Primary Injection Testing - Regional Differences

[mgtrp]

I am relatively new to North America, but have noticed a difference in approach to the importance of protection systems as compared to my home country - highlighted recently, when during commissioning of a generator, on-site staff decided to get rid of those pesky differential trips by turning off the step-up transformer protection... not as big a deal as you might think, as they'd never applied proper settings to it anyway!

Regardless, one specific item that I would like to know about is "to what extent is primary injection testing carried out on new and existing installations?"

I am used to seeing primary injection testing on all new installations to prove the ratio, polarity and wiring (although secondary injection to prove operation or the relays etc). On differential and REF elements, stability and operation would also be tested.

At the North American utility I am working with now, however, primary injection testing does not seem to be used at all. In fact, CT installations seem almost designed to inhibit primary injection, as they are often inaccessible bushing CT's in oil-filled transformers - a big no-no where I come from, due in part to the need to primary inject, but the feature seems to be ubiquitous in North America.

In the past year, I have seen situations where primary injection testing would likely have prevented several trips: the aforementioned transformer trips I suspect will turn out to be mis-wired CT's (trip occurs when the generator goes on load), and a case of a neutral CT being wired back-to-front causing several instantaneous operations of a transmission step-down transformer LV REF element for earth faults on the any of the connected MV distribution feeders.

Thoughts?

Thanks,
mgtrp

 

[ScottyUK]

mgtrp,


Somewhere on here I've previously described how to carry out a primary injection test on a transformer constructed as you describe. You can't use a primary injection set because of the winding impedance but it's entirely possible using a generator as the source. I'll post the sketch if I find it. Can't answer your main question though.


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

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

 

[alehman]

Some discussion here:

http://www.theiet.org/forums/forum/messageview.cfm?catid=226&threadid=29097&highlight_key=y

Alan
“The engineer's first problem in any design situation is to discover what the problem really is.” Unk.

 

[mgtrp]

I have no problems with "how" to carry out the test, it's more "whether" the test is carried out - it was where I come from, but doesn't seem to be with the utilities I have worked with more recently.

Is it common practice in North America?

 

[rcw retired EE]

This is not a common practice in North America. Most systems are not designed to do it easily.

We design it correctly, check it on paper, then check the CT polarity and relaying connections using several different methods. If the wiring matches the verified drawings, it is good to go.

Primary injection testing is expensive in terms of time and equipment. In my experience it is seldom done.

But we have done low level injection testing by connecting a 480V three phase generator on the incoming HV transmission line and applying short circuits at various places in the system to verify bus differential, transformer differential and other protection schemes.

In Asia and Europe, I have seen the requirement for primary testing at full current. (One customer wanted simultaneous full voltage, full current testing of all protections. We didn't like the idea of multiple three-phase bolted faults on a 400 kV system. The serving utility didn't either.)

I doubt primary injection testing will catch on in the US, except for low voltage (<1kV) breakers and systems.

 

[oldfieldguy]

I concur with rcwilson.

I've commissioned a wide variety of installations from power generation to transmission to industrial distribution. If you have a thorough systematic testing of CT ratios and polarities, secondary current injection (usually at the CT terminals or the terminal block closest to the CT where you accessed the wiring for your ratio and polarity checks) and verify proper installation in accordance with the drawing AND the needs of the installation, then primary injection is not needed.

However, I have had some clients who demanded it, and if they want to write the check, they get to say what gets done as long as it's safe.

In cases where we've had to do primary injection testing, I have been able to use single-phase current sources to inject current in two phase conductors with, as rcwilson says, shorting conductors applied at various points to route current through different devices. I then verified secondary currents by evaluating magnitudes and angles to be expected from the drawings.

The final arbiter for me, though, on "main-tie-main" or ring bus installations was to "split" two transformers, unbalancing the taps so reactive current would flow through the system, then evaluating magnitudes and phase angles from the circuits in view of the circulating current. This is tedious and makes a lot of people nervous, but it IS a primary injection test, too, although not in the classical sense of applying current from a test source.

old field guy

 

[rasevskii]

In the European and UK environments, and the countries of the world under their influence, primary testing is absolutely standard practice. It is relatively easy to do short circuit testing when commissioning a new generator. Absolutely necessarey to verify correcr CT connections and phasing for generator differential protection. The same for transformer differential protection, using a SC at the HV side, often an earthing switch can be used for the SC. Full current is not needed, about 20 to 40% current is sufficient.

