Fluke Meter Accuracy Explanation Needed

[K1JOS]

I got a Fluke 365 clamp meter as a gift.  I am a bit confused looking at the datasheet, the lowest resolution is 0.1 A and accuracy is 2% +/- 5 digits.  The maximal number of displayed digits is 4.  So what is the range of actual values if I am reading 0.1 A?  Does this mean it is in the range of 0.098 (+/- 0.00001)  to 0.1002 (+/- 0.00001)?   If so, when would the "5 digits" ever be relevant beyond the 2% accuracy limits?

Jerry 

[rsjsouza]

I think you will find good explanation and examples at: http://brianhoskins.uk/digital-multimeter-accuracy/

[retrolefty]

accuracy is 2% +/- 5 digits.

 I'm pretty sure the +/- 5 digit part really means +/- 5 counts of the least significant digit. Accuracy specifications can be quite confusing. Does the 2% part mean 2% of total range or 2% of meaured value?

 It best to turn this over to a true volt nut, of which this forum has several, that may educate us others. Or maybe it won't. 

[Performa01]

Accuracy spec. includes gain and offset error.

Gain error is always in % relative to the measured value.
Offset is an absolute error and can be specified either in % of the range or in digits.

It is of course not really an option to "measure" at the edge of the resolution, but just for fun let's calculate the specified error for 0.1A in the 200.0A range:

Gain error = +/-2% x 0.1A = +/-0.002A;
Offset error = +/-5 counts/digits = +/-0.5A; (since there are +/-2000 counts, it could also be specified as +/-0.25% of the range)
Total error = +/-0.5002A;

As can be seen, the gain error becomes negligible here and any display from -000.5A to +000.5A would still be in spec.

[dacman]

Fluke isn't the only manufacturer to use digits when they meant counts.  If I were testing the meter for accuracy with a 0.1 A input, to be in-tolerance, it would need to display between -0.4 to 0.6.  If the standards were adjusted for a 0.1 A display (on the Fluke meter), then the standards would need to be sourcing between -0.4 and 0.6.