Paralleling LV Transformer Solidly Grounded

You need an experienced power engineer for this sort of system. This is a lot of power and proper means of dealing with the assorted things such as different power factors on each load leg, and what happens with a fuse blown in one leg = the other fuse then sees double the load as you describe it and it blows as well. What about possible feed back/ground loops etc? I do not know enough to answer these, but at this power level there are risks

If you are asking the question of us then I would say no.

Why do you need parallel supply any way, is a single circuit potential load to high for one transformer alone to handle or are you trying to create a redundant supply system?

Is this a homework question, the circuit shown looks a little strange to me?

Yes, it's a redundant system. I want to switch the load from one to another, so I can maintenance one transformer.

There is no problem with paralleling the two transformers during the switching over from one to the other as the time they are in parallel is negligible. If the system is to be run in parallel for any length of time then only one neutral should be earthed.

The main problem to be dealt with is the impedance differences of the two transformers for no two transformers can be truly identical. I have paralleled transformers many times when switching from one to the other however the tap changers need to be on the same tap and the automatic tap changing disabled for the duration of the change over. At the end of the switching everything needs to be restored to the normal condition.

Have done this for up to 60MVA transformers in a substation

Normally only 1 transformer running and the other is standby. This parallel condition happen only during the switching over from normal transformer to the standby unit.

In the picture, the neutral of transformer is solidly grounding due its supply both 3 phase and 1 phase load. I believe your 60 MVA transformer using NGR/NGE right? Same here, in my experience I have done it up to 230 MVA transformer. But the problem happen when I want to parallel small transformer using solidly grounding. I hope I can give you the right description so you can help me.

Well you can do it easily two ways, one is offline where you shut the power down and switch the transformers over using overcurrent protected isolators, the other is with a source transfer switch or similar that switches over from one to the other with no break in supply.

There is a third way what would allow manual switching (remotely using motorised overcurrent protected load break switches) and paralleling live while you switch from one transformer to the other but this is more involved, risky and needs special safety systems, procedures, equipment, etc. Local standards and regulations may allow this but this is high risk unless you know what you are doing and have a good knowledge of the loads and required protection settings and standards.

I have to do it online. Due the load is pretty critical. I have done it online. But parallel condition is only may happen for 5 seconds. Due there is a 51G in both neutral ground of transformer will operate due its excess the settings.

The question is why in normal condition when only 1 transformer operating (Not parallel) 51G not sense any current in ground? The transformer using solidly grounding.

Are the two transformer earths and neutrals solidly connected to a common earth pit? If they are I wouldn’t think you will have any problems.

I’ve installed ten 11/.433kV and two 11/3.3kV transformers in one substation all with their star points connected to the same earth pit.

Yes, they have the same earth pit.

Does your transformer using solidly grounded? Without NEG or NGR? Do you parallel it both in HV and LV?

Probably your tie breakers are not 4 pole type / neutral bus bars of two power supplies are not paralleled!

If true, this could be causing circulating current between the neutrals through the earth connections.

Neutral potentials are unequal due to unbalance in loads supplied but, the tie connection between neutral busbars helps like a least resistance equipotential link between the transformer neutrals.

The tie breakers are 4 pole type. But I will check first about neutral bus bars.

Can you elaborate more about your explanation?

This transformer is using solidly grounding and supply both 3 phase and 1 phase load which make it unbalanced. When the transformer work independently the 51G wont operate. But when it parallel, 51G operate after 5 second delay. Is it safe to increase the setting to 10s?

If the circulating current through earth is more than the pickup threshold set for 51G, the relay is bound to operate whether you set it 5s or 10s delay. It is only delaying by another 5s and not preventing the trip.

Unless there is connection with earth of any of the phase wires, there shouldn't be any current through Earth. You need to identify what is causing the currents to circulate through earth.

Yes, I understand that it only delaying the tripping. But it is more than enough for switching of transformers. But does circulating current through earth will damage my transformer if I increase the setting?

I believe this circulating current through earth is happen due to unbalanced of the phases and I using solidly grounded transformer. Unbalanced load will create zero sequence current which flow through neutral. What do you think?

Because if there is a phase connect to earth, even when the transformer is not in a parallel condition (independent condition), the 51G shall have sense it and trigger. But It only pick up in parallel condition.

One possibility is that there is potential difference between the earth pits and once neutrals are interconnected at the bus coupler, the closed circuit is making the currents to flow.

Probably it is time to check the earthing system of both transformer neutrals.

