Member Login

Remember Me
Forgot Password?
Join Us!

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

Join Eng-Tips
*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.
Jobs from Indeed

Link To This Forum!

Partner Button
Add Stickiness To Your Site By Linking To This Professionally Managed Technical Forum.
Just copy and paste the
code below into your site.

rockman7892 (Electrical) (OP)
22 Nov 11 9:32

When looking at MV transformer protection I curious what is the correct method for determining weather or not the transformer damage curve is protected.

I know the NEC allows the primary of a transformer to be protected by a fuse up to 300% of the transformer rating, but by going by this 300% alone does this always guarentee adequate protection of the transformer.  Must we look at the transformer damage curve on a TCC in oder to determine if the transformer is adequately protected on the primary?  If so must the damage curve lie completely above the primary fuse on the TCC?

When we are looking at the damage curve are we only looking at it from a fault perspective or is it used for an overload perspective as well.  For example from an overload perspective if the damage curve falls below the primary fuse on the TCC but is still above the secondary protective device curve is the transformer considered adquately protected from an overload standpoint?  In this case how would we confirm weather or not it was protected from primary faults or through faults?
Helpful Member!  rcwilson (Electrical)
22 Nov 11 10:37
The primary fuse must be below the transformemr damage curve to protect the transformer on through faults.  If there is a fault between the transformer terminals and the low voltage breaker, the primary fuse msut clear it before the transformer is damaged.
So the primary fuse curve should be below the transformer curve on the TCC, after accounting for the delta-wye shift. (Primary current = 58% of secondary current for single phase-ground fault, 87% for phase-phase.

You are correct in assuming that we need to look at both overload and through fault protection and that one fuse may not do both.
rockman7892 (Electrical) (OP)
22 Nov 11 11:52
Thanks rcwilson

I'm not exactly following you with the shift.  I'm assuming this is becuase of the delta primary and wye secondary.  How does this come into play when looking at the transformer damage curve?  Are you saying that the fuse can still be above or even cutting through the damage curve if the shift multiplier cause the damage curve to fall above the fuse?

If the damage curve is above the fuse will this also gurentee overload protection or is this the primary function of the the secondary OCPD?  In this case do you need to make sure the that secondary device falls below the transformer overload pickup or only slightly above like maybe 125%?
rcwilson (Electrical)
22 Nov 11 14:13
On a delta-wye transformer with a grounded wye, a line-ground single phase fault on the secondary appears as a large phase-phase load on the primary.  If 10 per unit current is flowing in the one secondary phase, 5.77 per unit current is flowing in the two phases on the high voltage side (In phase A, out phase B). The 57.5% factor accounts for phase-phase voltage on the delta and phase-neutral on the wye. (Think kVA high side = KVA low side).

The transformer damge curve is based on current in the winding. The low voltage winding will have 100% fault current but the high voltage fuse or relay will only see 57.5% current. To protect that winding, the ANSI withstand curve has to move to the left 0.577.

You have the corect idea about overload.  Let the primary equipment protect the transformer from faults and use the secondary equipment to protect from overload.  The primary over current device still has to meet the NEC requirements, if that is the ruling code.
Helpful Member!  dpc (Electrical)
22 Nov 11 16:35
As a practical matter, it will be very difficult to provide full range transformer protection with a primary fuse on a delta-wye transformer.  The mechanical damage point at 2 sec is generally considered the most critical point on the damage curve.  Expulsion fuses generally can provide better protection than current-limiting fuses, due to the shape of the TCC.   

Also, for ANSI transformer, it must be determined if the "frequent" fault or "infrequent" damage curve will be used.   
rockman7892 (Electrical) (OP)
23 Nov 11 15:19
I looked in the IEEE Buff book and saw the frequent and infrequent damage curves dpc is referring to.  Since most of the applications I come across deal with secondary conductors enclosed in conduit or switchgear I'm probably more interested in using the higher infrequent damage curve.

I dont have the book in front of me now but I believe it stated that in this case the secondary protection could be below the frequent damage curve and then the primary protection would only be required to be below the infrequent damage curve as a backup?

Also it mentioned for coordination purposes that the primary protection device could cross through the damage curve but it recommended this crossing take place as low as a current as possible.  It sounds like from what dpc is saying that the most critical point on the damage curve is where the curve terminates at 2sec and this point absolutely must be above the protective device and any point below this 2sec point can be below the protective device but should be avoided as much as possible.

In software such at PTW is the damage curve plotted with the delta-wye shift accounted for if the transformer is entered as such?  Or is this a manual shift that must be looked at on the TCC?
dpc (Electrical)
23 Nov 11 16:50
It depends on the software.  These days, most analysis software can account for the delta-wye issue - this generally user-selected.  I'm pretty sure SKM software does this, but I don't have time to check right now.  

In the old days, the "2-second" point was all we worried about.  The "Z" curves for transformer damage were added about 20 years ago.  

These are all guidelines and goals.  Perfection doesn't exist in the real world and the transformer warranties don't last forever either.  

