How Close is too Close?

mbrooke

Batteries Not Included
Location
United States
Occupation
*
How close can a secondary fuse be near a transfomer's damage curve? How close to a primary fusing? Other secondary fuses? Is there a certain proportional ratio?

I have cases where E fuses touch a delta-wye damage curve and cases where a 140K fuses's "kink" comes close to the feeder fuse.

Obviously not likely to be put into practice- but as an extreme example:

 

NewtonLaw

Senior Member
This substation appears to be an older style that uses a single transformer tank with three separate transformer cores with double bushing on both the primary side and the secondary side allowing the customer to operate it as Delta-Delta, Delta-Wye, or Wye-Wye. The primary side is protected by S&C SMD-2 fuses it looks like. These provide protection for the transformer, low side bus and the three voltage regulators. The output of the voltage regulators feed a set of S&C SMD-5 fuses which protect the line. A one line diagram would be helpful. Also, are the damage curves you show for the power transformer, the voltage regulators or both? Would you also identify which fuse curves belong with which set of fuses?
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
This substation appears to be an older style that uses a single transformer tank with three separate transformer cores with double bushing on both the primary side and the secondary side allowing the customer to operate it as Delta-Delta, Delta-Wye, or Wye-Wye. The primary side is protected by S&C SMD-2 fuses it looks like. These provide protection for the transformer, low side bus and the three voltage regulators. The output of the voltage regulators feed a set of S&C SMD-5 fuses which protect the line. A one line diagram would be helpful. Also, are the damage curves you show for the power transformer, the voltage regulators or both? Would you also identify which fuse curves belong with which set of fuses?

All correct! SMD-2B Primary fuses and SM-5 secondary fuses.

The regulators are not part of the damage curve and omitted.

The pic is from S and C's coordinate. Far right fuse is the primary, the orange one to the left is the secondary.
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
This substation appears to be an older style that uses a single transformer tank with three separate transformer cores with double bushing on both the primary side and the secondary side allowing the customer to operate it as Delta-Delta, Delta-Wye, or Wye-Wye. The primary side is protected by S&C SMD-2 fuses it looks like. These provide protection for the transformer, low side bus and the three voltage regulators. The output of the voltage regulators feed a set of S&C SMD-5 fuses which protect the line. A one line diagram would be helpful. Also, are the damage curves you show for the power transformer, the voltage regulators or both? Would you also identify which fuse curves belong with which set of fuses?

Single line is like this -> SMD-2B fuse, transfomer, SM-5 fuse- all literally in series.
 

EmagSamurai

Member
Location
Alabama
Not sure if there is NEC guidance, but IEEE Std 242 says devices with a banded curve only require clearance between the curves for coordination. Though some manufacturers recommend a safety factor if the next device upstream is another fuse to prevent partial melting.

So if there is no upstream fuse, then a gap is enough for coordination. Though I can't say I've ever pushed it as close as you have shown, and probably wouldn't unless it was an existing installation or I had no other choice.
 

bwat

EE
Location
Western PA
Occupation
EE
I know of a PoCo that had the 'practice' of finding the fuse or setting that just barely fits in the curve, as close as absolutely possible, and then going 1 step/size further away and using that one. Within reason this was applied. It seemed to always make a reasonable coordination curve.
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
Are the damage curves the actual ones for the transformer or are they the library files? This alone could make a difference on the selection. Typically, the actual do not match the generic library data
Library data. By how much can generic vary in relation to actual?
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
I know of a PoCo that had the 'practice' of finding the fuse or setting that just barely fits in the curve, as close as absolutely possible, and then going 1 step/size further away and using that one. Within reason this was applied. It seemed to always make a reasonable coordination curve.
Thats what I've heard... guess it works in practice well enough?
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
Not sure if there is NEC guidance, but IEEE Std 242 says devices with a banded curve only require clearance between the curves for coordination. Though some manufacturers recommend a safety factor if the next device upstream is another fuse to prevent partial melting.

So if there is no upstream fuse, then a gap is enough for coordination. Though I can't say I've ever pushed it as close as you have shown, and probably wouldn't unless it was an existing installation or I had no other choice.
Can you give an example of this?

This whole scheme is a practice of limbo. :oops:
 

EmagSamurai

Member
Location
Alabama
I'm not exactly sure which part you're asking for an example of. When coordinating two protective devices, the gap between the curves is the Coordination Time Interval (CTI). I'm not sure if it's discussed in the NEC, but this is a margin between device characteristic curves that allows for things like breaker operating times, relay over travel, and tolerances. My interpretation of the 242 is that devices with a banded TCC (like a fuse) only require a "white space" between their curves as sufficient CTI. Something like a relay with a solid line can require .25 - .35 s CTI.

Regarding the manufacturer suggestion for additional margin between fuse curves, I believe this would be manufacturer specific. The band shown on a fuse TCC is the minimum melt along the left edge of the curve and the maximum clearing along the right edge (I believe, anyway. Haven't looked at this in a while). So what 242 is saying is you want to make sure the downstream fuse maximum clearing time is far enough away from the upstream fuses minimum melting time that you don't begin to melt that upstream fuses element before the fault is cleared.

EDIT: I'm no SME on this. The above is just my mental model, and it may not be 100% accurate. It's only worth what you've paid for it :)
 

NewtonLaw

Senior Member
I am not sure if your substation is a Utility station or privately owned but Utilities will add additional protection to the transformer if the damage curves are exceeded. This is especially true since Utility transformers are exposed to many fault conditions with the vast majority being unbalanced so the damage curve to the left in your diagram. If protection and coordination can not be met using only a fuse, we place a circuit breaker or circuit switcher in place of the fuses. As far as an operating gap goes we like to have a 10% buffer between curves but that can not always be met in all cases and we accept some miscoordination conditions to satisfy other criteria.

As I look at your coordination curves I can't help but wonder if the damage curves are for the transformer shown in the image since that type of transformer appears to be well over 70 years old in design. Also, the curves you show not only have black lines for the damage curves but also a series of black dots. If the dots represent a damage curve, looks like the transformer fails first followed by the fuse?
 

mbrooke

Batteries Not Included
Location
United States
Occupation
*
Dots are inrush.

Reclosers would do all the work today, breakers/circuit switchers in larger ones, but fuses are still economical and found in many existing installations.

I'm willing to assume a modern damage as these transformers are most likely to end being replaced... rather soon.

Typically primary fusing would not present much of a concern since there would be multiple feeders each carrying only percentage of the transformer's secondary FLA. Primary fuse can easily be 125% (give or take) of the primary FLA without much concern.

The issue here is that the secondary fuses must take into account all the secondary FLA, leaving rather grossly over sized primary fusing which may or may not be an issue.

If the primary fuse is say 200+% of the primary FLA a secondary bus fault (while rare) may not blow the primary fast enough to prevent damage. Consider these fuses just start at 220% of their rating, then add the fact a L to neutral fault on the secondary only produces about 57% current on the primary.

In other words the transformer is screwed.

But I'm willing to let someone convince me otherwise.
 
Top