A summary of the Nitrogen+Syngas article on lid, wall and floor failures in primary reformer tunnels.
Four failure mechanisms, each traced with finite element analysis.
And the design rules that follow from them.
Reformer flue gas tunnels fail mainly through thermal stress in the lids, creep and low strength in the chosen refractory, the weight of the brick resting on the insulating layer, and thermal growth that mortared walls cannot absorb. The 2015 Nitrogen+Syngas article by Blasch shows with finite element models that a lid breaks from a temperature difference of a few degrees, not from its span. A reliable tunnel therefore needs thin, light components that each take up their own expansion.
| Publication | Nitrogen+Syngas, issue 336, July and August 2015, pages 59 to 61 |
| Title | Improving flue gas tunnel reliability |
| Authors | Joe Quintiliani, Sales Engineer, and William Russell, Applications Engineer, Blasch Precision Ceramics |
| Topic | Failure modes of brick flue gas tunnels in primary reformers |
Flue gas tunnels sit on the floor of a top-fired steam methane reformer and carry the exhaust out of the radiant box. Most are built from insulating firebrick and firebrick laid in mortar, with half bricks left out as openings and large refractory slabs, often called coffin covers, as lids.
The authors start from the field observation that these tunnels have always caused reliability problems: unplanned shutdowns, lost efficiency, or at least high turnaround costs.
The article separates the failures into four causes and checks each one with a finite element analysis of a typical component at reformer temperatures around 1,900 °F. Static load, thermal gradient, material data and expansion behavior are looked at one by one.
Ten degrees of difference between the top and the bottom of a lid stress it 150 times more than its own weight does.
Thermal stress: a typical extruded lid at a constant 1,900 °F sees a maximum stress of about 10 psi from its own weight. With 1,910 °F on top and 1,900 °F underneath, the stress rises to about 1,500 psi, above the hot modulus of rupture of many refractories. Where many lids of one tunnel broke in the same campaign while the walls stood, this was the most likely cause.
Material: the hot modulus of rupture at service and upset temperatures needs a margin over the static stress, and the creep resistance should be as high as reasonably available. A lid that creeps sags in the middle and changes how it bears on the walls.
Mass: a brick tunnel with a solid lid loads the insulating firebrick below with more than 10 psi. Published ASTM data suggest that this layer deforms by a full percent within 100 hours at reformer temperatures. The furnace floor steel can deform as well. Cutting tunnel weight by 60% reduces the load on that layer by an order of magnitude, and a base component can spread it over up to five times the area.
Expansion: brick walls get fiber expansion joints every 6 to 10 ft, with mortared sections between them acting as one body. In the model, a mortared 10 ft wall section with 1,052 °C at the bottom and 1,038 °C at the top exceeded the strength of standard mortar. The same section built from separate 46 by 23 cm blocks with their own gaps stayed below about 65 psi.
The lid also grows more than the cooler base, by as much as 9.5 mm, and pushes the wall tops apart. The tunnel becomes trapezoidal and buckles more easily. When the whole tunnel grows longer than its joints allow, it snakes sideways, cracks the mortar and separates from lid and base.
The article ends with design rules rather than a product: thin walls, materials with high strength and creep resistance, every block managing its own expansion, no mortar, all parts interconnected, and a base of the same material and size class as the lid so top and bottom grow alike.
If your tunnel lids crack in the middle while the walls stand, look at the temperature history of the furnace before you order thicker slabs. A thicker lid stores more heat and sees larger gradients.
These rules became the StaBlox reformer tunnel system. The follow-up article from 2016 shows how the same design evens out the flue gas flow, and the 2019 field report covers the first two installations step by step. For the wider context of reformer furnaces, see hydrogen and ammonia production.
The article shows that weight alone gives only about 10 psi in a lid at 1,900 °F. A temperature difference of 10 °F between its faces gives about 1,500 psi, so the crack comes from thermal stress.
Mortar turns a wall section into one large body. In the analysis, that body built up more stress than standard mortar can carry, so the joints crack first.
Snaking is the sideways wandering of a tunnel that grows longer than its expansion joints allow. It cracks mortar and separates the walls from lid and base.
The full weight rests on insulating firebrick, which deforms under that load at furnace temperature. A lighter tunnel with a wider base keeps the foundation and the expansion allowances intact.
Tell us where your tunnels crack, lean or collapse, and a Blasch engineer will look at the cause with you.
