3 or 4 pole breakers at TX and G sides 1

[AusLee]

Hi,

I refer to the drawing uploaded on the following site:

http://files.engineering.com/getfil...c3-40b4-8417-87a906861f44&file=generators.JPG

Someone wants to put a 3P breaker at the transformer side, and 4P breaker at the generator side.

I say 4P breakers on Generator and on Transformer sides, because whatever reason for having a 4P on one source would probably be valid for the second source as well.

Notably, if the phases are not balanced and current flows in the neutral, then: if 3P breaker is used at the transformer, current will close the loop though this unbroken neutral, particularly if the connection between earth and neutral is lost for some reason.

Could you please tell me what do you think about this?

 

[jghrist]

If you had 3P breakers on the transformer side, then with the generators feeding the load, the neutral would have two ground points. Ground protection may be adversely affected because of stray ground currents.

 

[ausgen]

Most commonly in this part of the world, 3p/3p is used for switching at the board containing the MEN and 3p/4p used at DBs with a remote MEN.

4p/4p configuration remains acceptable under AS/NZS 3000:2007 but is not encouraged to the extent that it was in AS/NZS 3010:2005. It is true that a remote MEN under 3p/4p will provide a (hopefully high impendance) alternate path for return currents as jghrist indicates but I assume this is deemed to be less of a risk than an installation momentarily losing it's N-G reference.

I must confess that I've seen 500 kVA of data centre load fail during a 4p/4p ATS switching event as under UPS supply during changeover the installation somehow saw it's N-G reference climb to a dangerous level. (Loud noises from the racks, lots of smoke, VESDA operating, fire alarm, building evac, - great fun!)

As a side comment, reasons for switching sources differently can be valid. One source is typically very high impedance compared to the other so fault clearing can be problematic when on the standby source. Fault loop impedance needs to be as low as possible so switching standby source MEN at the ATS can be important if mains MEN is remote.

 

[waross]

I prefer one solid earth neutral connection and 3 pole breakers. Simple, dependable and cheapest.
If there is a reason to do otherwise (And some of the reasons for doing otherwise are valid.) then you will need two 4 pole breakers.
If you have a UPS carry over, then the neutral/earth connection should be in a common location and there is no need for 4 pole breakers.
My rule of thumb is 3 pole breakers. If you think that you need a 4 pole breaker tell me why and we will discuss that possible exception to the first rule.
I agree with ausgen's post.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

BTW

I say 4P breakers on Generator and on Transformer sides, because whatever reason for having a 4P on one source would probably be valid for the second source as well.

I agree. But unless you have a reason, use two 3 pole breakers.
The point is that one 3 pole and one 4 pole has no point.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ausgen]

I mostly agree, but often have a case in my designs where a smallish gen (500 kvA say) will be connected to a sub-board that might be a 50 m run from the MSB containing the MEN. With typical X0 values, the additional impedance in the fault loop along the earth conductor to the MEN and then back on the neutral to the DB with the ATS is sometimes the difference between protection being able to clear a fault or not.

Over here (aus) our NEC (AS/NZS 3000) seems to stumble it's way through the subject but essentially indicates that for standby source connections:

"The MEN connection shall be made within the installation at the switchboard to which the electricity generation system is connected"
and
"The incoming neutral to a MEN switchboard shall not be switched."
This second point seems to discourage 4p/4p switching but there is a later 'deemed to comply' reference to another standard (AS3010) which clearly accepts this option. (?!)

The downside of having the two MENs in circuit under 3p/4p is then treated by:

"When connected to a distribution switchboard and operating on the alternative supply, the submain neutral and protective earthing conductors may be operated in parallel through a remote MEN connection provided that—
(i) conductors are not overloaded by current sharing; and
(ii) conductors are suitable for the maximum calculated fault current; and
(iii) the nominal size of copper earthing conductors complies with the requirements of Clause 5.3.3; and
(iv) the current-carrying capacity of neutral conductors shall be not less than that of their associated active conductor."

 

[Handlin]

The issue of three or pole is shrouded by myth, with most myths focusing on issues with ground fault. If the wrong scheme is chosen there will be issues. Choosing the right system is determined by whether or not the generator neutral is bonded to ground at the generator. See attached file for illustration.

For small standby generators located close to the service entrance 3 pole or 4 pole schemes have been successfully implemented. For multiple generators (Paralleling Switchgear), the generator neutrals are almost always solidly bonded to ground or grounded through a grounding resistor. This application requires 4 pole scheme.

If you use a three pole scheme with the generator neutral bonded, issues with ground currents can cause problems. If you use a four pole scheme and do not bond the generator neutral, the neutral voltage will float while the load is on generator. Recently I was informed of a situation where TVSS failures occurred with a 4 pole ATS because the generator neutral was not bonded to ground.

As for the UPS application mentioned above, UPS applications should use a delayed (or closed) transition transfer scheme. In the delayed center "off" position, the ATS load neutral is floating. The best solution for the problem with the UPS is to fed the UPS with a 3 wire feed (no neutral) and bonded the inverter neutral to ground as per IEEE recommendations.

 

[Rodmcm]

Hi Hanlin
Could you direct me to the IEEE inverter neutral to ground reference please