We get calls every week that start with some version of: what torque should I use on a 6 inch Class 300 flange? The honest answer is that the question is incomplete. Bolt torque is a means to an end. The end is a leak-tight joint at operating conditions. Torque is one of the inputs that determines whether you get there, and on its own it tells you almost nothing.
The right question is: how do I produce enough clamp load on the gasket to keep the joint sealed across the full range of pressures, temperatures, and cycles the joint will see, given the bolt material, the lubricant, the gasket type, and the assembly procedure? That question has an answer. It just is not a single number.
The Clamp Load Equation
The standard relationship between applied torque and resulting bolt tension is T = K * F * D, where T is torque, K is the nut factor (a dimensionless friction term), F is the bolt preload force, and D is the nominal bolt diameter. The equation is approximate, but it captures what matters.
Torque is what the wrench applies. Bolt tension is what actually clamps the gasket. The K factor is the bridge between the two, and it is where most field problems live.
If you rearrange to solve for F, you get F = T / (K * D). A higher K means more of your applied torque is lost to friction in the threads and under the nut face, leaving less clamp load on the joint. Cut K in half, and you double the clamp load for the same applied torque. That is not a small effect.
The K Factor
K is the nut factor, and it is essentially a friction coefficient that bundles thread friction, bearing-face friction, and a geometric term. Typical values:
K = 0.20 for unlubricated, dry installation on plain carbon steel studs.
K = 0.16 for studs with a basic anti-seize compound.
K = 0.12 to 0.15 for studs with a high-quality lubricant like Never-Seez or molybdenum disulfide.
K = 0.10 or lower for studs with specialty lubricants designed for high-temperature service.
Going from K = 0.20 dry to K = 0.12 lubricated changes the bolt tension by 67% at the same applied torque. That is the difference between a joint that seats the gasket and one that does not. It is also the difference between a joint at 30% of bolt yield and one at 50% of bolt yield, which is a different problem at the other end of the curve.
Lubrication needs to be applied to both the threads and the bearing surface under the nut. Just lubing the threads leaves the nut-face friction high, which throws the K calculation off and produces inconsistent results across the bolt pattern. Field practice is to apply lubricant to the stud threads, the nut threads, the nut face that bears on the washer or flange, and the washer face that bears on the flange. All four surfaces matter.
Stud Material And Yield
The torque target depends on what stress level you are trying to put the bolt at, and the stress level you can safely target depends on the bolt material.
A193 Grade B7 is the workhorse stud material for general process service. Yield strength is 105,000 psi for sizes through 2-1/2 inch, dropping to 95,000 psi above that and to 75,000 psi above 4 inch. B7M is the lower-strength variant qualified for sour service per NACE MR0175, with yield at 80,000 psi.
A193 Grade B8 Class 1 is solution-annealed stainless (Type 304) at 30,000 psi yield. B8 Class 2 is strain-hardened to 100,000 psi yield in smaller sizes. B8M is the molybdenum-bearing equivalent (Type 316) for corrosive service.
A193 Grade B16 is a chromium-molybdenum-vanadium alloy for high-temperature service, with yield similar to B7 at room temperature but better retention of strength at elevated temperatures.
The yield number sets the ceiling. Standard practice is to torque to 50% of bolt yield for general service. Critical service may push to 60% or 70%. Going above 70% is rare because the margin against yielding under thermal expansion or operational transients gets too thin.
Target Stress
For B7 studs at 105 ksi yield, 50% of yield is 52,500 psi. The bolt tension that produces that stress depends on the bolt root area. For a 7/8 inch B7 stud, the tensile stress area is about 0.462 square inches, giving a target tension of roughly 24,250 pounds.
From T = K * F * D, with K = 0.15 (lightly lubricated), F = 24,250 lb, D = 0.875 inch, the target torque is 0.15 * 24,250 * 0.875 = 3,183 in-lb, or about 265 ft-lb.
Run the same calculation with K = 0.20 (dry) and you get 354 ft-lb for the same target tension. With K = 0.12 (well-lubricated), you get 212 ft-lb. Same target stress, 67% spread in applied torque depending on the lubricant condition. That is why a torque value pulled out of a table without a stated K factor is essentially meaningless.
Gasket Selection
Different gaskets need different seating stresses. The minimum seating stress (Y in ASME Section VIII Division 1 Appendix 2 nomenclature) and the maintenance factor (m) vary significantly across gasket types.
Compressed non-asbestos sheet (like Garlock 3000 or equivalent) seats at around 3,700 psi minimum, with m around 2.0 to 2.5. Easy to seat, but limited in pressure and temperature capability.
Spiral wound gaskets with graphite filler seat at around 10,000 psi minimum, with m of 3.0. They are the workhorse for Class 150 through 600 process service.
Ring-type joint gaskets seat by metal-on-metal contact in the ring groove. The required seating force is high (typically 20,000 to 30,000 psi at the ring), but the gasket then holds essentially regardless of internal pressure up to its rated limit.
