Pipeline Wall Thinning Formulae?

Only thing I can think of is perhaps the FEED Contractor has taken bend radius as 5D to the bend centreline, rather than inside the bend. This is more normal - e.g. a standard 1.5D bend has radius 1.5D to the bend centreline.

If so, in your case the inner bend radius is 4.5D and the formula gives 50/(4.5 + 1) = 9.1%. Not spot on but not a mile out.

Also, what does PD 8010 mean by pipe diameter - nominal bore, actual bore, or OD? If you take bend radius = 4.5*12" (nominal bore) but divide by 12.75" (OD) you get 50/(4.5*12/12.75 + 1) = 9.55%.

Cheers...........Codey

Thank you so much for your quick response. Is that OK if Contractor takes 4.5*12" (nominal bore) but divide by OD? The value of 9.6% i think perhaps they got from that way. Thank you.

Glad to help, but as to Is that OK if Contractor takes 4.5*12" (nominal bore) but divide by OD? - I've no way of knowing. You need to study PD 8010 to be sure what it says, and perhaps discuss with the contractor, including the definition of bend radius.

But the difference is unlikely to be significant. What is the thinning calc used for? If it's to check that the wall thickness is still OK for the pressure, it would have to be very near the limit if 10% reduction takes it above allowable stress.

Cheers.........Codey

Thank you sir!

% thinning in bends is defined as

= (R₀-R)/R₀

Where R₀ = Outside Radius and R = Centerline radius of the bend pipe.

ie

If we take D = Inside bend radius (ie bending mandrel radius) and

d = Tube Diameter (OD) then

R = D + d/2

R₀ = D+d

Then % Thinning

= (R0-R)/R0

= ((D+d - (D+d/2))/ (D+d)

= d/2 / (D+d)

= ½ x 1 / (D/d+1)

= ½ x 1/(n+1) by your terminology

which converted to % form becomes

= 50/(n+1)

if n = D/d = Inside bend radius/ Tube OD

Hence the contractor calculation of taking 12.75" is correct.

Thank you

Your calculation derives the formula t=50/(n+1) %, but it doesn't answer the OP's question.

If the FEED contractor had taken bend radius 5D (where D = pipe dia, not your Inside bend radius!) on same basis as PD 8010 i.e. inside bend radius, t = 8.3%.

Need to make some assumption such as mine in #1 to get somewhere near 9.6%. Assuming 5D is to bend centreline, inside bend radius = 4.5D which gives t = 9.1%, whether D is nominal, ID or OD, as it cancels.

To get closer to the 9.6% figure, I guessed that FEED contractor had taken inside bend radius as 4.5*D_nom, but divided by OD to find n. It might be a coincidence that this comes close to 9.6%, which is why it's a good idea for OP to discuss with FEED.

Cheers........Codey