faulty meter in Sub

[dtawse]

The volt meters in a lot of our subs meter only the red and blue phases and calculate the white phase. These are old electro-mechanical meters not electronic or transducer types. We lost white phase but the meter continued to give a soild (although not quite normal) reading. On paper we have looked at the possiblity of the red and blue phases being wired incorrectly but it does not appear to work. How would the meter appear to work correctly when everything is ok and then show volts on white phase when it was dead?

 

[DanDel]

How exactly are the voltmeters wired? Are PTs used?

 

[dtawse]

I'm sorry i made a mistake, it is an ammeter (getting the info second hand) and the CT's are on the red and blue phases and the white is derived from that. Not sure what the exact current readings where but the current on the white phase was showing slightly different but not enough to be of concern.

 

[jbartos]

Suggestion: What is your voltage system? Is it a single phase three wire system?

 

[dtawse]

it was in a 66kV/22kV substation. I believe the meter was on the HV side. at this point in the system it is three wire three phase.

 

[jbartos]

Suggestion: The current coils of kWhr meters connected in Aaron connection (Blondel's Theorem), i.e. two kWhr meters measure three phase power consumption, are in two phases only. If the third phase is lost, they still measure kWhr consumption.
Aaron connection will show:
kWhr=kWhr1 + kWhr2=Ea x Ia + Eb x Ib + Ec x Ic=Reading1 + Reading2=kWhra + kWhrb + kWhrc + kWhrcoil1 + kWhrcoil2
where
kWhrcoil1 is an energy consumption of kWhr1 potential coil
and
kWhrcoil2 is an energy consumption of kWhr2 potential coil

 

[busbar]

It’s not entirely clear from the description, but comments here are based on a scenario of a single ammeter and a three- (or four-)position ammeter switch.

See page 28 in

http://www.geindustrial.com/products/brochures/sbcat.pdf

Figures 11 and 12 deal with an ammeter switch with overlapping contacts to permit two CTs to monitor each phase in a 3ø 3-wire circuit, depending on switch position. In the problem case, it may be that the states represented in the contact logic has changed yielding the false reading. [The two-CT and ammeter-switch application will not work correctly in a 4-wire circuit with any neutral current.]

 

[jbartos]

Suggestion: The original posting refers to a calculation of the voltage in the white phase. The above posting refers to GE link that includes GE catalog for SB control and transfer switches. There is still an item to be accounted for, namely, which hardware does the calculation of the voltage in the white phase?

 

[DanDel]

Typically, for a system measuring only two current phases, the third phase is 'calculated' by summing the two measured phases and reversing the polarity(usually this is done with auxiliary 1:1 CTs). If there is no ground current, this is an accurate method of estimating the third phase current.

 

[peterb]

Actually, the third phase current measurement is made by simply inserting an ammeter in the common lead of the two CTs - this measures the vector sum of the two phase currents which in the absence of zero sequence currents is the value of the third phase current. This applies whether an ammeter switch + one ammeter, or three separate ammeters are used. As DanDel points out, this gives a valid reading for a 3-wire circuit with no neutral or ground return.
As far as voltages are concerned, the same principle applies for the open delta PT connection, where the third phase-phase voltage is not measured directly but is inferred from the vector sum of the two measured phases.

 

[dtawse]

The system is definently three wire three phase, no neutral conection. As DanDel pointed out the white phase is derived by taking the sum of the red and blue phase and reversing the result.
I am a trainee protection tester and have been asked to do some research on this. Ther powers that be basically want to know if there is a fundamental error in metering the current this way or if there is a problem at this particular substation.
Zero sequence current should not be caused by an open circuit on the white phase and when i draw a circuit diagram and try to follow it, it appears the white phase should read zero but i can't figure out why it did not.

 

[busbar]

Smeaton’s 1977 Switchgear and Control Handbook §14 13-14 contains basic information similar to the GE reference. For a 3-wire circuit, the middle phase current is measured by connecting the ammeter to the polarity terminals of CT secondaries on the two outer phases, with the non-polarity terminals of the CT secondaries connected together.

Common sense and the usual CT-handling procedures apply.

 

[DanDel]

dtawse, this system can be tricky if not connected correctly. Please explain the CT arangement, and are there auxiliary CTs or are they connected as peterb explained?

 

[peterb]

dtawse - in drawing out the circuit, remember that the red and blue phase currents must be added vectorially to give the remainder white phase. If the white phase primary is open circuited, then the red and blue currents will be 180 degrees apart and equal in magnitude, summing back to zero - which is the value that the white phase will read if it open circuited.

 

[jghrist]

As perterb said, the red and blue currents will be 180° apart with white open. If the meter on the white phase is connected to subtract instead of add the red and blue phases, you will get double the red or blue phase current instead of zero if white is open. With balanced loads, you will read 1.73 times the red or blue phase current if they are subtracted instead of added.

 

[dtawse]

I am working in another part of the state at the moment but will have to physically check the connections or get someone to do it for me and I will post the info ASAP.
As you pointed out when added using complex numbers or vectors the white phase current should show zero but it did not and it was not significantly off normal opperating load. We already calculated the 1.73 factor if the red and blue phase are added, this was ruled out by one of the senior testers.
as i said as soon as i can get the relevant information i will post it. your help has been much appreciated.

 

[busbar]

I apologize for the mistake, but I thought the installation contained one ammeter, an ammeter switch and two CTs on the outer phases. Are there 3 CTs involved?