Protection relays

[wolfie1a]

We are looking at upgrading some protection relays. We are looking at SEL, Siemens, ABB, Eaton....... Trying to determine who is the best? Looking for feedback. Any help would be greatly appreciated.

 

[Sn00ze]

For my province we stick strictly to GE and SEL for most of our subs. I'd say whatever is used most in your industry as swapping relays between buildings in a bind comes in handy as well as your supplier would have the relays more readily available.

 

[cranky108]

Scotty, you would see some things like when we read a manual written in British English. A little different, as I don't believe SEL has any manuals in anything other than American English. But they can be difficult to read for the first time (it is hard for me to tell as I have seen so many of them).

Start with a simple relay manual like the 551, to get the terminalogy, befor moving on to the more complicated relays (wish someone had told me that years ago). Although I assume other relays manufacturers are the same way.

Beware there are a few quarks, and some real difficult relays in there line up.

Maybe someone needs to develop a list of starter relays for young, and older, engineers to develop there understanding of manufacturers logic upon.

 

[slavag]

Make mix.
SEL with ABB, Siemens with GE, Beckwith with SE.
it will be interesting.
btw, we make line protection with REL650 as 21 and SEL311 as 87L.
In one project, Im used SEL2506, nice device

 

[ScottyUK]

In Stafford? That's the home of GEC's St. Leonards works... That can't be just coincidence. I wonder if their rep is an ex-GEC or ex-Alstom employee? I may do some more digging, although it would be hard to justify moving away from our installed base of Alstom relays without some fairly massive benefit. Thanks for the heads-up David.

 

[stevenal]

Here's a way to play with SEL relays without having any: [URL unfurl="true"]https://www.selinc.com/relayrack/
[/url]

 

[davidbeach]

"Hard to justify moving away..."

The way I put it is that we'll be using SEL relays until SEL forces us to make a change. I just can't see somebody else coming out with something so much better than what we're presently using to make the pain of changing worth while. And it would be a huge pain, pain for the protection engineers, pain for the relay techs, pain for the substation design engineers, bah. So, it's real easy to see why somebody else would want to stick with what they've already got installed.

 

[cranky108]

David, what did you use before SEL? Why did you change?

Yea I know where you worked before.

There was a reason many of us switched to SEL.
I tried Alstom relays years ago, because of one of there sales engineers. They weren't bad, but the tech's hated them, except for the bus diff. relays.

 

[ScottyUK]

Hi Cranky -

I find Alstom's relays are technically very capable and must be designed by some very smart people. The techs probably hate them because the manuals make them hard to commission, especially for the more sophisticated functions. I've been in this game a little while and I find the manuals are very hard work. My techs feel the same. Personally I hate the programmable logic editor for the Px4x relays because it is incapable of doing anything tidily. Nothing aligns, and nothing can be forced to align. It wouldn't have taken much effort to make it a good editor, and that's very frustrating. Visually it makes my work look like crap which bugs the hell out of me because I do take some pride in making drawings presentable. I have to make my own work look half-decent because I frequently send contractor's work back covered in red pen if it is untidy or badly presented: let he who casts the first stone and all that biblical stuff... ;-)

 

[davidbeach]

The SEL decision here was made long before I started. When, in the late 80's, it was time to do something other than the HZ or KD as line protection the early SEL relays were one of a few different things that were tried. We still have a bunch of SEL-121G relays and have only a very few other examples of the other things that were tried. By the time they moved onto the SEL-221G there was no looking back as far as transmission protection was concerned. It took longer to decide that brand-x distribution relays weren't worth the effort (at least a couple of different brand-x labels).

 

[cranky108]

It's true that the manuals must be very difficult to write. I think what was some of the bother with the Alstom maunals was one manual for several different relays. And if the truth be known, the only difference in the relay models might have been the programming. But the manuals should have been for the specific model of relay.

Some of the other things were the size of the manual, and the unfimularty with the symbols in the manual.

I think they liked the bus diff. relays because they were simple electromechinical, and very small because the MOV's were not in the same case as the relay. The MOV's could be placed at the bottom of the panel, in the area that one would not typically want to place much else.

