At Andy Marine, we believe that practical manufacturing knowledge is one of the most valuable assets built over time.
Inside our workshop, there are engineers, production specialists, and experienced technicians who have spent years working with stainless steel marine hardware — from material selection and casting processes to finishing and quality control.
Many of these insights are learned through daily production experience: understanding why a component fails, how different materials perform in marine environments, and what details truly matter when developing reliable hardware.
However, this practical knowledge is often shared only within the workshop and rarely reaches customers who are making sourcing and engineering decisions.
ENGINEER TALKS is our effort to organize these conversations and share the manufacturing experience behind marine hardware.
Through this series, we discuss materials, production processes, design considerations, and real-world lessons from the perspective of a marine hardware manufacturer.
Today's ENGINEER TALKS Ep.26 is presented by Engineer Li, written by me the editor.
When Resistance Climbs, the Useful Move Is to Stop
A stainless fastener that turns freely for the first part of its travel, and then becomes progressively harder to turn, is telling you something about the joint rather than about the torque you have applied so far. The instinct at the tool is to add force and get the fastener seated. That is the moment when a small extra input can turn a marginal assembly into a seized one — or into a fastener that is found impossible to remove at the next service.
The awkward part of thread galling is that it does not announce itself in an inspection certificate. Stainless fasteners that pass their incoming checks for thread form, material and mechanical properties can still fail to function together as a pair. Galling is not usually a defect sitting on the bench waiting to be measured; it is a condition created during assembly, between two surfaces that were each acceptable on their own.
So the useful response to rising resistance is a decision rather than a number: stop, read the evidence, and then decide whether the joint should be reassembled, reworked or re-specified. Before making that decision it helps to know which of several rather different conditions you are looking at, because they do not all call for the same action.
Four Conditions That Present as "It Will Not Turn"
Galling is often used as a catch-all term for any stainless thread that stops turning. At least four conditions present in similar ways at the tool, and they point in different directions:
| What the tool and the parts show | Check first | What it does not prove | Next step |
|---|---|---|---|
| Resistance rising sharply part-way through tightening, with torn, smeared or transferred metal on the flanks | Alignment, and whether the friction state matches the assembly specification | That the grade is wrong or the fastener is defective | Stop, back the fastener out, and examine both threads before anything else |
| A thread that did not enter cleanly from the first turn, with visibly offset flank contact | Whether the fastener was started square to the hole | That the material is at fault | Remove the fastener, clean the lead threads, and restart to the specified method |
| Grit, debris or dried compound sitting between the threads | Thread cleanliness and the assembly environment | That the joint design or the material is wrong | Clean both threads and reassemble to the specification |
| A fastener that turned freely when installed and will not come apart later, with corrosion products around the joint | Whether the assembly has been in service, and what deposits are present | Which of the two parts is responsible | Treat it as a seized assembly and assess it before applying further force |
These categories overlap in practice. Debris in the threads does not disappear because the parts are also slightly out of alignment, and contamination is frequently a precursor to galling rather than a separate explanation. A category is a direction for the investigation, not a diagnosis, and it does not tell you whether the joint is safe to put back into service.
What Is Actually Happening: Adhesive Wear That Accelerates Itself
Galling is a form of severe adhesive wear. Two metal surfaces in relative motion touch only at their high points — the asperities — and when a thread is tightened, those contact points are simultaneously loaded and made to slide. The chromium oxide film that keeps stainless steel passive is thin and re-forms quickly in air, but sliding shears it off at the contact points and exposes bare metal underneath. Where the contact pressure and the friction heat that accompanies it are sufficient, the exposed metal can weld at a solid phase, without bulk melting.
The consequence is not a gentle rise in friction. The welded junctions are frequently stronger than the parent metal on either side of them, so as the thread continues to turn, the failure happens inside the weaker parent metal: material is torn out of one surface and transferred to the other. The transferred lumps are harder than the surface they came from. They scrape the opposing flank, break down its oxide film, raise the local pressure, and create fresh contact points that can weld in turn. The damage feeds on itself, which is why galling tends to appear as a sudden change at the tool rather than a gradual warning.
