Selective coordination 5

[Goombah]

Situation:
Main control panel has a breaker rated at 1600A and several downstream 200A breakers with 100HP motor loads of 113FLA
The main breaker has Adjustable trips for Long, Short, Instantaneous and Ground fault.
The long, short and GF Have time adjustment as well
This breaker comes with everything set at minimum except the long trip set at max (1600A)
When a motor load was started, the main breaker tripped on instantaneous. The branch breaker did not trip.
The inst setting was set to a higher level and that solved the problem for testing purposes. What I really need to know is where to find documentation/ references which say this is alright to do. They are square d breakers and I can find the trip curves but no guidelines, or better yet, calculations for determining the LSIG settings to use for the main breaker. Any tips on selective coordination would be appreciated.

 

[dpc]

It's a somewhat complicated subject. Best references are actual books - IEEE Red Book and IEEE Buff Book.

For relaying, I'd recommend J.L. Blackburn's Protective Relaying book.

Coordination always requires a tradeoff between protection and coordination - in general these goals are directly conflicting.

In your specific case, a main breaker with instantaneous trip may never fully coordinate with the downstream devices.

 

[wbd]

To further complicate matters, the protective device settings that may be optimal for coordination, may not be optimal for protection against arc flash.
What really needs to be done is to have someone knowledgeable in protection and coordination look at your system.
Those are great references listed by dpc to get you started. There is a reason why it is known as "The Art and Science of Protective Relaying"

 

[HCBFlash]

Contact your local Square D supplier -preferrably the one that actually supplied the panelboard for your facility- and arrange to have a Square D representative visit you at your site. The Sq.D rep will be able to get you their guidelines for settings, and (if you can get through to an actual engineer of theirs), probably get you copies of some of the applicable sections of literature such as IEEE stds, papers and books.

Remember that your main C.B. has more to do than protect your motor, and other plant equipment may well be more important in determining what settings are acceptable.

Please let us all know how this works out for you! and don't forget faq731-376

 

[Goombah]

In case anyone wants to share in my frustration, The main breaker is a Masterpact NW20H with a Micrologic 6.0P electronic trip unit and a 1600 amp sensor plug. The Blower panel breakers are KH 36200. All of which are Square D. The loads serviced by the 200Amp breakers are 100HP blower motors and their control circuitry.
There are 2 main panels and one has 10 blower panels the other has 16(which brings another question to mind. Is the main breaker rated high enough if all blowers were to be on? according to my calculations, no. When I questioned this, I was told it was allowed as long as "provisions" were made to prevent that full load(per the NEC) I have not been able to locate this exception in the 2005 NEC)The likelihood of more than 10 of these loads running simultaneously is slim anyway but shouldn't it matter? Anyone know of this exception?
Thanks dpc,
"In your specific case, a main breaker with instantaneous trip may never fully coordinate with the downstream devices."
I read that the Instantaneous trip could even be turned off for the upstream breaker and still not lose the protection that the other trips provide.(See Square D data bulletin 0100db0403r5/o5)
Thanks HCBFlash,
If only our Square D rep and/or engineer had been forthcoming with that information. I had already gone over everything I could find on the subject of coordination. including the trip curves for both breakers I just wish I had access to the IEEE books mentioned.
I'm not an engineer yet but I'm learning!

Whatever you are, Be a good one!

 

[dpc]

The ability to turn off instantaneous trip may be an option depending on the classification and UL listing of the circuit breaker. If you can turn it off, this will improve coordination with the downstream feeders, but this also increases tripping time for main bus faults and will cause higher levels of arc-flash energy. That might be an acceptable trade-off - it just depends on the situation.

As I said before, it's always a compromise.

 

[HCBFlash]

This stuff is a blast to work out! Are you familiar with panel layouts, load characteristics, 1-lines, load schedules, etc like are used in an engineered set of construction drawings? I'm not in a good spot these days to recommend specific reading material with good examples. I've got plenty of good stuff, but it's in storage right now. I'm really hesitant to post my email address here, but I've got email notification of this thread going.

Your motor and load characteristics mean everything to determine starting current curves . If you're using freq drives, Y/delta, motor/generators, isolation or local transformers, etc, this will all help reduce your startup currents, but electronic drive do interesting things to the (true) running loads. Assuming that your blowers load their motors pretty well, you've got some long startup times due to the high inertial always present mechanical load.

