setra
Weighing Systems Division
   
Division Profile
Products
Resolution
Technology
Applications
 

Summary: When comparing scales it is important to remember that readability means resolution and displayed resolution is really the ratio of capacity to readability. Displayed resolution is a good criteria to use when evaluating scales in addition to sources of weighing errors, ease of use, and other factors. When looking at counting scales, internal resolution will be helpful in choosing the capacity that is best suited to your application as it can be an indicator of the lightest part that can be accurately counted. It is less useful in comparing counting scales of different makes, because manufacturers tend to define it differently. But one thing is quite clear, when using certain parts, counting scales with higher relative internal resolution will deliver higher counting accuracy for a given sample size.

The International Society of Weighing and Measurement (ISWM) offers the following definition of resolution:

RESOLUTION: The smallest change in mechanical input which produces a detectable change in the output signal.

The Scale Manufacturer's Association says, "resolution is the minimum change in the measured variable which produces response [in] the instrument." According to the complete Scaleman's Handbook of Metrology, "we can say that accuracy and resolution are independent concepts with completely different meanings; accuracy referring to a scales indication relative to a nominal weight value; resolution to the minimum effective (readable) weight change, one graduation or increment." The Instrument Society of America has defined resolution as "the least interval between two adjacent discrete details which can be distinguished, one from another."

Dictionaries provide a list of definitions, most of which are entirely unrelated to weighing, however this author has resolved to limit the following discussion to resolution only as it applies to scales.

Scale manufacturers use the term resolution to help describe the performance of the products that they make. They sometimes use it interchangeably with the term "readability". The Scaleman's Handbook speaks of a "minimum effective (readable) weight change, one graduation or increment." Readable here refers to the smallest weight increment presented to the display, and in today's electronic scales the display is almost always digital. However, readability or resolution is by itself insufficient for assessing differences in scales. A scale that has a capacity of 50 lb and is readable to 0.001 lb is considered a better performing scale than a scale that has a capacity of only 5 lb and a readability of 0.001 lb Although the smallest weight increment, or readability, is the same for both scales, it takes a more accurate scale to provide 0.001 lb readability when weighing loads up to 50 pounds than one that is limited to 5 pounds. This is because the larger load is being divided into more increments than the smaller load. There are 50,000 increments of 0.001 lb in the 50 lb capacity scale, and only 5,000 increments of 0.001 lb in the 5 lb capacity scale. To draw a distinction between the performance of two scales having the same readability, manufacturers specify the ratio of the capacity to readability, and call it "displayed resolution". For a scale with a capacity of 50 lb and a readability of 0.001 lb, the ratio is 50 / 0.001, or 50,000 / 1. For a scale with a capacity of 25 lb and a readability of 0.0005 lb, the ratio is 25 / 0.0005, or 50,000 / 1. For a scale with a capacity of 10,000g and a readability of 0.2 g, the ratio is again 50,000 / 1. Despite different capacities, readabilities and units of measures, these three scales have one thing in common, a displayed resolution of 50,000 / 1, often shortened to 1 part in 50,000, or abbreviated even further to 1 in 50,000. Displayed resolution is one measure of scale performance that can be reliably used to compare scales of varying capabilities.

The ISWM definition of resolution refers to a "detectable change in the output signal." In most scales the change is detected visually on the display and gives rise to readability, from which is derived "displayed resolution". Counting scale manufacturer's know that the circuits in their counting scales can detect even smaller changes in mechanical input than is presented to the display. Since counting scales are often required to count parts that weigh less than the readability of the scale that they are weighed upon, efforts to define just how light a part can be counted on a particular scale have given rise to the sometime confusing term "internal counting resolution," or "internal resolution." There is no unanimity among manufacturers when it comes to defining internal resolution. As such, internal resolution is a very poor basis for comparing different brands of scales. One useful way to define it would be to first define resolution in terms of parts. Borrowing from the ISWM definition, resolution pertaining to counting scales is:

  "The weight of the lightest part placed on a scale weighing element which produces a detectable change in the output signal."

