4-20 mA current loop

[Fatih]

How are lower (4 mA) and upper (20 mA) boundries determined? I have found that zero offset of the base current (4 mA) makes cable break detection simple. However, I did not understand why it is 4 mA in the first place? I guess it is a optimal value found after some calculations, and I want to understand the logic behind it. Same goes for 20 mA.

Thanks in advance.
Fatih

Edit: I do not want to be argumentative. I mean, I do not say why it is 4 mA or 20 mA, and not something else. What I am asking is not about mere numbers, but why it is preferable over other current values. I hope I made myself a bit more clear. I appreciate any further edit advice for clarifying the question.

[Rerouter]

it may have been a case of chicken and egg, as 1-5V signalling was used around the same time frame, which over 250 ohms becomes 4-20mA

[chibiace]

Four and twenty current loops are pretty popular in industrial 24v setups

[komet]

The lower bound was chosen to be enough to power a reasonable sensor. The upper bound was chosen to be low enough to not cause a spark, so that the interface can be used in explosive environments. Apart from that, standard values like this are never arrived at by calculation, but by compromise and tradition.

[SeanB]

There are other current loops as well, common being the 10-100mA ones sometimes used in process control, and of course the 600R loop used in telephony, which was actually accidentally standardised as it was the resistance of the first earpiece AG Bell wound to use in the experiments.

[4to20Milliamps]

Actually it has more to do with ohm's law than anything else:

http://www.bapihvac.com/content/uploads/2010/07/Designing-4-20-mA-Current-Loops.pdf< section on load resistor

[glatocha]

Probably worth to notice here is that standard allows to go slightly under 4mA and slightly over 20 for underrange and overrange.
In Simens S7 PLC series the analog inputs thouse values are something like ~3 mA and 22mA