IET logo
 
IET
Decrease font size
Increase font size
Topic Title: Let-through energy of cascaded fault current limiting breakers
Topic Summary:
Created On: 23 October 2012 11:38 AM
Status: Post and Reply
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 23 October 2012 11:38 AM
User is offline View Users Profile Print this message



Spark6

Posts: 17
Joined: 17 May 2012

Assuming you have 2 circuit breakers in series, both are Fault Current Limiting Type. The Upper breaker is being used to protect the lower breaker, which is installed at a location higher than it's native fault rating. Both breakers have peak current and let-through energy data.

The question is - how do you determine the let-through energy of the lower breaker? The let-through data is usually only provided up to the native breaker rating, which is exceeded in this case.

For an actual example:

Upstream Breaker is a Schneider NSX160N (TM-160D) rated at 50kA.
Downstream Breaker is a Schneider C60H (20A) rated at 10kA.

With cascading, the lower breaker can be installed at a location with a prospective fault level of 40kA.

To check what cable size can be connected to the downstream breaker, you need to know what it's let-through energy is at the 40kA level. The let-through energy graph ends at 10kA.

So how do we know what the let-through energy is for the lower device?

Just using the upper device let-through data to protect the lower device cable would result is a much larger cable than you'd think neccessary (10mm2 for a 20A circuit).

My guess would be to find the equivalent symmetrical rms current let-through by the upper device at 40kA, then use that figure to index up the lower device let-through curve.

To do this, use the peak current cut-off graph of the upper device to find peak asymetrical current at 40kA sym. rms. This gives 18kA peak Asymmetrical. Find the point where this value intercepts the non-limited line (stepped diagonal line on the graph - converting between sym and asym), then trace down vertically to find the symetrical rms equivalent = 10kA sym. rms.

Put 10kA into the lower device let-through graph and you get a let-though of 40000 A2s - easily handled by a 2.5mm2 cable.

This question would be identical for the case where the upstream device is a HRC fuse.

Does anyone know if this is how you work this out? I'm not sure if it is OK to do this with 2 fault current limiting devices in series, but even if it produces a worst-case answer, then that would be OK.
 23 October 2012 11:51 AM
User is offline View Users Profile Print this message


Avatar for MAXMIRA.
MAXMIRA

Posts: 78
Joined: 25 January 2011

Spark,
I know this is not much help but i would try contacting Schneider Technical to see if they can provide you with more data.

Dave
 23 October 2012 01:36 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19747
Joined: 23 March 2004

OK - do check this, but an NS compact backing up a C60H MCB has a cascading effect up to 40kA such that the MCB is protected at 10kA.

For real design, you may assume the downstream device plays no part in interrupting the fault, so you would size the MCB circuit based on the I2t coming through the NS compact breaker (at the fault level of the downstream circuit) but you'll get larg(ish) conductors

In order to size the cable supplied by the MCB you would use it's rating Icu ie 10kA, as your text shows and a 2.5mm2 will be fine

The test standard requires there to be a cable in circuit and for a 20A MCB that's a 2.5mm2 conductor. The standard however also allows the cable to be larger than that, so if it is, it will have to be stated by the manufacturers.

If you need I2t data then this is only available from the manufacturer.

generally though, it's pretty much as you've described

Regards

OMS

-------------------------
Failure is always an option
 24 October 2012 12:58 PM
User is offline View Users Profile Print this message



Spark6

Posts: 17
Joined: 17 May 2012

Thanks OMS.

Do you know what standard covers this situation? Do you know of any articles or references for this?

I found a good beama article that seems to cover what I am after:

http://www.beama.org.uk/downlo...4CE3-8AC60B8D0A030CD3

but other than that the internet is bare :-)
 24 October 2012 12:59 PM
User is offline View Users Profile Print this message



Spark6

Posts: 17
Joined: 17 May 2012

By the way - I reasonably sure the method I described above is NOT appropriate for 2 fault current limiting breakers. You need to take the worst case (Let-through at Icn of lower device) for values above Icn. Theoretically you might be able to land on a value lower than Icn using my method.
 24 October 2012 01:34 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19747
Joined: 23 March 2004

I think BSEN 60947 series is what you may be after.

If you want to deduce the current limiting effect of the downstream breaker in conjunction with the back up protection of the upstream breaker it is normally determined experimentally by the standards above - manufacturers will (may) give you access to the data in terms of I2t for the downstream device. Your method isn't wrong just as long as you recognize it may lead you right to the edge of the envelope. In practice, stick with the fault rating of the downstream device if (and it's a big if) the downstream and upstream combination has been tested by the manufacturer

That said, if you are then sizing cables on a much lower known I2t on this basis, you only need a minor change in selected breakers by say a contractor and you could really be in deep doo doo.

regards

OMS

-------------------------
Failure is always an option
 24 October 2012 04:31 PM
User is offline View Users Profile Print this message


Avatar for timothyboler                                      .
timothyboler

Posts: 230
Joined: 25 July 2008

This stuff isn't crystal clear to me when you have back-up protection where BOTH protective devices provide assistance in reducing the let-though energy - in effect a double limitation scheme.

The thing is you can have back-up protection with or without the downstream breaker providing assistance (by definition in the standard) and therefore providing a secondary limitation.

However BS EN 60947-2 Annex A.3.3 states that:

"For all values of over-current up to and including the short-circuit breaking capacity of the association [upstream breaker or fuse], C1 [downstream breaker] shall comply with the requirements of 7.2.5 of IEC 60947-1 [basically have the ability to make, carry and break short-circuit currents and exhibit i2t let-through energies in accordance with its intrinsic making/breaking capacity]..."

This seems to imply that whenever you have a cascaded back-up protection scheme the downstream breaker has to exhibit its rated characteristics at its intrinsic breaking capacity regardless of the breaking capacity of the upstream device.

However if you had back-up protection whereby the downstream breaker wasn't expected to break currents and just sit there and take the energy let-though (i.e. not providing assistance) it obviously wouldn't comply with A.3.3?

So in short, as Spark6/OMS said, I believe that you should in general just take the let-though energy as stated for the downstream device at its rated Icn.

Incoherent ramble over , Tim

-------------------------
Everyone loves a fireman - but hates the fire inspector.
 24 October 2012 04:43 PM
User is offline View Users Profile Print this message


Avatar for OMS.
OMS

Posts: 19747
Joined: 23 March 2004

However if you had back-up protection whereby the downstream breaker wasn't expected to break currents and just sit there and take the energy let-though (i.e. not providing assistance) it obviously wouldn't comply with A.3.3?


Well, it is going to try and open, and in many cases will fully open as in conjunction with back up protection you may need discrimination as well.

In that case, there will be further limitation but you can only prove that experimentally

regards

OMS

-------------------------
Failure is always an option
 25 October 2012 02:30 AM
User is offline View Users Profile Print this message



Spark6

Posts: 17
Joined: 17 May 2012

Thanks guys.
 25 October 2012 07:16 PM
User is offline View Users Profile Print this message



stableford

Posts: 64
Joined: 04 April 2006

I have come a cross this a few times, or at least some thing similar.
There are a couple a approaches I've seen done;
1, Built it, test it(PFSC) and pray.
2, Model it, Amtech or similar, blow in the supply impedance, and do a discrimination study, then build it, and then test it.

I prefer the second, because its less upsetting to the boss.

Ive found this issue occurs more on the industrial install's, due to the large supply sizing.

A similar situation also occurs on this side of the pond as well, except here they dont do the testing after building, so they have to engineer it up front.
Statistics

See Also:



FuseTalk Standard Edition v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.