Need help about lightning arrester/surge arrester

[narasura]

so I'm working on a project about the usage of multiple lightning arrester by paralleling it, with 2 or 3 arresters.
and my project mainly discuss about the discharge current ( current flowing through the arrester to ground)
so how would be the effect exactly??

the theory says, that arester is a resisstive component, and when you use multiple of it, the total resistive value would decrease, just like the equation (1/ I total) = (1/R1)+(1/R2)+(1/R3)+...+(1/Rn)
and with the resistive value decreased, the current value would be increased ( V=I*R)
so if I'm using 2 arresters, the discharge current would be doubled, and if I'm using 3, the current would be multiplied by 3, right?

but in my project, by using 1 arrester I get the discharge current of 119,884 Amperes
yet, when I'm using 3 paralleled arresters, the discharge current is only 182,1012 Amperes..

is the theory wrong? or is there anything that i missed that causes this??
please help me

 

[Baluncore]

There are two types of common anti-surge device. One is a voltage dependent resistor, usually made from zinc oxide. The alternative is a gas discharge device that contains an inert gas.

What type of device do you have? What is it's part number?

The peak current will also be dependent on the pulse generator used to trigger the device. What trigger are you using?

 

[narasura]

There are two types of common anti-surge device. One is a voltage dependent resistor, usually made from zinc oxide. The alternative is a gas discharge device that contains an inert gas.

What type of device do you have? What is it's part number?

The peak current will also be dependent on the pulse generator used to trigger the device. What trigger are you using?

i really appreciate your reply, good sir..
for the arrester, I'm using OHIO BRASS PDV100 Optima
it's the metal oxide one, with rated voltage of 18,5 kV written on the arrester.

for the impulse generator, the one I'm using in my lab is
High Voltage Testing Device, Type D250 (52), by Ogawa Seiki Co.,Ltd
and I'm still not sure about how big the current it produces when it simulates impulse(lightning strike)
here's the spec on its plate
· Power supply capacity : 5 kVA
· Input voltage : 220 V
· Output voltage : For AC test 50 kV, DC test 70 KV, impulse test 200 kV
Frequency : 50 HzI'm just thinking a bit, should i change the arrester to the ones with smaller discharge current capacity?
for example, the new arrester would break if apllied with impulse that has current of 30 ampere
so when I'm paralleling 3 of it, those arrester could withstand impulse with up to 90 amperes of current?
because the arrester I'm using right now, maybe had the current capacity up to 10kA in the spec sheet CMIIW.
and if that would be true, then my project had to break/destroy many arresters to complete :/

 

[Baluncore]

The resistance of a triggered MO varistor is non-linear. That is one reason why it does not perform as expected.

Depending on the geometry of the conductors near the arrestors, there may be transients that effectively distort the profile of the test pulse. It is also possible that one of the arrestors may be protecting another.

When testing you should keep track of the energy dissipated in each device as the characteristics and lifetime of a MO varistor is dependent on the accumulation of damage due to triggering.

Yes, it could get expensive.

 

[anorlunda]

What Baluncore said about testing also applies in the field application. Distortions and non-linearies may make the currents in the parallel arrestors unequal.

May I ask why you want multiple arrestors in parallel?

 

[narasura]

The resistance of a triggered MO varistor is non-linear. That is one reason why it does not perform as expected.

Depending on the geometry of the conductors near the arrestors, there may be transients that effectively distort the profile of the test pulse. It is also possible that one of the arrestors may be protecting another.

When testing you should keep track of the energy dissipated in each device as the characteristics and lifetime of a MO varistor is dependent on the accumulation of damage due to triggering.

Yes, it could get expensive.

how can i keep track of the dissipated energy in each device? whew I still get ton of things to learn..
so is the accumulation damage from triggering would cause the arrester to heat up and then makes the resistance to increase/decrease? and what are the factor that causes the nonlinearity in the arrester? sorry if I ask too much :(
i think that accumulation damage is one of the main problem aside from the peak current that the impulse generator can produce, because I didn't know about that, and just triggered it with only less than 15 second intrval, and i did it many many times :/

What Baluncore said about testing also applies in the field application. Distortions and non-linearies may make the currents in the parallel arrestors unequal.

May I ask why you want multiple arrestors in parallel?

oooh okay, thanks sir, I will sure to keep this in mind

it's about my undergraduate thesis, may professor offered me this theme as the research for the thesis
the goal is to find a way to replace arrester with higher discharge current capacity, by paralleling some arresters with smaller discharge current capacity
in case of lack of stock in the warehouse or something..

 

[Baluncore]

Another problem with prediction will be that the arrestors are chains of spark gaps and Silicon Carbide varistors. Once the spark gaps break down at the predicted voltage, the current is limited by the varistors.

Also there may be capacitors and/or resistors across the spark gaps to equalise the voltage across the series chain. There may be blowout magnets on the spark gaps to help recovery from the active state. There may be arc circulation magnets to increase the power handling capacity of the spark gaps.

 

[anorlunda]

I also suspect that the result may be non-repeatable. That is something you could explore in the testing environment. Non-repeatable results are a concrete reason to go back to the professor and say, "Sorry, it won't work because..."

 

[narasura]

Another problem with prediction will be that the arrestors are chains of spark gaps and Silicon Carbide varistors. Once the spark gaps break down at the predicted voltage, the current is limited by the varistors.

Also there may be capacitors and/or resistors across the spark gaps to equalise the voltage across the series chain. There may be blowout magnets on the spark gaps to help recovery from the active state. There may be arc circulation magnets to increase the power handling capacity of the spark gaps.

I also suspect that the result may be non-repeatable. That is something you could explore in the testing environment. Non-repeatable results are a concrete reason to go back to the professor and say, "Sorry, it won't work because..."

from what all of you just said, I think this theme is too wide for an undergraduate/bachelor degree thesis.
and my head just hurts with the many factor that causes the result would be unexpected
thank you for your responses, I'll go consult to my professor and maybe get simpler theme,
or maybe just proceed with this and said that the expected result could be done, but with more wider research, and that would be another thesis's job...

 

[anorlunda]

The choice is yours, but don't be too hasty. An engineering undergraduate should be able to design, set-up, conduct, and document tests. That does not mean that you must be able to explain all the reasons for the results.

Show your professor a printout of this thread and ask if he wants you to continue.

Edit: You could start another thread later on PF explaining your test plan and ask for comments.

 

[narasura]

The choice is yours, but don't be too hasty. An engineering undergraduate should be able to design, set-up, conduct, and document tests. That does not mean that you must be able to explain all the reasons for the results.

Show your professor a printout of this thread and ask if he wants you to continue.

Edit: You could start another thread later on PF explaining your test plan and ask for comments.

okay sir, certainly I will print this thread and ask for his comment, maybe things will get clearer
and again, thank you very much for answering my questions and for the advices