100 bar / 1451 psi sludge cake pipeline 1

[pch1]

I am in the process of speccing a carbon steel sludge cake pipeline for sizes 1/2" thro 4".
Given pressure is 100 barg (1451 psi), corrosion allowance of 0.04" and at ambient temperature (from table A-1 using an allowable stress of 20 ksi for pipe A106 Gr B).

For the 4" pipe, I've selected class 600 for group 1.1 materials from ASME B16.5, but when I do the calc for the wall thickness using ASME B31.3 I find it comes out at t(nom)=0.23" which is schedule 40 pipe.

When I checked this wall thickness against my 'The Piping Guide' (By David Sherwood and Dennis Whistance), I found for 4" schedule 40 pipe they have listed the design pressure as 604 psi. I realise 'the piping guide' bases the design pressure per ASME B31.1 with an allowable stress of 9000 psi, but I just get the feeling that my calculated results are a little too optimistic as I was expecting to find schedule 80 pipe mated to class 600 flanges.

Can anyone set my mind at ease, and agree/disagree that schedule 40 pipe is acceptable.

 

[zdas04]

My Cameron Hydraulic book (1988 edition) shows 4-inch Sched 40 pipe as having an MAWP of 1400 psig at 100F. I would call this inadequate for a 1451 psig pipeline.

Your ANSI 600 flanges look a tad marginal for the pressure also. My table shows them good to 102 bar below 38C which is a really skinny safety margin.

I would tend to use ANSI 900 flanges and Sched 80 pipe and sleep really well at night knowing that if the operation is really 120 bar after it gets started I'm not going to hurt anyone.

David

 

[dcasto]

Doing the detailed calculation I arrive at 1447 psi under B31.3 and 2206 psi maximum. The flanges would be rated at 1480 psi.

You used the word pipeline, so B31.8 would apply, but it is your choice.

You need to document your selection in any case for risk management. Sch 80 would be overkill and maybe going to .250 wall would be better. Same with the flange rateing, its your choice. Give that you have what appears to be a "mostly Harmless" (my appologies to Mr. Adams), you could stay with the lower line rating, agian your call.

 

[pch1]

It is process piping, pumping sludge cake from a ram pump to a silo in a waste water treatment plant.

Dcasto, when I did a check for pressure of the 4" pipe (P=2tSE/(D-2tSY)) I found a max pressure of 2199 psi (Using D=4.5", t=0.237" mill tolerance not taken into account, S=20ksi, E=1 for A106 Gr B).

There is one unknown - there is no design temperature for the pipe (That is why I originally assumed ambient), I think I may even select class 900 flanges as at 50 deg C the working pressure for class 600 flanges is 100.2 bar.

 

[BigInch]

0.25 wall OK, but I certainly don't understand why you would use 900#. I wouldn't normally even use them offshore for a 1440 psi system. High surge pressures? High temps and thermal stresses? High mechanical forces? I don't see it.

http://virtualpipeline.spaces.msn.com

 

[pch1]

BigInch

I have spoken with another engineer on the same project and he has confirmed that the temperature is to be ambient.

Initially I selected the class 600 (ASME B16.5:2003)which states from table 2-1.1 (Pressure-Temperature ratings for Group 1.1 materials) a working pressure of 100.2 bar at a temperature of 50 deg C. As this project is new and a lead mechanical engineer is still to be appointed, I was hoping to confirm my findings on this forum, so when my calcs are checked they will all be in order.

Biginch, can you confirm that 4" sched 40 pipe (0.237" wall thickness) is acceptable? If not then I may have to select from the next shedule up (i.e. Sched 80 - which would be overkill according to Dcasto)
From my original post the corrosion allowance is 0.04".

Thanks

 

[BigInch]

My B31.3-2002 says in pp 304.1.2
Since t<D/6 = 4.5/6 = 0.75",
Then use equation 3a,
P = 2 * t * (SE + PY) / D
Including a minimum weld quality factor of 0.85 (use 1 if seamless)
including a mill tolerance of 0.85
using "other material's" Y value of 0.4
using corrosion allowance of 0.04"
using wall thickness of 0.237"
Allowable stress 20000 psi
P = 2 * (0.85*0.237-0.04) * (20000* 0.85 + P * 0.4) / 4.5
starting with a P of 2000 and iterating for a final solution for P, I get 1579 psig

Nearly 10% higher than 1480 psig, the ambient limit for ANSI#600. Short term surge pressures up to the lowest value of either the ANSI#600 1480 psig + 10% = 1628 psig or the pipe's 1579 psig + 10% = 1737, so that's up to 1628 psig will not need any additonal wall thickness.

