Negative Sequence Overcurrent

[Jolted]

Help me understand negative sequence overcurrent elements.

I've recently taken over an account that uses methane from a landfill to generate power that they sell to the utility. The main breaker that connects this generation to the utility grid is protected with a SEL 351 protection relay.

They had this protection relay replaced a few months before we took over the account by another company. After a recent downstream fault, they were unable to bring one of their generators online as it would cause the negative sequence overcurrent element to assert and shut the whole facility down.

Working with a support engineer at SEL and digging through the manual, I discovered that a delay was recommended with this element to prevent transients from closure events causing nuisance trips. I added a 1.5 cycle delay and got them up and running.

A couple of weeks ago they called me again as they had experienced another trip event. The logs show that the reverse directional element was asserted. They were able to reset and be back in business, but want to eliminate these nuisance trips. SEL has been very helpful and tell me that these elements are set to default values and are working on getting recommendations for setting this up correctly.

I'm sure SEL will be able to help get this running, but this is new territory for me and I'd like to understand it better. Any of you utility or industrial guys able to explain how this works?
I get that it has to do with the voltage to current lead or lag, but I'm unclear on exactly how the values are derived and what typical setups look like. The manual is quite cryptic, any of you guys good at putting this in plain english?


Sent from my SM-A102U using Tapatalk

 

[paulengr]

There are lots of tutorials on sequence currents and sequence voltages. SEL has some good papers on it,

Conventionally we have three line to ground faults, three double line to ground faults, three phase to phase faults, and finally one three phase line to line fault. Typically we set up relays based on this concept with four sets of 50/51 protection functions. Technicians might not get into the details of precisely what kind of fault it is but they know based on which elements caused the trip where to look.

From an engineering point of view doing fault analysis using just the normal voltages and currents takes 3 sets of essentially identical calculations instead of just running the calculation just once for say “line to ground fault”. To simplify the math we can do a mathematical trick that converts the normal 3 (or 4) voltages and currents into a new set of 3 voltages and currents. Positive sequence currents is the normal current we expect. Negative sequence current comes from unbalanced currents such as in line to line faults. Ground fault current comes from line to ground faults. The math becomes very easy, particularly with ground faults. We can calculate fault voltages and currents just once using sequence voltages and currents instead of doing the math three times.

The problem though is this. So when the 351 trips with a negative sequence current fault, which phase do you look for the fault that needs repaired? It doesn’t tell you that! Not unless you look at the voltages and currents in the fault record. The engineering math makes it easy to tell the type of fault (line-ground, line-line, double line-ground, 3 phase line) but nothing at all about which phases were involved. This is where we hit an impasse between academics and engineers on the one hand, and operators and technicians on the other,

SEL has heavily promoted the idea of using sequence currents for trip functions because it is so easy to do in their relays. Negative sequence trips are indeed default functions. The idea is then you can look at the fault log to determine what happened. But it is impossible to explain to a technician what it does or how to fix it.

Turn that garbage off and just use the conventional 50/51 logic! There are some very exotic trip functions where it might be useful (special differential and directional elements) but if it’s one of those it should be fairly obvious what it is.

 

[Jolted]

Thanks Paul!

That's helps add some background and concept overview. I've been working through SEL whitepapers in my spare time (just don't have much spare time)

I like your thought of just going back to basics. The only thing this protects is the conductors, generators and step up transformer. The utility has their own protection scheme and breaker on the other side of the transformer. It doesn't help me a lot as far as reviewing logs. By the time I get to the site the operator will have reset and tried to sync at least once, if not multiple times. I pretty much just have to pull the logs to figure out what is going on.

 

[subtech_2]

You don't say the nature of the the event other than a reverse directional element was asserted
in the 351. Evidently that element supervises either an over current element, or possibly an impedance
(distance) element that looks back into your customers generation system.
(Assuming that a forward fault would be looking from your client's breaker through the transformer and into the utility system)
In a nutshell, negative sequence voltage quantities in the SEL relay are derived from imbalanced primary voltages,
and negative sequence currents in the relay are derived from imbalanced primary currents. The more significant the imbalance
in the primary system, the greater the negative quantities will be in the relay and the negative sequence vectors are reversed in sequence
when compared to the positive sequence vectors. For me, the best explanation I've found for how the negative sequence
directional elements work comes from the the instruction book for the old SEL-321 relay. It took me many hours of reading
and scratching things out on paper until I came to a full understanding of how these elements determine direction, but I stuck with it
until it was "cemented" in. I highly recommend you do the same. A few things that are critically important to
remember about the sequence quantities:

(a) A three phase power system with perfectly balanced voltages and currents contains only positive sequence quantities.
(of course, there is no such thing in reality)

(b) Any negative (or zero) sequence quantity that exists in one phase exists equally in the other two.

