flattening SS pipe properly

[rcmac21]

I am going to make a SSseamless pipe at 1.5" sch40 into a flow meter and I need to change the cross section shape. I intend to flatten pipe to about 1/2" height for fluid to pass through. When I do this, I think I am going to have a bit of a problem. I think the rating of the pipe for pressure will no longer be valid. Does anyone know of a method to do this and not loss pipe intregrity. I have not yet found a standard to cover this type of process hot or cold forming yet.

 

[JStephen]

One of the appendices to ASME B&PV Section VIII Div 1 covers rectangular vessels, and also covers "obround" cross sections, those being semicircular ends with flat sides. If your pipe cross section will look something like that, check into it.

The motivation to use round pipe in the first place is for strength, so if you change it to any other section, you're going to lose some strength. You could start with heavier pipe, or possibly a size larger and heavier-wall pipe, or perhaps machine it out of bar stock or other means.

 

[dhengr]

You probably want to make something like this as if it were a coupling, and only 4-6" long. Then weld it, flange it or thread it back onto the pipe at each end of the metering device. This way you can work on it from both ends in the forming process; which might involve pressing from two opposite directions from the outside, and into an outer die in the other two quadrants, while pushing some sort of a die into it from each end, all at the same time. Some sort of an up-setting operation. This might also be cast or made out of a thicker walled piece of pipe so it can be fairly finely machined to an inside shape, which I would assume you would want for a metering device, as JStephen s suggested. Trying to do anything like this mid-length in a longer pipe will give you more problems than its worth, and an uncertain interior shape, and a reduced bursting strength.

The pipe stress picture will change because with a perfectly round pipe we only have a nice clean hoop stress and a longitudinal stress, due to pressure, plus any beam bending stresses. Whereas what you are talking about may change the wall thickness and also the circumferential stress picture. You have a pressure vessel with flat sides and some radiused corners over some length, and then a transition back to round. This causes a much more complex stress picture both hoop stresses and bending stresses in the circumferential direction, plus the other stresses.

 

[berkshire]

Would it not be easier to machine your flow meter from a piece of plate,which is the outside diameter of a set of flanges ( remember the bolt holes.). then just insert it in between a pair of pipe flanges. This way if it is not what you want, you can take it out modify it and put it back. you also will not have to worry about loss of strength.
B.E.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[MikeHalloran]

How much pressure are you intending to contain?

Mike Halloran
Pembroke Pines, FL, USA

 

[hacksaw]

why not buy an engineered flow meter (wedge, verntui,coriolis, etc) and be done with it?