4.5" OD x 0.237" WT PIPE and ANSI#600 FLANGES at nominal ambient temps works very well for me.

http://virtualpipeline.spaces.msn.com

 

[pch1]

Thanks for the confirmation BigInch, and yes I'm using seamless pipe so E=1.

In case I'm not on this forum before the holiday season prevails, a big Merry Christmas to all.

 

[SteveWag]

I did two of these at 1000 PSI and 6-inch pipe, using Victaulic fittings and sched 40 pipe. Both have worked for over ten years. Sludge tends to be at ambient temps. The only problem was the LR ells tending to straighten; they needed to have reinforcing added. You may be able to lower the operating pressures by injecting polymer using an injection ring. These types of places tend to have polymer around.
Steve Wagner

 

[dcasto]

With 12.5% mill tolerance you get 1447 psi. I usually look at the mill and get some data that supports a smaller mill tolerance IF needed to keep from going to the next available thickness.

 

[zdas04]

Do you guys really design piping so close that the mill tolerance makes a material difference? I can see pushing limits in manufacturing where saving 12.5% of 0.237 inches (0.029 inches) on 40 gozillian parts is a major dollar savings, but in piping? The cost difference between Sched 40 and Sched 80 is less than $10k/mile for 4-inch. If lay costs are $75k/in mile, then you're looking at 3% cost increase or less--and pipeline jobs are by definition one-off. What is the motivation?

David

 

[BigInch]

On the other hand, why even bother to have a spec who's scope specifically includes determining suitability for this particular design case, specifies a particular cut-off point and yet produces a result that a number of you consider is either not safe, or so liberal that it warrents additional wall thickness to be arbitrarily included based on individual whim? I mean basically you can't find a situation where a spec can be applied in any more of a clear cut manner or any simplier than this, so if you can't believe what it tells you in this case, what can you believe? If one arbitrarily wants to call this a liquid pipeline, there's automatically another 12% more of an allowable stress available. Do you think Mother Physics knows the difference between B31.3 or 31.8? The secret of designing to the limits is first knowing when the limits apply. There is nothing that has been said so far to indicate this case is anything special at all, so I'd suggest that engineering judgement be saved for when its needed and otherwise just allow the code do its job.

http://virtualpipeline.spaces.msn.com

 

[BigInch]

Make that 15% more.

http://virtualpipeline.spaces.msn.com

 

[GregLamberson]

zdas04

As BigInch so eloquently pointed out, it has nothing to do with "designing it so close", it is simply following the code and then whatever dumps out dumps out and then you can decide which standard wall thickness to use or play around with the SMYS and optimize within the box that is defined.

Greg Lamberson, BS, MBA
Consultant - Upstream Energy
Website:

http://www.oil-gas-consulting.com

 

[BigInch]

Eloquently? Right. I see I was my usual self last night. Thanks for slapping my hand Greg. Appologies.

I've always had a sore spot with one engineer adding this bit of fat, one engineer adding that bit of fat, another painting it gold.. pretty soon you get a golden elephant with three legs and nobody knows what you really have underneath the hood. I also think, if you're using a code, you're never going to get very near a true limiting state if the product stays within the design conditions. Ex. a design pushing the burst limit in this case might be something like an internal pressure of 2 x 0.197 x 35000/ 4.5 /0.9 to get into the plastic range = 3064 psig. If the A106B is actually 42000, as it probably is, it might even be as high as 4085 psig. When accidents happen, its usually when some process variable has gotten out of hand, the cyclic life parameters, or the corrosion allowance was exceeded without replacement of the material, etc., cures for which are meant to be included elsewhere; not in "adjustments" to the piping codes. I'd just bet you that following a code to the letter will very seldom ever, if ever, be the source of a problem. At least I haven't ever found one single incident where I could have pointed to it. Have I been wrong all these years?

http://virtualpipeline.spaces.msn.com

 

[GregLamberson]

Actually, I was serious, I thought is was pretty well said. Eloquent, okay, maybe not, how about practical.

Greg Lamberson, BS, MBA
Consultant - Upstream Energy
Website:

http://www.oil-gas-consulting.com