Asperities in real contact - only the high points of the flanks touch
Sliding shears the protective oxide film and exposes bare metal
Local pressure plus friction heat welds the exposed metal (solid phase)
The weld is stronger than the parent metal: metal is torn out and transferred
Transferred lumps scrape the opposite flank - rougher, hotter, higher pressure
More bare metal, more welding: the cycle accelerates
Two practical consequences follow. First, where tightening is controlled by torque, a rising coefficient of friction means the torque achieved no longer represents the clamp load that was intended: the joint can look tightened while its preload sits below what the design assumed. Second, in the extreme case the resistance to further rotation can exceed the torsional strength of the external thread, and the fastener fails in torsion instead of reaching its seat. A torque reading that reaches its target is therefore not, by itself, evidence that the joint is correctly preloaded.
Why Austenitic Stainless Threads Are the Usual Setting
The majority of galling complaints involve both members of the pair being 300-series stainless steel, and the reason is not mysterious. Austenitic stainless steels are relatively soft and ductile, their passive film is thin and easy to shear during sliding, and they dissipate frictional heat less readily than carbon steel while generating more friction at the same time. On top of that, materials of the same type and the same hardness are generally more prone to galling than dissimilar ones — and a 316 bolt in a 316 tapped body is exactly that pairing. A finish or coating on one of the members tends to work in the other direction.
Material selection is a legitimate part of the answer, but it is not a lookup. Grades with higher hardness are generally reported to resist galling better, and microstructures with more than one phase, or with dispersed carbides, nitrides or sulphides, are reported to reduce cold welding between sliding surfaces — subject in every case to corrosion resistance being consistent with the service environment, which for marine work is the constraint that usually decides the question. Free-machining grades illustrate the trade: their sulphide inclusions act as a solid lubricant in sliding contact, and their corrosion performance has to be assessed separately.
Grade pairing is sometimes presented as a cure for galling. The commonly quoted example — a 304 bolt against a 316 nut — is not reliable, because the hardness difference between the two is not necessarily large enough to change the outcome. Property class is quoted in the same way, and there the reports disagree: some work indicates that higher-class 80 fasteners resist galling better than 70-class fasteners, while other reports hold that the cold work involved increases the risk. With the published evidence pointing in two directions, neither grade pairing nor property class is a basis for choosing a fastener on galling resistance alone. Where the question is critical to a joint, it belongs in the specification and can be settled by trial on the actual assembly.
The Variables You Can Control
Grade selection is one lever, and usually the least movable one once a product is in production. The variables below sit in the hands of the manufacturing, purchasing and assembly functions, and they are where most of the risk can be taken out.
Thread Quality and Burrs
Roll-formed threads are commonly reported to be less susceptible to galling than machined or cut threads, because the surface is smoother and the grain flow follows the thread form instead of being cut across it — though at least one stainless raw-material source disputes the benefit of cold forming in 300-series material, so this is a reported tendency rather than a settled rule, and it does not replace assessment of the actual failure.
More consistent across sources is the role of damage and geometry. Nicks, dents, seams, material laps and burrs form the high points that begin the cycle, which is why a thread with visible damage, or one that will not pass its go gauge, is a warning sign: the interference adds prevailing torque at the front of the tightening curve and the damaged geometry can start galling nearby. Fine threads are more easily nicked in handling and need more rotation for the same clamp load, which is why they are generally treated as more galling-prone than coarse threads. Surface finish also matters at both ends: finishes that are excessively smooth and finishes that are excessively rough are both reported to be less favourable than a controlled mid-range finish.
Handling and packaging belong to the same conversation. Heavy stainless fasteners dropped or thrown together in a container come out with nicked threads, and a nick is a galling stimulus. Incoming checks that a buyer can put in place are simple: wipe the thread with a cloth and note whether it snags, examine the flank under low-power magnification, confirm cleanliness, and apply the gauge and acceptance criteria that the purchase specification names rather than an improvised standard.
Alignment and How the Fastener Is Started
Mating parts should be lined up as well as the assembly allows before final tightening. Where alignment is poor, flank contact is restricted to two small arcs on opposite sides of the thread, so the same tightening work is concentrated into a fraction of the intended contact area and the local temperature and pressure rise accordingly. Hand-starting the fastener to make sure it has engaged square, before any power tool is used, keeps that condition from being built in at the first turn. Whether a thread may be used to pull a stack of parts together, and how the fastener is to be started, are items for the assembly specification; the reason those items exist is the contact geometry just described.
