Dear Arc Flash experts.
In IEEE1584 calculations of "reduced arcing fault current" is based on protective device type 1-14 if U<1kV.
-If type 0 is selected the arcing fault current is reduced to 85%
-If type 1-8 is selected the reduced arcing fault current is "not required" ( not reduced/100%)
- Type 9-14 is MCCB and outside my question
Protective device 1-4 is class RK1 fuses and 5-8 is class L fuses in size from 100 to 2000A.
How do i select protective device type if i use gG fuses according to IEC60269?
Best regards
Preben Jakobsen
Denmark

Welcome to the Arc Flash Forum!

There might be a bit of confusion here. IEC 60296 Parts 1 - 6 are for low voltage fuses.

The 2002 Edition of IEEE 1584 had the statement:
in 5.2 Calculate a second arc current equal to 85% of Ia, so that a second arc duration can be determined (see 9.10.4). n

An amendment was issued in 2004 IEEE 1584a clarifying 85% applies to "under 1000 Volts"
For applications with a system voltage under 1000 V, calculate a second arc current equal to 85% of Ia, so that a second arc duration can be determined (see 9.10.4).

The 85% multipler is independent of the type of overcurrent device, circuit breaker, fuse etc. It is about the arcing current. 85% was simply a sensitivity calculation since the arcing current calculation was approximate at best. The concern was what if the actual arcing current was less and it fell below an instantaneous or in your case, the current limiting region of a protective device.

It isn't really about selecting the device - although the calculations could result in someone making a setting change or perhaps select a different fuse if 85% Iarcing results in a larger incident energy because Iarcing fell below the instantaneous or current limiting region.

In the US, our National Electrical Code does have a provision for ensuring there is sufficient arcing current to trip the instantaneous of larger circuit breakers NEC 240.87 and this will be added for fuses in 2020 with NEC 240.67

In addition to Jim's comments I think you answered your own question. Device-specific equations were developed by a single manufacturer (Mersen) for two fuse classes. You can't use it for a different fuse or class because it hasn't been tested for that and a purist would argue that you can't even use fuses from a different manufacturer since there is no data to support whether or not fuses from competitors would give the same performance although pragmatically fuse accuracy is "loose" enough that an argument can be made that it probably wouldn't matter. These equations give MUCH lower incident energy values compared to say the empirical equation because it takes into account current limiting and dynamic resistance of the actual devices.