solar pv disconnection times??

[sparky1505]

Hi, there, just wondered if any one could clarify. With regards to the inverter disconnection. Correct me if i`m wrong - ``the inverter must disconnect the ac supply to the consumer unit within 5 seconds when the mains ac power is lost. The ac supply to the consumer unit from the inverter can be connected to a spare way on the RCD side of the consumer unit. The circuits on the RCD side of the consumer unit need a 0.4 second disconnection time. If the Rcd operates then surely those circuits could still remain live from the inverter for 5 seconds. ?
Cheers.

 

[ericmark]

You make a good point. Grid tie inverters may be required to disconnect in 5 seconds but as to what time they do take one would need to read the information that comes with the inverter. It may be less than the 0.4 second of the RCD?

I did try to download a technical spec PDF to look at disconnection times but they were not given. The problem with wind and solar is of course it can't be switched off so when the power is not required it's a case of diverting rather than switching off. Simple zener diodes may sink the power like used on old motor bikes but one worries about how long they can sink the power for?

I question how the inverter knows the power has failed? If of course because of the power failure there is an overload then that's easy but if the DNO for example came to change your meter then as they remove the fuse the inverter should auto close down but you may have well under the 1Kw the inverter can supply.

Reading Islanding is interesting. I look forward to see others comments.

 

[sparky1505]

Cheers for the reply, Its a trickey one, so far each installation i have done i have connected the inverter onto the double pole side of the consumer unit or where there is no room provided a new consumer unit for this reason. When i did the logic course with Grant they wern`t sure either.
I wonder if the protection is provided by electrical seperation as the inverter has an isolating transformer so the disconnestion time may be irrelevant?

 

[ericmark]

If you read the link seems some countries allow 60 seconds disconnection time and really that should be good enough.

It talks also about the problem in detecting the grid power has failed. There are it seems inverters that allow Islanding and instead just isolate the incoming supply.

All inverters I have worked on were designed to ensure there was no power failure and had three power sources. Single phase which was normally routed straight through and powered all the PC's and coms equipment if this failed then there was an independent 3 phase power to recharge and supply output and it that failed then went to battery power.

It used electronic switching which was very fast and auto sync when power returned.

But what I have been talking about with my son was installing a inverter from a battery bank then adding to that power with a grid tie inverter. However all our efforts to find out how the anti-islanding worked so as to know if we could add inverters in this way failed.

Boat now up for sale so no longer required but still interested in how it all works.

 

[sparky1505]

very good link makes interesting reading thanks, but as Grid-tie inverters are designed to quickly disconnect from the grid if the utility grid goes down. And is an NEC requirement that in the event of a blackout( as Islanding can be dangerous to utility workers, who may not realize that the building is still powered even though there's no power from the grid). distributed generators must provide anti - islanding i.e detecting islanding and immediately stop producing power. The question remains How can it be the case that the requirement of the 5seconds can apply to the inverter disconnection?