Platform Scale

What is a Platform Scale

A platform scale is an electronic weighing instrument with a flat, rigid deck under which load cells transmit the weight of the object to a measuring circuit. Pressing down on the deck deflects the spring elements inside the load cells by a few micrometres, the bonded strain gauges change resistance in proportion, and a weighing indicator converts the resulting electrical signal into a displayed weight. Because the operator places, reads, and removes the load by hand, a platform scale is formally a non-automatic weighing instrument, the category defined by OIML R 76 and by NIST Handbook 44 in the United States.

Structurally a platform scale has four parts: (1) the deck or platform, a steel or stainless-steel plate stiffened by a channel frame that distributes the load and resists deflection; (2) the load cells, almost always four, mounted at the corners on rocker pins or feet that decouple horizontal forces; (3) the junction box, a summing card that combines the four cell signals and provides corner-trim adjustment; and (4) the weighing indicator, which supplies excitation voltage, amplifies and digitises the millivolt signal, applies the calibration, and drives the display and data outputs. When the deck is small enough to sit on a workbench the same architecture is called a bench scale; when it is large and low enough for a pallet jack it is called a floor scale.

The history of weighing runs from the equal-arm balance of antiquity to the platform mechanism patented by Thaddeus Fairbanks in 1830, whose system of compound levers let a heavy load be balanced against a small poise weight and made the floor-standing platform scale practical for commerce. Mechanical lever-and-poise scales dominated for over a century. The decisive shift came with the bonded foil strain gauge, refined in the 1950s, which allowed force to be turned directly into a precise electrical signal. By the 1970s electronic load-cell platform scales with digital indicators had displaced mechanical mechanisms in industry, and microprocessor indicators in the 1980s added tare, counting, checkweighing, and serial communication.

In scale terms a platform scale spans a very wide range. Bench units start near a 3 kg capacity with 0.1 g readability for parts counting, general platform scales sit around 60 to 600 kg, and heavy floor and pallet scales reach 3,000, 5,000, and 10,000 kg, with truck weighbridges, a separate category, extending to 60,000 kg and beyond. No single deck covers this range, so engineering selection is the act of mapping the load, the deck footprint, the required readability, and the legal-for-trade obligation onto a specific capacity, division, and load-cell rating.

Four metrics determine whether a platform scale fits its duty: maximum capacity (Max), the verification scale interval or division (e or d), the accuracy class, and the ingress protection and material grade of the deck. Together these set both the legal admissibility of a reading and the total cost of ownership. A cheap deck with a coarse division and a painted mild-steel frame is fine for a stockroom, but it cannot be used for trade, will corrode under washdown, and will need replacement long before a stainless legal-for-trade scale specified correctly the first time.

Scale Types and Classification

The platform-scale family is usually divided by deck size, capacity, and how the load arrives at the deck. The three main members, bench, platform, and floor, share the same load-cell architecture, so the labels describe scale and mounting rather than a different operating principle. Choosing the wrong member is the most common selection error: a bench scale overloaded by a dropped pallet, or a floor scale specified where a bench scale would give ten times the readability. The table below sets out the practical envelope of each type.

Bench scales are compact platform scales sized to sit on a workbench, with decks from roughly 200 by 200 mm to 500 by 650 mm and capacities from a few kilograms to about 300 kg. Their advantage is readability: a small deck on small-capacity load cells can resolve 0.1 g to 5 g, which is why bench scales dominate parts counting, packing, and laboratory-adjacent weighing. They are normally placed at working height and loaded by hand, so eccentric-load error is small but still tested for legal-for-trade use.

General platform scales stand on the floor on adjustable feet and carry decks from about 400 by 500 mm up to 1,250 by 1,500 mm, rated roughly 60 to 600 kg. They are the workhorse of production weighing, taking drums, sacks, and cartons that are too large for a bench but lighter than a full pallet. A column or wall bracket usually carries the indicator at eye level. Capacity and division are chosen so the heaviest routine load sits below about 80 percent of Max while keeping useful readability for lighter items.

Floor and pallet scales use large decks, typically 1.0 by 1.0 m to 1.5 by 2.0 m, rated 600 kg to 10,000 kg, and are built to take pallet or forklift loading. They divide into two installation styles. Pit-mounted floor scales sit flush with the floor for drive-on access but require civil work. Low-profile or ramp scales, only 35 to 90 mm tall, install directly on the floor and are loaded over an approach ramp, avoiding a pit entirely. Heavy floor scales use structural-steel channel frames and diamond treadplate decks to resist forklift impact and minimise deflection, which in turn protects corner-load accuracy. Where the deck must reach truck or vehicle scale, the device crosses into the weighbridge category, which uses many load cells and a reinforced concrete or steel platform.

Load-Cell Sensing Technology

The load cell is the heart of every platform scale, and its construction sets the accuracy, capacity, and environmental rating of the whole instrument. Four cell forms dominate industrial platform and floor scales, distinguished by spring-element geometry and by whether the output is left analog or digitised inside the cell. The table below compares the engineering envelope of each form.

The strain-gauge principle underlies almost all of these. Thin foil resistors are bonded to a metal spring element, usually aluminium for bench cells and alloy or stainless steel for heavier cells, and connected in a Wheatstone bridge. The indicator applies a regulated excitation voltage, commonly 5 to 15 V, across the bridge. When load deflects the element, two gauges stretch and two compress, unbalancing the bridge so it outputs a small differential voltage proportional to force. Sensitivity is rated in millivolts per volt of excitation, typically about 2 mV/V at capacity, so a 2 mV/V cell on 10 V excitation produces roughly 20 mV at full scale. That signal is far too small to use directly, which is why the indicator carries a high-gain amplifier and a 24-bit analog-to-digital converter.

Single-point or platform load cells are built so a single cell can carry an off-centre load on a small deck without corner error, using a parallelogram spring element that rejects bending moments. This makes them ideal for bench and small platform scales up to roughly 635 kg, where one cell replaces four and simplifies construction. They are commonly rated OIML C3, meaning 3,000 verification intervals, with higher-accuracy C6 versions for fine counting scales.

Shear-beam and compression cells handle the heavier ranges. Four shear-beam cells at the corners are the standard for floor scales because they are short, rugged, and tolerant of side load. Above roughly 5 t per cell, rocker-column compression cells self-centre under their own load and are used in heavy floor scales and weighbridges. In both cases the four cells feed a junction box that sums their signals and provides corner adjustment, so the total reading is, ideally, independent of where on the deck the load sits.

Digital load cells move the analog-to-digital conversion and a microprocessor inside each cell, outputting a calibrated digital value over a bus such as RS-485. Each cell is individually corner-calibrated in firmware rather than trimmed with shunt resistors, which speeds commissioning and allows per-cell drift, overload, and tilt diagnostics. They cost more per cell but reduce service time on large multi-cell platforms and are increasingly standard on premium floor scales and weighbridges. Whether analog or digital, all of these cells are governed by OIML R 60, the international recommendation for load cells, which defines the class designations C3 and C6 quoted on the cell label.

Accuracy Classes and Legal Metrology

Because a platform scale usually determines the value of a transaction, its accuracy is not a marketing claim but a regulated category. Two frameworks govern most of the world. OIML R 76 is the international recommendation for non-automatic weighing instruments and underlies European and most national approvals. NIST Handbook 44, administered through the NTEP program of the National Conference on Weights and Measures, governs legal-for-trade scales in the United States. Both grade an instrument by the number of verification scale intervals it can resolve, not by a percentage error alone.

The key concept is the verification scale interval, written e, the smallest legally recognised step of the scale, and the number of such intervals n = Max / e. A scale that reads 0 to 600 kg in 0.2 kg steps has n = 3,000 intervals. OIML R 76 defines four accuracy classes by how many intervals they permit; the table below summarises the limits that matter for industrial platform scales.

Class III is the home of nearly all industrial platform and floor scales. It allows between 100 and 10,000 verification intervals and a minimum capacity of 20 e, which is why a typical legal-for-trade floor scale is offered as, for example, 3,000 kg by 1 kg (n = 3,000) or 600 kg by 0.2 kg (n = 3,000). The maximum permissible error scales with the load: for Class III it is plus or minus 0.5 e from zero up to 500 e, plus or minus 1.0 e from 500 e to 2,000 e, and plus or minus 1.5 e from 2,000 e to 10,000 e, measured at initial verification. Doubling the number of intervals roughly doubles the cost of the load cells and indicator, so over-specifying resolution is a common and expensive mistake.

NTEP and Handbook 44 follow the same division logic but issue a separate certificate. An NTEP-approved scale carries a Certificate of Conformance with a CoC number and a marked class. Most platform and floor scales are NTEP Class III with up to 10,000 divisions; very large or vehicle scales fall under Class III L, which permits a larger minimum division for heavy capacities. A distinction worth noting on US spec sheets is between d, the displayed division, and e, the verification division: a legal-for-trade scale must have e equal to or larger than d, and for trade the value of e is what the law recognises.

A practical consequence is that a scale legal for trade in one region is not automatically legal in another. A scale exported between the European Union, the United States, and China commonly needs an OIML or EU type approval, an NTEP CoC, and a national pattern approval, each with its own verification and re-verification cycle. For purely internal weighing, where no transaction depends on the reading, none of this is mandatory and a non-approved deck is acceptable, but the accuracy-class concept still gives a clean way to compare two instruments objectively.

Key Specification Parameters

Reading a scale spec sheet well prevents most field disappointments. A platform scale data sheet may list two dozen lines, but a small set of parameters drives the decision: maximum capacity, verification interval and resolution, accuracy class, eccentric-load error, deck material and IP rating, temperature range, and the indicator output and protocol. Each is explained below.

