Span Specification Justification

[Xrayvissle]

When validating specifications on electronic equipment thats has a span what's the best route? Will the span double because of positive/negative values (i. e.  -10 to +10 = 20 for span) or will it be absolute value of (i. e.  +/-10 = 10 for span).  Span for phys/D and analog meter readings are of course cake, just getting mixed vibes on this, and want to see what the rest of the PMTMDE field thinks.

[scottbp]

It doubles. Say your meter reads 10 volts plus or minus 0.002 volts. The acceptance limits on your data sheet should be 9.998 for the low limit, and 10.002 for the high limit. That's a span of 0.004 volts. Hope this helps. 

[msrichmond]

Agree with scottbp; you want to use the absolute value for span, not the algebraic value.

[John Treekiller]

I think the original question was along the line:  If the meter reads -10 to +10 Volts, and has a specification of lets say +/-0.3% of span.  Do you use the span of 20 or the value 10 to calculate the limits?  I have run into this problem recently, but at least one of the manufacturer's specifications did indicate that it was of the Full Scale span.  Sometimes we just have to contact the manufacturer and ask their intent with the specification.  As a default, I would use the full span of 20, unless someone could prove otherwise.

[Hawaii596]

That's what I do.  I agree with what I think is alluded, that it is a confusing and ambiguous issue.  SPAN, to me, represents "End-to-end" value.  If it is an analogue meter with a -10 > 0 > +10 (let's say, DC Volts), then the span would be from -10 to +10, for a total span of 20.