Instantaneous trip X 10

[27hillcrest]

They other day we were bringing online a small data center. During start up and assurance testing the main tripped do to the fact the instantaneous setting on the breaker had not been set properly. We dialed it up to 10 and everything held as planned. Now my question is: What is the calculation needed to figure out how high to set the INST? The data center will be adding 480/120-208 PDU's down the road and I don't want the main to trip when I turn them on.

Any thoughts?

 

[ron]

Generally, the setting gets chosen when plotting the characteristics of the protective device upstream and downstream and trying to make sure they don't overlap (selectivity).
If you set it on 10 (probably the highest setting), then it will "hold", although you might trip a breaker/fuse upstream if the characteristics overlap.

 

[brian john]

In a properly designed and engineered electrictrical distribution system with the added cost of adjustable trip overcurrent protection devices, one would hope a coordination study was completed and as part of acceptance testing all OCP's were set per the coordination study.

Many a data center (and other critical facilities) have crashed due to improperly set LSIG (Long Time, Short Time, Instantaneous, Ground Fault) on circuit breakers.

 

[zog]

If you have an arc flash study completed you may have just increased the arc flash hazards for everything downstream of that breaker.

What makes you think it was set improperly? The fact that it wasnt cranked up all the way?

 

[jim dungar]

In a properly designed and engineered electrictrical distribution system with the added cost of adjustable trip overcurrent protection devices, one would hope a coordination study was completed and as part of acceptance testing all OCP's were set per the coordination study.

Many a data center (and other critical facilities) have crashed due to improperly set LSIG (Long Time, Short Time, Instantaneous, Ground Fault) on circuit breakers.

I am amazed at how many data centers pay hundreds of thousands of dollars for redundancy and not a dime to make sure the components are coordinated. We will have to see if this changes with the NEC 2008.

 

[kingpb]

Personally, I would like to see the NEC require a short circuit and arc fault analysis for every installation. Sorta like CA and their energy code calcs. Jim is dead center on the redundancy/coordination effort or lack thereof.

Besides it would make more money (I mean work) for us poor starving engineers.

Common practice for so long was to not even have Instantaneous on the main because you wanted downsream devices to trip first. With Arc Flash the whole coordination changes dramatically.