CT kneepoint determination (IEC ratings) 5

[redox]

Hi all

Is there any way how you can determine the kneepoint of a protection CT with the nameplate values?(IEC ratings) To determine the kneepoint with a voltage source and a ammeter is no problem - I want to know if it is possible to find it mathematically.

Thanks

 

[jbartos]

Suggestion: Geometrically or mathematically: Reference:
1. J. Levis Blackburn "Protective Relaying Principles and Applications," Second Edition, Marcel Dekker, Inc., 1998,
page 137:
The knee point or effective point of saturation is defined by ANSI/IEEE Standard as the intersection of the curve with a 45deg tangent line.
However, the International Electrotechnical Commission (IEC) defines the knee as the intersection of the straight lines from the nonsaturated and saturated parts of the exciting curve.

 

[ajit1960]

Knee point voltage is the function of magnatic material and number of turns. Genaraly for M4 grade C.R.G.O. toroidal core it is observed that the knee point is at 1.42 Tesla at 50 Hz. If you know the cross sectional area of core and number of turns on CT secondary you can use standard transformer equation of V=4.44 *Ag *Bm *f * T where v is voltage, Ag is gross area of core, Bm is flux density, f is friquancy, and T is turns to find out the knee point voltage.
For other magnatic materials you should have B/H curve to know the knee point.

 

[scottf]

I believe his question is how to figure the knee-point voltage from the nameplate ratings...

You can use the namplate data to get an approx. knee-point voltage. In the IEC world, an example of a protection rating would be 5P20 - 30 VA, which means, with a 30 VA max. connected burden, the CT will maintain an accuracy of 5% up to 20 times over current. So, if this is a 1A rated secondary, that means that the maximum connected burden is 30 ohms (I^2xR). 1A through 30 ohms develops 30 V. Since the CT is rated with a 5% accuracy up to 20 times rated current, you can assume the knee-point is somewhere around 600 V (20 x 1^2 x 30).

In reality, the knee-point defined by the tangent to the non-linear part of the excitation curve is a little lower than this number, but it'll get you close.

In the IEEE world, we use ratings like C800. For a typical C800 rated core, the defined knee-point is normally somehwere around 700 V or so.

 

[jbartos]

Suggestions: Visit

http://banesassociates.com/newsletter3.htm

for:
1. C800 voltage taken 800V
2. Knee point voltage in 3 standard methods:
10%-50%
IEEE-30
IEEE-45
3. ""A hysteresis curve is hand plotted from all this data and the saturation point (knee) is determined and a tangent (knee) is drawn with a straight edge. This is tedious and time consuming.

The Vanguard EZCT is an excellent piece of test equipment because it performs all of these tests automatically in about two minutes.""

 

[scottf]

jbartos-

I'm not sure I follow you post...

The question is concerning IEC ratings.

Also, for IEEE rated CTs, it is not necessarily true that the knee-point voltage is the same as the C rating. The C rating is generally a bit higher than the knee-point, at least using the definition of knee-point voltage under IEEE.

What do you mean by 10-50%, IEEE30 and IEEE40?

 

[redtrumpet]

The point on the excitation curve where a 10% increase in secondary exciting voltage results in a 50% increase in secondary exciting current is the British Standard definition of knee-point voltage.

The IEEE 45% and 30% definitions apply to Class C cts and refer to the point on the excitation curve where the tangent is at 45 (or 30) degrees to the abscissa. 45 degrees is used for nongapped cts and 30 degrees for cts with gapped cores.

These are the three definitions that the Vanguard EZCT can use to calculate knee-point voltage, what jbartos is referring to as 10%-50%, IEEE-30, and IEEE-45.

In my experience the IEEE definition will result in a lower knee-point voltage than the British Standard definition. The IEEE knee-point voltage is definitely less than the rated secondary terminal voltage, as mentioned by scottf.

