PS Class CT Knee point voltage. 2

[sakaran51]

Hi,
The CT requirements for the electrical system are to be specified in the initial detail engineering stage itself for any project.For differential protection for equipment like Generator,Transformer,Bus etc.,the required knee point voltage based on system fault current and relay manufacturer recommendation is to be determined and mentioned in PS Class CT spec.But the Rct value is not known and not readily available from the CT manufacturer at that stage and the knee point voltage is calculated mostly on assumed value of Rct, which may not be achievable by the CT manufacturer.
I would like to know from the forum members how to sort out this issue?
Thanks.

 

[veritas]

First of all what CT standard do you specify your CT's to?

 

[sakaran51]

Standard followed for CT Spec.will be IS Standard which is line with IEC only.

 

[lume7006]

Hello,

The rct value is needed to determine the total burden that the CT will have to support.
What you can do is asking the supplier (or some other CT manufacturers) to provide you the whole data of CT's design under IEC standard
and they would provide:
CT Ratio (the one you are taking into account)
PS Class
Rct
Vknee
In this way, you will have information available in order to be able to compare your calculations

However, based on experience, I can tell you that leads resistance is typically the most demanding factor
to take it into account to your calculation.

Besides that, it is important you to know the kind of relay to be used, currently digital ones are almost the rule, however
in some cases, if the relay were of the olds ones, the burden is very , very high.

Best Regards

 

[scottf]

The best is to state the Vkp as a function of a variable Rct, since your required Vkp will change based on the Rct of the CT.

That also gives the CT manufacturer the most flexibility and could end up saving you money over specifying an arbitrary Rct value and forcing the manufacturer into that figure.

 

[sakaran51]

Yes Scottf.That is the only way to go about when the CT secondary resistance is not known.
Thanks everybody!

 

[slavag]

No problem with CT mnf.
You can calculated Ukp according to protective terminal CT requirements, calculate current loop and ask CT mnf for approx. Rct <= of needed value. Mnf send to you some value, you recalculated parameters and if needed ask mnf again for other parameters, few times ping pong with mnf. and you found optimal data.

 

[davidbeach]

So, as I read these various CT specifying threads, pretty much all coming out of IEC land, I have to ask - do people seriously try to specify the "optimal" CT for every application?

I'm much more familiar with generally "over specifying" and then not needing to reevaluate the CTs for every system configuration change. It seems that "optimizing" results in lots of different CTs in use and the risk that a breaker can't easily be reassigned. Heck, we issue a blanket spec for breakers for the next several years, complete with CTs, long before the use of each and every breaker is determined. But experience has shown that we can get acceptable CT performance in pretty much every application with the same CT specification. Sure there are places that we might get by with a lesser CT, but only at the cost of a life time of reduced flexibility.

 

[slavag]

it's depend on:

Utilities is usually used same oversizing CT's for decades, for lines.

but, for special application, gas, steam turbine, step up, step down, metering, revenue metering, CT must be calculated every time.

 

[veritas]

sakaran51 - since you mentioned diff protection I assume you are dealing with a Class X CT as per IEC 60044.1? To take Scottf's suggestion a step further, Ekp may be expressed as follows:

Ekp = A*(Rct + B)

A = KSSC*Ktd
KSSC = rated symmetrical short-circuit current factor = If/CTR where If is the symmetrical fault current
Ktd = rated transient dimensioning factor (most often taken as X/R + 1. I sometimes reduce this as it is the most onerous)

B = total external loop resistance = leads + relay resistance

You will find more detail regarding the above in IEC 60044.6

 

[veritas]

Would like to add following as well:

For high impedance differential applications such as REF, high Z buszone, etc. a value for Rct is required when calculating the stability voltage. Rct can then be approximated using 5mohms/turn for a 1A CT and 2.5mohms/turn for a 5A CT. I would recommend that this be discussed with the CT manufacturer as the calcs may need to be revisited and fine-tuned if required.

 

[sakaran51]

Hi veritas,

"Rct can then be approximated using 5mohms/turn for a 1A CT and 2.5mohms/turn for a 5A CT. I would recommend that this be discussed with the CT manufacturer as the calcs may need to be revisited and fine-tuned if required."

That is exactly the point.What I want to know is whether that approximation is reasonably holds good to start with? Can it be applied for all the CTs? BTW,the value 5mohms/turn and 2.5 mohms/turn for 1A and 5A CT respectively, are actually 500 milli ohms(0.5 ohm) and 250 milli ohms(0.25 ohm)?

 

[veritas]

I have done this calculation for many Class X CT's throughout the years and I can assure you 5milli ohms/turn is indeed a very good rule of thumb (in my experience that is). So for high Z applications I use this to specify Rct and cannot recall it ever being a problem.

 

[sakaran51]

Sorry Veritas,
I made a mistake in a hurry.You are correct.I should have mentioned that it is 0.5 ohm and 0.25 ohm for 100 turns for 1A and 5A CTs respectively.

 

[scottf]

Just a point of clarification, all of the references to an approximation of ohms/turn seems to be assuming a window-type CT, i.e. where the number of primary turns is known to be 1 turn. In medium-voltage and high-voltage applications, you would not be able to know the exact number of primary turns used in the CT design without asking the manufacturer.

For example, a 100:5A, 15 kV CT might have 4 primary turns, so there would be 80 turns on the secondary and not 20 (if you assumed 1 primary turns).

 

[veritas]

scottf - the CT's I deal with typically are from 11kV to 500kV. Yes, most are window type with a single bar primary. Does it really matter if there is an extra turn here or there since my rule of thumb is based on declared performance specifications from the manufacturer.

 

[waross]

If it is a primary extra turn or two it does matter. A lot.

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

 

[veritas]

I was not talking about extra primary but secondary turns.

 

[scottf]

Yes, the primary turns matter a lot.

For example, a CT with an 800:1A ratio.

1 turn primary = 800 secondary turns
2 turn primary = 1600 secondary turns
etc...

Point being is that if you use an ohms/turn rule of thumb, you need to be sure of the number of turns.

Veritas...unless we're mixing terms, there is no such thing as a 500kV window-type CT. Once you get above 46 kV or so, they are going to be exclusively post-type/live tank (where the cores are at the top and the tank is at HV) or hair-pin-type/dead-tank, where the cores are at the bottom and the tank is at ground potential.

For ratios above 2000:1 or 5A, you can normally be pretty confident that the CT will have a 1-turn primary. For ratios below that, you have no way of knowing without talking to the manufacturer or looking at the schematic of the specific CT...and normally a schematic will not show the exact number of primary turns.

You said you weren't talking about extra primary turns but extra secondary turns. I'm not sure I follow that comment. If the ratio is set, then the relation of primary turns to secondary turns is set. There are no real "extra" turns.

 

[davidbeach]

Wouldn't a bushing CT be considered a "window-type" CT? I guess, though, it wouldn't really be a "500kV" CT even if applied on the bushing of a 500kV breaker or a 500kV transformer winding.