Some of the SEL manuals are also difficult to read, and what comes to mind is the 387 manual and the difficulty I am having with consultants and the trip equasions. The issue is the method eluded to in the manual will not produce the correct phase targets on the relay (It displays all three phases). And while it makes little difference which phase on a three phase transformer is faulted, the people in operations want to see them (I'm guessing they want to over analize the problem).

I know stupid issue.

 

[wolfie1a]

Appreciate all the replies.

 

[MKFPE]

I have adopted SEL only for our system.

Their customer support is absolutely the best of any company I have dealt with and I have been in this field since 1975.

You should seriously consider their high-interrupting capacity output contacts for your trip outputs.

On our system we had several cases of older medium voltage air-magnetic breakers opening slowly and melting the Basler relay output contacts which are not rated to interrupt trip coil current.

Relays had to be replaced and obviousy we went to work on breaker maintenance, and in the end upgraded all of them to new vacuum breakers.

But if this happens in the future, the high capacity contacts will save a relay replacement.

 

[GherlanC]

Hi all,

In the couple of minutes I have read a really nice discussion that you composed here.

wolfie1a, can you tell us what is the application where you want to upgrade relays? I could help you with an advice for appropriate SEL relay and arrange to loan the SEL device.

 

[DTR2011]

I'll throw in my $0.02 about SEL relays.

I like them. Have dealt with them for 20 years. I started with SEL, so I'm familiar with them. All positives about service, support, in my experience are all true. I have also found the pricing to be more than competitive.

The only thing (as a commissioning guy), that drives me nuts is how things can change from relay to relay - series to series.

Example would be naming conventions for 51 (51P / 51S) elements. Commands used in the terminal mode to accomplish the same thing. Opto Inputs that on some models that look like a coil (voltage testing wise) and others that appear open (high impedance).

With all of that said, I am on a project where there are 20 year old SEL relays used for bus protection, while the feeders are still EM. We had a feeder lock out the other day. Via smart programming of the bus relay (extra OC elements used to trigger event reports), I was able to plug into the 1994 relay (and settings) to help the customer identify phase of fault (relay trigger an event). Additionally SEL relays can always communicate via a terminal emulator program (Hyperterminal, etc) to access settings, SER, history - everything. There are better ways to do this, but this method always works.

I have had experiences with other brand relays, where there was a program for settings, one for logic, one for event reports. It was old 16 bit stuff and if not installed in the correct order, nothing worked. Many still keep an old, separate computer with these programs installed, just for these reasons. The terminal emulator rule was a well though out idea 30 years ago.

 

[davidbeach]

DTR - interesting comments. The fact that everything that needs to go to the relay or come from the relay can be in text format is very useful. We have developed a setting process (presently in excel, being rewritten in VB) that produces a text file that makes its way to the relay without further human editing. That eliminates a lot of places where things can go wrong between setting development and settings in service in the relay. But every time they change the name of something we get a bunch of errors when the old setting names aren't recognized by the new version of the relay. It would also be nice if they'd gone with the C37.2 definitions of N and G, but oh, well, life goes on. But the support is great and the product warranty can't be beat.

 

[DTR2011]

I will add another feature of the SEL relays that I find very useful in commissioning and relay testing.

SEL calls it a "Relay Word Bit". It is a logical element in the relay (an input (IN101), an ouput (Out101, TRip, 51PT, 32QF, etc). If the relay word bit is listed in the book, this can be displayed logically (show me this bit "n" times in the terminal display), or freely mapped to a contact output, or displayed via an LED.

Testing procedures for EM relays often called for blocking a contact open or closed (with a $20 bill to remember to remove). A SEL relay can be programmed to act a same way (via an unused output contact (Testing)), to either replicate existing testing standards / procedures, or for troubleshooting & commissioning, keeping the main tripping contact logic intact. This is extremely useful when trying to get ones head around the logic of both the relay algorithm and the settings applied. I have found that for both expediency and clarity, being able to "see" what the relay sees is very useful.

