On a shear beam the sensing element is a web machined into the middle of the body. Bolt one end down, hang the load off the other, and that web is sheared between the two. Gauges sitting on it register the shear, the bridge puts a voltage on it, and a scale factor makes it pounds.
Simple enough as a component. The reason process weighing has a reputation for being harder than it looks is that the sensor is almost never the problem. In a tank, hopper or platform installation, the question that decides your accuracy is not how well the cells measure — it is how much of the load reaches them at all.
Everything solid that connects the vessel to the outside world is a second load path. Pipes, conduit, ladders, handrails, stay rods, guy wires, a cable tray resting on a bracket — each of them can carry a share of the weight around the cells instead of through them. Published vessel weighing guidance states the requirement bluntly: ladders, pipes and check rods should not be allowed to shunt the weight that ought to be resting on the load cells.
And a parallel path is worse than a plain error, because it moves. If a pipe took a fixed fraction of the load, you could calibrate it away and forget about it. It does not. Pipe stiffness changes with temperature and with what is inside it, a ladder carries a person occasionally, a bracket settles, insulation is added. The share of the weight going around your cells varies with the weather and with the shift, which is precisely the kind of error a calibration cannot remove.
Thermal expansion turns a rigid restraint into an enormous force. The numbers are startling to anyone meeting them for the first time. A one inch diameter steel stay rod four feet long, rigidly held at both ends, can develop in the order of nine thousand pounds of force per bracket from a sixty degree Fahrenheit temperature change. A ninety-six inch stainless vessel grows about seventy thousandths of an inch across its diameter over an eighty degree rise. Restrain that movement with something stiff and the force has to go somewhere, and where it goes is into your load cells.
Which is why the mechanical arrangement, not the electronics, sets what you get. Published guidance puts a weight indicator's contribution to system error at around one hundredth of one percent, with the load cells and the mechanical installation accounting for essentially all of the rest. A sensor holding three hundredths of a percent inside a badly piped tank will not give you three hundredths of a percent, and no amount of instrument specification recovers it.
Sensors of this type are built into industrial and platform scales, conveyor belt scales, tank and hopper weighing, silo and bin monitoring, batching and mixing systems, automated filling machinery, material handling equipment and process weighing generally.
The bridge is conventional; the interesting parts of this design are how much signal it produces, how it must be mounted, and what its environmental protection actually claims.
Signal, and there is a great deal of it. Rated output is 3.0 mV/V with a tolerance of one percent, from a 350 ohm bridge, and excitation runs at 10 VDC with 20 VDC the maximum. Run at the top of that range and full scale is around sixty millivolts — roughly three times what a conventional two millivolt per volt sensor at ten volts will give you, and the largest signal of anything in our catalogue. That matters more here than on a bench instrument, because process weighing usually means several cells summed together and long cable runs back to a control room, both of which spend signal. Starting with three times as much is the cheapest resolution you will ever buy.
The output tolerance is tight, and that is worth noticing. One percent on rated output and one percent on zero balance are unusually well controlled for an economical sensor, and they make life considerably easier when cells are used in multiples: units that are closely matched share load more predictably and need less trimming to bring a system into agreement.
Mounting a cantilever shear beam has specific requirements. Published installation guidance is consistent on four points, and all four cost accuracy when ignored. The mounting surface must be flat and level, and the block it bolts to must be thick enough for proper thread engagement. The corner over which the beam cantilevers should be a hardened surface, because a soft edge deforms under the concentrated reaction and changes the effective geometry. Mounting bolts should be grade five or better. And the load must be introduced vertically, through the centreline of the load hole — a pull at an angle is reported as force like any other strain.
The load introduction has to be flexible, because the beam moves. Full scale deflection on this series runs from fifteen to twenty-five thousandths of an inch, which is real movement. Guidance for this class is explicit that the load introduction mechanism must provide enough flexibility to avoid transmitting extraneous forces and to tolerate the unavoidable deflection of the cell itself. A load applied through something rigid enough to resist that deflection is fighting the sensor rather than loading it.
Environmental protection is a defined middle tier. Nickel-plated steel construction, water-resistant potting and a compression-sealed cable entry make this an environmentally protected sensor — suited to normal indoor installations and sheltered outdoor ones. It is not a hermetically sealed cell with a published ingress rating, which is the specification washdown and corrosive service call for. The compensated band also runs wider than our metric weighing products, reaching to a hundred and twenty-two degrees Fahrenheit with a safe range extending considerably further, which suits process equipment that runs warm.
On a vessel or platform system the sensor selection is the quick part. Our application engineers spend most of their time on the arrangement around it, and that is the conversation worth having early.
Work out the load per cell, then add generous headroom. Divide the total by the number of supports, but do not stop there — load never shares evenly across three or four points, and vessels get overfilled. Published guidance recommends choosing cells with capacity fifty to one hundred percent in excess of the calculated nominal load per cell, and at these capacities that margin costs very little compared with what an overloaded cell costs.
Remember that the vessel is most of the weight. On a tank or hopper the empty structure frequently outweighs the contents you actually want to measure, so a large share of your capacity is consumed before anything is added. That is not avoidable, but it must be in the arithmetic, and it is why a system that looks generously specified on paper can end up resolving the contents poorly.
Plan the piping before you plan the scale. This is the single highest-value thing you can do. Use the smallest acceptable pipe diameter, run the longest practical unsupported horizontal length before the connection to the vessel, use flexible connections rather than rigid ones, and support the pipework independently of the weighing structure. Piping designed after the fact is the commonest reason a good weighing system underperforms.
Design restraints to allow movement, not prevent it. Vessels expand and contract, and a stay rod that rigidly holds both ends converts that movement into very large forces. Restraints should be arranged so thermally induced movement is accommodated rather than resisted, and check rods can be fitted loosely where their job is safety rather than location. Where a system uses several cells, they should be level and the structure stiff enough not to redistribute load as it flexes.
Be honest about the environment. Water-resistant potting and a sealed cable entry cover normal indoor and sheltered outdoor service comfortably. Direct washdown, submersion, chemical exposure and corrosive atmospheres are a different specification and worth raising with us at the enquiry rather than after installation.
Then choose the instrumentation to suit a summed system. Multiple cells feeding one reading need summing and trimming so the total is independent of where the load sits. An amplifier signal conditioner module gives a voltage or current output for a controller, while a digital display with setpoints handles batching, filling and level alarms directly — on process equipment the setpoints are usually the point. Cal-Teds plug and play puts calibration data on the sensor itself, which is worth having where cells are replaced individually in a multi-cell system.
Describe the vessel or platform, what it weighs empty, what you need to measure, how many supports it has and what is plumbed into it, and we will work through the capacity and the arrangement with you. Every capacity above is a stocked item, and educational and research orders are discounted — mention it when you enquire.
SBL Series Load Cell Applications.
The Transducer Techniques SBL Series heavy duty shear beam force sensors are ideal for process weighing and for use in low profile industrial scales, and address a wide range of industrial applications.
- Industrial Scales: SBL Series load cells are used in industrial scales of all kinds.
- Platform Scales: These load cells are used in low profile platform scales.
- Conveyor Belt Scales: SBL Series load cells are integrated into conveyor belt scales for in-motion weighing.
- Tank and Hopper Weighing: These load cells are used for tank and hopper weighing applications.
- Silo and Bin Monitoring: SBL Series load cells provide silo and bin level and inventory monitoring.
- Batching and Mixing Systems: These load cells are used in batching and mixing systems.
- Automated Filling Machines: SBL Series load cells are integrated into automated filling machinery.
- Material Handling Equipment: These load cells are used in material handling equipment.
- Process Weighing: SBL Series load cells are used throughout process weighing applications.
- Load Monitoring: These load cells provide load monitoring in industrial installations.
- Heavy-Duty Environments: SBL Series load cells are suited to heavy-duty industrial environments.
The Transducer Techniques SBL Series shear beam force sensor's robust construction, environmental protection, and ease of installation make it a preferred choice in various industries for ensuring accurate and dependable weight measurement and process control.
Frequently Asked Questions
What is a shear beam load cell and how is it mounted?
The sensing element is a web machined into the body of the beam. One end bolts down to a rigid block; the load is applied at the other end through a threaded hole, and the web between them is sheared. Because the reaction is concentrated where the beam cantilevers off its mounting, that corner should be a hardened surface, the mounting face flat and level, and the bolts grade five or better. The load must arrive vertically through the centreline of the load hole — anything applied at an angle is strain like any other and is reported as weight.
What capacities does the SBL Series cover?
Ten, in pounds: 500, 1,000, 2,000, 2,500, 3,000, 4,000, 5,000, 10,000, 15,000 and 20,000 lb. The range is built on three body sizes — the 500 to 4,000 lb models share one, the 5,000 and 10,000 lb another, and the 15,000 and 20,000 lb the largest, with the load hole thread growing from 1/2-20 to 3/4-16 to 1-14 as it goes. Worth checking which body your capacity falls into before designing a mounting block, since the footprint changes with it.
How much signal does the SBL produce?
More than anything else we make. Rated output is 3.0 mV/V against the 2 mV/V that is standard across most load cells, and the excitation ceiling is 20 VDC rather than 10 — so at the top of the range full scale is around sixty millivolts, roughly three times a conventional sensor at conventional excitation. That headroom matters in process weighing specifically, because summing several cells and running long cable back to a control room both cost you signal. Starting with three times as much is the least expensive resolution available to you.
How accurate is the SBL Series?
Nonlinearity is 0.03% of rated output, with hysteresis and nonrepeatability each 0.02%, zero balance within 1% and temperature coefficients of 0.002% per °F on both zero and output. Those are strong figures. The honest caveat for anyone building a vessel system is that the sensor is rarely the limiting factor — published guidance puts a weight indicator's contribution to system error at around a hundredth of a percent and attributes essentially all the rest to the load cells and the mechanical arrangement, with well-installed systems reaching a quarter of a percent and ordinary ones half a percent or worse.
How much does an SBL deflect at full load?
Between fifteen and twenty-five thousandths of an inch depending on the model — considerably more than a precision test load cell, and by design. That movement matters mechanically: installation guidance for this class states that the load introduction must be flexible enough both to avoid transmitting extraneous forces and to tolerate the unavoidable deflection of the cell itself. A load applied through something rigid enough to resist that deflection ends up fighting the sensor, and the reading suffers for it.
How many load cells do I need under a tank or platform?
Three or four is usual, and the choice is structural rather than electrical. Three supports are inherently determinate and share load predictably on an uneven floor; four suit rectangular vessels and platforms but need the structure to be stiff and the mounts level, or the load redistributes as things flex. Whichever you choose, size each cell on the load it will actually see rather than the total divided evenly, because it will not divide evenly. Tell us the vessel, the supports and the floor and we will work through it with you.
What capacity should I choose per cell?
Published vessel weighing guidance recommends selecting cells with capacity fifty to one hundred percent above the calculated nominal load per cell. That allowance covers uneven load sharing between supports, overfilling, and the shock of material dropping into an empty vessel. Remember also that on a tank or hopper the empty structure is often heavier than the contents you want to measure, so the dead weight is consuming capacity before anything useful is added — include it in the arithmetic rather than sizing on the contents alone.
What is the SBL made from, and how well is it protected?
Nickel-plated steel for corrosion resistance, with water-resistant potting and a compression-sealed cable entry. In weighing industry terms that is an environmentally protected cell, suited to normal indoor installations and sheltered outdoor ones. It is a defined middle tier rather than a compromise: above it sits hermetic construction with a published ingress rating, which is what direct washdown, submersion and corrosive atmospheres require. Describe your environment to us and we will say honestly which side of that line it falls.
What temperature range is the SBL compensated over?
14° to 122°F compensated, with a safe range extending to 176°F — wider at both ends than our metric weighing products, which suits process equipment that runs warm and installations in unheated buildings. The coefficients are 0.002% per °F on both zero and output, which are among the tightest we publish. Bear in mind that on a vessel system the temperature effect on your reading is usually dominated by what the tank and its pipework are doing rather than by the sensor.
Can I mix SBL capacities in one weighing system?
