Earth Leakage

[colohara]

I have building that keeps tripping its main ACB, I have spoke to the manufactures who have advised to monitor the earth leakage.
Is there an acceptable limit for earth leakage from a building back to the transformer.
The breaker is and MG 1600amp with a str58 u control.
The earth leakage for the complete building is 2.5 amps maximum the building pulls in 400amps on each phase.
the building is used as offices and also has a small datacenter.

 

[roydm]

Should you really have earth leakage for a whole building?
I have seen this before where a fault to ground (10 Amp fuse) tripped a whole factory. The engineer finally admitted it was a design mistake.
Roy

 

[Zogzog]

What type of breaker and trip unit?

 

[ScottyUK]

The trip unit is an STR58U, a fairly common Merlin Gerin part. You probably see them with a Square D badge. It fits a MasterPACT ACB, which was quoted as 1600A.

I agree with Roy - a low E/F setting at source is troublesome in this kind of application. It is impossible to achieve coordination between the E/F element and the MCBs or fuses protecting final circuits unless the E/F level is raised to a higher level.

We have a building which originally had a similar E/F scheme: the cheap & nasty luminaires often failed to earth and the fault was cleared at the incomer, not the final circuit MCB. From memory we set the E/F protection on the incomer to 50A or so with a definite time and have had more problems. We meet all the clearance times required by the UK regulations. The designer on that job also admitted it was an error to have set the E/F so low.


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

If we learn from our mistakes I'm getting a great education!

 

[Kiribanda]

colohara,

From your post it seems to me that your LV installation is "SOLID GROUNDING". So I should say, this problem is one of the disadvantages of solid ground systems comparing to high resitsnace ground systems.

As ScotyUK pointed out, unless you have a good "coordinated" earth fault protection scheme with the downstream feeders for the main ACB, you will continue to have this nuisance tripping. So if you cannot implement at this stage such a coordinated earth fault protection scheme for the whole LV network, I too support the idea of increasing the earth fault setting of STR58U trip unit to a "considerbly high value" to avoid tripping the main ACB.

Hope this helps.

Kiri

 

[ScottyUK]

Oh, I wish there was an edit function...

"...and have had more problems" should read "...and have had no more problems"


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

If we learn from our mistakes I'm getting a great education!

 

[ausgen]

Earth leakage through the SMPSs on the office PCs and data centre rack equipment might be a big part of the problem. Along with increaingly leading pf comes the leakage through all these capacitors. If you have small UPS throughout the building these add up to 5 mA each also. In a large building, it's easy to build up a few amps.

Sometimes it's just not feasible to include EL protection at the incoming MSB. From memory, the Schneider gear only lets you range to around 1A(?)

 

[davidbeach]

Must be one of those IEC/ANSI things but I can not imagine even trying, or why anybody would want to try, to have service switchgear ground fault set less than several hundred amps. If every single breaker on the whole service had ground fault, maybe, but even then it would need to be many amps at the service entrance. Ground fault on everything would just add a tremendous cost for very little benefit.

 

[ScottyUK]

Hi David,

My opinion is that it is a function which is present on most of the mid- and high-end tripping units for ACBs and the larger MCCBs, and when a function is present the designer feels some irrational need to enable it. The manufacturers are partly to blame too because they give a limited range of adjustment on the integrated tripping unit which can result in unrealistically low setting of E/F when a medium sized breaker in the 630A - 1000A range is used as an incoming device. The better solution would be to use a shunt trip and an external E/F relay with a more suitable setting, but the self-powered tripping units are _so_ convenient.

Of course that doesn't excuse a designer adding a CBCT and relay over and above the integrated tripping module as happened in our case: that was just a bad design!


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

If we learn from our mistakes I'm getting a great education!