Trade Certified Belt Weighers

Belt Weighers can be approved for trade use. This is a two part process where the equipment is initially tested by the approving authority, and then later each application must be certified by representatives of that authority. In Australia, the approving authority is the National Measurement Institute (NMI).

Pattern Approval of Belt Weighers: Initial Testing Phase

Belt weighing equipment is tested in the NMI laboratory. The procedure includes temperature stability tests, electromagnetic Interference testing and testing for various metrological characteristics. These tests are defined in OIML Recommendations R50-1 and R50-2.

Once the equipment has passed the laboratory tests, one field installation is tested by the NMI (in accordance with OIML R50) before final pattern approval is granted.

After this, other examples of the same equipment can be certified in the field under Australian state government regulations. Australian state governments have now all passed legislation to enable a “Uniform Testing Procedure” so that the rules that govern trade approved belt weighers are the same in all states. The rules for certification are now spelled out in NMI V7, a new document which is an expanded and illustrated version of the OIML R50 document.

In each state, installations may be certified by inspectors of the appropriate government department or by other registered certifiers working for private companies. CST's Managing Director Ian Burrell is registered in most states of Australia to carry out these certifications.

The fact that a belt weigher is pattern approved can sometimes be used in a sales situation to give the impression that the pattern approved unit is superior to other non pattern approved equipment on offer. This is not necessarily the case. Although it is possible to purchase two-idler belt weighers certified for trade use as a Class 1 Device, the certification in itself does not ensure that the weigher can deliver the accuracy required. A two-idler belt weigher, though Trade Certified, may not be the best choice for the application. 

It has previously been claimed that a two idler Class 1 belt weigher was equivalent to a four idler, four load cell belt weigher by other manufacturers. This represents a misunderstanding and misapplication of the pattern approval system.

Class 0.5, Class 1 and Class 2 Belt Weighers

There are three classes of trade approved belt weighers, Classes 2, 1 and 0.5 in order of increasing quality. The errors which are permitted with each Class are shown in the table below. The MPE is the Maximum Permissible Error.

For many years, only Class 1.0 and Class 2.0 systems were mandated in Australia. In the last few years, it has been possible to purchase Class 0.5 Belt Weighers which, as shown in the above table, allow trade use belt weighers with a tested accuracy to 0.25% on the day of verification. Given that 0.25% belt scales have been offered on the market for many years, it is strange that Class 0.5 trade use belt weighing was not a legal possibility at an earlier date. It is perhaps an indication of the well earned skepticism with which users have held the accuracy claims of belt weigher manufacturers over many years.

When to use each Class of Trade Approved Belt Weigher

The three classes of trade use belt weighers which are available, come without any particular guidance about when it would be appropriate to use Class 0.5, 1.0 or 2.0. Apparently, trade approved belt weigher users can choose for themselves to trade legally at accuracies of 0.5%, 1.0% or 2.0%.

In some countries (Canada) the law is that only ‘cheap materials’ can be legally traded on a 0.5% certified scale. If the material is not “cheap” it must be traded on a 0.1% certified scale. This is certainly an attempt to provide some guidance about when a more accurate scale is needed; however, CST would like to suggest an alternative perspective.

A million dollars worth of dirt still costs one million dollars. Considering this fact it becomes apparent that the justification for the accuracy of a scale has much more to do with the total value of the material being purchased than to do with the cost in dollars per kilogram of the material. Take for instance, a power station which uses 5 million tonnes of coal per annum, purchased at conservatively $50/tonne. The power station will spend $250 million on coal per year and this amount is probably determined by a trade certified belt weigher.

If some money could be spent to improve the accuracy of measurement by 0.25%, the station could justify spending 0.25% of $250 million, in fact $625,000.00 which should buy a very impressive belt weigher indeed. It can be readily appreciated that when such large sums of money are involved, only the best accuracy will do. Coal might be considered to be a ‘cheap’ material when compared to say copper/gold ore; however, if your business is purchasing enough of any ‘cheap’ material, you still need the best possible accuracy from your weighing systems.

Since most organisations already understand this basic economic fact, most of the trade certified belt weighers CST have recently produced and serviced are Class 0.5 devices.

There are a few requests for Class 1.0 belt weighers, but this seems to be due to a lag in specifications writing. Specifications still in use by engineering companies which specify Class 1.0 belt weighers have not been updated to reflect the new, more accurate Class 0.5 devices. There may also be a misconception that “Class 1.0” means “the best class”, not realising that it is just a statement of the in-service accuracy of the unit.