The total bolt force has to exceed the gasket seating force plus the hydrostatic end load with margin. On a 6 inch Class 300 flange with a spiral wound gasket, the required bolt load is on the order of 50,000 pounds total distributed across 8 studs. On the same flange with an RTJ ring, the required seating load is closer to 80,000 pounds. Same flange, same studs, different torque target.
Assembly Sequence
Even with the right torque target, getting there in the wrong sequence produces uneven gasket compression and joint leakage.
Standard sequence is a cross pattern (sometimes called a star pattern), with bolts tightened in opposing pairs across the flange. For an 8-bolt pattern, that means 1-5-3-7-2-6-4-8 (numbering opposite bolts as pairs).
Four passes is the field standard. First pass to 30% of target torque, all bolts in the cross pattern. Second pass to 60%. Third pass to 100%. Fourth pass at 100%, this time circumferentially in order (not cross pattern) to verify that no bolt is loose.
Pause between passes is important. The gasket relaxes between passes as it compresses. Without the pause, the early-tightened bolts shed load as later bolts compress the gasket further. The four-pass procedure with pauses gives the gasket time to settle and produces more uniform load across the pattern.
ASME PCC-1 (Guidelines for Pressure Boundary Bolted Flange Joint Assembly) is the document to reference for formal procedures. It covers everything from joint inspection through final torque verification.
Temperature Effects
Bolt stress relaxes at elevated temperature. Carbon steel bolts at 700 degrees F can lose 20% to 40% of their initial preload over the first 24 to 72 hours of service. Chromium-molybdenum alloys like B16 retain preload better but still relax.
Differential thermal expansion compounds the problem. A B7 stud and an A105 flange have similar thermal expansion coefficients, so the differential is small. A B8 stainless stud on a carbon steel flange has a significantly higher expansion coefficient in the stud, which means the stud lengthens faster than the flange thickness grows. The result is increased clamp load on heat-up, sometimes pushing the bolt past yield if the cold torque was already aggressive.
Re-torque after the first heat-up is industry practice for high-temperature service. The procedure: assemble cold to the target torque, bring the system up to operating temperature, let it stabilize, shut down, let it cool to a safe inspection temperature (typically below 200 F), and re-torque to the original target. The re-torque catches the relaxation that occurred during the first thermal cycle.
When Not To Use A Torque Wrench
Above about 1-1/4 inch bolt diameter, the torque required to reach target preload exceeds what a hand-operated wrench can comfortably deliver. Hydraulic torque wrenches help up to a point. Beyond that, the right tool is a hydraulic bolt tensioner.
Tensioners stretch the stud directly with a hydraulic cylinder. They measure preload by hydraulic pressure (force = pressure * piston area), which is more accurate than torque-based methods because it bypasses the K factor entirely. The tensioner stretches the stud, the nut is run down to the flange, and the hydraulic pressure is released. The remaining elastic stretch in the stud holds the preload.
For very critical service (nuclear, large compressor casings, primary pressure vessel closures), ultrasonic bolt elongation measurement is the gold standard. An ultrasonic gauge measures the actual stretch of the bolt as it is tightened, giving a direct readout of preload regardless of friction conditions. It is slow and expensive, but it produces preload accuracy in the single-digit percent range.
Common Torque Mistakes
Over-torquing on the assumption that more is better. Over-torque pushes the bolt past yield, which permanently deforms the stud and reduces its load capacity. The next thermal cycle finds the weakened stud and either snaps it or relaxes the joint.
Under-lubricating, which leaves the K factor high and the bolt tension low. The joint seats apparently fine and leaks under operating pressure.
Using a torque value from a chart without knowing the lubricant condition or the gasket type. The chart value was developed for a specific case that may not match yours.
Skipping the cross pattern. Tightening bolts in order around the flange circumferentially produces a gasket compression that is high near the starting point and low at the finishing point. The gasket leaks at the low-load region.
Re-using studs that have been torqued to yield. Once a stud has yielded, its elastic range is reduced. The next installation cannot reach the original preload reliably.
Pulling It Together
Torque is one input in a system. The system has to deliver enough clamp load to seat the gasket and maintain it through all operational conditions. Getting there requires the right stud material at the right grade, the right gasket for the service, a correctly characterized K factor, a documented assembly procedure with cross pattern and multiple passes, and where appropriate, a re-torque after the first thermal cycle.
Our bolt torque calculator walks through the calculation for standard B7, B7M, B8, and B16 studs across the common sizes and pressure classes, with K factor inputs for dry, lubricated, and heavily lubricated conditions. For deeper background on stud dimensions and material selection, the bolt-dimensions reference on texasflange.com is a good companion piece.
For studs, nuts, gaskets, and the technical support to specify them, call (281) 484-8325 or email sales@texasflange.com. We carry A193 B7, B7M, B8 Class 1 and 2, B8M, and B16 across the common sizes, with NACE-qualified material when sour service applies.