Bottom tank plate - edge settlement or deformation?

[Eduardo1982]

Hey guys.
In my workplace we have a storage tank that works around 350 °C. During an outage we drainaged the liquid and started to perform a inspecion (internal and external too removing the insulation). We have noted that the bottom plate (edge side) has been deformed as shows in pictures attached (brown lines)

The bottom tank shape is flat (center) and conic (in edges).The theory of edge settlement indicates that the settlement could be occured in the opposite side of the current status of my tank.
We dont´know so much why the bottom edge could have changed on this way. Do you guys have ever faced a similar situation before?
Any idea or info I will higly appreciate.
Best regards.

 

[MJCronin]

Flat bottomed storage tank operating at 350C ? .... Hmmmmm.......Does it store molten salt in a solar power plant ?

Can you post any pictures or plans, datasheets or design drawings ?

We cannot help you without more information ....

What are the shell and bottom materials of construction ? Were they ASME/ASTM certified materials or A36/A283 structural plate ?

What was the Code of Record and who is the engineer of record for this unique tank ? ....( My copies of API-650 and API-620 do not permit usage at such an elevated temperature )

Was the tank chime insulated and were there anchor bolts ? Were the anchor bolts pulled out by the expanding flat bottom ?.... Can you post pictures ?

Finally, and most importantly, How many other owners of 350C flat bottomed tanks do you think are out there ?

I believe that your problems may be linked to the design limitations inherent in a FB tank configuration.

Regards



MJCronin
Sr. Process Engineer

 

[Eduardo1982]

Yes.. it storages molten salts.

 

[JStephen]

Possibly due to bearing under the plate, especially if supported on sand/soil rather than concrete.
Could be due to thermal expansion- the outer edge stays cooler, tank expands more than the chime, and flips it up some.
Some of the effect may have been weld distortion in the original construction.

 

[HTURKAK]

- WHICH THEORY ? WILL YOU PLS SHOW ANY REFERENCE ?..


- APPARENTLY YOU ARE HESITANT TO SHARE SOME DETAILS ( MOC, DATA SHEETS, SOME DESCRIPTIVE SKETCHES, FOUNDATION DETAILS ..)

IN THIS CASE, ANY RESPOND WILL BE A GUESS ONLY .. MY GUESS IS;

- THE ANNULAR BOTTOM PLATE THAT PROJECTED OUTSIDE ( 100 mm ) IS OVERKILL FOR ELEVATED TEMP. TANK . WHAT WAS THE REASON FOR 100 mm ?... APPARENTLY THE CANTILEVER PORTION EXPERIENCED MOMENT MORE THAN PLASTIC RESISTANCE OF THE MATERIAL .( THE REASONS COULD BE , WHEN THE TANK EMPTIED THE SHELL WAS COOLER THAN THE BOTTOM , AND THE SHELL VERTICAL BENDING MOMENT TRANSFERED TO ANNULAR PLT AT STEADY STATE CASE ) ..

- I WILL SUGGEST YOU TO CHECK THE STRESSES ( FEM WOULD BE THE TOOL) FOR STEADY STATE AND WHEN THE TANK JUST EMPTIED.

- ANOTHER REASON COULD BE IF THE TANK EXPERIENCED SUBSTANTIAL SEISMIC FORCES..

YOU MAY TRY TO POST MORE DETAILS IF YOU WANT MORE SPECIFIC RESPONDS ,

 

[Eduardo1982]

sorry guys, I did not reply on time.
I provide some info (some others are still on investigation or recovering).
bottom plate material is ASTM A516 GRADE70.

Construction of tank was built following API 650 recomendations. NDT was more rentringent. As far as I know, there are not code or standard for molten salt tank construction (350°C-550°C).
This tank does not have anchor bolts, just it supports on its foundation (Mix of refractary material, sand and bricks).

 

[IFRs]

Clearly the refactory/sand/brick support under the shell was not capable of supporting the weight of the tank shell, and it settled. Is there any real question of what happened?

 

[MJCronin]

I disagree with IFRs ....

He states:

"Clearly the refactory/sand/brick support under the shell was not capable of supporting the weight of the tank shell, and it settled. Is there any real question of what happened?"


Yes there is .......

I think that the distortion is due to the massive thermal expansion of the floor of the flat bottomed tank AND the restricting effect of the shell on said expansion. The tank was sublect to thermal cycling. Flat bottomed tanks are not suitable for this high temperature service !!!!! . Another, proven configuration (such as a horizontal cylinder) should have been selected by the Plant Engineer. This is clearly a significant design error by the designer, IMHO ..

Both flat bottomed tank design Codes, API-650 and API-620 specifically exclude flat bottomed tank at this high temperature

Where are the engineering stress calculations for these tanks ?..... Were they even done ? .... Did a group of Architects and MBAs design this plant ?

MJCronin
Sr. Process Engineer