Trackrig wrote:
I believe all scales used for commercial purposes have to be certified by the state they're in so they should be pretty close.
Bill
It depends on the state. It is also important to understand that the term "certified" is nothing more than a piece of paper that says on a certain date, a known weight was applied to the scale and the scale's displayed weight was within the range specified.
The intervals at which the scales are checked varies. But it is usually no more than a year. In many places, it's a much shorter interval.
As someone who works on scale systems, I have found one of the easiest ways to check for a problem is to check for repeat-ability.(more on that later)
The actual device that measures the weight is fairly accurate, it does not usually lose its "calibration". Usually the on;y thing that affects it's readings is large temperature variations. (which is why it is a good idea to have the scales calibrated every quarter)
In a sense, these scales nothing more than mechanical devices designed to transfer the weight applied to the deck to one or more load cells. The load cells put out (actually reflect) a voltage based on the deflection of the strain gauge. Depending on the type of array, the voltage signal from many load cells are combined at the summing board. That signal is then sent to the scale.
As with any mechanical device, it needs maintenance. That is where variations come up.
So the best way to rule out a general problem, is to reweigh. Most accuracy problems with the scales start with some sort of mechanical bind between the scale deck and the load cells. Simply pulling completely off, then pulling back on at a slightly different location on the scale deck will expose binding issues. If there is a bind, you will not get the same reading. If the reading is the same, it is likely a mechanically sound scale.