Generator Rated kVA & Parameters at Various Ambient Temperaure

[sykimk]

Hello,

I have 14.24 MVA rated synchronous generator driven by gas turbine and it is rated at 42 degree C and all parameters are given based on this temperature.

If this generator will be running at 0 degree C. of site ambient temperature, the generator can cover more kVA e.g. let say 16 MVA as an example.

I think I should use the maximum possible generator MVA at lowest site ambient temperature (e.g. 16 MVA at 0 degree C.) for short circuit current calculation. If so, how about generator parameters? I guess reactance values (X"d, X'd, Xd, etc.) would stay at the same values regardless of ambient temperature, but resistance values (Ra, etc.) would be slightly lower than that of 42 degree C. This would affect generator X/R ratio for short circuit current.

If so, do you think I can use the following equation used for cable resistnace correction?

R’ = R ( 234.5 + Tc )/( 234.5 + Tb )

R = 0.00213 p.u.
Tb = 42 deg.C
Tc = 0 deg. C
R' = 0.001806 p.u. (corrected at 0 deg.C)

This will increase generator X/R ratio and DC component value of short circuit current.

Would you guys please give me advices whether my approaches are appropriate for the maximum short circuit current calculation?

Thanks in advance.

 

[davidbeach]

The stator temp will remain at a fairly constant, and much higher, temperature than ambient when running near full load. Ra values are corrected to design operating temperature not ambient temperature.

 

[dpc]

I think you are worrying a lot about very little. The slight changes in X/R ratio will have little impact on the magnitude of SC current. It's a second or third order effect.




David Castor

http://www.cvoes.com

 

[rcw retired EE]

The generator parameters are +/-10% at best and some are +/-15%. The minor change in resistance is negligible compared to the "noise" or inaccuracy in the generator parameters. (It's like measuring with a micrometer to cut with an axe).

The initial short circuit values have nothing to do with the turbine MW output. An unloaded generator delivers about the same initial short circuit current as a fully loaded unit. From a short circuit perspective, the MW output increase due to ambient air temperature is no different than increasing load by opening the throttle.

MW loading has secondary effects on short circuit currents, but if your system is that close to the ratings, you are in trouble already.

Just use the ISO ratings and nameplate values in your calculations. If you want to get conservative, use 85% of the given impedances because ANSI standards allow a 15% variation in estimated and actual impedance. That is similar to adjusting the transformer impedance to 92.5% x %Z since standards allow a +/-7.5% variation from the specified value. Also, run the short circuit calculations at 105% nominal voltage. Anyone of these variations has a greater effect on the calculated short circuit current than ambient temperature.

Sizing of the generator transformer and generator cables/bus is where the ambient temperature effect on gas turbine output needs to be considered.

 

[sykimk]

Dear davidbeach,

Thanks for your quick advice and it makes me more comfortable. Do you think it would be good idea to use the max. generator ouput (16 MVA)at 0 degree C with the given parameters at 42 degree.C for my short circuit current calculations?

Dear dpc,

The reason why I am worrying a lot was that I will have four (4) generators directly connected to 11kV Switchgear (IEC), and IEC standard breaker (Std. time constant = 45ms) would not suit for my case due to high DC component especially for outgoing feeder breakers but I don't have any problems with peak and breaking currents. I will specify higher time constnat for %DC component.

In addition, if I would have TRV peak problem, as far as I know, a possible solution would have breakers with higher rated voltage (e.g. 15kV, 17.5kV, etc.). Would you have any recommended alternative solutions to lower TRV peak voltage?

But I don't expect I would have RRRV issue since generator surge capacitors will increase the rise time to TRV peak voltage.

Do you have any more advices regarding DC component and TRV concerns with my configuration?

Thanks in advance

 

[davidbeach]

If you're trying to tweak generator parameters to solve a breaker rating problem, you are in way too deep. Stop. Take a deep breath. Now increase your breaker ratings so that you have at least a 15% margin between what they have to contend with and their ratings. At best you know the generator parameters to +/- 10% and you are looking to make much smaller adjustments for temperature?

 

[dpc]

I agree with David - I would never be comfortable resolving a short circuit rating issue by tweaking generator impedance constants - as said, they are not that accurate to start with.

You need to increase the breaker SC ratings, add some additional impedance somewhere, or run one less generator.



David Castor

http://www.cvoes.com

 

[davidbeach]

Reminds me of a rule of thumb from earlier in my career (don't do that kind of work anymore):

If the inspector has to get out the tape measure to check the required clearance, you don't have enough clearance. Doesn't matter what the measurement is, if it has to be checked there isn't enough clearance. Same thing applies here, if there is a need to massage the numbers you are starting with the wrong numbers.

 

[sykimk]

I really appreicate all of your advices and I agreed I was too worrying about it.

Just for your information, I have 50kA breakers and its calculated DC component was about 33 kA_dc at this time while breaking current is about to 22 kA. I cannot increase SC rating of breaker since I cannot find out 63kA breaker for 11kV.

I just need to specify higher time constant for %DC component. i.e. t4=120ms. All other time constant does not suit for my case when breaker is delayed by protective relays.

I am now just wondering whether higher time constant (120ms) for breaker will affect breaker price.

Thanks for all of your advices again.

 

[sykimk]

By the way, do you have any recommendations for TRV concerns especially for controlling TRV peak voltage value?

Thanks in advance.

 

[dpc]

In the ANSI world, the C37 has a procedure for dealing with SC ratings for generator breakers and breakers near generators. We don't generally deal with the TRV issue directly. That is left to the breaker designers.

I take it you are using IEC ratings and procedures.

You might want to get Conrad St. Pierre's book on SC calculations. It covers both ANSI and IEC methods.

David Castor

http://www.cvoes.com

 

[waross]

Don't confuse asymetrical short circuit current with Available Short Circuit Current.
Asymetric current includes the DC offset current.
Available Short Circuit Current is calculated from the impedance without regard to X:R ratio.
However, breakers are often rated in Available Short Circuit Current. But, a breaker that is rated at 20,000 Amps Available Short Circuit Current will be tested at a current level and phase angle corresponding to an expected XR ratio.
A breaker rated for a given Available Short Circuit Current will be tested and will be able to safely interrupt the corresponding actual asymetrical short circuit current in all but extreme installations.
Be sure that you are working with the correct values.
It is not appropriate to calculate the asymetrical current and use that value to select a breaker rated for Available Short Circuit Current.
It is appropriate to determine the X:R ratio corresponding to the breaker testing and compare that with the X:R ratio of the source. Further calculations may not be needed.

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

 

[davidbeach]

Bill, you basically describe the ANSI system, except it isn't really all that difficult to exceed the test X/R. When the test X/R is exceed the standard provide a formula for calculating the adjustment factor. The IEC world seems to go about things differently and the asymmetric current is needed.

 

[ScottyUK]

Siemens and ABB both have generator breakers available for this duty, but they aren't particularly well known even within the sales force belonging to those organisations. I'll try to find some type numbers later on.

You're right about the high X/R and the long time to first current zero, but please don't consider what some cheap & nasty designs by certain EPC contractors have done, which is to incorporate a timer to delay breaker opening for a few cycles until the DC component has died down a bit. The only reason for doing this is so they can use a cheap distribution breaker inlieu of a larger and more expensive generator breaker. My own opinion is that cost saving is a poor excuse for intentionally leaving a fault on the system when breakers are available which can clear it quickly.


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

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