wind interface circuit destroyed

[electricavenue]

Hi all,

Having destroyed 2 commercial wind interface controllers - PVI Wind Interface from Power One, I was wondering if more experienced members on this site could shed some light on the problem.

I am not entirely sure about the operation of the circuit itself. I was hoping to save a bit of money by replacing components within the circuit myself but I am not sure how to test faulty components.

I found some really useful information on a posting on this site at http://www.fieldlines.com/story/2008/11/20/186/13008

However I was looking some more detailed information and an explanantion of how the circuit works.

Today I was feeding my new controller with 25Vac RMS, 7amps and getting around 1400Ws on my inverter with a 20Vdc output from the controller. I think that peak outputs from my turbine at night are causing the controllers problems when switching to the dump load for these peaks.

Any comments or thoughts would be greatly appreaciated.

Many thanks in advance.

moved to the proper section.

Not much to go on.

[electricavenue]

I think that the paralleled GTO thyristors are faulty. Has anyone tried replacing these with power mosfets?

[electronbaby]

im confused as to what you are trying to accomplish. You list no turbine type or manufacturer, and no electrical characteristics.

No inverter brand, or type of use (grid tie, off grid).

basically all you told us, is that its not working...  ?

[electricavenue]

Hi roy,

Thanks for your reply, hope the following helps. I am feeding the interface with a 10kW turbine and its feeding a 9kW dump load and the mains through a power one inverter - 6kW.

Hope this helps a little.  Apologies for leaving out anything, not very confident when it comes to electrical specifications.

[electronbaby]

No, it doesnt really help. You didnt tell me what voltage you are using from the turbine into the interface, and therefore into the controller. You didnt tell me what 10k machine you are using. Im assuming it might be homebrew?? What voltage did u wind your stator for? Whats your alternator IV curve look like?

My first guess is that obviously, your dump is sized too small for that turbine...10k turbine on a 9kw dump load. Also, your inverter is sized too small, at 6kw. Who did your MPPT programming? or are you using that function of the inverter?

need MUCH more info.

[electricavenue]

Voltage from turbine to interface controller is ~25Vac, ~7amps. I'm using a homebrew PMG.This is rectified and the output to the inverter is about 120Vdc.

Your right about the dump load as I feel that it should be 15kW minimum-or 1.5 times turbine size but for smaller inverters I was advised to use 1.5 times the inverter rating.

I need to research MPPT programming a bit more. Think my dump load is taking too much power for a prolonged period of time and the interface controller is not rated for handling this power.

Are there any suitable inverters commercially available that you could recommend.

Many thanks for your patience with my novice electronic experience.

[electronbaby]

It would be nice of you to include a picture for the rest of us. Im kinda curious to see your setup. Normally for a 10kw system I would use two PVI-6000.

Your dump load should never come on if the inverter is working correctly, or unless you have a grid failure.

You could be doing much better with the MPPT programming, but for this, the inverter needs to track frequency from your turbine. You need to know the power curve.

It also is a little weird that you are running such a low voltage. 25vac is normally not enough to even turn that inverter on unless you change it from its default setting. Maybe you need a higher voltage.??

Whats the highest voltage you have seen in gusty conditions?....also while furled?

[electricavenue]

thanks for sticking with me on this one. I'll compose a picture and get it up soon to show you what I am trying to do-live in UK and its quite late now so...

My issue is that when there is a grid failure the dump load takes the feed from the PMG but for some reason the wind interface keeps getting destroyed when feeding the dump load. I'm not sure of what components I could change in the interface which may overcome this problem.

In gusty conditions I was seeing around 48V but while furled ~100V.

[electricavenue]

Think i may have found the fundamental fault. The interface max output is ~7kW, while the dump load is rated at 9kW.

Do you know of any similar interface circuits capable of feeding this load?

[electricavenue]

in response to your earlier question about volatges during gusty conditions and while furling - i get about 400Vac and 1000Vac respectively open circuited.

Also the interfeace is only rated for 400Vac max input so I need voltage regulation I think but not sure of any circuits which could perform such regulation.

[electricavenue]

i was reading one of your posts about a windygirl, think this may actually solve the problem. Any further ideas would be great.

[electronbaby]

My first question is whats your swept area, and second, how do you know that its capable of 10kw?

It seems to me that you might want to set the furling early, such as maybe, 7kw.

1000vac is way too high a voltage open circuit. You may want to rewind your stator for less voltage...possibly ~400v open circuit on the dc side (after interface / rectifier). You then should research using the MPPT for wind. Im guessing that you are maybe running the PV version ov the Aurora. They make one for wind also that has special programming. These are the only ones I am familiar with. My experience is that the interface and the inverter work great, BUT you have to stay within the design limits or you are asking for trouble. It might be possible (after you decrease your turbine output slightly) to use two of the 4000w interfaces, and connect them to each inverter input. You are aware that the PVI-6000 has two inverters working in tandem? You may need to have isolation built into your stator, but thats getting more complicated now.

