Discrimination Study

[timbobelfast]

Can anybody explain the discrimination overlaps in the attached protection study?

For example, a Live to earth fault in the downstream OCPD and how this may play out.

I know the current settings are not ideal, but this is a legacy issue restricted by cable sizes etc!

 

[pc1966]

Can anybody explain the discrimination overlaps in the attached protection study?

For example, a Live to earth fault in the downstream OCPD and how this may play out.

I know the current settings are not ideal, but this is a legacy issue restricted by cable sizes etc!

I think "very poorly" is how it would play out!

But the plots don't quite look right to me. For example:

• The HV fuses would be transformed from 100A at 6.6kV to 400V (presumably) at 1650A equivalent, but the plot shows them coming in at around the 10kA point.
• The first ACB is set to 1440A but seems to be starting its trip region around 8kA
• The 800A BS88 fuses at around 4kA
• The second ACB seem to trip at roughly the same threshold, but has a very aggressive time/current setting, probably an attempt at selectivity

It might be the log-scale plot is just out by some factor, as relative to each other the plots look believable.

What is not clear though is the ACB let-through and if they can possibly stop the upstream fuses from going, as most mechanical breakers take a few milliseconds to initiate opening and that normally lets through a whole lot more energy than a HRC fuse might for the same sort of conditions.

Same applies for the ACBs for any down-stream fuses, though the #1 seems to have been set with a bit of a delay to try and remain selective with the down-stream 800A fuses, but as you can see on the plot (ignoring and scale errors for now) the brown and blue curves cross each other at multiple points so there are various fault currents that would lead to having, or not having, selectivity there.

 

[pc1966]

If the curves are to believed, and the fault currents to scale, then downstream of the Wallascetown ACB should be OK, it appears to disconnect before any of the others, but it is not clear what happens in the region below 100ms or so. It might well have too much let-through.

Otherwise the bad news is the HV fuses are probably going to go before anything else! At least for faults above the 3kA-ish sort of range (guessing the horizontal scale).

Hopefully @Julie. will be along soon, this is her sort of area!