Lightning Protection on a Wind Farm 1

[shattenjagger]

Hello All,

I am currently working on a wind farm and we are running into some real issues with getting our towers grounded to the level required to keep a warranty with the turbine manufacturer.

To combat this, I have proposed that we run copper at the same level as the warning tape beginning 300ft from each side of the tower as we trench in our UG collection system. Unfortunately, another engineer is suggesting that lightning could induce dangerous current levels in the 34.5kv direct burial cable as it is dissipated through the grounding grid. The collection system already calls for the concentric neutrals on the cable to be connected to the turbine grounding grid inside the transformer, so it would seem to me that if any current were to be induced, it would be happening already through the concentric neutrals.

So my question is, Is there any merit to my colleagues argument? This new plan will cost between 35% and 10% of the original plan for this issue.

Thanks for your assistance,
Grounding Monkey

 

[sslobodan]

First there is no 100 % protection from lightning, and you should know that. But there is one trick that can help you. Instead of grounding rood from each tower You should consider that Lighting rod pillar is actually protecting as a cone, so sticking rod high enough to get few towers high enough should decrease costs and should save you with the trouble of thinking of the inducted currents (from lightning) in lines.

% %Tower
% 0 % 0 Grounding rood
%
This is usual farm setting so you can increase wind farming capability so I guess you have similar situation on hand.

Now there is always a way to do a more certain protection but I believe it will increase costs of facility
You make small grounding pillars on edges of facility app. 20% higher than the tower height. And if you have space in middle You should make a bigger pillar in middle with app. 40 - 50% of the towers height. Now connect the pillars outside in a continuous line one for each next one |--|--| So you get outside "circle" (the edge of the field is connected) and connect it all with one line per pillar with main pillar in the middle. With this you will get a grid with 99.9% chance of deflecting lightnings. Of course this all is limited with the size of the field and your budget so first solution would look as an cost effective.
And if you have thin line of wind generators you can always do a outside pillars high enough to protect Wind towers without middle pillar.
Generally in all calculations you want to make you should use that Length of protection cone in 2xheight of pillar carrying the Protection rood. so you get cone with dimensions H, r=2H s=H*sqrt(5)=2.24H and that if you put grounding too close to a wind tower you will have induction from the line that is carrying lightning voltage to a ground into a network of a wind powered generators. And no need for copper it will increase price of a system. Cr ZN stripe from pillars and bare steel wire around for interconnection like on HV pillars that are protecting the long interconnection lines

 

[shattenjagger]

Thank you for the outside the box solution to this issue. However my application requires that the resistance measured
using the fall of potential method falls below 5ohms at the 62% distance mark. With out certified documentation of this fact, the wind turbine mfg warranty is null and void.

I do like the idea and may propose it on a future addition to the site. From what I understand this has some potential, however the terrain in many of these wind farm areas are particularly menacing and large. Considering this, what is the minimum spacing on the outside ring of pillars for this setup to be effective? Also, what is the maximum distance between the main pillar and the outer ring pillars?
When you are calculating your efficiency in deflecting lightning are you calculating that the cone of protection covers 99.9% of the turbine surface area?

Thanks,
Grounding Monkey

 

[jghrist]

If the 34.5 kV cable is jacketed so the neutral is not in contact with the earth, the surge current travelling in the neutral will be limited. Any surge current entering from the ground system at the struck tower will have to go all the way to the next point where the neutral is grounded. The impedance to high frequency surges will be very high and little surge current will follow this path.

There will be coupling between a counterpoise wire directly above the cable and the concentric neutral, but the self impedance of the neutral will still be high, so I doubt if a dangerous amount of surge current will be induced. Without a full blown transient analysis, though, I'm not sure if you can show this convincingly.

Your warranty requirement of a measured 5 ohm resistance might be met by a one or two 300 ft horizontal wires, but this would not provide very good lightning protection. Radials are good for dissipating lighting surges, but they are only effective for about 50 feet. Copper tape provides lower surge impedance than wire.

I'd suggest several 50' radials of copper tape at each tower, with a couple ground rods along each radial plus one at the end. A detailed ground system analysis with CDEGS, WinIGS, or similar program would calculate the expected 60 Hz ground resistance to be measured. You could probably meet the warranty requirements less expensively if you did not worry about providing actual protection.

 

[umrpwr]

I have built several wind farms and am quite familiar with the difficulty you are encountering regarding grounding of wind turbines. The GE specifications in particular require I believe require that the resistance to be below 10 ohms, but require additional surge protection and 5 ohms or below without any additional surge protection. The main issue are the overvoltages created by a direct lightning strike of a turbine. Unfortunately wind turbines tend to be located in areas that aren't always condusive to good grounding like in mountains or on solid rock. So getting a good low impedance ground can be almost impossible. The issue with using a copper conductor as counterpoise in the same trench as the collector system cable is probably not feasible. The projects I worked on GE would not accept that as a part of their acceptance and warranty. You may want to verify that though. What we did on the projects I worked on was running a copper conductor counterpoise in a separate trench that was filled with a conductive soil cement. That worked although the lengths required were pretty extreme on the order of a mile or more. The alternative though of having to add additional surge protection is much less desirable based on my experience. I would also mentioned that you should look at the IEEE papers since there are some very good papers that cover this.

 

[MikeDB]

You did not say what the soil resistivity was or how deep the water table was. Depending on the soil type and how much moisture is there, deep driven ground rods or grounding wells may be more effective than counterpoise. It may take a combination of radial counterpoise and deep grounds.

Are you using the concrete foundations as part of the grounding system? They are deep, have a lot of surface area and a lot of steel. They will be part of the path and you are paying for them anyway. You may need to bond the anchor bolts to the rebar.

We have used the copper tape referred to earlier for connections from the tower to the ground grid. I don't using the tape in the ground grid will make economic sense or lower the resistance.

Good Luck.

 

[shattenjagger]

Thank you UMRPWR

That is the exact issue I am running into.
The lengths are pretty long on some of the sites and are almost unfeasible at times. We tried deep earth grounding as well, but the GEM entrenched radials seem to accomplish nearly the same effect for about half the price.

You are correct, the engineer is not accepting the solution of connecting a running ground to the grid and then following the collection trench at 2 feet directly above the cable.
I was pretty much just wondering if anyone had run into this same issue and found a way to convince the engineers that because of the extreme length found in some of the cases, the radials are forced to run almost parallel to the collection system down the roads anyway.

Thanks for everyone input,
Grounding Monkey

 

[sslobodan]

That towers should have been grounded before setting up the equipment. That is completely different from lightning protection. And addition should not increase resistivity but lover it (additional rods in ground). There are some issues about current spreading in such system due to an 2 or 3 phase short circuit (when it happens) but that is a different issue. Now as far as distance there is one thing what you should know. When lightning leader is formed from ions it will create shortest path to a ground (law about natural balance). That means if your lightning system resistance path is higher than path that can connect leader and ground than the one trough your equipment (it is also grounded and because of that suitable to take lightning to a ground) it will discharge over your gear.
I have found some data about studies of Lightning protection in US you can find it at

http://www.nrel.gov/docs/fy02osti/31115.pdf

My contacts with wind turbines were on much smaller models.
and here you can see one serious calculation that might help you with your problem

http://www.hsa.ei.tum.de/Publikationen/2004/04_ki_2.pdf

Hope links will be helpfoul