Glass Fiber-Reinforced Piping Design 1

[Kevin Batalha]

Good day all,

I was wondering if someone was able to help / guide me in calculating the hoop stress of a thin-walled steel liner that is REINFORCED with a glass fiber material:

We are currently in the midst of performing a feasibility study for high-pressure piping being reinforced with a high tensile strength glass fiber material (E-glass, S-glass etc...). The objective is to achieve an optimal balance between the amount of steel pipe thickness and glass fiber reinforcement thickness. The main idea is to reduce weight, and overall cost of steel. This design will not yield for high volume manufacturing.

- High pressure (internal) for this application is 7,500 PSI.
- Atmospheric Pressure on the outside of the reinforced pipe.
- The inside diameter of the finished pipe will be approximately 4.0 inches.
- Since we want to minimize material, we are going to use a high alloy steel (AISI 4140 HTSR). The amount of piping should not influence the overall cost of the design.

AISI 4140 HTSR: UTS = 150,000 PSI ; Sy = 130,000 PSI ; E = 30,000,000 PSI​

- As a fiber reinforcing material, we are looking at 2:

E-Glass: UTS = 282,000 PSI ; E = 10,442,700 PSI ; Density = 162 lb/ft³​

S-Glass: UTS = 681,700 PSI ; E = 12,473,200 PSI ; Density = 155 lb/ft³​

The above listed above are our constraints for the time being.

Now, for an internal pressure of 7,500 PSI, I would like to accurately determine the maximum HOOP STRESS seen by the steel pipe with the reinforcement included. What we do know, is that the fiber reinforcement will only contribute to load sharing when the steel pipe expands due to the internal working pressure.

Question:

- Is there a relationship that I can use between the two materials that would allow me to determine how I much stress the steel liner is reduced to once the reinforcement is added?​

---------------------------------------
I've done some general calculations below, but I believe it may be more complicated...

Steel Pipe Hoop Stress WITHOUT reinforcement (assuming a thin-walled pressure vessel):

Pi = Working Pressure (psi) ; ri = Steel pipe Inside Radius (in) ; ts = Steel Pipe Thickness ; σ_h = Hoop Stress

σ_h = Pi * ri / ts -------- Since all variables on the right are given, I am easily able to determine the un-reinforced hoop stress of the steel pipe.​

Since the reinforced fiber material will only see stresses due to the elongation of the steel, I am assuming that the elongation in both materials are equal to determine my hoop stress :

ε = Strain ; Δr = change in radius due to elongation ; r_glass = mean radius of glass ; r_steel = mean radius of steel

ε = Δr / r = σ_h / E --------> Δr = (σ_h * r / E)_steel = (σ_h * r / E)_glass​

Solving for the hoop stress of the glass material then gives me:

σ_{h glass} = (E_glass / E_steel) * (r_steel / r_glass)* σ_{h steel}​

To recap:

- I solved for the hoop stress of the steel liner assuming it has no reinforcement.​

- Using the stress-strain relationship, I determined the hoop stress of the reinforced material assuming that both the steel and glass fiber reinforcement stretch the same amount.​

The Dilemmas:

- I am quite certain that it would not be accurate for me to determine the hoop stress based on the steel elongating at its maximum amount. The thickness of the fiber reinforcement will take up some of that elongation, and so I feel that this approach would be way to conservative. Given that the reinforcement will take up the elongation, it will also take up some of the hoop stress seen by the steel that I calculated above.​

- Is there a way that I can solve for the reduced steel hoop stress given the information above???​

Any help on this would be greatly appreciated!

Regards,

Kevin Batalha

 

[LittleInch]

You might be surprised to find that this has been done before.

Ameron developed their SSL - Steel Strip Laminate doing exactly what you suggest.

Las time I went looking for it for a project in concept I couldn't find it on their website so I think it has been withdrawn. However they might have a patent on it.....

See attached which notes that the design required complex 3D FEA.

http://www.compositesworld.com/arti...e-handles-high-pressure-oilfield-applications

note the date ( 2004)

"The actual load sharing is more complicated than these simplified calculations and must balance eight equations and eight unknowns, recognizing the hoop and axial modulus of both the steel and glass/epoxy layers, the Poisson's ratio in both directions, plus a force balance in both directions," says Friedrich. "But the result is very similar."

Ameron are now part of NOV and don't appear to offer this anymore.

http://www.nov.com/Segments/Completion_and_Production_Solutions/Fiber_Glass_Systems/Oil_and_Gas.aspx

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MJCronin]

I am having trouble understanding where such a pipe construction would be an advantage ...

WEIGHT ---- No

COST -------No

Corrosion resistance ----- No

Reliability --- No

Before anyone performs the mathematical gymnastics necessary,

Explain the use of such a construction .....

MJCronin
Sr. Process Engineer

 

[pennpiper]

Kevin,
You have asked a bunch of interesting questions. You have included a lot of data,factors and formula all for the finished product under design and operating conditions.

The thing you left out is what the finished product will look like and the process for constructability.
- What size will the outer pipe be (if the inner pipe ID is 4")?
- How do you plan to line these pipes?
- How do you makes joints in the FRP sleeve?
- How do you plan to make the joints in the steel pipe without damaging the FRP sleeve?
- Will there be connections to tanks and other types of equipment?
- How to you make these connections?
- Will there be Valves?
- What type of Valves?
- What size Valves?
- Will these Valves also be lined?
- How will they be lined?

Hopefully you have already thought about all of this. Please share the answers to the questions and a simple sketch of the configuration so we can help you get the constructability right.

Sometimes its possible to do all the right things and still get bad results

 

[LittleInch]

I can't find anything relating to the ameron system after about 2007

It was expensive ( + 40% compared to steel ), the joints relied on O rings and interlocking groove rings and other than the odd 10" line installed by Aramco appears to have bitten the dust.

It's advantage of larger diameter and higher pressure than standard GRE with the benefits of no corrosion as it included a couple of fibre layers on the inside before the steel band seemingly didn't find favour from the oil companies.

The fibre materials might have very high UTS and yield strengths, but the composite with the epoxy is, from memory, much lower than steel, hence why GRE pipe is, size for size, thicker than steel for the same pressure rating.

Hence the steel expands, but there needs to be much more GRE to get a similar expansion for the same internal pressure.

the use of very high strength steels always falls down at the how do you connect it part. any cost saving from using thinner steel just evaporates when coming into contact with reality.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MikeHalloran]

You haven't mentioned the nature of the process fluid, or the fluid temperature or the ambient temperature.

The marine exhaust shop where I used to work does a small side business in slightly flanged stainless steel sleeves to be glued inside GRP muffler terminations, which get soft when hot, and yield under compression from silicone tubing retained by double or triple hose claps.




Mike Halloran
Pembroke Pines, FL, USA