(c) Some references show negative sequence vectors labeled a-b-c and rotating in the opposite direction of the positive sequence vectors.
Other references show negative sequence vectors as a-c-b but continuing to rotate in the same direction as the positive sequence vectors.
Either is ok, but the most important thing for you to remember is that electrical systems do not rotate. Only the machinery that connects
to the system does. What matters is the sequence in which the vectors pass a fixed point of perspective.

(d) If you are working in an ABC system (check the relay nameplate or in the settings) positive sequence is ABC and negative sequence is ACB.
Vice versa if you are working in an ACB system.

Addtionally, we'd all like things to be simpler, easier, quicker, and more convenient.
But I think you and I both know that on this planet that just isn't going to happen.
I gather from your writing that you would never turn off elements in a relay to make things easier on yourself.
I recommend that you stick with that. An old line crew foreman told me once, "Right after Mr. Hurry_Up comes Mr. Disaster."
Boy, was he ever right.

Finally
Your customer is not syncing his machines to the power company system properly.
The evidence of this is the reverse negative sequence element picking up and subsequent tripping.
This can be the fault of the operator or relays/controls.
Your client may pressure you to relax the settings in the relay to the point where it will tolerate sloppy syncing.
Don't.

I'll be watching this post. Feel free to ask any question or comment back whenever you'd like.

Best wishes

Mike
Relay technician
Interim system protection engineer
and
First assistant to lead custodian

 

[Hv&Lv]

Help me understand negative sequence overcurrent elements.

I've recently taken over an account that uses methane from a landfill to generate power that they sell to the utility. The main breaker that connects this generation to the utility grid is protected with a SEL 351 protection relay.

They had this protection relay replaced a few months before we took over the account by another company. After a recent downstream fault, they were unable to bring one of their generators online as it would cause the negative sequence overcurrent element to assert and shut the whole facility down.

Working with a support engineer at SEL and digging through the manual, I discovered that a delay was recommended with this element to prevent transients from closure events causing nuisance trips. I added a 1.5 cycle delay and got them up and running.

A couple of weeks ago they called me again as they had experienced another trip event. The logs show that the reverse directional element was asserted. They were able to reset and be back in business, but want to eliminate these nuisance trips. SEL has been very helpful and tell me that these elements are set to default values and are working on getting recommendations for setting this up correctly.

I'm sure SEL will be able to help get this running, but this is new territory for me and I'd like to understand it better. Any of you utility or industrial guys able to explain how this works?
I get that it has to do with the voltage to current lead or lag, but I'm unclear on exactly how the values are derived and what typical setups look like. The manual is quite cryptic, any of you guys good at putting this in plain english?


Sent from my SM-A102U using Tapatalk

The other two posts explained the negative sequence, I’ll not repeat...
Before you turn them off, make sure you pull the HIS and EVE from the terminal window of the 351 and investigate first.
You also should look at the event in Syncrowave.
Negative sequence comes from somewhere, try to find out where.. it could be there is a loose or arcing connection, a misoperation of one pole of the breaker, maybe even a problem with the generator rotor.
From a generator the loads should be balanced. If they trip on negative sequence there is an unbalanced condition or it could be a momentary open conductor. You did mention a reverse element. If it’s grid tied, and there is a loose connection behind the relay, I could see a possibility of negative sequence and reverse tripping.

 

[paulengr]

Just to reiterate there are 3 currents, not just two. That bit about positive and negative ABC currents is incorrect.

In phasor (real) currents we have Is, Ib , Ic.

In sequence currents we have I+, I-, and I0. They usually get other names but I want to emphasize the nature. Equipment has three impedances (Z+, Z-, Z0). Usually Z+ and Z- are identical since loads should be balanced while the ground impedance is a little different and hopefully infinite under normal conditions.

In normal current flow, I- and I0 are zero. Three phase faults are also the same.

A ground fault appears with equal sequence currents. I+=I-=I0= V/(Z+ + Z- + Z0).

A phase fault creates a negative sequence current.

I+=I-=V/(Z+ + Z-)

A double phase to ground mixes the two. See here:

Electrical Fault Calculation | Positive Negative Zero Sequence Impedance | Electrical4U

Before applying an electrical protection system, it’s crucial to understand the conditions of the electrical power system during faults. This knowledge helps in deploying the right protective relays at various locations in the system. Information regarding values of maximum and minimum fault...

www.electrical4u.com

From an engineering point of view the math for fault calculations becomes quite easy and we don’t need to even concern ourselves with which phase(s) are involved. So if I0 is close to zero we know ground is not involved. If I- is also near zero it’s a three phase fault. Otherwise it’s phase to phase. If I0 is present and I0 and I- are almost equal, it’s a ground fault. Otherwise it’s a double line to ground fault. See how easy this is?

Ok, but which phase is it on (is, In, Ic)? No idea...sequence currents make it so we don’t care. But the repair crew needs to know where to look. This is where phasor faults (50/51) tell us useful information but negative sequence faults do not. Ground faults sort of do but again we don’t know on which phases.