Maximum capacity (Max) is the largest load the scale may weigh within specification. It should be chosen so the heaviest routine load sits below roughly 80 percent of Max, leaving headroom for the tare container and dynamic impact. Safe overload is typically rated to about 150 percent of Max, the point beyond which permanent zero shift or mechanical damage can occur; heavy floor scales add bumpers and overload stops to protect the cells from dropped pallets.

Resolution appears as two numbers that beginners confuse. The display division d is what the screen shows, while the verification interval e is the legally meaningful step. A non-trade scale may have a fine d for convenience, but a legal-for-trade scale must satisfy e greater than or equal to d. The internal resolution, the number of A/D counts behind the scenes, is usually much finer than e, which gives the indicator room to apply tare and motion detection without losing a legal count.

Eccentric load error, also called corner or shift error, is the change in reading when the same load is moved from the centre of the deck to a corner. It is the single most important accuracy parameter unique to platform scales, because the load rarely sits in the centre. OIML R 76 and Handbook 44 verify it with the eccentricity test: a load of about one third of capacity is placed at the centre and at four off-centre positions, and every reading must stay within the maximum permissible error. It is corrected at commissioning by corner adjustment in the junction box or in digital-cell firmware.

Material and ingress protection determine durability and hygiene. The table below maps cleaning regime to the practical minimum specification.

Temperature range on legal-for-trade scales is commonly specified from minus 10 to plus 40 degrees Celsius, the band over which the accuracy class is guaranteed; outdoor or cold-store duty needs an extended-range or heated version. Indicator output and protocol is the interface to the wider system and is the second most common selection oversight after capacity:

• RS-232 / RS-485: the long-standing serial links for printers, remote displays, and PLC reads of the weight string.
• Analog 4-20 mA / 0-10 V: a retransmitted weight value for batching controllers and chart recorders.
• Ethernet TCP/IP and industrial fieldbus: PROFINET, EtherNet/IP, PROFIBUS, and Modbus TCP for direct integration into a PLC or MES.
• USB and wireless: for data logging to a PC, and Wi-Fi or Bluetooth on portable and forklift scales.

Two further parameters round out the sheet. Excitation and cell count tell you how many load cells the indicator can drive, typically four 350-ohm cells in parallel for a platform scale, which bounds how large a deck a given indicator supports. Stabilisation or settling time, the interval before the reading is declared stable, matters on high-throughput packing lines where a one-second settle is the difference between meeting and missing the line rate.

Selection Decision Factors

To turn the preceding five chapters into a specific model, follow the decision sequence below. Most selection mistakes come not from one wrong answer but from deciding a later step before an earlier one is settled, for example fixing on a deck size before confirming the legal-for-trade obligation. These eight steps double as a fixed RFQ template.

1. Capacity and division: Identify the heaviest routine load including its container, set Max so that load sits below 80 percent of capacity, then choose e so the lightest item you must resolve is read with adequate steps. Confirm n = Max / e stays within the 10,000-interval Class III limit.
2. Legal-for-trade requirement: Decide whether the reading settles a transaction. If yes, the scale must carry OIML R 76, NTEP (Handbook 44), or the relevant national approval, which constrains division, class, and verification. If purely internal, a non-approved deck is acceptable and cheaper.
3. Scale type and deck footprint: Choose bench, platform, floor, or low-profile per Chapter 2, sizing the deck so the load is fully supported with margin, and decide pit-mounted versus ramp loading for floor scales.
4. Material and ingress protection: Match the deck and cell construction to the cleaning regime per the Chapter 5 table, from painted mild steel at IP54 to stainless 316L at IP69K for high-pressure washdown.
5. Load cell and accuracy class: Select single-point, shear-beam, compression, or digital cells of the right capacity and OIML R 60 class (C3 or C6), remembering that each higher accuracy tier raises cost.
6. Indicator, output, and functions: Choose the indicator for its protocol (RS-485, Ethernet, fieldbus), its functions (tare, counting, checkweighing, batching), its display visibility, and its own IP rating, which must match the deck.
7. Environment and certification: Confirm operating temperature range, vibration and shock exposure, and any hazardous-area requirement (ATEX or IECEx), plus sanitary design (EHEDG or 3-A) for food and pharma duty.
8. Total cost of ownership: Add purchase price, installation (pit civil work where applicable), periodic calibration with certified weights, statutory verification fees, and the cost of downtime if the scale fails a transaction audit. A correctly specified stainless legal-for-trade scale routinely outlasts two or three under-specified decks.

One dimension teams routinely overlook is serviceability and calibration logistics: the availability of certified test weights up to one third of capacity for the eccentricity test, local accredited verification bodies, spare load cells and junction-box cards, and indicator firmware support. A scale must be re-verified after every relocation because gravitational acceleration varies with latitude and altitude and changes the span, so a unit moved between cities or floors needs recalibration even if nothing else changed. Mettler Toledo, Rice Lake, Minebea Intec, Avery Weigh-Tronix, and Ohaus all maintain calibration and spare-parts networks, which is why they remain the default choice for legal-for-trade installations where audit failures carry real cost.