One further note about SEL relays and instruction books. In the SEL 251 series instruction manuals, there is an application note on how using both a bus and feeder arrangement of relays (via different setting group selections), a bus and feeders can be protected for a failure of a feeder relay (via 2 sets of self fail alarm contacts), can tell the bus back up relay there is a failure (change settings and become more sensitive) and via Aux relays and wiring (no more complicated than a lock out relay scheme), to trip the feeder (and only that feeder) that has a failed relay. With additional wiring to this scheme (a "Feeder Relay in Test Mode" Switch, this also allows for feeder relay testing / maintenance while still providing adequate protection with the feeder in service. I commissioned one of these schemes, as outlined in the SEL 251 manual when I was still a greenhorn. Years, later I see the beauty of this as it also allows the maintenance of a feeder relay in service. This scheme could be applied to any digital relay with proper design (and 2 alarm outputs).

As has been mentioned earlier, if one has not been exposed to a SEL relay, it can at first seem intimidating. When written out, it is an Algebraic math argument (TR = 51PT +(OR) 51GT *! (AND NOT) 50L). Algebraic order of operation matters.

I've seen the GE Brick system and challenged a qualified GE applications engineer to really show me how it works. It does and the hardware for the brick (ADC - 61850) is very robust, the fiber patch system is logical, etc. GR UR series relays do have some very compelling features. I think with a brick type solution, for example, it would be possible to add 87B (Bus Diff) to nearly any circuit and that would help to solve the arc flash problems for many.

At least here in the USA, there is still a reluctance to take that next step. One reason may be that when a mis-operation occurs, there is hell to pay for the folks in the field (and others Im sure). With current technology (non 61850), we have test and isolation switches that are (hopefully) available to make checks for trip signals, volts, amps, etc. On a new project where everything could be tested via primary injection, confidence could be gained. On upgrade/brownfield projects, where it is a 40+ year old station (with old, unreliable drawings) this could be a problem. If, for example we already have issues with settings and drawings now, how about with a non standard implementation of 61850 that was not well documented ( in the field) 40 years from now.

Watt and Var consumption (mostly real power) has occured

 

[Electic]

As great as SEL is, I do wish their manuals were (substantially) less than 500 pages. Being on the troubleshooting side of things, we only look at these during panic outages.

 

[HamburgerHelper]

The only things I don't like about SEL relays are that Compass the autoupdater is terrible, too many differences in terms and operations among the different kinds of relays, and too much flexibility for the user. Most applications don't require the amount of depth that the relays offer and this leads to the possibility of mistakes being made. I think the SEL-411L has an option of blocking off portions of the relay that are not going to be used based on the (basic, medium, or advanced use setting). The relay should only as complicated as it needs to be because a tech or another engineer will have to review and modify your settings at one point.

The most stupid thing that I came across in their relays though was the non-explicit output that is dedicated fast differential tripping on one of the differential relays. Instead of showing explicitly in ACCelerator that the fast trip bit is OR'ed with whatever logic you have in place for that output, you only come across this when going through your manual. The relays are great but most misoperations are due to bad settings. SEL's goal should be to only make the relays as complicated as necessary. They have a template maker thing you can buy, I haven't used that but maybe that would help.

 

[davidbeach]

While I can feel your pain, I completely disagree with the bit about "too much flexibility". I wish their default settings were everything off so that the user didn't have to go through and turn everything off before building from there, but I wouldn't want to have to deal with more versions of the relays than already exist. We have our own "template" process for developing line relay settings with the goal of applying one set of standards to a variety of relays; the SEL templates (which they aren't pushing nearly as much as they used to) are specific to one firmware version. That makes it even worse in my opinion.

 

[wolfie1a]

A few have asked what we are doing and why. I have a couple of 15KV to 4160 subs we are looking at upgrading the relay along with adding a remote open/close feature. I think the subs are 20 to 25 years old. The relays 50/51, 51GS and 87T’s are no longer manufactured or supported. We are hoping we can spec out relay that has multiple function, using the same software, easy to program and test with a good 10 to 15 years of service support. We have done some brainstorming in house as to what we want in a relay and weighted each feature. Hopefully as we go thru our relay evaluation we can determine if one mfg has advantages over the other. The open/close feature will probably be a touch screen mounted outside the arc flash zone allowing us to safely operate the breakers.