I believe ANSI standards define "Frequent" fault duty as more than 10 through-faults during the expected service life of the transformer.  Using the Frequent curve is always conservative, so if you can protect that, it is always better.
rockman7892 (Electrical) (OP)
28 Nov 11 14:09

Im trying to understand why the primary side will only see 57% current.  So if there is a line to ground fault on the secondary side and 10per unit current is flowing then this same 10per unit current will be flowing on the delta winding as well.  Since this delta winding current is made up of the line currrents on the delta which are 120deg out of phase is this why the line currents are only at 58%.  Aren't the line currents 173% of the phase currens in a balanced condition?  

I understand the concept here I'm just trying to wrap my head around the math.

Helpful Member!(2)  jghrist (Electrical)
28 Nov 11 15:42


Im trying to understand why the primary side will only see 57% current.  So if there is a line to ground fault on the secondary side and 10per unit current is flowing then this same 10per unit current will be flowing on the delta winding as well.  Since this delta winding current is made up of the line currrents on the delta which are 120deg out of phase is this why the line currents are only at 58%.  Aren't the line currents 173% of the phase currens in a balanced condition?
It's because there is only current flowing in one winding.
Take an example:
12470 delta - 277/480 wye, 1000 kVA
Voltage ratio - 12470/480 = 26
Turns ratio - 12470/277 = 45
Secondary FL current - 1000/.48/sqrt(3) = 1203A
Primary line FL current - 1203/26 = 46.3A
Secondary Ø-grd fault current = 10 pu = 12030A
Primary winding current = 12030/45 = 267.3A  
The winding current is the same as the line current in two lines - it flows from one line, through the winding, and out the other line.
Primary line current in pu = 267.3/46.3 = 5.77
rockman7892 (Electrical) (OP)
29 Nov 11 9:44

I see it now Thanks!

So I was also reading in the Buff book where it was talking about the transformer primary protective device and its settings in relation to the transformer let-through.  It said that instantaneous elements on the primary of the transformer should be set higher than the maximum through fault on the secondary and that the operating current of the primary device should be less than the short-circuit current of the transformer as limited by the source and transformer impedance.

I interpret this as saying that for primary protection the use of an instantaneous element should only be used to protect for faults on the primary or within the transformer.  So for that the instantaneous element should be set higher than the maximum through fault current that would occur on the secondary?  Is this maximum through fault usually represented somewhere on a TCC?

As far as the operating current mentioned above it is stating that it should be set less than the short circuit currrents of the transfomrer as limited by source and transformer impedance.  So this is is esentially saying that the overcurrent element of the primary protective device should be set less than the transformer damage curve especially at 2sec in order to provide backup protection?  Is this correct?

Are faults on the primary winding of the transformer typically protected by setting the primary protection below the non-frequent damage curve?

Also for applications where one breaker is feeding several transformers and allowed to be sized up to 250% by the NEC does this this primary breaker usually not provide any protection and strictly relys on the secondary device for protection?
rcwilson (Electrical)
29 Nov 11 10:44
Primary instantaneous overcurrent setting has to be above the transformer inrush current which is usually close to the maximum through fault.  In my experience, it is difficult to set the primary instantaneous to reliably allow inrush and yet be low enough to trip on a secondary short circuit. Setting it above the secondary fault level assures coordination with down stream devices.
It is useful to show the major fault levels on the TCC curve. Put a tick mark on the X axis for  primary and secondary fault currents. Put a tick mark at the top for 100% or 125 % FLA. Those are the benchmarks for transformer protection, along with the damgae curve and the inrush point.

You have the correct idea on transformer protection.  Set the overcurrent under the damage curve.

For multiple transformers on a single overcurrent device, overload protection is by the secondary overcurrent device, transformer fault protection is by the primary fuse/breaker.

With multiple units on a single breaker or fuse, it is probably difficult to get the fuse under the individual damage curves.

Save money on the distribution, possibly spend it later on replacing a transformer. That still may be a good design for smaller units.
rockman7892 (Electrical) (OP)
29 Nov 11 12:00
At what size transformer do we typically become concerned with protection and how it relates to the through fault curve?  For small 150KVA systems and such are we concerned with protection and how it relates to the damage curve as much as we are simply making sure the protective devices follow the rules established in the NEC?

Rcwilson mentions the secondary protection device providing overcurrent protection.  Is this usually done at 125% for most transformers?  I believe I saw 125% as being the "rule of thumb"?  So as long as the secondary device is below 125% mark for transformer it is considered protected form overloads.  I guess it would be a rare case that the primary protective device would fall under the 125% mark and therefore provide overload protection for the transformer.

So from what I gather you are saying about multiple transformers protected by a single breaker or fuse it may be allowed by NEC but it may prove difficult to fit the primary device curve under the transformer damage curve?  Even if the secondary device on these transformers falls below the damage curve we still want the primary device to be below as well in order to protect the transformer for faults on the primary?  I'm assuming that the through fault damage curve also applies to faults on the primary of the transformer and thus is the reason why we want primary protection to be below the damage curve?
dpc (Electrical)
29 Nov 11 12:35
The damage curve is for through-faults on the secondary side of the transformer.  If the fault is in the transformer or on the primary side, the damage curve has no bearing on anything.  Even if there is a secondary main device, there is still the possibility of a fault between the transformer secondary bushings and the load side of the main device, such as a fault in the main secondary breaker itself.  It may be unlikely, but it represents a worst-case for transformer through-fault damage.  

