Minimum SC calculation on UPS distribution

[PrebenJakobsen]

When calculating minimum SC value on the output side of a UPS I don't think we can use the normal IEC60909 approach.

Inverter current will be limited:
When bypass is not available the inverter alone shall supply the SC.
If SC occurs close to the UPS the impedance in the loop is low, and the inverter will limit the current by reducing the voltage. The inverter is an active component and the internal impedance is un-linear and will adjust according to SC conditions.
If the SC occurs at the end of a long cable the current will be in the range where the inverter is able to supply it, and the output voltage will be kept at nominal level. The inverter internal impedance will be zero.
(Above considerations is valid for the first e.g. 150ms where the inverter can supply e.g. 230% of nominal)

Load current to other consumers will remain:
UPS systems are typically used for computer equipment with power supply that will operate on e.g. 50% voltage. During SC the voltage is low but the entire load in fault free branches will remain consuming power.

Both phenomena's will reduce the current available to trip the MCB in the faulty branch which is critical due to discrimination.

Does anyone know literature that describes how to calculate minimum SC on a UPS system?

Preben Jakobsen
Denmark

 

[YuriB]

For the UPS to continue to supply power while there is a short circuit in one of the branches the line in that branch should be kilometers long.

 

[PrebenJakobsen]

Hi YuriB
If the UPS is 200kVA and a SC is made at the end of a 50m, 2,5mm2 cable the voltage will be nominal on the UPS distribution panel!
Preben Jakobsen

 

[YuriB]

Hi PrebenJ. If the SC means short curcuit, if the UPS output voltage is AC, let s say, 230V, then the protection device not tripping at the current caused by a SC at the end of the 50m long 2.5 mm3 copper line should lead to fire.

 

[YuriB]

Resistance of 2.5 mm2 100 m long wire = 0.9 Ohm

 

[sibeen]

Preben, this is one of those situations in engineering that isn't black or white and sometimes there's not much you can do about it. It's always been fairly well known that the output from a UPS or Inverter is severely current limited and the chances of this current being able to open anything but a fairly small circuit breaker was close to zero; but, you also have to look at the chances of this occurring.

How many times does the bypass supply actually fail on a UPS (mains failure)? How many times is there a downstream fault?

Both are fairly small numbers and both have to occur at the same time for your scenario to arise. There is also the compromise that would need to be reached between the high fault currents that are normally involved when the bypass (Reserve) is available and the small fault currents available the relatively few times when only the inverter is able to supply current.

For what it's worth, when I complete a study like this I take into account the fault currents that can be produced on the bypass supply and discount the very unlikely scenario that you've considered.

 

[PrebenJakobsen]

Sibeen -
I agree with you that the fault is unlikely, but I find it strange that the issue is not addressed by vendors or in literature.

In my case i have an offshore installation where discrimination is a damand in safety critical systems, and protection against indirect contact is always required.
We have discussed the issue with the customer, who don't just want to fulfill legal requirements, but also looks at the cost of several 100k$ if a trip initiates a production shutdown.

Preben