Weighing and Scales FAQs

Yes. There are numerous OIML Recommendations that cover the wide range of content in the scale-relevant sections of NIST Handbook 44. 

Please see a complete crosswalk between NIST Handbook 44 and OIML Recommendations here.

When selecting a scale for a particular application, several aspects need to be considered:

• Suitability.  Most importantly, the scale must be suitable for the product that is being weighed. For example, a scale for weighing fertilizer should be corrosion resistant.
• Accuracy. The accuracy class of the scale must be appropriate for the application. The typical accuracy classification for various weighing applications are provided in Table 7a “Typical Class or Type of Device for Weighing Applications” of NIST Handbook 44, section 2.20, requirement UR.1.1.
• Capacity. The maximum load that can be expected to be weighed should not exceed the capacity of the scale.
• Minimum Load. The minimum load that can be expected should exceed the amounts noted in Table 8 “Recommended Minimum Load” of NIST Handbook 44, section 2.20, requirement UR.3.1. As a result, this minimum load defines the required verification scale interval, e.  For example, if a class III scale will be used to weigh loads of 0.25 lb or more, the e should be no greater than 0.01 lb.  If the load on a class II scale will be no less than 100 g, then the value of e should be no greater than 2 g.
• Usability considerations. Any other desired functionality for the scale’s usability, such as price calculation, network capability, etc., will need to be considered on a case-by case basis.

Scales used for commercial applications must comply with the laws of the state in which it will be used as a legal metrology device. Since all 50 US states have adopted NIST Handbook 44, the most important aspect is that the scale has an active NTEP Certificate of Conformance as proof that it is compliant with NIST Handbook 44. Note that NIST OWM does not recommend or endorse any particular scale types.

Before a scale can be used for commercial applications, it must be inspected and approved by an authorized body of the state it is used in. Click here for a listing of state and territory weights and measures offices and contact information.

Commercial applications are defined in NIST Handbook 44, section 1.10 General Code, G-A.1 as applications in which a weighing instrument is being used or employed:

1. in establishing the size, quantity, extent, area, composition (limited to meat and poultry), constituent values (limited to grain), or measurement of quantities, things, produce, or articles for distribution or consumption, purchased, offered, or submitted for sale, hire, or award;
   
   For example, a scale at the deli in the grocery store.
    
2. when assessing a fee for the use of the equipment to determine a weight or measure;
   
   For example, a vehicle scale at a truck stop allows drivers to check the weight of their trucks.
    
3. in determining the basis of an award using count, weight, or measure; or
   
   For example, the scale used at a fishing contest.
    
4. in computing any basic charge or payment for services rendered on the basis of weight or measure.
   
   For example, a postal scale or the luggage scale at the airport.

The scale division, d, is the actual resolution of a scale (analog or digital). The verification scale division, e, is a measure of the accuracy of the scale and is used to define the scale classification and to determine the tolerances. Every scale (that falls under the Scale Code, section 2.20 of NIST Handbook 44) has a verification scale division, e. But not every scale has an indication (e.g., a balance), and therefore, not every scale has a scale division, d.

As the verification scale division, e, relates to the accuracy of the scale, all tolerances defined in the Scale Code, section 2.20 of NIST Handbook 44, are expressed in e. Some requirements relate to the scale division, d, but those requirements do not define any tolerances.

For scales of accuracy classes III, IIIL, and IIII, (used in most applications), the scale division, d, must be equal to the verification scale division, e (i.e., the resolution of the indication must be approximately the same as the accuracy of the scale). Only scales of class I and II may have a scale division, d, smaller than the verification scale division, e. But that does not mean that the scale is more accurate. To avoid a false sense of accuracy, the digit indicating the smaller scale division, d, must be clearly differentiated from the other digits.

Note:  Item SCL-23.3 on the 2023 NCWM S&T agenda aims to correct inconsistencies regarding the use of the terms “scale division” and “verification scale division” in NIST Handbook 44.

Click here for additional information.

In a weighing operation to determine the net product content, the packaging material of a commodity is called tare. The gross load of the commodity consists of the tare load (packaging) and the net load (product content). On a digital display, these values are rounded to the value of the scale division and may not add up mathematically.

E.g., A commodity consists of a net content of 2.85312 kg, and the packaging material is 0.323 kg. The total (gross) weight of the commodity is 3.17612 kg (Gross = Net + Tare).

But on a digital indication with a 0.01 kg scale division, these values would be rounded and displayed as 2.85 kg, 0.32 kg, and 3.18 kg, respectively. In this case, there is no mathematical agreement of Net, Gross, and Tare, as 2.85 kg + 0.32 kg equals 3.17 kg, not 3.18 kg.