 

[Jolted]

You don't say the nature of the the event other than a reverse directional element was asserted
in the 351. Evidently that element supervises either an over current element, or possibly an impedance
(distance) element that looks back into your customers generation system.
(Assuming that a forward fault would be looking from your client's breaker through the transformer and into the utility system)
In a nutshell, negative sequence voltage quantities in the SEL relay are derived from imbalanced primary voltages,
and negative sequence currents in the relay are derived from imbalanced primary currents. The more significant the imbalance
in the primary system, the greater the negative quantities will be in the relay and the negative sequence vectors are reversed in sequence
when compared to the positive sequence vectors. For me, the best explanation I've found for how the negative sequence
directional elements work comes from the the instruction book for the old SEL-321 relay. It took me many hours of reading
and scratching things out on paper until I came to a full understanding of how these elements determine direction, but I stuck with it
until it was "cemented" in. I highly recommend you do the same. A few things that are critically important to
remember about the sequence quantities:

(a) A three phase power system with perfectly balanced voltages and currents contains only positive sequence quantities.
(of course, there is no such thing in reality)

(b) Any negative (or zero) sequence quantity that exists in one phase exists equally in the other two.

(c) Some references show negative sequence vectors labeled a-b-c and rotating in the opposite direction of the positive sequence vectors.
Other references show negative sequence vectors as a-c-b but continuing to rotate in the same direction as the positive sequence vectors.
Either is ok, but the most important thing for you to remember is that electrical systems do not rotate. Only the machinery that connects
to the system does. What matters is the sequence in which the vectors pass a fixed point of perspective.

(d) If you are working in an ABC system (check the relay nameplate or in the settings) positive sequence is ABC and negative sequence is ACB.
Vice versa if you are working in an ACB system.

Addtionally, we'd all like things to be simpler, easier, quicker, and more convenient.
But I think you and I both know that on this planet that just isn't going to happen.
I gather from your writing that you would never turn off elements in a relay to make things easier on yourself.
I recommend that you stick with that. An old line crew foreman told me once, "Right after Mr. Hurry_Up comes Mr. Disaster."
Boy, was he ever right.

Finally
Your customer is not syncing his machines to the power company system properly.
The evidence of this is the reverse negative sequence element picking up and subsequent tripping.
This can be the fault of the operator or relays/controls.
Your client may pressure you to relax the settings in the relay to the point where it will tolerate sloppy syncing.
Don't.

I'll be watching this post. Feel free to ask any question or comment back whenever you'd like.

Best wishes

Mike
Relay technician
Interim system protection engineer
and
First assistant to lead custodian

Thanks Mike, this is extremely helpful.

I've been dealing with a breakdown on a production line all weekend, but will get back into this soon. This will really help to interpret all the math in the papers and manuals.

It oversees an overcurrent element.

I don't think this is a syncing issue, possibly a generator issue. The synchronizing controls were correctly reading and bringing the generators on in phase. The generator that was causing the issue had a head failure shortly after I was there the last time and has to get re-built.

SEL tells me that the reverse directional element is set up default and probably needs changed. They're working on a recommended set up, I want to understand what I'm doing before I implement it.

I'll start studying when I find the time and post back once I hit a wall or have an aha moment.

 

[Jolted]

The other two posts explained the negative sequence, I’ll not repeat...
Before you turn them off, make sure you pull the HIS and EVE from the terminal window of the 351 and investigate first.
You also should look at the event in Syncrowave.
Negative sequence comes from somewhere, try to find out where.. it could be there is a loose or arcing connection, a misoperation of one pole of the breaker, maybe even a problem with the generator rotor.
From a generator the loads should be balanced. If they trip on negative sequence there is an unbalanced condition or it could be a momentary open conductor. You did mention a reverse element. If it’s grid tied, and there is a loose connection behind the relay, I could see a possibility of negative sequence and reverse tripping.

Thanks, I appreciate the practical application of how this works. At the moment I'm more interested in the broader picture of what's going on and how to set this up properly, as well as an understanding for the next time I run into this, but I've attached a screen shot of the last trip event for to help explain what's been happening.

 

[ron]

When trying to come up for an acceptable setting for negative sequence overcurrent elements in advance of actual operation is a crap shoot in my experience.
The selection of the setting is almost always left at the default value or some value that the engineer or technician has used in the past that has worked.
There are good reasons to use or not use the protective function as mentioned, but you almost always need to see what the hairy edge of what normal operating conditions are, before deciding what the "right" value is IMHO.
Sort of like what differential protection. Ideally it should be zero, but there are lots of normal things that occur that have to be taken into account when deciding what zero in/out should really be.

 

[Hv&Lv]

Thanks, I appreciate the practical application of how this works. At the moment I'm more interested in the broader picture of what's going on and how to set this up properly, as well as an understanding for the next time I run into this, but I've attached a screen shot of the last trip event for to help explain what's been happening.

Sorry,..
Here then..

https://cdn.selinc.com/assets/Literature/Publications/Technical%20Papers/6710_ManMadeFaults_JL_20160208_Web4.pdf?v=20180608-173402