be an extra margin of safety.

 

[tbuelna]

Benbarca7,

Unless you can show by analysis that all of the fastener bodies have a proper shear fit with the holes in the clamped parts, then you have to assume that any transverse loads across the joint are reacted purely by friction in the clamped interface.

If there is any clearance between the fasteners and holes, however small, then once the joint transverse loads exceed the static friction at the interface the parts will move. This will initially result in two fasteners taking the load in shear. And after they yield, the other fasteners will begin to load up. But since this condition means a plastic failure of the hole edges or fastener bodies, it is not usually acceptable by an analysis.

As for the CTE mismatch (Ti and Al) between the clamped parts, this will result in strains within the two parts until the clamped static friction force is exceeded. Depending upon the size, stiffness and temperatures of the two clamped parts, this thermal mismatch strain can be a big problem unless some sort of strain relief is provided in the design.

Hope that helps.
Terry

 

[bf109g]

I remember ugly discussions when the shop wanted to make the screw holes larger and go wiyj floating nutplates on a gen av prop spinner. They claimed it would make it easier to assemble, which it did. Then we started getting complaints about failures. Turned out that less than half of the screws were actually touching the spinner or bulkheads.

We ended up reworking the drill tooling to tweak the locations of the holes AND made the holes a little smaller. Engineers like me just don't know what they are talking about.

 

[RichChwa]

Ben,

Unless you can proof by test which you have to convince the FAA/DER your test plan, test methods and numerous test samples you need to develop. Which will cost a lot of money and you still may not get the right answer. There is no methods or allowable and no FARs that I had known. It depends on the loads that you need to show and number of fasteners (nutplates)and you also may have to show what is the load if contacted on one or few fasteners and each fastener load will be able to carry by NAS1097AD3 or AD4 rivet shear/bearing load allowable etc... That is why no stress engineers use friction methods and show the substantiation.
You tell the management you want to do the expensive test or take the liability and fool the DER buy out the analysis or design it right and show the correct method of substantiaiton.
Good luck!