So, ideally, the primary device is fully below the transformer damage curve.  

Protection against long-term overloads is not really the job of the overcurrent relays.  You need to comply with NEC if applicable, but for a large transformer, overload protection should be provided by temperature monitoring in the transformer.   
rockman7892 (Electrical) (OP)
29 Nov 11 12:48

Yes I read that one of the reasons for ensuring that primary device is below damage curve is to protect for "worst-Case" faults occuring on line side of secondary main breaker as you mentioned.  It referenced this primary protective device being below the "infrequent" damage curve.

So what is used for protection for faults on primary side of the transformer since as you mentioned the damge curve is only relevant for thorugh faults?  How do we ensure transformer is adquately protected for such faults?  Is this where transformer "differential relaying" comes into play?

dpc (Electrical)
29 Nov 11 12:54
If we are talking about faults outside of the transformer on the primary side, the transformer is not at risk, since it doesn't see any fault current, unless there is a source of power on the secondary side.  Primary faults ahead of the transformer are a system protection issue, not a transformer protection issue.

rockman7892 (Electrical) (OP)
29 Nov 11 14:36
My apoligies dpc I was refering to faults inside the transformer on the primary windings, or even on the secondary windings for that case?  Are these faults only protected with differential protection?

Primary faults ahead of the transformer are system protedtion issue as you mentioned but I believe are usually protected by the primary protecttive device depending on the primary device location.
dpc (Electrical)
29 Nov 11 14:59
Faults in the transformer must be cleared by the primary protection.  These can be cleared by overcurrent relays or primary fuses.  

Differential protection is always better than overcurrent protection for any transformer.  

rockman7892 (Electrical) (OP)
29 Nov 11 15:44

Is there any guideline for determining if primary protection is adequate for clearing faults inside the transformer?  (Some sort of reference point or curve?).  Or is it just one of those things that you need to try to size the primary protection device as low as possible while allowing proper operation and coordination?

Thanks for the help.  I've learned a great deal from this discussion!
dpc (Electrical)
29 Nov 11 16:27
It's always a compromise.  People like to be able to overload transformers when convenient but it's a big drag when they catch on fire.   Generally, primary protection that meets the NEC should eventually clear an internal transformer fault.  

If you assume a bolted fault on the primary, your primary protection needs to be able to clear that, obviously.  A turn-to-turn fault - not likely to be detected until it morphs into something more exciting.  

rockman7892 (Electrical) (OP)
29 Nov 11 17:30
O.k. yes but a transformer protected on the primary at 250% will not necessarily protect the transformer from overload but rather may be protected from overload by the secondary device.  So I guess its plausiable that that what you are saying is that if the primary is protected according to the NEC it will eventually clear internal transformer fault, while overload protection can be provided by the secondary protection device at 125% of transformer FLA?

When you talk about the primary device needing to be able to clear a bolted fault on the primary I assume that you mean that the protective device is set below (and not above) the avaliable fault current on the primary of the transformer?
dpc (Electrical)
29 Nov 11 17:57
Overloads are not faults.   You seem to keep worrying about overloads.  The NEC overcurrent protection requirements provide very minimal protection against extreme overloading conditions.  The intent of the overcurrent protection is mainly to prevent fires, not to eliminate all transformer overloading. Utilities routinely overload their transformers up to 200% for short periods of time because it saves them money.  

rockman7892 (Electrical) (OP)
5 Dec 11 15:59

I see your point about confusing overloads with fault.  I agree they are two seperate issues.  

Is it safe to say that typically the primary device does not provide the overload protection for the transformer so this usually fall on the secondary device?

Are there more stringent overload guidelines to follow besides those in the NEC which are minimum as you mentioned when providing overload protection of transformer?  In other words the NEC allows secondary protection up to 300% on MV transformer secondaries.  I assume we should strive to protect lower than this.

It sounds like setting for overload and through-fault protection are not related at all since one occurs in the long time region and the other occurs in the short time region.  So therefore selecting protection based on one should have no effect on the other correct?
dpc (Electrical)
5 Dec 11 17:14


Is it safe to say that typically the primary device does not provide the overload protection for the transformer so this usually fall on the secondary device?

No, because sometimes the primary protection is all there is.  

I would say that for long-term overloads, the overcurrent relays will provide some protection for serious overload, but that it may or may not protect the transformer from all damage.  Most relay curves don't even start until 1.5x or 1.25x pickup.  And the TCC stops at 1000 seconds.  So we are looking at a limited range of events when we coordinate using TCCs.  Also, transformers can carry more load when the ambient is lower, and the overcurrent relay (usually) knows nothing about that.  

The best protection against long-term overloads would be internal winding temperature monitoring.


Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!

Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close