Probably this standard is the most referred one world wide in drives power quality requirements specs. But are drives really responsible of h < 11? Aren't they very much influenced by the "upstream" grid characteristics? I mean: Of course in filter dimensioning every hi order must be considered. But ¿Aren't drive dimensioning and filter dimensioning not exactly the same thing? ¿are drives somehow over dimensioned in h < 11 current distortion limits?
I will appreciate your comments on this, thank you!

IEEE 519 is probably one of the most misapplied standards. It is a system standard that defines the requirements for harmonics at the point of common coupling - i.e. the point at which your load and others can be connected to the common source. This point can be on the low voltage side of the transformer in cases where several users share the same transformer or on the high voltage side , if you own your own transformer. It is also to be noted that the reference current is the maximum total 15 or 30 min load current. While it is true that a single rectifier can have relatively high harmonics ( depending on the design) this does not make it non - compliant - a small drive of say 1 kW can have a THD of 100 %, but compared to the rest of the load ( say 1 MW ) on a 1,2 MVA trafo this has no influence on transformer heating or vaoltge distortion. In order to apply IEEE 519 correctly, yiou need to know the whole load characteristics. Some suppliers say their products meet IEEE 519 at their input - this is a wrong application of the standard - but it does make the customers life easier, as you rarely need to perform system analyses then.

I would also like to point out that many people seem to associate IEEE519 as meaning less than 5% TDD, but in reality that is not always the case. Depending on the size of your transformer compared to your total load, you could actually have a limit of 20%

yes Anthony, that's true. But what is also true is many systems that have less 5%, don't meet 519 because of the invidual levels set for different harmonics. This case is especially true for modern AFE systems that better 519 below h<50 but eceed it on higher harmonics. Fortunately, the utilities usually don't care above the 50th.

Hi all,
Thanks again all for your answers.I think that there is one step missing: the global analysis of harmonics.

It is very difficult to find serious harmonic studies of complex systems (different equipments connected together through a quite complex internal grid). It is also very difficult to find serious studies about grid voltage characterization at PCC. This makes that everybody is basically focused on systems individual current distorsion limits. And here is where IEEE519 is widely used, it has max limits defined in a very detailed way.

But I think that this simple approach is not too bad for high order harmonics but that it is wrong for low order harmonics (hi < 11).