Double-Ended Beam Load Cells

Bending strain grows with the distance from the load to the gauges, which is workable where a bolted platform fixes the loading point and a liability at tens of thousands of pounds, where the load arrives through a bearing whose contact point shifts as the structure moves.

Shear strain does not vary along the web, so the reading survives that wandering. Tolerance of an imperfect load path, not raw accuracy, is why heavy weighing is built on shear geometry, though the accuracy is there too, at ±0.03% of rated output for nonlinearity, hysteresis, repeatability and creep after 30 minutes.

The body is alloy steel with a nickel-plated finish for corrosion protection. That is a coating, not a hermetic seal, and the datasheet quotes no IP rating, so outdoors the water management belongs to the pit, the conduit and the junction box.

Temperature is specified twice and the two figures differ. The safe range is -10 °C to +70 °C; the compensated range, over which output and zero each drift no more than 0.002% per °C, is only -10 °C to +40 °C (14 °F to 104 °F). Above that the load cell keeps working and stops being characterised, which matters on a sun-heated deck.

A summing junction box parallels the outputs, and parallel outputs average. Eight load cells under a deck do not produce eight times the signal: at full system capacity the box still delivers the rated 3.0 mV/V, which at the rated 10 V excitation is 30 mV for the whole scale, however many sit under it. Asking the indicator for more divisions than that span supports adds digits, not information.

What does change is the load on the excitation supply. Input impedance is 775 ohm ±20 ohm, so eight in parallel present roughly 97 ohm and four about 194 ohm. Check that against the indicator's minimum bridge resistance, and keep excitation at or below the 15 V maximum. Corner trimming happens in the box, and has to be redone whenever a load cell is replaced.

The store sells 2,500, 3,000, 5,000, 10,000, 20,000, 25,000, 30,000, 50,000 and 75,000 lb. The datasheet series is wider: it is specified from 1,000 lb, and 1,000, 1,500, 2,000 and 15,000 lb appear on the sheet without being stocked. Those are built to order.

Bolting the ends down is correct. Bolt the deck hard to the centre as well, on a structure that moves with temperature, and the load cell stops being a sensor under a structure and becomes a member of it. When the deck expands it cannot slide, so the growth is reacted as a horizontal force at the ends. That force acts at a different height from the shear webs, so it arrives as a moment, and it is not symmetric (one end always grips harder than the other), so the bridge does not cancel it.

The result is not a broken load cell. It is a scale that zeroes first thing and reads high in the afternoon, does not return to zero after a heavy load has sat on it, and cannot be calibrated straight because the error is not constant. Every load cell in it tests fine on the bench.

The load cell is characterised with its supports at defined positions and the load at the centre between them. Move the supports inboard or outboard and the geometry between the reaction points and the gauged webs changes, and so does the output at a given load. Nothing warns you: the scale still calibrates, to a span it was not certified on. Use the mounting hardware designed for the load cell rather than supports fabricated to fit an existing frame, and before accepting the installation check that both end supports are level and coplanar, that neither is twisted relative to its partner, and that the centre bearing sits square.

A self-checking mount handles small movement, not a vehicle braking on the deck or a vessel being knocked, so the structure gets check rods and bumpers to limit travel. Their job is to carry horizontal force at the limit of movement and no vertical load at all, ever.

Set the clearance with the structure on the load cells and loaded, not empty and not on the bench. A bumper that touches all the time is a load path around the load cell, and the share of the weight it steals changes with temperature. It is the second most common reason a heavy scale will not hold calibration.

The connection is four-wire (red and black for excitation, green and white for signal) with no sense lines, so lead resistance sits inside the calibrated system. Never shorten one: in a paralleled installation that changes the load cell's share against its neighbours, and the corner trim will fight you. Order the length that reaches the box with a service loop, and strain-relieve to the foundation rather than to the moving structure.

Outdoors, fit a grounding strap across the mount so welding current and lightning surge have a path that is not the strain gauge bridge. Insulation resistance is specified at ≥5,000 M ohm at 50 V DC, and that is the measurement to repeat when a scale drifts after rain.

By the structure, not by the load. A single-ended beam needs one rigid mounting face and is cheaper to install, so it suits a frame that does not move much. A double-ended beam needs two supports and a centre bearing, and earns that back in capacity for the same body depth and in a structure left free to expand. If the deck is long, outdoors, or carries rolling load, the double-ended answer is the sound one.

Yes. The DBS bolts down at both ends through its mounting holes. Where the structure above it grows and shrinks with temperature, bring the load into the centre through a rocker pin or a link rather than a rigid joint, so that movement is not reacted through the load cell as drift that tracks the weather.

No NTEP certificate is quoted for it. Where the reading has to be a legal weight, the SBS single-ended shear beam is certified under CoC 95-146 from 1,000 through 10,000 lb. Above that range, tell us what the certification has to cover.

Regularly. The series is specified from 1,000 lb, so capacities below and between the nine stocked ones are built to order, and cable length outside the standard 6 m, 10 m and 20 m is routine. The DBS is also supplied with its mounting hardware as a weighing module kit.