NMI V7 and Certifying Belt Weighers

The move to enable Uniform Testing Procedures for trade approved weighing systems has encompassed the Australian State Governments to enact appropriate legislation; and the National Measurement Institute (NMI) to produce the NMI V7 document, “Uniform Test Procedures for the Verification, Certification and In service Inspection of Continuous Totalising Automatic Weighing Instruments (Belt Weighers)”.

NMI V7 is based on the OIML Recommendation 50. It should be noted that OIML R50 is a document aimed at the testing of belt weighers for initial pattern approval, and includes “in situ” testing only in the context of being part of pattern approval. So, while NMI V7 is based on OIML R50, its focus is specifically at the on-site certification of new, and the re-certification of existing, belt weighers. Belt weighers may not be certified for trade use unless they are first pattern approved by the NMI, and bear an approval number. Belt Weighers have a number of the form 6/14D/X. The number for CST’s pattern approved product is 6/14D/13.

Testing belt weighers according to the requirements of NMI V7 involves using live material only. Trade certification is not possible using any other simulation of live conditions other than the live loads of material. Belt Scales cannot be trade certified with electronic calibration, static masses or calibration chains. The NMIV7 (and OIML R50) testing regime involves ten individual live load tests, arranged as five pairs of tests at three different flow rates. This extremely thorough testing regime tests (i) with real material, (ii) repeatability and (iii) linearity. Each test must be equal to, or greater than, the “minimum totalized load”.

The Minimum Totalised Load (MTL) must be the greater of either (i)Two percent of the load delivered in one hour at full scale (ii) A Minimum of 800 scale divisions – for Class 0.5 and (iii) Equal to at least one belt revolution , i.e. equal to the amount which would be delivered at full scale flow in the time it takes for one complete belt revolution. The test loads are weighed on a ‘Control Instrument’ such as a bin or a weigh bridge, which must have three times the precision of the belt weigher. Some times the MTL must be scaled up to become a large enough quantity to allow the control instrument to weigh with sufficient accuracy.

One of the beneficial features in OIML R50 and NMI V7 is the fact that the MTL does not need to be at least equal to one belt revolution at full scale, if the totaliser updates only once per belt revolution. This is a boon for trade use belt weighers on long conveyors. CST recently certified a weigher with only 80 tonnes as the MTL, the usual amount of material required for one belt revolution is 300 tonnes but CST was able to use a CST designed feature which allows totaliser updates only once per belt revolution. This advance in technology also allows smaller totaliser increments, so that the only practical determinant of the MTL is the requirement that the MTL be equal to at least 2% of the flow delivered at full scale in one hour.

Overseas Weights and Measures Testing

CST also has experience with overseas trade certification. Our equipment is trade certified in Canada (Measurement Canada) and in the USA where it is certified under the “National Type Evaluation Program” (NTEP) which uses a document called Handbook 44 (HB44) as the equivalent of what we now have in NMIV7.

Australians should be very pleased that we have one of the best trade certification procedures in the world. When compared to HB44, we have to do more tests, but we can use test loads of any size that suit the criteria while HB44 tests must always be with 200 tons.

The main deficiency of the US and Canadian systems is that they only test at one flow rate. This is quite an unwise policy because more recently, as they have tested some of their installations over a range of flow rates as a part of their consideration of moving HB44 more toward OIML R50, they have seen errors at half scale in the order of 2 to 4 times the supposed accuracy of the certification. The technical facts of the matter are that an Australian OIML R50 compliant Class 0.5 system is up to 5 times more accurate that an equivalent HB44 certified 0.25% unit.

The good thing about the NTEP system in the USA is that it employs “permanence testing” as part of its type evaluations. Permanence Testing is, as the name implies, a test of the ability of the system to be as accurate as it was certified to be, in six months time. There is no advantage in certifying an instrument today and sealing it for trade use, if tomorrow it has errors which are outside specification.

CST's Managing Director Ian Burrell has been in discussion with the National Measurement Institute regarding the adoption of Permanence Testing. At present, under the new Unified Testing Procedures, belt weigher installations are to be re-certified every 12 months. The CST proposal is that testing be done before the seals are broken to see if the unit is still within specification, and that these tests be reported to the state authorities or the NMI. The idea is: if a unit does not meet its performance requirement (+/-0.5% for a Class 5 instrument), the period between re-certifications should be halved. An unsatisfactory unit would have its certification period first halved to 6 months, then 3 months, then 45 days, and so on. This course would prevent sub-standard units from remaining in service, as the cost of maintenance would become prohibitive. On the other hand, good units could have their re-certification period stretch to two years if all was well, and it would not be necessary to continue with full certification tests. (OIML Comments and Suggestions)