Fastener spec vs joint spec — know which number you're holding
This is the definition that decides whether this tool applies. Every torque figure in marine work is one of two things, and they are not interchangeable. Confusing them is how castings get stripped and gaskets get crushed.
Fastener spec — what this tool computes
A value derived from the bolt itself: diameter, thread pitch, grade or alloy, and whether it's dry or lubricated.
It assumes the fastener is the weakest thing in the joint, and the goal is to preload the bolt to a target share of its own strength.
Any bolt of that size and grade, in a joint strong enough to take it, gets the same number. That's why a chart can produce it.
Joint spec — what the OEM manual publishes
A value chosen for the whole assembly, set by whatever is weaker or more sensitive than the bolt:
• threads in an aluminum casting — the housing fails first, not the bolt
• a gasket, O-ring or seal that must crush a controlled amount
• a bearing that must end up at a specific preload
• sealant or thread locker changing the friction a calculation assumes
• a sequence or pattern that distributes load across a joint
Same bolt, different number — because the bolt isn't what's being protected.
The rule, when both exist: the joint spec wins. Always. Use this tool for general fasteners where the manufacturer publishes nothing and the joint is plain steel into steel. The moment the manual gives a value for that fastener, the manual's value is the value — on drives, gearcases, powerheads, transom assemblies, and anything threading into a casting.
On a sterndrive, assume joint spec by default. We classified every Tier-1 torque value we hold across MerCruiser Alpha One Gen II and Volvo Penta SX/DPS — roughly 35 fasteners. About 94% are joint specs, and not one is a clean fastener-spec match. That isn't an accident of sampling: sterndrive and gearcase housings are aluminum by design, so almost nothing on a drive is a steel-into-steel joint. Preload nuts (pinion, bearing carrier, retainers) aren't clamping anything at all — they set a bearing's running position, which a chart has no concept of.
So on a drive: use the manual, not this tool, for anything drive-side or gearcase-side. This tool's proper role there is accessory hardware, and the honest fallback of "no OEM number exists, give me a reasoned starting point."
Spark plugs are neither — don't use this tool for them. Plug torque is chart-derivable, but from the plug manufacturer's table by thread size and seat type (taper vs. gasket), not from a bolt-grade calculation and not from a bespoke engine joint spec. Use the plug maker's figure or the engine manual's. And the ESI standard: if you pulled a plug on gas, or an injector on diesel, you run a compression test before it goes back in. The access is already paid for — take the reading and record it against engine hours.
The tell. A published value noticeably lower than what this tool calculates for that size and grade is your signal that you're holding a joint spec, not a fastener spec. That gap isn't an error in either number — it's the casting, the gasket or the bearing telling you what it can actually take. It's also exactly the case where reaching for the chart does real damage.
If a published value is higher than the chart figure, stop and re-read before pulling on anything — check the units (lb-in vs lb-ft is a 12× error), the grade, and the lubrication state.
Quick bolt ID — head markings
How these are calculated. Torque = K × D × clamp load, with clamp load = 75% of the grade's minimum proof strength × the thread's tensile-stress area. Friction factor K = 0.20 dry / 0.15 lubricated. Stainless (304/316) values are calibrated to published anti-seize charts. These are accurate general reference values — when the engine or component manufacturer gives a torque spec, that spec wins. Verified against Portland Bolt (Grade 5/8) and stainless torque-tension charts.
ESI Service Tech reference tool · confirm final torque to OEM spec
