Voltage Transformer @ 25 Hz

[flyboyz]

Hi all

I had a request to supply VT drawer (for an old mining site) connected as open delta, 13.8kV/120V, operating at 25Hz. Accuracy required is 0.3Z with thermal burden 400VA ea.

My questions:

1) Is it possible/feasible to derate a normal 60Hz VT
for use in my application? If yes, is there any
calculation to do that?
2) Does the primary fuse amp rating be decrease/increase
from the 60Hz FLC rating?

If anyone have experience on this, please let me know.
Thanks in advance.

 

[scottf]

flyboyz-

In short, no...you cannot simply derate the VT's performance by a frequency ratio.

You'll need to have a VT specifically designed for 25 Hz operation. You will most likely need to reduce the burden requirements. Do you really need 0.3Z performance (Z being 200 VA) and a thermal burden of 400VA? Unless you're using electromechanical relays, it is unlikely that you'll have anywhere close to 200 VA on the VT.

 

[jbartos]

Suggestion: Electrically powered trains used to have overhead catenary with 60Hz and in some section 25Hz. There was a change-over switch that had to switch the transformer winding to be able to obtain proper voltage on the transformer secondary when entering a different frequency power supply from the catenary. The posted 60Hz VT will not have the option of winding taps on the primary for 25Hz supply voltage. However, if you use 60Hz VT and connect it the same voltage 25Hz, the output voltage will be lower than 120V. Notice that the transformer voltage is governed by:
E=4.44 x f x N x FIm x 10^-8, in Volts.
If the secondary voltage is low, then a boost transformer on the secondary will be required to increase the voltage to 120V.

 

[scottf]

jbartos-

Since VTs are used for measuring voltage, in that ratio error and phase angle error, it is not possible to use an aux transformer to fix the accuracy issue this simply.

Also, and more a matter of system safety, the lower frequency also changes the saturation characteristics of the VT. If a VT designed for 60 Hz usage is energized on a 25 Hz system, you will run into serious saturation problems. If the system is delta, then there will be big ferroresonance problems. Also, the VT will heat-up excessively and fail relatively quickly.

 

[jbartos]

Suggestion to the previous posting: There is an open Delta mentioned in the original posting.
Eddy current losses will decrease with the frequency decrease.
The Steinmetz remagnetization losses also decrease with decreasing frequency.
The train transformers did not have problems on 25Hz.
The accuracy issue is somewhat more difficult to resolve. An exception in accuracy will probably have to be requested.

 

[davidbeach]

The locomotive transformers would have been designed for 25Hz operation. Using a 25Hz transformer at 60Hz is a far different situation than using a 60Hz transformer at 25Hz. Besides, as scottf points out, there is an expectation of accuracy with a VT that would not exist in a locomotive.

 

[scottf]

Jbartos-

I'm sorry, but for a VT design for 60Hz operation, when energized at 25 Hz, there will be severe saturation...there is no doubt about that. Take the formula for fluz density and work out the effects of lowering the frequency.

As for what has worked in past train application, I agree with davidbeach's post...a unit designed for 25 Hz can be used at 60 Hz, given the correct turns adjustments on the primary side, to compensate for the accuracy changes you noted above.

For a fixed core size, a 25 Hz VT will have to use many more primary turns than a 60 Hz unit to keep the fluz density below the saturation point. By using that many more turns, the accuracy rating suffers, which I why I originally notes that the burden expectations will have to be lowered substantially if a 0.3 accuracy is required.

 

[ScottyUK]

jB,

The principles which scottf is referring to are common to any transformer constructed using a magnetic core. There is a basic and well-known relationship between frequency and flux density: halving the frequency doubles the flux density; applying a 25Hz supply to a core wound for 60Hz operation results in a flux density 2.4x higher than design. For any normal transformer, this would be well in to magnetic saturation, a condition which would hopefully cause primary side fuses to blow, or alternatively would result in heating of the core and winding until they are damaged.





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

Start each new day with a smile.

Get it over with.

 

[jbartos]

Comment: I am familiar with 60Hz to 25Hz power supply transfers that were used on the electrically propelled rail cars, e.g. AMTRAK. One transformer served both supplies.
When it comes to the PT without taps on the primary and design for 60Hz and 25Hz, it is evident that the parallel magnetizing branch in the transformer equivalent circuit would have much lower impedance at 25Hz causing the transformer to overheat. Once the complete circuitry is designed or projected a voltage drop resistor or voltage divider could be considered for the transformer primary to reduce the voltage on the PT primary sufficiently low so that the transformer will not overheat. This will also lead to a reduction of PT VA rating. I think that this is what the original posting is inquiring about.

 

[scottf]

jbartos-

If I follow your post correctly, are you proposing to use a voltage divider to step down the voltage on the input of the VT to such a degree that it could be used at 25 Hz?

If so, this would only really work if only a minimum output from the VT was required...in the range of 0-2 VA maybe. Resistive voltage dividers can only handle a small current through the divider, as the voltage (accuracy) changes dramatically as a function of applied burden. Such a set-up would not be advisable for metering applications and many protection applications.

IMO, the only option here is to order a VT designed for 25 Hz.

 

[ScottyUK]

Longshot:

You might get away with a 33kV VT from the European markets. The operating flux density would be comparable, but as I'm sure scottf will confirm, the accuracy will be outside of the original design parameters. Might be ok for your application.

Keep away from resisitve dividers: the attenuation of the overall arrangement becomes burden-dependent, which is exactly what you don't want.



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

Start each new day with a smile.

Get it over with.

 

[Skogsgurra]

jb,

What are you saying???

What about phase shift? What about voltage drop under different loads? What about saturaion at no-load?

 

[busbar]

IEEE C57.13-1993 states simply, Instrument transformers shall be designed and rated for operation at a frequency of 60 Hz.

The classic GET-97D instrument-transformer manual states with respect to PTs, A decrease in frequency increases the flux density, the exciting current, and the load current, while it decreases the reactance. However, again depending of the design of the transformer, saturation of the core may be reached with a relatively small reduction in frequency. At this point, very large errors and considerable heating will take place.

 

[flyboyz]

Thanks to everyone who provided inputs to my inquiry.
Looks like I would have to go with scottf on VT specifically
design for 25 Hz, instead of derating the 60Hz VT.

One more question, does the primary/secondary fuse amp rating requirement will be affected by the reduction in frequency? In my mind, I think there might be a different since we have AC and DC rated fuses. But I can't explain if less than nominal freq. would affect the fusing requirement.
Inputs would be appreciated.

 

[scottf]

Flyboyz-

That's really a question for your VT manufacturer. Knowing how we would design such a VT, it would not need different fusing, but it really depends on the design philosophy....mainly in regards to the inrush current.

As for a fuse behaving differently for the same current magnitude at 25 Hz vs. 60 Hz...I'm not really sure about that. It may depend on the type of fuse used.