A tare operation on a weighing instrument is an operation to determine a net weight by the scale by taking a tare weight into account during the weighing of a gross load, i.e., during a single weighing operation.

That depends on the configuration.

If a computer calculates the difference between the weight of the loaded truck and the weight of the empty truck, then the scale performs two individual gross weighments, and the calculation of the net load is nothing but a mathematical deduction of two gross weights. The weight values of the two gross loads are produced by the scale and, therefore, subject to the requirements in NIST Handbook 44. The net weight is not determined by the scale and is not subject to the requirements in NIST Handbook 44.

If the weight of the empty truck is stored and then entered into the scale as a stored tare or keyboard tare (in OIML referred to as preset tare) while weighing the loaded truck, then this is considered a tare operation, and all indications (gross, net, and tare) are subject to the requirements in NIST Handbook 44 related to stored or keyboard tare operations.

Please note that even when the receipt or invoice speaks about the net weight, tare weight, and/or gross weight of the vehicle, it does not mean that a tare operation has been performed by the scale involved.

Weights and Measures jurisdictions interpret NIST Handbook 44 as requiring that these values are in mathematical agreement.

• Although there is a widespread belief that mathematical agreement is required, there are no paragraphs in Section 2.20., Scales that require mathematical agreement. S.1.2.1. Digital Indicating Scales, Units includes this exception: “The requirement that the value of the scale division “d” be expressed only as 1, 2, or 5, or a decimal multiple or submultiple of only 1, 2, or 5 does not apply to net weight indications and recorded representations that are calculated from gross and tare weight indications where the scale division “d” of the gross weight is different from the scale division “d” of the tare weight(s) on multi-interval or multiple range scales.“ While this doesn’t require the net indication to be in mathematical agreement (mathematically correct), it does allow it.
• T.N.2. Tolerance Application states that the tolerance shall be applied to the Net value. However, S.1.2.1. allows the Net value to be calculated from the rounded Gross and rounded Tare values. This can lead to a deviation in the Net value that can be as large as the acceptance tolerance, or even more in the case of multi-interval instruments.
• S.1.2.1. allows the scale division of the Net value to be a value other than a submultiple of 1, 2, or 5. This contradicts S.5.3. Multi-Interval and Multiple Range Scales, Scale Division Value, which states that the scale division, d, must equal the verification scale interval, e, while the verification scale interval is always a submultiple of 1, 2, or 5.
• Allowing mathematical agreement conflicts with the requirement in the General Code G‑S.5.2.2. Digital Indication and Representation, which states: ”A digital value coincides with its associated analog value to the nearest minimum graduation.” and “A digital value “rounds off” to the nearest minimum unit that can be indicated or recorded”. Enforcing mathematical agreement between Net, Gross, and Tare introduces a deviation in one of these values, which, therefore, is no longer rounded to the nearest minimum graduation for that value.
• Mathematical agreement conflicts with S.1.1.1. Digital Indicating Elements, which requires Net zero to be within 0.25 e. Calculating the Net value from the rounded Tare and rounded Gross values leads to an offset of the Net zero that can be as much as 0.5 e. 

• Mathematical agreement introduces a deviation in either the Net value, the Gross value, or the Tare value, depending on how the tare operation has been implemented by the manufacturer. This leads to a non-uniform approach to tare operations, which conflicts with the primary goal of NIST Handbook 44: harmonization of specifications, tolerances, and other technical requirements.
• When a device calculates the Net value from the rounded Gross and rounded Tare values, more uncertainty is introduced into the measurement process. The net value should be determined from the measured gross and tare values and rounding of the Net value should then be the last step in the calculation.

• Most scales don’t indicate or record all three values simultaneously
• Test weights are applied in scale intervals when determining the accuracy of tare values during testing (e.g., 2.000 kg or 500 g with a 1 g scale division) are an exact multiple of the scale division, and don’t disclose the effect of rounding on the measurement
• There is no test procedure in HB44 or Pub 14 to determine the accuracy of net zero after activating tare
• Testing in high resolution does not reveal the problem

Possibly. It is conceivable that manufacturers implement tare operations the same way across all models they manufacture. Even if a scale does not indicate all three values, it is very well possible that the Net value is internally calculated from the rounded Gross and rounded Tare values. If so, then the Net zero still has an offset. Without a proper test during inspections or type approval, this issue will remain unnoticed.

No. OIML Recommendations for weighing instruments require that the measured (internal) values of Gross, Net, and Tare are rounded to their nearest value expressed in the applicable scale division when displayed or printed. This may lead to a weighing result where the Gross value does not equal the sum of the Net value and the Tare value.

For example, a Class III scale has a 15 kg capacity and an e = d = 0.005 kg.