How else can one properly verify ratios, polarities, and correct phases of CTs and matching CTs for transformer differential? Even more fun for a three-winding transformer...

In the USA environment, time and speed are paramount. Labor and professional hourly costs on site are prohibitive. But I have seen or heard of spurious differential trips of units once having been loaded up on line, because things were not properly done beforehand.

The same is true for voltage related protections. One US client that I once had, did not even see the need for generator overvoltage protection, and it did not exist. They were used to old mechanical voltage regulators and not modern static exciters, where, when and if it goes bonkers, it will reach ceiling excitation in a second or two, and something had better trip...

rasevskii

 

[mgtrp]

I think that the last three posts match my observations to date of the differences between the IEC/IEEE worlds.

The gentleman who was perhaps my country's most well-respected commissioning engineer liked to describe the Americans who came over from time to time to commission equipment as "cowboys" in terms of their approach.

What I've seen to date here would only serve to reinforce that view. In fact, I found out yesterday that in commissioning the aforementioned generator with the switched off step-up transformer protection, they've also done away with significant parts of the generator protection too! This after they ignored a series of over-temp alarms (there were a lot of alarms coming up, hard to keep track of a pesky few in amongst all the others!) and ended up activating load-shedding when they tripped the generator.

That being said, I suspect (hope) that this is not representative of all North America! I've no doubt that good pre-commissioning practices could make primary injection seem unnecessary in a lot of cases. I still favour it however.

Thanks,
mgtrp

 

[rasevskii]

Well, yes, I did not want to use the word "Cowboys", but I can well remember, a couple decades and more ago, when a 31 MVA startup transformer for a large synchronous condenser, nearly exploded, having had 18Kv connected at the same time to the 18KV primary and 3150V secondary, due a recloser not having been disabled on the MV side. The backup protection on the EHV side of the main bank cleared the fault in 10 cycles, otherwise we would all not be around to tell the story, it would have been a fireball less than 50 feet from the control house in which all the commissioning people on site were in...

Not only oil platforms are dangerous places to be....

rasevskii

 

[rcw retired EE]

Sound like you primary current injected the transformer and found the problem.

 

[ScottyUK]

Rasevskii,

I'll happily use the word 'cowboys', certainly in relation to the engineers from a collapsed energy giant formerly based in Houston. A generator transformer and two station service transformers, all connected to the 275kV system, all with reversed neutral CTs in the restricted earth fault scheme. One was not only reversed, but the connection was to the wrong CT.

It has to be easier to commission on a plant during construction than to do it on an operational plant, but that wasn't the way things were done back in the early 90's. Every one of these problems was picked up following protection misoperation, and were proven without doubt by an injection test. Primary injection on a generator transformer takes a bit of doing.


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

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

 

[rasevskii]

Yes, the transformer in question was complete scrap afterwards. The electromechnaical forces from the fault had caused the core laminations to fly apart, cutting the windings. There was no smoke or fire as the fault was cleared in the 10 cycle period. Even the oil still tested good, from outside the tank it looked almost normal, just open circuit on all phases on both windings...

rasevskii

 

[rcw retired EE]

Scotty - Those Houston cowboys were always in a hurry- no time to do it right but time to fix it later when it took twice as long. I had to follow them around fixing things after multiple trips. The best design is still no good without competent commissioning engineers and techs.

For many generator & GSU combinations it is possible to jumper across the generator windings in teh termainal enclosure and connect a 600V portable generator on the high voltge side of the GSU. With the generator windings shorted and all breakers closed the transformer impedance limits the current but there is enough current to check most CT connections and protections. Swapping CT polarities or clamping shorts on bus verifes tripping and proper polarities. It can usually be done fairly quickly and is easier than trying to get everyone to run the turbine generator for a day or two for relaying tests.

 

[Milliamp]

We use primary current injection throughout our electric utility any time we install equipment.

 

[bacon4life]

We use a single phase primary injection source on circuit breakers and stand alone CT's. For transformers we do polarity and ratio checks with a single phase CT tester. We have not used low voltage generator.