This discussion is locked.
You cannot post a reply to this discussion. If you have a question start a new discussion

UPS Fault Level Calculation

Does anyone know if there is a standardised way of calculating fault levels at the output of a large UPS like IEC60909 does for transformer and generator supplied faults?

I imagine things will change between manufacturers but a good rule of thumb would be very useful as it seems that no matter the installation, downstream of the UPS the breaker settings have been poorly thought out and my concern is that these downstream circuits will not actually acheive ADS due to the fault level being too low to trip the downstream breaker because the trip settings are too high.

Parents
  • I agree constant voltage for big UPS and small faults - where the fault loop resistance is high enough that the UPS just sees the fault as a load that is within its capability. BUT, as Andy alludes above,  unless the UPS is massive and supplies a great many very long thin circuits, that will not cover the case of most credible faults. The problem is that at some overload level, be that 120%,. 150% or even 200% of the maximum load, the UPS will go into current limit, and the voltage will collapse, perhaps maintaining a constant current, perhaps folding the voltage back more steeply than that. At this point you do not have the energy to operate the fuse or breaker promptly, and the other connected equipment sees a brown out - and if SMPS based, may even try and draw more current to compensate for the falling voltage, leading to even sharper voltage collapse and even less power available to perform the disconnection. This is a state to be avoided if at all possible, being bad for the UPS and bad for the other connected equipment as well.
    So unless all the instant trip currents of all the breakers or fuses are well within the UPS range of normal loads, you need to consider how the UPS system can recover from this. As above, the earth fault relay or on smaller systems the humble RCD is your friend, as this can be set to a small fraction of the load current of the circuit it protects, typically 0.1 % to 1% depending on the load type and operates promptly even when the UPS is in collapse, allowing UPS power to be restored to all but the faulty branch, hopefully.

    Mike

Reply
  • I agree constant voltage for big UPS and small faults - where the fault loop resistance is high enough that the UPS just sees the fault as a load that is within its capability. BUT, as Andy alludes above,  unless the UPS is massive and supplies a great many very long thin circuits, that will not cover the case of most credible faults. The problem is that at some overload level, be that 120%,. 150% or even 200% of the maximum load, the UPS will go into current limit, and the voltage will collapse, perhaps maintaining a constant current, perhaps folding the voltage back more steeply than that. At this point you do not have the energy to operate the fuse or breaker promptly, and the other connected equipment sees a brown out - and if SMPS based, may even try and draw more current to compensate for the falling voltage, leading to even sharper voltage collapse and even less power available to perform the disconnection. This is a state to be avoided if at all possible, being bad for the UPS and bad for the other connected equipment as well.
    So unless all the instant trip currents of all the breakers or fuses are well within the UPS range of normal loads, you need to consider how the UPS system can recover from this. As above, the earth fault relay or on smaller systems the humble RCD is your friend, as this can be set to a small fraction of the load current of the circuit it protects, typically 0.1 % to 1% depending on the load type and operates promptly even when the UPS is in collapse, allowing UPS power to be restored to all but the faulty branch, hopefully.

    Mike

Children
No Data