Flashover on a newly installed 230/13.8 KV transformer cable box 5

[TechARA]

Recently we have experience a flashover on a newly installed 100/133 MVA, 230/13.8 KV transformer cable box. The flashover happened two days after the transformer was energized in one high voltage (230 KV) oil filled cable box. An internal high energy arcing occurred in the transformer cable box between the box cover and top shield holding points.

My question: what is the optimum radius around the conductors (insulation distance between current carrying conductor and box wall) that will ensure cable box safety? Is 350mm is enough?

Also, what is the adequate paper insulation thickness around the conductors? Is 3mm is enough?

Is there any pressure relief device that can relief the generated pressure by high energy arcing flashovers?

Thanks

 

[7anoter4]

hi TechARA
First of all you may consult the article from this link.
"Oil-filled cable pothead with venting means and tool therefor"

http://www.patentgenius.com/patent/4273951.html

I'll try to calculate in a simple way the insulation solicitation.
The voltage to earth will be 230/sqrt(3) =133 kV. The test voltage may be 212 KV [shock wave]. The insulation is paper –air in series [or paper-oil]
For paper 3 mm is good. For air 350 may be fair, but better will be 500[as for air maximum permissible electric field intensity will be 32 KV/cm in normal conditions, but for the warm condition 20 KV/cm will be possible.]
For oil 150 KV/cm and for paper 100KV/cm will be maximum permissible.
For oil 100 mm will be enough [if oil will be instead of air]
For electric field intensity approximate formula is:
E [air] =U*epsrpaper/(epsrpaper*thickair+epsrair*thickpaper) [KV/cm]
epsrair = relative permittivity of air =1
epsrpaper=relative permittivity of paper=2.3
epsroil=relative permittivity of oil=2.5
thickpaper =paper thickness =0.3 cm
U=212 KV
There are many types of safety relief valve. I think that this may be useful:

http://www.wittgas.com/filepool/Datenblaetter/englisch/armaturen/av619_es_uk.pdf

Never the less you have to consult the manufacturer prior to purchase the valve.
Best Regards

 

[7anoter4]

Never the less if we take the air gap as 350 mm and if I use a more accurate formula
Eair=epsrpaper/R2*U/(epsrpaper*ln(R3/R2)+epsrair*ln(R2/R1))
For a minimum radius of copper of 35 mm the maximum air Electric Field Intensity will be 24 KV/cm and this it seems to me tolerable. See attachment
Regards

 

[TechARA]

Valuable information, thank you very much.
Is it more practical to add additional pressure relief device or move the top one to be closer to cable termination. I think this will be more effective to prevent cable box flashovers. In case of two devices, can both devices be of the same size and relief capacity knowing that cable box expansion factor is very small compared to transfomer main tank that makes handling the pressure generated by arc is very difficult(could reach 20 bars).

 

[ScottyUK]

Based on a very hard lesson at our site, a PRD isn't guaranteed to vent oil fast enough to relieve the pressure from gas evolved during a catastrophic fault. In our case the flange between the lid and tank ruptured. The idea of using a PRD to prevent flashovers is quite strange - the PRD is the last line of protection and will normally trip the transformer. Prior to that a Buchholz relay might trap gas being evolved more slowly if the transformer has a conservator. Are you really looking for a bleed valve to vent air reliably from the cable box rather than a PRD to vent overpressure from a fault?


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

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

 

[prc]

Such a cable pot cable connection is not designed for 230 kV ,but for 395 kV or 460 kV AC test voltage.Permissible stress levels mentioned by you seems not correct.Paper can always stand much higher stress levels than oil.Usually the oil filled cable box with bushing to pot head connection is tested at factory by putting an oil to air bushing in series to oil to oil bushing.

As Scotty said,a pressure relief device will only save blasting of the cable box that too if you are lucky the fault occurred very near to the PRD.

The patent could not be understood as the figures are not seen with it.

