low voltage equipment for injection of CT primary current 4

[odlanor]

For comissioning , there are 2 methods to inject primary current in CT for checking the Differential Protection.
The test arrangement varies with the application.
1- On network power transformers and asynchronous machines, a low-voltage test is preferably conducted,
where low-voltage current source is used to energize the protected object, which is completely disconnected
from the network. A short-circuit bridge, which is capable of carrying the test current, is installed outside the protected zone and allows the symmetrical test current to flow.

2- For power station unit transformers and synchronous machines, the tests are performed during the current
tests, with the generator itself supplying the test current .

We use 2nd method because we always have power plant. But, we want to join 1st method for more flexibility.
Anyone have any suggestions for how to size 1st method equipment to apply on 2nd arrangement?

 

[DTR2011]

If I had more time to discuss, I would.

This is a good paper from SEL on the basics. (1st item)

http://www.selinc.com/search/SearchPage.aspx?searchtext=primary injection

I have used 480 V Generators quite a bit 100 kW, or so, depending on the impedance of the GSU, etc.

Newer clamp on meters have very good accuracy in amplitude and phase angle.

Please feel free to ask questions and I can answer later.

One other thing to consider is using an off mains frequency for injection. Omicron CPC 100 can go as low as 15 Hz....

 

[slavag]

We use a very simple method, of course you need calculate current.

400V/480V 3 phase connected to HV side of transformer and put 3-ph short-circuit wire on the LV side.
Its work for lot of yaers.
Next option, we connect emergency disel to low side of transformer.

 

[ScottyUK]

Works well with MV gensets too. It is useful if you can get direct control of the generator field using a DC source because you can then adjust the injection current accurately. You may need to disable some of the protection functions, especially if your tests cause a highly unbalanced load condition. Best to use a rental set.


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

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

 

[odlanor]

ScottyUK
Are you talking about genset with DC generator to supply generator field ?
What about 3-phase short circuits? Can you generate it with DC source?
Can you explain better?

 

[DTR2011]

Now I have time.


A 3 phase genset is the temporary source. Here in the US, 480/277 is a common voltage. My rule of thumb is ~100kW.

I will give an example of the last time I performed this test at a wind farm from the 34.5 kV Metalclad switchgear, through the 34.5kV/230kV 80 MVA GSU (10%Z), through the HV breakers.

I use a few quick calculations in my head, to determine sizing, and have an excel sheet that does the calculations, which I also use for documentation.

First is the question of where to connect. There are debates on HV side, vs. LV side. I connect where it is easiest and safest, so that changes from site to site. This for me is a commissioning test, so I usually see what is available in terms of wire, available genset, available spare cables, lugs, etc. For my last test, the MV SWGR had a ground / test breaker available, so I choose to use that, as I could test the MV Low Z 87B scheme as part of the test (and test each of the 5 Feeders in the scheme).

It is important that all relays are tested first, all settings are in place, all CT's are tested, all secondary injection has been performed. Up until now, each test has been component based, with overlap as possible. We don't want an open circuit CT, or IED settings not correct to cause problems or confusion. Now we test the entire scheme.

Some words on terminology. Some people call this Primary Injection, some people call this Stability, some call it a Through Fault Test. These are all the same test in my words.

In an ideal world we have huge gensets, plenty of time and unlimited budgets. I guess I missed that world, so I am luck if I can convince the powers that be to perform the test, it is safe and placing a direct short circuit on a system connected to a $30k genset will cause no damage. And yes, if all goes well, I'll have it complete in about 2-4 hours (depending on size of system and scheme).

Back to some theory. In my case, I inject from MV breaker through GSU, through HV breaker(s). Three phase current flows. Checks 87 circuits and any other metering, etc. What about 51N, REF, etc. We'll get to that in a minute. I also have 21 Protection, BF, 87B on HV SWGR, etc.

The GSU is an impedance. The nameplate says Z=10%. What does that mean. In simple terms, under (full voltage) short circuit conditions, I will have ~10x FLA (full load amps). If I had 5%Z, I would have 20x FLA, and so on. My GSU is Wye/Wye (230kV-34.5kV). My genset is 480/277. I have the ratio of 277/19200 applied voltage = 0.014, which is my expected current multiplier on the LV side of the GSU. Say FLA = 1000 and Isc=10000A, multiply by 0.014, to get 144.27A on LV side of GSU. Say for simplicity I have 100:1 CT, I have 1.44A on CT secondary. I am very happy with my set up so far. Now I am concerned about the current on HV side of GSU. The GSU turns ratio is 6.91. My expected GSU HV current is 20.87A, my CT ratio is 50:1. I will have .418A secondary on my CT circuits. Good enough for me, but you can check your relays and phase angle meters for accuracy with relay test set. The rule of thumb is ~0.5A on SEL relays (which are popular in my region). I'm sure ABB/Siemens, etc are equally accurate. Follow these examples out for all CT circuits in the test circuit.

As mentioned earlier, I injected from the 34.5 kV MV breakers, so I used grounding cables on the HV side of circuit, as the busbar had provisions to connect cables. I choose to connect every breaker possible in circuit (in my case a breaker and a half scheme). I leave all lock out relays reset(but the breaker trips open from LOR), so if there is a problem, I can identify it, but not interrupt the test in progress. I can check 51N or other single phase faults, by shutting down the genset and removing 2 phases of the ground.

If I connect the 277 V voltage from the genset to my phase angle meter (A phase (L1)), I now have a phase reference. I word about expected current phase angles. The GSU is largely an inductive load and I am placing a bolted short on the HV side. The GSU is Wye/Wye, so things are fairly straightforward. In normal conditions, my CT secondary currents are 180 degrees out of phase. Negating resistive losses, the HV current is going to lag by 90 degrees.

If interested in more, I will finish when time permits.

 

[odlanor]

smallgreek,
I hope that your time will allow you to end the explanation

 

[ScottyUK]

odlanor,

Most small gensets (up to a few MVA) can have the field winding driven by a standard laboratory-type variable DC power supply, so there's no need to get involved with DC generators. It is useful to be able to control the output voltage if the standard generator output voltage isn't particularly convenient for the combination of transformer ratio, impedance, etc that you have to deal with.

I've attached some notes from a previous job which might help (or confuse).

smallgreek has done a great job of explaing the test setup. Star for you sir.


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

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

 

[jghrist]

Scotty,

Since you've used this for a previous job, I must be wrong, but it seems to me that the neutral CT secondary current direction is wrong.

If the primary current flows out of P1, the secondary current should flow into S1, but it is shown the other way. If it flowed the other way, the differential relay would not operate.

 

[ScottyUK]

jghrist,

You're correct.

The S1 and S2 terminals are shown connected to the wrong bus wires. That has been wrong for a long time - I'll have to correct that (or delete it if I can't find the original CAD files!)



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

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

 

[jghrist]

But if the S1 and S2 terminals are connected as shown, there will be no current into the differential relay and it won't operate. Since the test is essentially an external fault, isn't this the correct result?

 

[ScottyUK]

Actually the CT polarity is correct, it's the arrows and currents shown in red from the neutral CT which are incorrect for a stable scheme. The generator connection simulates an external earth fault on the system (i.e. out of the protected zone) and no trip should result.

The red arrows and currents may well be what happened during the test. We found a mis-wired (or mis-marked) CT terminal box on the transformer which effectively swapped S1 and S2 of the neutral CT at the CT as far as the scheme was concerned. It's a long time ago and I should have read through that note it before posting it.


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

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