Help specifying bolts for a hydraulic manifold flat seal face

[ScottAC]

Hi all,

Not sure if this is in the right section, apologies if not.

I have been tasked with checking a hydraulic manifold design is adequate and I would like to check the bolts details.

See attached image. The manifold is bolted to a flat face (blue) via three bolts (black) and has two ports which will be carrying pressure of around 300 bar (red) - there will be two o rings (yellow) around the ports to seal. I have suppressed a lot of features from the manifold to keep the image simple.

My understanding is that there will be a force trying to push the manifold off which will be resisted by the bolts. I would like to find out what that force may be so I can then work out if the bolts are suitable and then try to determine a torque value.

I have tried searching but I am not sure what the correct definition is for this problem..flange sealing?, clamping force? etc - any guidance in the right direction would be much appreciated. I have no doubt this is probably a simple question and I am overthinking it.

 

[MintJulep]

Force = Pressure x Area

What's the area that the pressure is acting on?

 

[LittleInch]

The hardest part will be the unequal forces on the sealing box / manifold meaning the clamping thing will have a bending moment within it trying to lift the top right corner. Not good or easy to solve. Why no bolt in the top right corner??

But in a simple way this is a flange calculation. You need to know the minimum sealing force required for the O rings to not get pushed out of their grooves. work on the basis that they will be squashed and the two surfaces are touching. So the min clamping force is the force required to squash the o rings and achieve mating of the two surfaces. then add a bit.

After that the separation forces are those from the surface area of the manifold exposed to the fluid pressure and the force from the bolts holding it tight.

Even once you have the bolt force, converting this to torque is a risky business as much depends on the friction of the bolts which is hugely variable depending on materials, cleanliness, lubrication etc.

 

[JStephen]

Assume both items are rigid, gasket is flexible with force proportional to deflection, and it's relatively simple to calculate gasket pressures and bolt forces.
Justifying or removing those assumptions- considerably more complicated.
Showing that the gasket won't leak and the bolts won't loosen- also more involved.
One of those issues where experience with very similar items would help a lot.

 

[desertfox]

Hi ScottAC

In the first instance you need to multiply the sealing area by the pressure to obtain the force trying to throw the block off and in your case the force will be acting on both ports.
The force obtained from the above will have to be resisted by the bolts so this will give you some idea of the preload for the bolts. Now because the bolts are not symmetrical around the ports and the ports are not central within the block you may encounter bending forces in the bolts due to the block trying to cantilever from the far end of the block and in order to help you further we would need size of bolts, sealing area size and the x and y dimensions of the block, position of the ports and position of the bolts from the far end of the block.

 

[3DDave]

The bolts are unlikely to see bending. They certainly will likely not see similar loads. Because two bolts bridge the rough average center of pressure the third bolt (lower left in picture) is doing nothing but squeezing metal in a way that does not resist the pressure, though there may be some other forces applied that need to be considered.

Without knowing the stiffness of the cover it's not possible to determine if any of this will work at all.

The torque value is independent of all of this. Once you have selected a fastener with enough strength you need to have enough elongation in the free portion to get to about 80% of yield. I'd recommend low, seating torque followed by a sufficient turn to produce the correct elongation. The initial torque should be only enough to make the surfaces contact and squeeze the o-rings. Lubrication and roughness affect the torque-to-elongation too much to be as reliable as amount of turn.

 

[Alistair_Heaton]

There is an ASME pressure vessel code for it.

It's not as simple as it first looks apparently. There is also fatigue issues.

I left the industry 20 plus years ago. But do know it was there as a fellow research assistant was working on related items.

Nuclear inspectorate also commissioned research on the subject. Jim Boyle and Tom Grey might be worth a paper search on the subject