 

[scottf]

redtrumpet-

Thanks...I just wasn't thinking when I asked about IEEE45 and IEEE30...never seen it written that way.

Interesting fact...the angles listed for determination of the knee-point are based on normal silitsium-steel core matertial, which is the norm for normal protection cores. The angles change if a different material is used...this sometimes comes into affect for special type protection cores.

 

[minimalibu]

Hi Guys

would a ct rated as 5P20F20 be a 20VA CT then?
What does the F stand for?

 

[scottf]

Not sure...the F is not an IEC terminology that I am aware of.

 

[jbartos]

Suggestions to scottf (Electrical) Dec 28, 2003 marked ///\\jbartos-
I'm not sure I follow you post...
///It is easy to follow. Visit

http://banesassociates.com/newsletter3.htm

for an easy to read and understand website.\\
The question is concerning IEC ratings.
///Yes. I am aware of it. I provided the first reply to the original posting where I addressed the IEC approach.\\Also, for IEEE rated CTs, it is not necessarily true that the knee-point voltage is the same as the C rating. The C rating is generally a bit higher than the knee-point, at least using the definition of knee-point voltage under IEEE.
///Yes, agreed.\\What do you mean by 10-50%, IEEE30 and IEEE40?
///Visit my posted link above, namely:

http://banesassociates.com/newsletter3.htm

for the answer there. Incidentally, the IEEE Standard 45 should be considered not IEEE40.\\\

 

[scottf]

jbartos-

I understood your post fine, I was just commenting that it was not relevent to the original post, since it was asking about IEC rated CTs and your post was geared to IEEE rated units...was just a comment.

 

[jghrist]

I suspect the F is the instrument security factor F[sub]s[/sub] and indicates that the core saturates at 20 times rated current. However, it is not normal for a protective core to have an F[sub]s[/sub] and metering cores usually have an F[sub]s[/sub] of 5 or 10.

 

[scottf]

Jghrist-

I doubt that it is an instrument security factor, as this is only for metering application and counter to any protection application/rating.

Fs means that it must saturate before the value, ie 0.2Fs5 means it must saturate before 5 time nominal current.

The "P20" means that it must saturate above 20 times rated current, so it could not be P20 and Fs20 at the same time.

 

[Deansharafi]

F stands for Factor, and refers to Rated Accuracy Limit Factor. When it is not mentioned it is regarded to be 20, otherwise it can be 5, 10, 15 or 30.

 

[scottf]

Deansharafi-

F does not stand for accuracy limit factor under IEC. In a protection class CT, the number following the 'P' is the accuracy limit factor. So, 5P20 means 5% composite error up to 20 times rated current.

Also, there is never an assumption about the accuracy limit factor...it is always listed for protection cores. If it is not listed, you cannot assume it is 20. Check section 12 of IEC44-1 for more details.

If I had to venture a guess, I would guess that by writing 'F20' someone is trying to list the rated burden as 20 VA, although that's really not the way to do it. If there is no other VA rating listed, then that's probably it.

 

[jbartos]

Suggestion to ronaldoalmeida (Electrical) Jan 23, 2004
Please, would you clarify:
"""The impedance reflected into the secondary of main CT will be Z’b = Zb / (Ip/Is)²"""
since generally:
Iprim/Isec=Nsec/Nprim
Zprim=(Nprim/Nsec)^2 x Zsec
Zsec=(Nsec/Nprim)^2 x Zprim = (Iprim/Isec)^2 x Zprim

 

[Bung]

The F20 accuracy limit factor is used in the now superceded AS1675 Australian Standard. A 5P50 F20 CT has a secondary output voltage of 50V and 5% accuracy at 20 (=ALF). The Australian 5P20 is not the same thing as an IEC 5P20! The Australian standard does not specify a VA rating at all.
The Aust Std has now been aligned with IEC.


Bung
Life is non-linear...

 

[scottf]

I believe Bung is correct...based on the original post, I was thinking only in terms of IEC ratings.