Using the digitally displayed count as a measure of the output signal, and considering the detection of a single part as a detectable change in the output signal, then resolution for counting purposes could then be defined as the weight of the lightest part that can be counted in increments of one part. The ratio of the scale's capacity to the weight of the lightest part that can be counted in increments of one part is then the "internal resolution." A 50 lb capacity scale that can count parts as light as 0.0001 lb one part at a time would then have an internal resolution of 50 / 0.0001, or 1 part in 500,000.

In theory, a scale buyer can determine the lightest part that can be reliably counted on any given counting scale by back calculating from the capacity of the scale and its rated internal resolution. This is important in selecting the appropriate capacity for a counting application. Knowing the lightest part that must be counted, the buyer can compare this weight with the ratio of published capacity to published internal resolution.

One would expect a 5000g capacity scale with an internal resolution of 1 part in 500,000 to be able to count parts a light as 0.01g (5000 / 500,000) 1 part at a time, a 12,500g scale should count parts that weigh 0.025g (12,500 / 500,000), a 25,000g scale should count parts as light as 0.05g, (25,000 / 500,000), etc. Knowing a scale's internal resolution should enable a buyer to make the optimal trade-off of capacity, based upon his/her heaviest bin of parts, and resolution, based upon the lightest part that is stocked. Keeping in mind that manufacturers define internal resolution differently, such comparisons are more practically done on an intra-brand, as opposed to an inter-brand basis.

When counting fairly uniform parts the limiting factor for accuracy is almost always the total sample weight. The larger the sample size, the greater the total sample weight, and the better the counting accuracy. With certain items, usually parts that are light relative to the scale's capacity, it becomes impractical to count out a sample by hand that is large enough to provide the accuracy needed. One solution is to use a scale with a high internal resolution. Scales with high internal resolution develop the same counting accuracy as other scales, but require a smaller sample weight to do it. The ability to resolve a much smaller change in mechanical input, i.e., a much lighter part, means that the total sample weight, and therefore the Average Part Weight (APW) computed with the total sample weight, will be less subject to digitizing errors than the APW's computed by scales with lower internal resolution. The APW is divided into the weight of the bulk load placed on the scale to determine the total count. The more accurate the APW, the higher the accuracy of the count. For example if the true average part weight for an item is 0.6714g, a 50,000g capacity scale that has low internal resolution (1 part in 100,000) might calculate the APW as 0.65g when a 10 part sample is used, internally resolving to 0.5g. A scale with five times better internal resolution (1 part in 500,000) would resolve to 0.1g and calculate the APW as 0.67g. The resulting count on the lower internal resolution scale will be 96.8% accurate, while the count on the higher resolution scale will be 99.8% accurate. Therefore, two scales with the same capacity, using the same sample size to count a particular part, can easily produce widely varying results based only upon differences in the internal resolution.

When comparing scales, it is important to remember that readability means resolution, and displayed resolution is really the ratio of capacity to readability. Displayed resolution is a good criteria to use when evaluating scales, in addition to sources of weighing errors, ease of use, and other factors. When looking at counting scales, internal resolution will be helpful in choosing the capacity that is best suited to your application, as it can be an indicator of the lightest part that can be accurately counted. It is less useful in comparing counting scales of different makes because manufacturers tend to define it differently. But one thing is quite clear; when using certain parts, counting scales with higher relative internal resolution will deliver higher counting accuracy for a given sample size.

   
top of page  

|Division Profile| |Products| |Resolution| |Technology| |Applications|

|Site Map| |About Us| |Contact Setra| |Employment| |Home Page|

   
setra                              159 Swanson Road, Boxborough, MA 01719-1304
                               Tel (978) 263-1400 · Toll Free (800) 257-3872
     
   

1998 Setra Systems, Inc. All rights reserved. All names and logos are  trademarks. All other names, logos, and trademarks are property of their respective owners.

Web site design by DreamLight Incorporated: DreamLight.com