UG99b hydrotest according to Note 36

[freschif]

Dear ALL,
considering a case where a vertical vessel have different operating pressure at different levels (gas pressure constant along the height plus a material pressure variable along the height - material is having a higher density respect to water).
We define therefore for this vessel different design pressure at different heights. Suppose 8 barg at top (where only gas is present), 8,5 - 9,4 - 10,8 - 11,5 - 9,3 - 8,4 from top course to bottom one.

When determining the hydrotest pressure to be used (with the vessel in horizontal position) we have doubt which is the correct pressure to be used:
1) if we use gas (Design) pressure times 1.3 + water head (given by the diameter of the vessel when in horiz. position - suppose 5 mt) could not be sufficiently a proof of tightness and resistance (it could happen that some section during hydrotest stays at a pressure lower than operating)
2) if we use gas (Design) pressure times 1.3 + water head + additionl pressure only to correct water head level for eventual site hydrotest vertical condition: could still be the case above, even if more reliability on the test is present
3) if we use the max Design pressure (gas + solid) *1,3 + water head could be too cautionary and could require plate thickness increasing

Is there some Code indication which specifies better our case?

 

[SnTMan]

freschif, I didn't read your entire post, but if your vessel is likely to be hydro'd full of some fluid in the vertical position, it must be designed, constructed, stamped to consider the hydrostatic head at the lowest elevtion. Don't overcomplicate it.

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[freschif]

Mike, thank you
Not clear your answer (maybe also my question was not so clear), anyway seems that you suggest to use my option 1 which as said is not sufficiently "safe": as said it could be that under test some course could be less stressed than in operating/design condition.

P.S. it is not full of fluid, but bulk material.

 

[SnTMan]

freschif, all Note 36 is telling you is you may base test pressure on design pressure rather than a calculated MAWP.

You have to, somehow, base your test pressure on the 11,5 barg design pressure plus the allowance for static head in the test position, horizontal or vertical. Will your solid material be present? Is it credible that different zones of the vessel can be at different test pressures? Will it someday be tested in the field, vertically, corroded? If so you have to make allowance for that. Will the 8,4 DP + static head at that zone be more or less than the 11,5 DP + static head at that zone?

You have to base the test pressure on the greatest combination of design pressure and static head applied in the test position, both in your shop and in the field.

You've got seven (7) design pressures. How many nameplates you planning on?

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[Geoff13]

For me, UG-99b is the standard test pressure formula that works fine for "standard" pressure vessel configurations.

However you seem to have a unique vessel with granular material loading in addition to the design pressure. I presume the odd pressure distribution is from Janssen formulas, or a similar method. In addition you want to test both horizontally and vertically. I believe all of this puts your vessel outside what UG-99b is best suited for.

I'd be doing UG-99c checks in both test positions. You may conclude that two different test pressures are required. From the calculations you will know what test pressures meet Code, but also if those pressures seems to test all areas adequately, or if any areas seem over-stressed. During a hydrotest I'd suggest something around 0.8Fy to 0.9Fy as an over-stress limit for a standard carbon steel vessel, but you need to decide what your risk tolerance is. It is possible that the lower area may have to be thickened only for the hydrotest so that you can adequately test the thicker areas.

I'd think carefully before making the lower rings thinner than the mid-section. Granular pressures are more art than science. How certain are you that the pressure will be lower under all fill / withdrawal / thermal variation / pressure variation / etc. conditions? How accurate are your granular properties? Will the process send consistent granular properties over the life of the vessel? Is the pressure from a gas or liquid, and if liquid does this alter the granular properties through lubrication or bonding?