Seeming that no one in the Spanish forum has the knowledge to answer this question I will post it here
To get the resistor scales (E12, E24, E48...) you use the following formula:
10^(i/n)
Where n is the scale number (12,24,48...) and n is a natural number from 0 to n
The problem is that when getting the E12 resistor values, the first ones come ok, but when arriving to higher values like 4.7 Ohms, the formula gives 4.64 so even with rounding you can't get the 4.7 value
The wikipedia article E series of preferred numbers - Wikipedia implies that your formula is true for "Renard numbers" (R5, R10), but that the resistor series has numbers that were derived some other way (no formula given, but there's a Standard that has the list of values.)
It all started with the now very old 20% resistors. The values were chosen so that going from one value to the next one would always result in an increase in resistance even if the first value was 20% up and the second value was 20% down. With such a wide tolerance it makes no sense going for values more precise than two places.
When 10% and then 5% and 1% resistors came in there were opportunities to "fill in the gaps" hence the different series. The formulae is an actual description of the process but doesn't reflect the fact that only two significant places were used. In other words 4.64 is 4.7 when rounded up for two places.
Rounding up so that the precision of the number is expressed by, at the most, two numbers followed by one or more zeros.
In the specific case of 4.64 then the value of 4.6 is too low to prevent the overlap so in order to have a monotonic series the next value has to be rounded up to 4.7. To leave it at 4.6 would be to round it down.
If we want to get the E12 it has a 10% tolerance
From the formula we get 1.78 rounded to 1.8 and the previous one was 1.5.
Now we search for overlapping:
1.51.1=1.65 <- 1.5 plus the 10%
1.80.9=1.62 <- 1.8 less the 10%
I think your problem is that you are trying to apply a formula to the values that were not derived for the formula in the first place.
Have a read of this site:-