As you mentioned, if it is only changeover that you are intending, there is no harm in temporarily increasing the time delay. What is the pickup threshold for 51G? There could be scope for increasing the pickup, if not the time delay which makes more sense if the existing setting happens to be too sensitive.

Dear Mr. Raghun

Can you explain why you think there is potential difference between the earth?

Its setting is 0.6 A. Do you have any reference which can explain the standard setting of 51G in solidly grounded transformer?

Earth pits are all supposed to be interconnected in a given location.

If it is not done or if the interconnection is broken, there is always possibility that the two pits are not at the same potential (as it is the interconnection that is supposed to maintain them at a common potential).

The 51G is to be coordinated with the incomer earth fault protection, hence cannot be any standard setting.

It is rare in the US to have a 51G on a solidly grounded system, since it really only protects the bus between the sec. main breaker and the transformer secondary windings. It is often omitted on LV systems 2000kVA and below, 480V. Typically the phase over current protection will also operate on a ground fault.

These 51G relays cannot be coordinated with primary protective function for ground fault, assuming the typical delta-wye winding configuration. The 51G should trip the primary breaker if it is to be useful, and in many cases in the US, that would trip off a whole string of load center unit substations.

You could measure your actual current in the neutral. I can’t imagine it would double if you paralleled another system capacitance, and did not share the charging current equally. I’m surprised a 0.6amp pickup is adequate, unless that is the relay pickup and you have a 500/5A CT in the line...

I agree the 51G needs to trip the primary breaker.

Interesting to know that 51G is not common in IEEE world. In IEC world, this is quite common. For 2MVA transformers, Restricted Earth fault protection is also specified by some clients.

I agree the utility of this protection is limited as the incomer in the switchboard any way has over current and earth fault protection.

0.6A setting is 12% with 5A secondary neutral CT. This is not too sensitive, considering the capacitive currents in LV distribution will not be much unless there are capacitors connected in the system.

Yes, but 12% of what would be interesting to know. High resistance grounding is far more common in industry, but still lots of legacy solid grounding I found in metals, for instance.

The charging current is not actual capacitors, but distributed capacitance in the conductor system, between phase conductors and ground. That charging current is what makes high resistance grounding at high voltages 2.3kV + difficult to construct.

All you need to worry about is:

1. The voltage transformation ratio of the both transformer must be same. same pri taps
2. The polarity must be same. i.e. subtractive or additive Delta-Wye
3. The percentage impedance should be same.
4. The vector diagram and the phase displacement must be same, as in 2..
5. The phase sequence must be same, obviously.

The unbalanced loads are all line to line loads, assume this is not a 4 wire system, also there is not a lot of non-linear load as a percent of total. The distributed capacitance of the combined system may result in flow in the neutral/ground conductor, there is no reason to change the time delay from original, the pickup might need adjustment.

It is not possible to see what instrumentation there is, nor have you said what the current setting of the neutral overcurrent relay is, or if there are line O/C relays.

However, at 380V (& 220V to neutral/star earth) you can use a portable voltmeter(s) & currents may be measurable with portable clamp meter.

Transfos appear to have same 3.3 kV voltage at H.T.

Other posts have advised you about need for same ratios, polarity, tap-setting, sequence - so I presume you can parallel the secondaries, load breakers open.

The impedance should be about 4% - so even a 4% voltage mismatch [unlikely!] would only circulate half load. There will be currents in neutral & lines.

The circulating current should give an idea of the no-load voltage matching. You can even de-energise, change HV taps & judge the matching at several currents.

As I understand it, transformers to be paralleled would usually be identical & from same manufacturer to minimise differences. This is not essential, but any circulating currents must be tolerable.

Yes I dont put any instrument in the picture. But in real installation there are OCR on both lines and there are Over Current Ground Relay (51G) in the neutral-ground of the transformer.

I forgot to mention, this transformer's neutral is directly grounding (Solidly Grounded). From literature I read, any unbalanced load will create a zero-sequence current flowing in neutral of the transformer. Is it have any connection with circulating current which detected by 51G relay?

Unbalanced loads line to neutral, not line to line. Do you have line to neutral loads?

Yes, that is the source of the unbalance on the phases. It is almost impossible to balancing all the load, don't you think?

Yes, that does make it difficult. So your 51G relays are set very high, similar to line rating of transformer, so should not pick up when paralleling transformers. You should not have a problem. The 51G is back up protection for main breaker and main protection for a fault between the transformer and the secondary main breaker, so it should be set high and long time for good selectivity/coordination.