[Flux]

I an glad Roy is sticking with this as I can't follow any of it.

One minute we are talking about 25 to 48v next minute we are talking about 400 -1000 v.

What is a wind interface anyway. What have you got? what are you trying to do?

One minute it looks like a low voltage home built generator for something like nominal 48v.

For grid tie you normally have the option of charging a battery and using s battery grid tie inverter or more usually you have a high voltage inverter which for uk voltages will probably run in the 300 -600v region.

If it is a furling turbine then you need to limit voltage to the inverter with a clipper. Is this what you call the wind interface.

Unless you can explain what you have and are trying to do I doubt that anyone can help you and you will blow up a lot more kit. It's not easy trying to do grid tie with home built generators and unless you know what you are up to there is little chance of success.

If you are trying to clip 1000v with the potential of having 10kW available then it's a major task .

Flux

[Rob Beckers]

Hi Electric,

These are PVI-7200-WIND-INTERFACE boxes? Normally it's one or more NTC resistors that blow due to overvoltage. They are part of a crowbar circuit with an SRC, intended to keep the electrolytic caps from exploding. The crowbar activates very close to 600V DC on the bus (the diversion load comes on at 530V DC, though this can be changed with the little dip switches at the top-left side of the PCB). Look for three dark brown round disks with "NTC 10" written on them. One of them is probably blown to bits.

That by itself shouldn't kill the PVI-WIND box, but the SRC may be dead and failed shorted out. It's an BTW69-800, found on the little auxiliary PCB at the left side, says "Magnetek VMID BULK" at the top of the circuit board. You also checked the 3 fuses, right?

Hope this helps!

-RoB-

[electricavenue]

Hi Rob,

Thanks for the great help. Yes I am using the PVI-7200 wind interface.

Have you expeienced any issues with this kit similar to my problems? At one point the interface was actually sharing the load between inverter and dump load!!! Power one/magnetek couldn't explain over the phone so I was at a dead end until posting here.

I will act on your advice and keep you posted on how I get on.

Many thanks

[electricavenue]

Hi Roy,

In reference to your point about the PVI 6000 unit having 2 inverters in tandem - does that then mean that one of these inverters feeds the mains during normal conditions but during gusts one of the inverters feeds back throught the interface(on its DC bus) to dissipate this power into the dump load? I was of the thinking that the inverter's microcontroller switched the crowbar circuit(as rob mentioned) within the interface box, hence the inverter wasn't actually seeing any power and thus keeping itself safe.

I have uploaded a photo here which shows the system I am using. This is courtesy of power ones datasheets/brochures.

http://www.otherpower.com/images/scimages/12003/wind_system_setup.jpg

thanks Roy

[electricavenue]

Hi again Rob,

Just buying a new SCR at the minute Rob, think that has failed - no output from it anyway.

Do you have a schematic of the interface circuit by any chance, so as i can understand it further?

My understading so far has come from an excellent post you made on a similar issue which stated :

"To make your own would be quite a project. I think do-able, but not all that easy to get right. The voltages and energy involved make it tricky. You could follow the same formula as Power-One: There is an IGBT that switches on the diversion load. An SCR takes care of the crowbar circuit. Then there's the usual electronics to derive a stable power supply for internal use, some 293's for voltage comparators, zeners to drive the SCR, driver transistors, lots of capacitors and a handful of MOVs for surge protection. It's not rocket science, but getting it wrong can be dangerous and/or expensive (as in blown inverters)."

Think I would like to try to build this circuit.However, I wouldn't be sure of component specifications - for a max input to such an interface of 400Vac and a max output of 600V DC to the inverter.

Thanks

[Rob Beckers]

The diversion load is switched parallel to the output (to the inverter). So in other words, it is supposed to come on concurrently with the inverter. The output to the inverter is not switched off when the diversion load is on. What you were seeing (inverter and diversion load both on) is normal behavior.

Yeah, we have experience with blowing up PVI-WIND boxes... :-(

That's why I was speculating that your issue likely is the same we saw happen with our boxes. In case of overvoltage that SCR crowbar switches on and blows in the process. Power-One told me that the circuit was put in at the request of the UL examiners, to prevent the capacitors from blowing (while not really dangerous unless you have your face right above them that would be quite a bang!). It was not meant to self-destruct in the way that it does, but rather, blow the fuses. I would have hoped that they would design the box to handle overvoltage a bit more gracefully. Still, I suppose it is better to blow a PVI-WIND box than a PVI-6000 inverter. Those cost a whole lot more.

-RoB-

[Rob Beckers]

Please contact me by E-mail (Rob_at_solacity_dot_com). As this is somewhat of a sensitive issue. I have no NDA with Power-One, nor have they ever supplied me with schematics (what I have is from reverse engineering), but I am one of their dealers.

-RoB-