There are many, many examples where the nominal value of a parameter is the minimum, or the maximum expected value of that parameter. Insulation properties of materials, leakage currents, current transfer ratios of transistors, diameter of threaded fasteners, etc, etc. Even in your list, the average weight of, say, a bag of sugar is usually greater than the nominal weight.
That's true, and I have already discussed that issue, so maybe I should not have written quite what I just did. As I've said, there are many situations in which interest is 'asymetrical' - as in the examples you quote, including bags of sugar. However, in such situations it is common to quote a minimum or maximum figure (or a range), without or without a 'tolerance', rather than a 'nominal' one.
In terms of bags of sugar etc., I think that, in practice, what I wrote is very close to being true. Although the requirement is that the average weight of a bag of sugar "e-marked" as, say 1 kg, must be 'no less than 1 kg', in practice the average will be very close to 1 kg. If the average were 1.1 kg, it would be foolish of the manufacturer to label it as (and charge for) "1 kg".
The same is at least partially true of characteristics of electronic components such as you mention. Such components (resistors and capacitors etc., as well as semiconductors) are commonly 'graded' on the basis of measured characteristics (e.f. h
fe of a transistor) - so a 'nominal' (more likely stated as 'minimum') hfe of, say, 50 does not mean '50 or greater' but, rather, 50-75 or 50-100 - since beyond those ranges it would be assigned a different nominal/minimum value.
Whatever, these discussions about detail are a bit irrelevant. Not only does the 230V 'nominal' electricity supply voltage in the UK not correspond to the 'expected'/'intended'/'observed' voltage, but
nor does it correspond to the minimum, maximum, any sort of average (mean, median, mode or whatever) or any other summary statistic of the voltages which are actually supplied. I therefore think that I
do challenge you to come up with other examples in which a declared 'nominal' value has so very little (essentially no) relationship to any numerical feature of the observable quantity concerned.
Let's face it, it is surely the
range of permitted voltages which actually matters. Would it make any difference if, rather than 230V ±10%, the 'target' figure was stated as 240V -1
4%/+5.5%? Both represent (approximately) 2
07V - 253V, and any voltage within that range would be acceptable - so what is the significance in one having a 'nominal' value of 230V and the other a nominal value of 240V?
Kind Regards, John
edit: arithmetic corrected!