Concept model of a transfer chute passing through multiple floors of a tower

Engineering Against a Plant That Moved

The job was a new transfer chute in a coal plant tower built in 1976. The first thing our team asked for was the drawings. The answer came back fast.

“There’s no drawings for that,” Jason Schwartz recalls being told. “And if there’s drawings, they’re all wrong. We added three stories to that. We ripped out the staircase. We tore out the elevator.”

That is brownfield existing conditions engineering in one conversation. The structure you are designing into is not the structure anyone drew. It grew three floors, lost its stair and lost its elevator, and every one of those changes moved loads, openings and access somewhere new. So we started laser scanning the tower to build our own engineering backgrounds, because there was nothing else honest to design against.

The plant was supposed to be done by now

A lot of coal units carrying forty-plus years of changes were planned around a retirement date. Those dates are moving.

In September the Department of Energy ordered two units at NIPSCO’s R.M. Schahfer station and CenterPoint’s F.B. Culley Unit 2 to stay available through December 18, 2026 — nearly a year past planned retirement, and the fourth set of such orders since December 2025, as the Indiana Capital Chronicle reported. A week later DOE issued an order for Craig Unit 1 in Colorado (DOE 202(c) orders). In Missouri, Ameren’s new 20-year plan now runs all four Labadie coal units until 2042; two had been set to retire in 2036 (St. Louis Public Radio).

A plant that was going to be run out now has to be engineered for another decade or more. Conveyors, chutes, towers, piping and supports that were being nursed to a finish line are getting capital work instead. And that work lands on a plant that has been moving for fifty years.

Why record drawings drift

Nobody did anything wrong. That is the part that makes this hard to fix with a records project.

A tower gets more stories. A staircase comes out. An elevator is removed. Each change is made for a real operating reason by competent people, usually under schedule pressure, often during an outage. Very few of them come with a budget line to redraw the original set.

So the drawings stay as they were, and the building keeps changing around them. Twenty or thirty years later, the drawing room holds an accurate picture of a plant that no longer exists.

Schwartz describes what happens when an engineer walks into a plant like that: “My first question as an engineer is, can I have some drawings? And then they would just bust out laughing.”

The laugh is not a joke at anyone’s expense. It is the plant staff telling you, honestly, that they know. And it matches the pattern he sees at the start of nearly every retrofit: “The pain point always starts with a lack of information at the beginning. And they search for it for hours and hours and hours, and they still come up with nothing.”

“Field verify” moves the risk, not the problem

When drawings do exist, they tend to arrive with a condition attached. Jason Adkins has seen it on owner drawing packages for years: “You have to field verify all of these because you cannot use them for design, or if you do, it’s on you.”

That sentence is worth reading slowly. It does not say the drawings are wrong. It says the owner will not stand behind them, and whoever designs from them owns the difference. The note transfers the risk to the engineer without transferring any of the information needed to carry it.

On a plant that is staying online past its planned retirement, that is a lot of risk to inherit from a disclaimer.

What changes in the calculation when the basis is verified

A verified basis does not make the engineering more complicated. It makes the engineering true. On a tower like the 1976 one, the difference shows up in specific places.

The load path. Three added stories put load on columns and connections that the original drawings describe for a shorter building. Checking a new chute and its supports against the original framing confirms that the arithmetic is right for a building that is gone. Checking it against the measured framing tells you what the structure is actually carrying and where the load really goes.

The openings and the envelope. Where a stair or elevator came out, the floor could have been patched, reframed or left open, and the drawing will not say which. The chute geometry, the drop height, the liner access and the support points all depend on what is actually there. Designed from the record set, the chute fits a space that exists only on paper.

The tie-ins. Every connection to existing steel, existing conveyor and existing head pulley structure has a real position, a real elevation and a real orientation. Those are measurable. They are also the dimensions most likely to have moved.

What stays standing during the work. This is the one people miss. Adkins tells the story of a transfer chute job where the crew had conventional demo drawings. “They took out all the chute work, took the head pulley out. Next thing, they’re cutting out all this structure. It’s like, whoa, dude. No — now we’ve created a huge issue, because we’ve got to tie the structure back to the structure in the floor below.”

The demo drawing showed what came out. It did not make clear what had to stay, or why. On a verified basis, the engineering knows which members carry the floor below and the floor above, and the demo and install sequence can say so step by step. That job is why our demo packages became step-by-step 3D sequences — remove the head chute, then the head pulley, then the bearings — with the install run as the same sequence in reverse.

Brownfield existing conditions engineering for a plant with no end date

When a unit had five years left, a lot of plants could live with a drawing that was mostly right. Crews worked around it, the field fixed the gaps and the unit ran to its date.

A unit running to 2042, or held available one emergency order at a time with no clear finish, is a different engineering problem. Every retrofit done now becomes the starting condition for the next one. If this project is designed against the old set, the next engineer inherits that set plus one more undocumented change.

The better sequence is simple to state:

  • Measure the conditions the engineering depends on before design starts — the structure, the openings, the tie-ins and the access.
  • Engineer against those measurements, and say plainly what the design is allowed to assume and what it is not.
  • Plan the demolition and installation from the same basis, so the crew knows what stays as clearly as what goes.
  • Leave the plant with a record that matches the building, so the next retrofit starts from truth instead of from a disclaimer.

We covered the drawing side of this in Nobody Will Sign the P&ID, and the equipment side in why new equipment still lands in an old plant. The tower is the same lesson at structural scale.

The 1976 tower was not unusual. It was a normal coal plant building that had done fifty years of honest work and changed along the way. Plenty of the plants now staying online look a lot like it. The drawings stopped in 1976. The plant did not.

Engineer the retrofit around reality.

→ Industrial retrofit engineering · Scan-driven engineering


Acensium is an industrial engineering firm for complex brownfield retrofits — bulk material handling, high-energy piping, and equipment replacement in operating facilities. Founded in 2003, 1,300+ projects delivered. We connect verified field conditions with practical design, analysis, fabrication control and installation planning.

More insights

View all →
Loading…

Put this into practice

Have a project like this?

Tell us what you're working on. Our engineers will help you scope reality capture, modeling, and design support for your facility.

Discuss your project →