Not sensibly. Cells summed into one reading should be the same capacity and ideally closely matched in output, because a system sums millivolts per volt rather than pounds — mixing capacities means each cell contributes a different weight for the same signal, and no amount of corner trimming makes that consistent. The tight one percent tolerance on rated output helps here, since units from this series are well matched to begin with. If you have an unusual arrangement where supports genuinely carry very different shares, talk to us before ordering.
Questions From The Field
My tank system reads low and the error changes with the weather.
That combination points almost certainly at something shunting weight around the cells rather than at the cells themselves. Pipes, conduit, ladders, handrails, guy wires and stay rods can each carry a share of the load, and because their stiffness changes with temperature the share varies — which is exactly why it tracks the weather and why a calibration does not fix it. Walk the vessel and list every solid thing connecting it to the building or the ground, then work through them: flexible connections on pipework, independent pipe supports, restraints arranged to allow movement rather than prevent it.
Can rigid pipework really affect the reading that much?
Yes, and the arithmetic surprises people. A ninety-six inch stainless vessel grows roughly seventy thousandths of an inch across its diameter over an eighty degree Fahrenheit rise. Restrain that with something stiff and the force generated is substantial — published guidance gives the example of a one inch steel stay rod four feet long, rigidly constrained at both ends, developing in the order of nine thousand pounds per bracket from a sixty degree change. Against a load cell rated in the low thousands of pounds, that is not a small perturbation. Flexible connections and restraints that permit movement are not refinements; they are the design.
How should pipework be arranged on a weighed vessel?
Four rules cover most of it. Use the smallest pipe diameter the process allows, since stiffness rises steeply with size. Run the longest practical unsupported horizontal length before the pipe reaches the vessel, so the run itself can flex. Use flexible connections at the vessel rather than rigid flanges. And support the pipework from the building independently of the weighing structure, so its weight and movement are not carried through the tank. Getting this right at the design stage costs nothing; retrofitting it around installed pipework is expensive and often only partly successful.
The reading is unstable at very light loads.
Check what you are asking of the system before suspecting a fault. Weighing guidance advises against attempting to weigh a load smaller than about twenty display divisions — on a scale reading in half pound steps, that means not trusting weighments below roughly ten pounds. On a vessel system the problem compounds, because the empty tank may already be consuming most of the capacity, leaving a small live range spread across a large one. If you need fine resolution on small additions to a heavy vessel, that is a system design conversation rather than a sensor swap.
One corner of my platform reads differently from the others.
Three usual causes, in order. The mounting is not level or not flat, so that cell carries a different share and its beam is pre-stressed before any load arrives. Something is touching or fouling at one corner — a conduit, a guard, debris under the deck. Or the system has not been corner-trimmed, which is a normal commissioning step on any multi-cell arrangement and makes the total independent of where the load sits. Work through those before replacing anything; a genuinely faulty cell usually shows up in its resistance readings rather than as a corner error.
Do I need a hardened plate under the mounting end?
Where the beam cantilevers off its block, yes — published installation guidance calls for a hardened surface at that corner. The reaction there is concentrated over a small area, and a soft mounting block deforms slowly under it. The consequence is not a dramatic failure but a gradual change in the effective geometry of the installation, which shows up as a slow drift in calibration that nobody can account for. Along with that, make sure the block is thick enough to give the bolts proper thread engagement and use grade five fasteners or better.
Can I run the cells at 20 volts to get more signal?
Yes, up to that maximum, and it is a sensible thing to do on a summed system with long cable runs. Output is proportional to excitation, so doubling the supply doubles the signal — taking full scale to around sixty millivolts. Two provisos. Make sure the instrument is actually configured for the voltage it is supplying, or the reading will be out by the ratio. And note that several cells in parallel draw proportionally more current from the supply, so check the instrument or power supply is rated for the total rather than for one cell.
How do I tell whether a cell in a multi-cell system has failed?
Disconnect the suspect cell and measure it on its own. Check resistance across the excitation pair and across the signal pair against the specification, and check insulation resistance between the circuit and the body — a healthy cell reads very high, and a figure that has collapsed indicates moisture reaching the gauges through a damaged cable or a compromised entry seal. Compare the unloaded output against the zero balance figure as well, since a shifted zero that has stayed shifted points to overload. If all of those are sound, the problem is in the installation, the summing arrangement or the wiring rather than in the sensor.