Gear Operator Ratio: Calculating Valve Stem Torque

A differential pressure torque test on a 12" Class 900 slab gate valve at 2160 psig produces one number that matters: the torque required at the stem to break the gate away from the seat with full DP across it. Everything measured at the gear operator handwheel is that number divided by the gearbox ratio and multiplied by the gearbox efficiency. Get either factor wrong and the reported stem torque is off by a factor of two or more.

Ratio Shortcuts That Produce the Wrong Number

Counting handwheel revolutions from fully closed to fully open and calling that the gear ratio is the most common substitution, and it is not a ratio at all. That count folds three independent quantities together: the gearbox reduction, the stem thread lead, and the gate travel. On a rising-stem gate valve the relationship is N = i × (Δh / L), where N is handwheel turns, i is the gearbox ratio, Δh is stem travel and L is stem thread lead. A full-bore 12" gate strokes on the order of the gate height with overtravel, and the stem lead is a fraction of an inch, so a stroke count in the hundreds of turns is normal on a gearbox whose true reduction is a small fraction of that. Using N as i can overstate the reduction by an order of magnitude.

Taking a torque multiplication factor from a gear operator catalog is closer but still wrong for this test. Catalog factors are published at rated input or rated rim pull and at running conditions. Breakaway on a pressurized slab gate is a near-static stiction event, and worm gearset efficiency at zero speed under high thrust is lower than the running figure. Applying a running factor to a breakaway measurement understates stem torque.

The handwheel rim pull specification is a design limit, not an instrument. It tells you what the manufacturer expects a person to apply, and reverse-calculating output torque from it describes an assumed operator, not the valve under test.

Stripping the gear operator off and putting the digital torque wrench directly on the valve stem is the correct measurement plane, and it is also the one that hurts people. With 2160 psig on one side, the stem carries the unbalanced pressure load through the stem nut and thrust bearing that live inside the gearbox. Do not remove the operator and leave the stem restrained only by packing and a hand wrench.

What the Turn Count Actually Encodes

Rearranging the stroke relation turns the useless number into a usable one. Measure stem rise over a counted number of handwheel turns and the ratio falls out directly:

i = (N × L) / Δh

All three inputs are measurable on the valve as it sits. N is counted at the handwheel. Δh is read with a dial indicator or scale against the rising stem or the stem protector. L is the thread lead on the stem, which is the axial pitch multiplied by the number of thread starts. Measure the pitch on the exposed thread and count the starts at the thread end face; multi-start ACME stems are standard on valves this size and a single-start assumption is a frequent source of error.

Turn enough of the stroke to make the measurement meaningful. Twenty handwheel turns against a rise you can resolve to 0.01 in gives a ratio good to better than one percent. Two turns does not.

Quantities, Limits and Where to Read Them

Quantity	Value or typical range	Where to read it
Gearbox ratio i	Ratio of input to output revolutions	Gearbox nameplate; or measured as N·L/Δh; or worm wheel teeth ÷ worm starts
Worm set efficiency η	Below 0.5 for any self-locking worm set; lower at breakaway	Gear operator data sheet, or bench calibration at test load
Stem thread lead L	Axial pitch × number of starts	Measured on exposed stem thread; valve GA drawing
Stem travel Δh	Closed to open, plus overtravel	Dial indicator on stem
Test differential pressure	2160 psig, one side pressurized	Test procedure / valve MOP
Seat mean seal diameter d_s	Not the nominal bore	Valve general arrangement drawing
Seat friction coefficient μ_s	Greased metal-to-metal, typically 0.10–0.25	Valve manufacturer qualification data
Maximum allowable stem torque (MAST)	Valve-specific	Valve manufacturer datasheet
Maximum gearbox input torque	Operator-specific	Gear operator nameplate or catalog

Measuring the Ratio and Converting to Stem Torque

1. Read the gearbox nameplate first. If a ratio is stamped, record it and still verify it by measurement — replacement operators get fitted without nameplate changes.
2. Mark the handwheel and set a dial indicator against the rising stem or stem coupling with the valve unpressurized.
3. Turn the handwheel a counted N revolutions in the opening direction. Record Δh.
4. Measure the stem thread pitch and count the starts to get L. Compute i = N·L/Δh.
5. Cross-check by opening the gearbox inspection cover and counting worm wheel teeth. For a single-start worm, ratio equals the tooth count; for a two-start worm, half of it. The two methods should agree within the resolution of your rise measurement.
6. Establish η for the actual gearbox rather than assuming it. Apply a known torque to the gearbox output shaft on a bench, measure input torque with the same digital wrench used for the test, and compute η = T_out / (i · T_in) at a load near the expected breakaway.
7. Convert the field measurement: T_stem = T_handwheel × i × η.

Define the measurement plane before quoting a result. On a rising-stem gate valve the stem nut and its thrust bearing usually sit inside the gear operator, so the gearbox output torque is what turns the stem nut against thread friction, thrust bearing friction and packing drag. That is a different quantity from the torque an actuator manufacturer will ask for. State which one you are reporting.

Verifying the Result Against the Physics

The measured number has to agree with the load that produces it. On a slab gate, DP pushes the gate against the downstream seat, and the resisting force is friction across that seat contact:

F_seat = P × (π/4) × d_s²
F_friction = μ_s × F_seat

Using the 12" nominal bore as a stand-in until d_s is read off the drawing, the area is 113 in² and the gate load at 2160 psig is roughly 244,000 lbf. At μ_s = 0.20 that is about 49,000 lbf of seat friction, before adding the unbalanced stem load P × (π/4) × d_stem² and packing friction. Convert total thrust to stem torque with the power screw relation using the stem pitch diameter, lead, thread friction coefficient and the ACME half-angle, then add the thrust bearing term. If your gearbox-derived stem torque and this thrust-derived estimate differ by more than the spread in μ, the ratio or the efficiency is wrong.

This is a mechanical energy balance, not a logic problem: the work you put in at the handwheel rim, minus what the worm mesh turns into heat, is the work delivered to the stem. An efficiency below 0.5 is not a defect — it is the condition that makes the gearbox self-locking so the valve does not back-drive under load.

Recurring Traps on Pressurized Torque Tests

Backlash inflates the first reading. Take up lash in the opening direction before zeroing the digital wrench, or the breakaway peak is contaminated by the torque needed to close the mesh.

Ratio is direction-independent but efficiency is not the same number opening as closing, because thrust reverses across the bearing and the seat load path changes. Calibrate in the direction you intend to test.

Check the computed stem torque against the valve's MAST before the pressurized run, and check the handwheel torque you plan to apply against the gearbox input rating. Exceeding MAST during a test twists stems and shears keys; exceeding the input rating destroys the worm.

Efficiency also degrades with grease condition and mesh wear. A gearbox that has sat outdoors will read lower efficiency than its data sheet, which is exactly why the bench calibration is done on the specific unit going onto the valve.

Stop and go to the valve and gear operator manufacturers when the nameplate ratio and the measured ratio disagree, when no efficiency data exists for the operator and you cannot bench-calibrate it, or when the derived stem torque approaches MAST. Ask the valve OEM for the seat seal diameter, stem thread data and MAST in writing, and ask the gear operator OEM for the ratio, efficiency and maximum input torque for that serial number. If the test is being run for acceptance against API 6D or a purchaser specification, confirm the required measurement plane and reporting basis with the certifying party before pressurizing.

Frequently Asked Questions