Are these two situations the same when setting an overcurrent RELAY to protect a transformer(s) ?

[bdn2004]

Attached is a 3 page sketch showing (2) situations: the first page is the single line. The next two are with the secondary shorted on the transformer.

Per the information I've been looking at...it's best to short the secondary of the transformer - determine the primary current in that faulted condition, multiply that value by 10% and set the instantaneous trip of the primary relay to that value. The reasoning being the primary current can never be higher for a through fault - so you can trip instantly for any value above that indicating a fault has occurred on the primary side.

My question is on the second page.... What if the primary circuit feeds multiple transformers...Would you use the same reasoning ? - and that it's only likely that one transformer would fault - and set the instantaneous to that same value as above? Or do you need to add the FLA of the other transformers to the equation? It seems the initial conditions would be higher amps in that circuit. I realize the long term setting would likely be higher in this situation. I'm only referring to the instantaneous setting.

 

[stevenal]

I would add the full load of the others and see where that gets you in relation to the primary short circuit current that you wish to reliably trip on instantaneous for. In a similar situation, I used the average of the primary fault and through fault values.

 

[dpc]

Your logic is sound and this is common practice. Don't forget about the asymmetrical current. Faults can have initial current much higher than calculated symmetrical current. Most microprocessor relays IN THEORY will filter this out. Electro-mechanical IT units and typical molded case breakers do not, and will respond to the dc offset current. In olden times with E-M relays, we typically used 1.6 x symmetrical fault current to account for the dc offset and than another 10 or 15% as a safety factor. Also mind the transformer inrush current.

 

[bdn2004]

In looking at one of my old books...(Industrial Power Systems by Beeman, pg 551) it states:

"The miniumum setting of the instantaneous relay in a high-voltage feeder with several transformers on it will likely be dictated by the need for riding over their total magnetizing current inrush. As stated, this will be in the neighborhood of 10 to 15 times normal, which should be high enough to clear any probable condition such as a short circuit on the secondary of one or two units and normal current on the others."

In my example on the second page - all 5 transformers on the circuit will be energized at the same time. That's the concern. What's the magnetizing current ? In my example the FLA of transformers at 12470V is 46A. 46A * 5 = 230A, so per Beeman - the instantaneous setting should range between 2300A and 3450A. That makes more sense imho.

In my example I calculate the fault current is 850A with the secondary shorted on one of the transformers. I was theorizing to add the 850A + 46*4 = 1034A - that would be way too low. This secondary shorting reasoning works good with one transformer - but not more than one on the same circuit.

 

[dpc]

Inrush depends on the transformer type and multiple variables. For oil-filled transformer, Beeman's 15X is extremely conservative in my experience. You need to look at the maximum secondary side fault current as well as the maximum inrush and set IT pickup above both. The logic doesn't change for multiple transformers - it's just that with 8 transformers, maximum inrush will exceed the through fault current for a single transformer.

 

[bdn2004]

Looking at the auto-coordinate function of our software - it uses the [total FLA of the connected transformers] x 8 ... slightly above that value at .1 sec as the magnetizing current for multiple transformers on a feeder. And another Company standard I have says to use 12X the connected FLA. Doesn't seem to be a real consensus but they are in the ballpark.

These are all dry type transformers we are dealing with, and using the 8X FLA inrush point to set the relays to not trip on start-up has apparently worked with no issues.

 

[dpc]

Dry-type inrush can be significantly higher than oil-filled. If 8X is working, then stick with it. But I've had to go as high as 14X for large dry-type transformers. I proved that 8X, 10X, and 12X didn't work.

Inrush current is not the same each time. It depends on voltage phase angle on energization as well as flux state in the core. And re-energizing entire feeders is probably not a common occurrence, so don't be shocked if at some point in the future, you experience a nuisance trip on energization. It's a good thing - proves the relay can trip the breaker. If you have a digital relay, you can configure the event recording to give you event data, perhaps with oscillography each time the feeder is energized.

 

[prc]

dpc, I am not sure whether inrush current will be more for dry type transformers. I remember on earlier occasions also you mentioned about it. The inrush current mainly depends on the air core reactance(reactance of winding when removed from core) of the winding under energization. For the same MVA rating, air core reactance of dry type cannot be more than that of oil filled type. May be reactance slightly more for dry type (ie less inrush) due to larger diameter from more clearances in air. So what you experienced may be due to other factors that you explained.

Inrush current( as times full load current) will be more if energized from inner winding, less with increase in MVA, less with lower working core flux density,high with lower % impedance, less with auto-transformer (esp if switched from outer winding), less if source impedance is high, less or negligible if switched at peak of voltage wave.

But let me bring up a topic slightly not relevant to the subject. Tripping of VCB under inrush can cause severe transient over voltages in dry type units that will deteriorate the insulation permanently and even cause failures. So at the time of commissioning new dry type units, protection engineers should not do trial and error tripping to cause tripping of transformer with inrush. Transformer manufacturers should give calculated inrush and % second harmonics so that protection engineers need not do guess work.

 

[dpc]

I'm not sure why either, but it is reality. Data from manufacturers regarding inrush - I've found it to be not terribly helpful - assuming any information is provided. I understand the transient voltage issue - arresters should be provided - both at the breaker and the transformer. I did not mean to imply that it was a good thing to intentionally trying to get the relay to trip on inrush. At the same time, if we set it at 20X and the maximum is 6X, a lot of protection has been lost. My point was that the goal is to set the IT to operated just above the maximum inrush transient. Sometimes this means a trip will occur but that should not be thought of as a failure but rather something that can occur during commissioning and testing.

 

[VTer]

i just want to add that the transformer inrush cannot be higher than the available fault current at the primary terminals of the transformer. Keep that in mind when using the inrush multipliers on multiple transformers. For single units I have been using the IEEE standard of 12X the FLA at 0.1 sec with positive results. Most of my experience is with liquid-filled units and smaller dry type units.

"Throughout space there is energy. Is this energy static or kinetic! If static our hopes are in vain; if kinetic ù and this we know it is, for certain ù then it is a mere question of time when men will succeed in attaching their machinery to the very wheelwork of nature". û Nikola Tesla

 

[bdn2004]

A question I have is what is the point of reducing the setting from 15x, 12x, 8x....if it’s proven per the software that your fusing or relay setting is protecting the transformer from mechanical and thermal stresses and the cable is protected - evenat the 15x setting. So why reduce it? Isn’t what we are trying to do?