gfi

[Shaun Burgess]

hi all . last night my problem came to a head. before when i turned turbine on the gfi light would come on and inverter would turn of then restart perfect . well now its turning on for 30seconds doing its checks and gfi light comes on turns of and keeps turning on and of until i brake turning then turns off .. the big question is were to find the problem i have 360mtr of 3 core 16 square swa under the ground is there any way i could check theses .. could i try hooking up a petrol generator at turbine and send down ac power either 110v 0r240v on 2 of the wires then change to the last wire . could i then turn the inverter on and see does it power up without gfi then i would no its the turbine theres a juntion box on turbine or it could be the stator .. will still be using recifer so the inverter will be getting dc power just an idea i had ..i tryed doing a omhs check at turbine theres about 13mtrs of cable up to head' could not get multimeter to give a reading the figures just kept jumping about turbine was stopped and disconnected from the wires down field.. thanks for any help sb

[Rob Beckers]

Shaun, I've attached Power-One's guide for troubleshooting ground faults. Hopefully that helps. For the wind inverters I have never seen a case where the inverter erroneously gave this error (though in honesty there is one case I know about where the installer never resolved the cause). In other words, the odds are that this error is real, and not an inverter defect.

One thing that is fairly easy to test is section 3.4 in the document, to see how much is leaking to ground while the turbine is running. Just make sure the inverter is not connected to the grid when you do this!

I've seen a few cases where the inverter will start up with a ground fault, while later in the day it will work fine. The cause in those cases can be condensation on exposed conductors (in the turbine, wiring boxes etc.).

If you can get your hands on it, a "Megger" is what's needed to measure insulation resistance. Power-One calls it a "MegaOhmmeter" in their document. It puts a high voltage on the line (in this case Power-One suggest not going above 500V) while it measures resistance. It is far more accurate than a multi-meter, which measures at just a few Volts from a little battery. Megger is actually a brand name, but there are many brands making them and everyone refers to them as Meggers.

Using a generator to isolate the turbine itself and test the wiring is certainly possible. Make sure though to change the MPPT table (you can save your existing table into a file, using AuroraInstaller), and write an MPPT table that puts just a small load on the generator.

If all else fails, it is possible to add an isolation transformer to the 240V side of the inverter, and switch off the ground-fault detection in the inverter. That will make the turbine and inverter work even if there's a ground fault. However, those transformers are not cheap, and they use energy even when there is no wind (you can switch it on/off based on wind by using a measurement relay and contactor, I can send you a wiring diagram, but it gets complicated that way).

-RoB-

[Shaun Burgess]

thanks rob today at the point were power from turbine came into the garage . i rewired from that point throw a different controller into inverter and bypassed the old system . and am still getting ileak above 30mla.. am told should be under 20mla.. so nearly has to be wires in field or turbine .. will have to get digger going and drop turbine . its funny when mill is down the wind is blowing strong.:laugh:. will you please send the wiring diagram to over ride the ground fault detector could be cheaper then trying to replace 360mtr of wire.. thanks sb

[Shaun Burgess]

well i megger the stator and wires today .. . 1 of the wires on the run to inverter is grounding to the steel wire on cable .. so the fun has started and trying to find over 360mtr .. if there a devise that could find this fault or is it a matter of cutting cable in middle and working from there thanks

[Rob Beckers]

Hi Shaun,

Actually there is equipment that will tell you where shorts or discontinuities in a cable are; electrical companies use it to find out how far down the line things are broken (I'm from the Netherlands and remember seeing this being used 35 years ago by the electrical company my dad worked for, to find a break in the underground wiring, so they would know where to start digging, rather than to rip up a whole street). No idea though were you would be able to get your hands on it, or even who makes this.

It works by sending a high-frequency pulse into the line, and measuring the time it takes for the reflection to bounce back, reflecting off a discontinuity. Time translates to distance, that's how they know where to dig. If you have a HAM operator amongst your friends who has a scope and pulse generator you may be able to rig this up yourself.

For a wiring diagram to allow you to switch off ground fault detection in the inverter, see this link: http://www.solacity.com/AuroraStack.htm
While this is about stacking wind inverters, it requires the same thing. To stack inverters all but one needs to isolated with a transformer.

-RoB-

[Shaun Burgess]

hi rob . am back up and running . i found a small cut in the cable one phase was touching to the outer steel wire in cable . the problem should have been seen early but i never grounded the outer steel wire each end . should i ground this cable at turbine and down at inverter . . at min inverter is showing around 11ma (ileak) is this ok shud it be lower. and also i think this has cured my powering down problem .. happy days thanks sb

[Rob Beckers]

Shaun, 11mA leakage current is fine. This is in the middle of the values I've seen over the years. Even if your insulation is perfect there will be a leakage current, because the wires act as a capacitor to ground.

For armored cable (sounds like that's what you have) the steel mantle should be grounded for safety. The risk is that it gets energized by something like your cut in the insulation, and then passes that voltage on to other parts, which you can potentially touch (electrocuting yourself).

Glad to hear it's working again!

-RoB-