Kauud,
Most of the members of this board are most familiar with standards of practise as used in North America. To move forward on this discussion, we will first need to find some common ground in terminology.
SDS: a 'separately derived system' is a source (generator, transformer secondary, etc) which doesn't have some sort of built in 'galvanic' connection to your reference system. An ordinary delta-wye transformer is an example of an SDS; the system created on the secondary side does not have an electrical connection to the primary side. Unless intentionally being used to create an ungrounded system, SDSs require bonding and grounding.
MGN: 'multi grounded neutral': this is a utility conductor that is the neutral of the utility distribution system; it is regularly and frequently connected to earth electrodes. The MGN is used as the grounding point for the secondary of utility distribution transformers, and is thus electrically tied to the neutral conductor supplying a building.
In North America, distinction is made between a 'service' and an 'SDS', even when they are conceptually the same. The reason is that different electrical safety codes are applied to the utility distribution system and the building electrical system. These electrical safety codes are developed by different standards organizations, and are not always consistent with each other.
You have been using 'IEC' grounding terminology. Could you please confirm that this article:
http://en.wikipedia.org/wiki/Earthing_system is reasonably correct in its description of the terminology?
If the wikipedia article is correct, then most 'services' in North America are type 'TN-C' with an additional earth electrode at the customer.
I do not see a significant life safety issue with the lack of 'system earthing' in your figure 1. It seems to be a 'TN-C' system with an earth electrode moved from the source to the customer, but otherwise equivalent.
The key question to ask: are all enclosures _bonded_ (electrically connected) to the system neutral by some metallic path. It is this bonding which allows fault current to flow in the event that an energized conductor contacts the metallic frame of the equipment, and it is the fault current which causes protective devices to open. The connection to a soil electrode is necessary to provide protection from high voltage events, such as leakage from the transformer primary. The single soil electrode shown would probably provide sufficient protection, although it may not be sufficient to meet code requirements.
The TT system shown in post 11 has a very significant life safety problem, and would not be permitted in North America. In the event of a fault to the equipment chassis, the only flow of fault current would be via the soil electrode path. This is not considered a sufficient path for operating protective devices. (I believe that such TT systems _are_ used in some parts of the world, however they use some form of residual current detection that will detect the low fault current through the soil path.)
If, as 76nemo surmises, you are talking about individual pieces of equipment, rather than entire buildings, then the system as described still has proper 'bonding' and thus has a fault current path. However a consequence is that each equipment enclosure would be subject to the voltage of its respective neutral. Different pieces of equipment in close proximity could be at different voltages, thus presenting a touch hazard.
-Jon
