Depropanizer Flooding vs. Fouling: The Reboiler Steam Reflex
Your 42-tray depropanizer has run 14 months at 250 psig on a 100 lbmol/h feed containing 40 mol% propane. It makes 40 lbmol/h of 98% propane overhead, and bottoms must stay below 0.5 mol% C3. At 06:00 the column differential pressure climbs from 7 to 13 psi in 20 minutes. Feed rate, feed composition, overhead pressure, and reboiler steam flow are all unchanged. The bottoms sample comes back at 3.1 mol% C3, and the overhead accumulator level is falling even though reflux is on flow control at R = 2.5. The board operator wants to raise reboiler steam to "push the heavies back down." The column is at roughly 82% of its flooding vapor velocity, and the trays are valve trays on 24-in spacing. The thermosiphon reboiler has a fouling history and the last turnaround was 14 months ago.
1. Diagnose the most likely cause of the simultaneous dP rise and C3 breakthrough, and explain what the shape of the dP trend tells you about flooding versus fouling.
Read the worked answer
The onset is too fast for fouling. Fouling develops over weeks and, on a thermosiphon reboiler, shows up first as falling boilup and falling dP, not a 6 psi jump in 20 minutes at constant steam flow. A step change at constant feed, constant pressure, and constant boilup points to a hydraulic limit inside the column: flooding, downcomer backup, or foaming triggered by a feed contaminant.
The separation data confirm it. Flooding degrades trays across the whole column rather than at one end, so light material is carried down and heavy material up. That is exactly the pattern here: 3.1 mol% C3 in the bottoms, and an accumulator level falling because liquid is being entrained overhead and the condenser cannot knock it back.
Distinguish the mechanisms with a boilup test. If dP falls sharply and nonlinearly when you trim 10-15% of the steam, you are on the flooding curve. If dP stays near 13 psi at any boilup, suspect a plugged or damaged downcomer or tray deck and plan an inspection.
2. Quantify the operator's proposal. Distillate is 40 lbmol/h at R = 2.5. What does raising boilup do to vapor load, and what would cutting reflux to R = 2.0 buy?
Read the worked answer
Vapor load is V = (R+1)D = 3.5 x 40 = 140 lbmol/h. At R = 2.0, V = 3.0 x 40 = 120 lbmol/h, a 14% reduction.
Raising steam by the same 14% would push the vapor rate to about 140 x 1.14 = 160 lbmol/h. At a fixed flooding velocity, the approach goes from 82% to 82 x 1.14 = 93.5%. Entrainment normally becomes significant well before 100% of flood, so this does not solve the problem, it guarantees it: more vapor means more liquid carried overhead, a lower effective reflux ratio at the trays, and worse rather than better C3 rejection to the bottoms. The proposal is exactly backwards.
Cutting reflux to R = 2.0 reduces vapor load but also reduces the liquid-to-vapor ratio on the trays, so the bottoms C3 will rise further. It trades a hydraulic problem for a spec violation. The correct move is to turndown feed and boilup together while holding R, which preserves both the hydraulic margin and the separation, then find the contaminant.
3. What single test would separate flooding from a tray or downcomer restriction, and what would you do in the first 30 minutes?
Read the worked answer
Run a controlled boilup turndown. Reduce steam in 5% steps, holding reflux ratio constant, and log dP and overhead temperature after each step settles. Flooding gives a steep, strongly nonlinear response: a small steam cut produces a large dP drop. A mechanical restriction or a plugged downcomer gives a nearly flat dP that barely responds to steam, because the resistance is fixed geometry, not vapor load.
First 30 minutes: cut feed and boilup by 10-15% together, hold R at 2.5, and confirm the accumulator level recovers. Pull a feed sample and check for surfactants, caustic carryover, or a slop stream that could be foaming the trays; foaming mimics flooding and clears when the contaminant is removed. Verify the overhead pressure and reboiler steam flow transmitters are reading true, since a failed steam flow reading would explain the constant-boilup assumption in the first place. Do not open the steam valve. Prepare a bottoms rundown to slops if C3 stays above 1 mol%, and line up an inspection if the boilup test shows a flat dP.