Shell capacity vs working storage capacity for crude tanks

Every analyst has made this mistake at least once: pulling a tank's nameplate capacity off a terminal spec sheet, multiplying by the number of tanks in a farm, and treating that as the ceiling for how much crude could show up in next week's print. It's wrong, and it's wrong in a predictable direction.

Shell capacity is the volume between the tank floor and the top of the shell wall. It's the number stamped on the engineering drawing, the one you'll find in an API 650 design spec or a terminal's regulatory filing. It describes the steel, not what the operator will ever put in it.

Working storage capacity is what's usable day to day. It sits below shell capacity for two structural reasons, and the gap between the two numbers is exactly the volume that trips up analysts trying to back into inventory levels from tank counts and farm capacity tables.

Where the volume goes: heel and freeboard

Tank heel is the crude that stays at the bottom and never gets pumped out under normal operation. On a floating-roof tank, the roof itself has legs, and when the roof lands on those legs (a tank running low enough that the roof can't float anymore) there's still crude below it that the suction line can't reach. Depending on the tank's design and the viscosity of what's stored, heel can run anywhere from a few inches to several feet of product across the entire tank floor. That volume sits in every capacity calculation whether the tank is "full" or not, and it never counts as deliverable supply.

Freeboard works the opposite end. It's the empty space operators deliberately leave at the top of the shell, above the maximum operating level, so the floating roof has room to rise without crude overtopping the rim during a fill or a thermal expansion event. Terminal operators size freeboard to their local fire code and to how fast they expect to receive crude, and they don't run tanks to the rim as a matter of course. A tank farm operating at what looks like "95% full" on a shadow read may still be well inside its working range.

Net it out and working capacity for a floating-roof crude tank typically lands somewhere in the 85-90% range of shell capacity, though the exact number depends on tank diameter, roof design, and the operator's own safety margins, which aren't published anywhere public. That's the number that should anchor your math when you're trying to figure out how much more crude a farm can physically absorb before an operator has to either slow receipts or start moving barrels out by pipeline or rail.

Why this matters between agency prints

The weekly EIA number gives you a stocks level once a week, after a reporting lag, aggregated to PADD. It doesn't tell you where in its working range a given farm sits, and it doesn't tell you how much more room Cushing or a coastal hub has before storage gets tight. If you're trying to call a contango steepening or a run on tankage ahead of a refinery turnaround, size your math off the working-capacity ceiling. That's the constraint that bites, well before shell capacity on the spec sheet ever comes into play.

That's also why floating roof position is a more honest read on inventory than most people treat it. The roof rises and falls with the fill level inside working capacity, which means a day-by-day change in roof height tracks draws and builds independent of whatever the next agency release says. We built a daily shadow-based read of floating roof levels across covered tank farms for exactly this reason, as a way to see the draw or build happening in real time rather than waiting for Wednesday's print. If you want a sense of what that looks like at the facility level, the Oil Inventory Index landing page walks through the daily CSV series by region and facility.

Next time a capacity table crosses your desk, run the heel and freeboard math before you treat the shell number as anything other than an upper bound nobody actually operates against.