- Flange leaks are more often caused by incorrect bolting practice than by gasket failure alone, and the tightening sequence used matters as much as the final torque value applied to each bolt.
- A star or crisscross tightening pattern, applied in multiple passes at increasing torque, distributes gasket load evenly around the flange face, while a sequential around-the-clock pattern reliably produces uneven compression and localized leak paths.
- Bolt lubrication changes the relationship between applied torque and actual achieved bolt tension dramatically, and using a torque value calculated for one lubricant with a different or absent lubricant on the actual bolts routinely results in significantly under- or over-tensioned joints.
Torque Is a Proxy, Not the Real Target
The actual engineering objective of flange bolting is achieving a specific, even clamping force, or bolt tension, that compresses the gasket enough to seal without exceeding the gasket’s or flange’s mechanical limits. Torque is simply a convenient, measurable proxy for that tension, related to it through a friction-dependent relationship that varies with bolt condition, thread lubrication, and surface finish. Two identical bolts torqued to the same value under different friction conditions can end up with meaningfully different actual tension, which is why torque specifications are only as reliable as the lubrication and surface condition assumptions they were calculated against.
Why Tightening Pattern Matters as Much as Torque Value
Tightening bolts in simple sequential order around the flange, sometimes called a clockwise or around-the-clock pattern, causes the gasket to compress unevenly: the first bolts tightened compress their local area of gasket, which then relaxes slightly as adjacent bolts are tightened and the flange faces flex, and by the time the pattern reaches the last few bolts, earlier bolts may no longer be at their intended tension. A star or crisscross pattern, moving to roughly opposite points around the flange with each successive bolt, distributes this flexing effect much more evenly, and performing the full pattern in multiple passes, each at a higher percentage of final target torque, further evens out the compression before any single bolt reaches full target load.
Multi-Pass Tightening Procedure
A typical multi-pass procedure runs the star pattern at roughly 30 percent of final torque, then again at 60 percent, then at 100 percent, and finally a full circular pass at 100 percent to catch any bolt that relaxed slightly during the tightening of its neighbors. Skipping straight to full torque in a single pass, even using a correct star pattern, tends to leave more variation in final bolt tension than the multi-pass approach, because the flange and gasket haven’t had the chance to settle into their final compressed state before each bolt reaches its target load. For critical, high-consequence joints, some specifications call for a final verification pass to confirm every bolt is still at or above target torque after the sequence is complete.
Lubrication and the Torque-Tension Relationship
The friction coefficient between the bolt threads and nut, and between the nut face and the flange or washer, absorbs a large fraction of applied torque as friction rather than converting it into useful bolt tension, often 80 to 90 percent of the applied torque is consumed by friction rather than stretching the bolt. Anti-seize compounds, molybdenum disulfide lubricants, and even simple oil all reduce this friction coefficient differently, meaning the same torque value applied to a well-lubricated bolt produces meaningfully higher tension than the identical torque applied to a dry or lightly lubricated bolt. Torque specifications should always state the assumed lubrication condition, and field crews need to use exactly that lubricant, in exactly the way the torque value was calculated for, or the specified torque number stops meaning what the engineer intended.
Gasket Type and Compression Requirements
Different gasket materials have distinct minimum and maximum compressive stress requirements to achieve a seal without either failing to seat properly or crushing the gasket beyond its useful compression range. Spiral-wound gaskets typically need higher seating stress than compressed fiber or elastomeric gaskets, and this required seating stress, combined with the gasket contact area, is what actually drives the torque calculation in the first place. Substituting a different gasket type at the same torque value without recalculating for its different compression requirements is a common and avoidable source of both under-compressed leaking joints and over-compressed crushed gaskets.
Bolt Condition and Reuse
Reusing bolts across multiple maintenance cycles is common practice for cost reasons, but bolt condition degrades: thread damage, corrosion, and residual lubricant contamination all change the effective friction coefficient in ways that are hard to predict without inspection. Bolts that have been overtightened in a previous cycle can also have exceeded their elastic limit, meaning they no longer return to their original length when unloaded, which changes their tension response the next time they’re torqued. Visual inspection of thread condition, and replacement of any bolt showing visible damage, thread deformation, or corrosion, is a cheap insurance policy against joints that don’t behave as the torque calculation assumes.
Bolted Joint Assembly for Critical Service
For high-consequence services, hydrogen, high-pressure steam, and highly toxic or flammable process fluids among them, many organizations now specify tensioning methods beyond simple torque wrenches, such as hydraulic bolt tensioners, which directly stretch the bolt to a target elongation rather than relying on the torque-friction relationship at all. Where torque wrenches remain the method of choice, calibrated, regularly verified torque wrenches, documented lubrication procedures, and a documented multi-pass star pattern executed by trained personnel meaningfully close the gap between simple torque wrench tightening and the more expensive tensioning methods, at a fraction of the cost.
Field Checklist
Confirm the torque specification matches the actual gasket type and lubrication being used, inspect bolt and thread condition before reuse, follow a star or crisscross pattern executed in at least three ascending-torque passes, and perform a final full-circle verification pass. None of these steps requires expensive equipment, but skipping any one of them is a well-documented path to a flange leak that a correct torque value alone was never going to prevent.