It sounds like you've got quite a bit of loads there on that system. The number of loads operating typically and those running intermittantly however apparently don't add up to much of a steady load though, or you'd be facing tripping quite often. The "NEC" (NFPA70) can be a useful document to figure out your loads, as far as how they need to be wired and protected adequately, but only you can know what your connected equipment operating cycles are, both typical and during times when your facility is really busy. NEC addresses this somewhat in defining and having articles regarding intermittant loads. A great online resource is at

http://www.lmphotonics.com

Please let us all know how this works out for you! and don't forget faq731-376

 

[oldfieldguy]

There's an old rule of thumb that I've used for years that says "You can't coordinate instantaneous overcurrent devices"

It's true, too. I had to respond to a client who was completely befuddled as to why he'd lost a whole motor control center instead of the branch circuit which actually had the fault. I pointed out that there was an instantaneous trip setting on the power circuit breaker that fed this MCC, and that it had flagged on the trip. It was set at four times rating, 3200 amps. He pointed out that the instantaneous trip setting of the branch breaker was 10 times nominal, 1000 amps. Subsequent tests at his request showed that both devices operated properly.

The situation, though, is that his fault out on the branch circuit was sufficient to operate BOTH instantaneous elements, and the upstream breaker was just a bit faster.

The moral of this sad story is that if you do set an instantaneous setting on an upstream device, make sure it's above the available fault current of devices downstream of the next devices with instantaneous settings.

Incidentally, GE's newer microprocessor relays muddy up the "instantaneous" definition by allowing a definite time delay to be set on instantaneous overcurrent protection. The precision inherent in newer protective devices allows some variation in traditional coordination applications.

old field guy

 

[ScottyUK]

Sounds like that would change the instantaneous trip into a definite-minimum-time function - have GE re-invented the wheel?


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[oldfieldguy]

ScottyUK--

That's what I thought. The definition I always used for "instantaneous" in system protection was "no intentional delay". After all, there were time-delay overcurrents with fixed delays. Now GE makes it a capability of their "instantaneous" elements.

I'm not saying it's a bad thing. I've used them myself. But it does cloud the definitions.

old field guy

 

[dpc]

Scotty,

It's a terminology problem. It's called an "instantaneous" element because that is the term commonly used in the old electromechanical relays. Then an option is provided to add a time delay to the instantaneous element. Same as definite-time, but that terminology isn't widely understood or used in the US.

But with low voltage breakers, it's not so simple since the breaker standards require instantaneous tripping in many cases, such as typical molded-case breakers. These must have instantaneous trip.

 

[ScottyUK]

dpc / oldfieldguy,

Thanks for the clarification.

 

[davidbeach]

Any molded case/insulated case breaker and many of the newer power circuit breakers all has instantaneous trip, even if you can order them without instantaneous or you can turn the instantaneous off. What they have is a fixed instantaneous override that may be at the interrupting rating of the breaker, but the instantaneous is there. Any time you can set the instantaneous higher than the available fault current, you essentially don't have an instantaneous trip. That used to be a great coordination tool, but doing so certainly complicates the arc flash considerations.

The definition of instantaneous has become muddied in recent years, starting with numeric relays. In the relays, many instantaneous (50) elements have an optional time delay that turns them into definite time elements. When instantaneous was a magnetic action, whether in a LV circuit breaker trip mechanism or in an electromechanical relay, the timing was what it was, and could have some inverse relationship to current. Now there are more options and the terminology hasn't necessarily kept up.

 

[tommom]

Another potential issue to weigh in on this is whether the breakers are intended to function as a series-rated system. With a current-limiting larger device protecting a lower-rated downstream device, there are situations where the upstream breaker is SUPPOSED to trip first to protect the downstream circuit breaker.

 

[Goombah]

Thanks everyone, You've been a really great help.
I did find in the literature from the main breaker maufacturer that if the instantaneous trip can be turned off it should be (for coordination purposes) The short time pickup and delay is still in place. Of course the delay would have to be kept to a minimum for arc flash considerations. Now the ground fault pickup is determined by first getting the available short circuit current available right? and then just match that up with the breaker trip curves.

"Whatever you are, be a good one!"-Abraham Lincoln

 

[dpc]

Goombah- everything related to selective coordination is a compromise. The "safest" system is one with everything set at minimum.

Ground fault settings are complicated by the fact that most breakers downstream will NOT have any ground fault detection, so you risk tripping an entire switchboard (or similar) for a ground fault in a downstream motor.

Get the IEEE Red Book or Buff Book and run through the examples in there on low voltage coordination.

 

[Goombah]

I don't think that would be a problem in our application. I think I'd prefer it if only the downstream tripped but at least that would be safe. Most of what I've read seems to point toward fuses having the best coordination capabilities but you go with what the man asks for.

"Whatever you are, be a good one!"-Abraham Lincoln

 

[dpc]

Fuses have some advantages over molded case breakers in coordination for high level faults. However, they do not provide sensitive ground fault protection and can single-phase your motors.

There are many trade-offs between fuses and breakers.