Assembly Speed and Friction Heat
Most of the work put into a threaded joint goes into overcoming friction rather than into clamp load, and at the flanks that energy is deposited into a small volume of metal. The bulk of the fastener may stay cool while the loaded flank surface gets much hotter than the part as a whole, and higher installation speed raises both the heat generated and the risk of over-tightening. This is the basis of the standard workshop advice to drive stainless fasteners more slowly than carbon steel fasteners of the same size, and to watch the seating torque rather than the clock. Locking features deserve the same attention: prevailing-torque elements add axial load as they engage, which adds a tail of extra work at the front of the tightening curve and changes the friction state of the joint.
Approved Lubrication and Friction Control
Friction control is a design decision, not a consumable choice. Adding a lubricant, or changing a coating or plating, changes the torque-tension relationship of the joint, and the specified tightening torque has to be re-evaluated whenever that happens — a dry-thread torque figure does not transfer to a lubricated thread, and applying a fixed percentage reduction to the original figure is not a substitute for re-determining it.
Which lubricant or coating is acceptable depends on the service medium, the temperature, the materials in contact, galvanic effects and the fastener manufacturer's approval. Two cautions are worth recording for marine work: graphite-bearing compounds can promote galvanic corrosion of stainless steel and also have high-temperature limitations, so their use needs to be justified rather than assumed; and a grease that dries out, cracks or fails to wet the surface can leave a pocket where water collects, which is a crevice condition in its own right. Supplier-applied coatings — wax, PTFE or molybdenum-disulphide systems — are widely used on stainless fasteners and are easier to control in production than lubricant applied at the bench. There is no single anti-seize product that suits every joint, and this article deliberately does not recommend one.
Torque, Friction and Preload
There is no universal tightening-torque table for stainless fasteners, and the reason is structural rather than a gap in the literature. ISO 3506, the stainless fastener property-class standard, gives minimum breaking torque rather than tightening torque; published reference tables for stainless fasteners are calculated on stated assumptions about surface condition and lubrication, and are described by their publishers as a guide rather than a specification, with trials indicated for critical applications. The assembly torque for a given joint therefore has to be established against the fastener, the joint geometry and the friction state that has actually been specified for it — and re-checked when any of those change.
When to Stop and Investigate
- Stop when resistance rises before the parts are face to face. Do not add torque, and do not reach for an impact tool to drive the fastener through the obstruction. The rise in resistance is information about the joint, not an obstacle to be defeated.
- Back the fastener out and look at both threads. Note torn, smeared or transferred metal, grit or debris, dried compound, and corrosion products. Check whether the fastener started square and whether the parts were aligned.
- Sort the finding into the categories above, and keep the overlap in mind. A seized assembly that has been in service is a different case from a fastener that jammed during first assembly, and neither tells you on its own which member of the pair is responsible.
- Treat the damaged parts as damaged. Any fastener or body with galled threads should be assessed against the manufacturer's guidance and by qualified personnel. A thread that has been dressed, filed or otherwise reworked on site should not be returned to a critical load-bearing connection.
- Record what was found, with the conditions. The friction state, the tools and speed used, the alignment condition and the part numbers are what let the next assembly avoid the same result. A galling event without that record tends to be repeated on the next batch.
What to Put in the Purchase and Assembly Specification
Most of the risk described above can be moved from the assembly bay into a document. For stainless hardware with threaded connections, these are the items worth recording:
- Thread form, diameter, pitch and tolerance class, stated for both members of the pair rather than for the bolt alone.
- Material and property class of both members, including the hardness relationship between them where a joint is known to be galling-sensitive.
- Thread quality and acceptance criteria — the gauge or inspection method, the cleanliness requirement, and the burr and damage limits, agreed before the order is placed.
- Packaging and handling requirements that protect finished threads in transit, storage and kitting.
- The intended friction state: dry, plated, waxed or coated, which lubricant or coating is approved, and who applies it.
- The tightening method: drive speed, seating torque, sequence and number of stages as defined by the assembly specification, together with the basis on which the torque-preload relationship was established.
- Traceability, so that a seized assembly can be traced back to a batch and a specification change.
None of this is a specification for a particular supplier. The point is that thread quality, cleanliness and the friction state can all be written down and verified, and that leaving them implied is what converts an ordinary assembly question into a warranty argument.
Discussing Threaded Fastening Requirements
If you are specifying or sourcing stainless marine hardware with threaded connections, send us the component drawing, the assembly it belongs to, and the friction and torque conditions your specification defines. We can review the thread and material requirements against the application and discuss what needs to be recorded on the purchase order so that the joint behaves as intended at assembly and at the next service.