 

[7anoter4]

You are right Scotty, as the PRD will be not useful for an arc flash in the transformer tank, inside. The transformer is provided with Buchholtz relay-in 2 stage- and an expansion pipe c/w a safety membrane in order to release the gas pressure. But, if I well understood, the pot head is located outside the transformer tank. The pot head enclosure is a hollow cylindrical box [see the sketch] and a paper insulated cylindrical copper bar is located in the center of the box. The air volume is relatively small, so may be the PRD will be useful in the case of gas expansion.
The paper insulation of the transformer winding is indeed more solicited and the stress may reach the 80 KV/cm or more as prc said. But still I have to believe that this pot head is outside located and for this configuration my appreciation is correct.
Best Regards

 

[ScottyUK]

Hi 7anoter4,

Thanks for the sketch - I think I see where this is going now.


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

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

 

[TechARA]

I am looking for a good mechanism to vent sudden pressure in case of high energy arcing as a worst case scenario. I learned from your replies that PRD's do not handle sudden oil over pressures since high energy arcing creates over pressures exceeding preset pressure thresholds and the rate of generated gases in oil exceeds the detection speed of a Buchholz relay. So, as you mentioned Scotty that a safety relief valve probably is the right solution to vent more reliably from the cable box. Could you please suggest one and how they differ from Sudden Pressure Relay? thanks

 

[slavag]

Hi.
TechARA, diff (87T) was operated?
Best Regards.
Slava

 

[ScottyUK]

TechARA,

A PRD may be the best you can do. A major arcing fault evolves gas a such as colossal rate that it is unlikely that any practical RV will deal with the volume, especially as you will have mixed flow of liquid and gas. The basic answer is that under major arcing fault conditions your transformer is going to suffer damage. The PRD will, given time, at least prevent the tank from becoming a highly pressurised vessel which would pose a serious risk to structures and personnel near it. I have never personally seen a PRD fitted to a cable box.

To give you an idea of just how energy is let loose during a fault, this is the wreckage of a 48MVA transformer on my site which suffered a massive internal failure:

http://img95.imageshack.us/img95/2821/transformertopisophaseduct3jt3.jpg

http://img223.imageshack.us/img223/9747/unittransformerisophaseductterminalcubicles1on0.jpg

The large PRD operated correctly, it was simply overwhelmed by the rate of gas evolution.


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

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

 

[prc]

HV &EHV Oil filled cable boxes in transformers will have the following accessories. One -an oil to oil condenser bushing for taking out connection from HV winding to cable box.A cable pot head terminating external cable(XLPE ,earlier days PILC) in to oil filled cable box.An insulated lead connection with supports, for connecting bushing to pot head .Normally cable box oil will not be communicating with oil in the transformer tank.A separate conservator(or you can use common conservator) with connecting pipe and gas operated Buccholtz relay in it.A pressure relief device ( similar to one used in main transformer tank,but smaller in size,100 mmor 80 mm size)to avoid cable box rupture in case of a flash over.Different design concepts-eg:elephant trunk arrangement -are adopted for such cable box arrangements.

Usually PRD contacts will be actuated after the gas opertaed relay contacts are operated.You cannot be sure of PRD operation if the arc point is slightly away from the PRD location.Purpose of sudden pressure relay is same as Buccholtz relay.SPR is used in nitrogen sealed transformers as Buccholtz relay is not feasible in that case.

 

[slavag]

PRC and Scotty.
Thanks a lot for the your great explanation.
Best Regards.
Slava

 

[TechARA]

Is it common to use oil rupture disk to depressurize power transformer cable box that is isulated from main transfomer tank? Thanks

 

[slavag]

Hi Folks.
I read now about some intresting idea: used arc protection into trafo tank, based on the fiber optic.
On this moment are few problems.
withstand of connector: +80 deg C and vibration.
As protective relay used relay like to ABB's REA.
Regards.
Slava