Drain Dry System for conical floor tank.

[FTank]

Has anyone had experience with a "drain dry" system for a conical floor storage tank. I believe the idea is to slope the tank floor to the center and run an outlet pipe from the side of a sump which would be installed at the low point of the cone, thru the tank pad (sand & gravel construction) and terminate in a 150# connection flange.
The tanks are 191' Dia. so the piping would have to run approx. 95' horizontally thru the pad. My concerns are:
1. Centre column support for roof rafters would be located inside the sump. (or sump c/b located off center to avoid this.
2. Loadings on the horizontal, buried pipe running thru the tank pad.
3. onstruction of sloped gravel/sand tank pad.
4. Etc., etc.

I checked API 620 & this is not addressed there.

API 650 Appendix "O" - covers under-bottom connections for flat bottom tanks so I'm thinking I could use this as a guide. But I'd rather borrow on others' experience if you've run across this before.

 

[TankDude]

I have seen this used and have actually been involved in projects with this type of system. You would want to use a water draw line in conjunction with the suction line. Also:
1. You could bridge the sump for your center column support, or you could consider a geodesic dome roof. If you are at a coastal location the geo-dome may be a bad idea though.
2. The under-bottom pipe should be capable of handling the overburden load, however you could consider an underbottom trench constructed of heavy plate to house the piping. I have seen these used with ringwall foundations mostly.

 

[IFRs]

You could also use an off center sump as long as the bottom sloped to it from every direction. This avoids the column support issue. At lease one major US pipeline company uses center sump as a standard detail, they are careful to design the ringwall penetration and sump details for leak detection, strength, durability, etc. One nice thing about their system is they have no shell nozzles other than manways and design their floating roofs to land 6" off the floor at the shell. If you use the sump for inlet you will have to design the sump with a diffuser to limit turbulence, static generation, scouring, etc.

 

[LeonEarle]

Another alternative might be to consider a hillside foundation. We have built a number of 150' - 220' tanks with hillside foundations sloping 12" from high to low. In these tanks, the drain-dry sump is installed at the shell. Some of the advantages are that you do not have a long length of pipe under the tank. In almost all cases, the drain-dry sumps are installed into a formed vault built into the ringwall. The sump itself is supported off the bottom of the vault and the piping and underside of the sump are exposed for easier observation and inspection.

 

[JoeTank]

Undertank piping is a terrible idea unless there are minimal foundation settlements expected.

Joe Tank

 

[MJCronin]

Based on the tank reports that I have read, I agree with "Joetank".

The drainage scheme that you are describing is commonly called "cone-down" configuration. Both the "cone-up" and "cone-down" configurations are described in the book by Phillip Myers "Aboveground Storage Tanks".

Most evidence is anecdotal, but it is my understanding that "cone-up" tanks are preferred because of resistance to corrosion (distributed in the edge, as opposed to concentrated in the tank center) and resistance to leakage, if the tank settles in the ground.

My opinion only...

Anybody else have experiences or references to share ?

-MJC