Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Modern protective relays 2

Status
Not open for further replies.
This is a general discussion for the fraternity to ponder and put their views.

In the era of digital technology, the development of relays have gone far beyond what is really required. As many features are being added to the relays they are getting too complicated. Some of the relay manuals are now reaching to more than one thousand (1,000) pages! In the internet age where the attention span of the people is getting reduced, it is serving the purpose by publishing such a huge instruction manuals?

The protection relay itself is loosing its focus from the basic functionality. In my opinion the time has come for the relay manufacturers to adapt different strategy. Make the protection relay simple and easy to understand the set. The relays can be handled by an average qualified technician. This would benefit all in the industry.

As the technology is getting cheaper by the day, it has become a trend for the relay manufacturers to make it more fancy and sell with these features.

Any thought on this view or anyone who share these views?
 
I happen to mostly agree- but also somewhat disagree.

In some POCOs/substations relays are called to do a number of very complex functions which necessitate a plathora of features and logic sequences being used as offered by modern relays. Add "smart grid" to that equation and the relay is using it all.

On the other hand in most other applications (especially retrofit) I think modern relays are grossly over featured. It is those excessive features and complex user manuals that IMO create reliability risks because the odds of something being set incorrectly increase- not to mention through programmer confusion.
 
Only enable and set what your application requires. Sure, our standard line relay has an IM that runs to 1300 pages. But many of those 1300 pages are things I'll never need to read. Compared to line relays 20 years ago, the relay has a wealth of features that allow us to do things in the line relay that either couldn't be done or required separate relays; often those external functions didn't have any redundancy. For every one of those great to have new features there's a handful that I have no use for but other people do. If I ever need any of those features, they will be there in the relay ready for use.

There's also something to be said for working with as few different relays as possible. If I use just a handful of functions in a relay I've very familiar with I'm less likely to make mistakes than if I try to "optimize" the relay selection and work with an unfamiliar relay that is a "better" fit for the limited application.
 
The line relay is one of the most complex relays that protections engineers use. And granted it is complex, but I only need to purchase one relay, and configure it for the application I need.

If you want simple relays, they can be found, just don't specify the top of the line relays that were made for the EHV applications, which is likely to use more of the relay functions.
Does the quarter or half cycle faster really matter for a 100kV line. Maybe a faster breaker would make up for that (if you can justify that).

And david is right, turn off what you don't need.

Simpler over current relays do exist.

 
I agree, but Cranky108- you said it. In order to get that faster processing speed I need a top of the line relay. Why can't an SEL311 have the same response speed as an SEL421?
 
I believe the math processing is different. Which SEL states removes one of the common mode failure concerns, if you use one of each.

So it is not the processor speed alone that makes a faster relay, but also how it processes the information.

If you look at the tables on the relay speed, you can see typical times for different faults.

If you want a little faster speed, order the high speed contacts.

It should be possible to setup a speed vs cost chart, but I don't of anyone who has done that.

But some people have turned to line differential because they want to avoid the setting complexity issue.
 
@JG282, The SEL387L is meant for extremely short line sections such as those between a switch yard and generating facility. It is not practical for line protection as it has no basic functions like auto-reclose, sync check, breaker failure, over current, step distance, ect. But of course I do not know what the OP is dealing with so that may be the right choice after all.


@cranky108, Thanks for elaborating on this. My understanding is that the SEL400 series relays are always faster than 300 series for any given fault, particularly a zone 1 fault, but I could be wrong on that.


Line differential looks good for high speed clearing, but in any case it is always a good idea to set step distance and over current elements into the relay as though you had no differential to begin with. Even redundant fiber optic paths can fail (heck you have people putting the splice boxes near ground level) and Zone 2/3 is needed in the event both relays fail at the remote substation.
 
Have you looked at the 311L? It has fast line differential, and line distance, ground over current, reclosing, sync-check, etc. It is a combo of the 387L and the 311C, with fast solid-state outputs.

I have tended to avoid the 400's because of the complexity issues, not just for my self setting them, but also the training of the tech's, and the testing.
As I said, if you don't need the horsepower, don't buy the top of the line relay.
 
We still supervise our 311L 87 pickups with a Z2 21.

It sure is a slow moving industry....
 
@cranky108: what is the processing speed of the 311L compared to the 421L? I'm with you though, the 300 series are far easier to set.
 
I believe you might be asking about the speeds of the 311L vs the 411L.

SEL does not give the processing speeds, that I know of. They do however give the typical speeds for reaction to different types of faults.

This also depends on the fault current of the fault.

The 87L functions do operate faster than the 21 functions.
 
Zone 1 distance on either the 311L or the 411L will generally beat out the line diff. Mid-line fault so that it gets zone 1 from both ends will generally find the 87L trip occurring between the distance trip and when the breakers open.
 
Thank you and I did not know this. Do you know why that is?
 
@Cranky, yes, 311L vs 411L, but also the 311C vs the 421. I do not know where I read it (and it looks like I might be wrong on this) but I have always assumed that the 421 and 411 were always 1/2 a cycle faster on zone1 and differential. But again, just an assumption.
 
Mbrooke,

I've always thought that the delay you want to be concerned about with a 311L 'type' device is comm delays, not processing... I believe this is also the reason a Z1 21 trip will go faster than an 87 trip (from a comparable relay.
 
The zone 1 trip occurs entirely in the relay. Anything with the 87L has some communication latency in between initial recognition and action; that comm delay may well be the entire amount by which the zone 1 is faster.
 
Makes sense.
 
Coming back to the complex relay subject, you a right when it comes to line differential relay or distance relay, the complex functions are desirable. But when it comes to the high impedance earth fault relay (64 or 87N) or just an earth fault (51G) relay or a LV feeder relay having 51, 51N protection, such a complex relay is not really required.

The problem is that the suppliers are going for "one fit for all" solution. This mind set should be changed. The electrical industry invests lot of money for many unknown future provisions. Majority of the time they will be scrapped untouched along with the one which is in service at the end of the life span.
 
Sorry, but I disagree. We've found that using the smallest number of relays, with the fewest number of designs, results in the lowest life cycle cost. One line terminal design to maintain, one transformer, one distribution feeder, etc. Pick the one relay the has the minimum necessary feature set for the most complex configuration and run with it. That's why we'll install the SEL-411L in locations where we'll never have line differential; the maintenance of odd-ball designs, over the life of the installation, will cost more than the initial cost of the "more than necessary" relay. Part of it for us may be our setting process; line relay settings are generated in a rather complex excel file that writes a text file that can be imported into the setting software with the complete relay settings. It is possible to generate a complete set of settings without doing any editing in the setting software. A lot of upfront work, several years ago, mean that setting a 421 or a 411L is really not much more work than setting a 311C. Today, that is; the first few took many times longer to get the logic right, but now that it works and nearly all of the bugs have been worked out the type of relay really makes no difference. Just because the relay has an overwhelming feature set is no reason to be overwhelmed.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor