Question about Measurements and sig. figs.

[junior_J]

This is a question about practical physics ... if its posted in the wrong section my sincere apologies !

How do i decide the least count of a measuring device such as an ammeter (analog) ?
I know that the least count is the smallest division of measurement ... but why do people take the measurement to the nearest half division ?
Does it have something to do with the pointer(needle) of the ammeter ?? may be its thickness ... i have the same problem with understanding why a measurement made by a metre rule with a least count of 1mm is quoted as say (1.25+/-0.05)cm .

 

[vanesch]

A good instrument normally has a specification of the precision of the reading, but it for sure isn't better than the last digit (otherwise the manufacturer would have added digits !). Also, most serious manufacturers don't make a fool of themselves by providing totally meaningless digits.

So if you have no specific indication, then guessing that one unit of the last digit is in the ballpark of the error on the measurement is often, with lack of better knowledge, a good idea. But for any serious error study, you should look at the detailled specifications of precision of the instrument, as given by the manufacturer.

You consider that the display of an analog instrument is digitized by the finest grading scale you are using.
If the error is 1 scale jump, then you can model that by +/- 0.5 a scale jump. Remember, in any case you're guessing here, and an order of magnitude for the error is the best you can get this way. If you want to have a better estimate of your error, you should consult the documentation of the instrument.

 

[junior_J]

So if you have no specific indication, then guessing that one unit of the last digit is in the ballpark of the error on the measurement is often, with lack of better knowledge, a good idea.

Im sorry i don't understand ... ! Can u elaborate a bit more . what last digit ??

Ive skimmed through some textbooks and websites ... What i don't understand is why some quote the uncertainty for a regular metre rule as
+/-0.05mm and others as +/-0.1mm ... and what's the difference between the two ? The same thing with the analog ammeters ...

Thank u !