300A Supply Side Interconection

iceworm

Curmudgeon still using printed IEEE Color Books
Location
North of the 65 parallel
Occupation
EE (Field - as little design as possible)
Well, what kicked off this discussion (300A on the supply side) is hardly a residential PV system. :D
Yes, I didn't see the connection. The PV is 70kW, the gen is likely in the 50kW range.

The biggest inverter gen I've seen is 3kW, The biggest I've heard of is 7kW. **

And I don't know why an inverter gen would be "cleaner" than one with copper, iron, and a rotating magnetic field.

many modern generators are DC generators coupled with an inverter; there's no reason for the power from them to be "dirty".
I'm thinking "dirty" involves millisecond to sub-millisecond transients. Slowly changing voltage or frequency swing don't tent to cause any problems. "Slowly changing defined as more than 100milliseconds for a variation not exceeding spec. Which makes me think I need a definition of "dirty". In this context please, I promise I did take a shower this morning.

the worm

**Not exactly true. I did see one with a design capability of 40MW for 1/2 hour. But maybe that was a UPS not a generator - it was battery fed. Still, absolutely impressive.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
By dirty, I was primary thinking of frequency.
At any rate, I know from experience that a Sunny Boy will recognize at least some generators' output as grid power. In the Sunny Boy/Sunny Island/generator system I used to build for a course I taught, when the batteries were low and loads were engaged the SI would charge the batteries from the AC bus with both the SB and generator (a Generac) running.
 

pv_n00b

Senior Member
Location
CA, USA
So, you are saying the PV inverters are always connected to the utility. The transfer switch disconnects the utility and PV inverters when the generator is on line.

Question: I'm not familiar with PV inverter systems. So why not allow the PV inverters to be on-line when the generator is running? I can see that one would have to deal with some sync issues.

There are several reasons why PV inverters designed for grid tie can't be islanded with a generator.

1) It used to be that generators did not provide quality power and the inverters would either not start or cycle on and off. Many newer generators, that are not the cheapest on the shelf, will provide quality power though so this is less of a problem.

2) Inverters designed for grid tie do not modulate the output based on the load. They take the full output of the array and put it out as AC. So if the load is less than the PV output that energy has nowhere to go and the inverter has no controls to reduce the output. That will result in damage to equipment and unreliable system operation.

The only way to use grid tied inverters in an island system with a generator is if the PV system is sized to provide maybe 20% of the minimum expected load at any time the PV system is producing. The generator picks up the remaining 80%. These are the numbers SMA uses for its diesel assist PV systems in industrial projects. It's not reasonable to design a system to comply with this in a residential or commercial environment because it would limit the real reason for the PV system, to make energy to offset usage from the grid.
 

iceworm

Curmudgeon still using printed IEEE Color Books
Location
North of the 65 parallel
Occupation
EE (Field - as little design as possible)
There are several reasons why PV inverters designed for grid tie can't be islanded with a generator.

1) It used to be that generators did not provide quality power and the inverters would either not start or cycle on and off. Many newer generators, that are not the cheapest on the shelf, will provide quality power though so this is less of a problem.

2) Inverters designed for grid tie do not modulate the output based on the load. They take the full output of the array and put it out as AC. So if the load is less than the PV output that energy has nowhere to go and the inverter has no controls to reduce the output. That will result in damage to equipment and unreliable system operation.

The only way to use grid tied inverters in an island system with a generator is if the PV system is sized to provide maybe 20% of the minimum expected load at any time the PV system is producing. The generator picks up the remaining 80%. These are the numbers SMA uses for its diesel assist PV systems in industrial projects. It's not reasonable to design a system to comply with this in a residential or commercial environment because it would limit the real reason for the PV system, to make energy to offset usage from the grid.
Thaqnks. However, except for the 20% SMA spec, ggunn and jag already explained this.

Except for item 1. I don't know where that comes from. AC power generation principles have not changed much in the last 135 years. If you are discussing hardware store grade generators, I seriously doubt they have improved any in the last twenty years. Maybe you could define "quality power" and give some context for the type of generator you are discussing. Here are some specs that come to mind:
Frequency limits and rate of change
Voltage limits and rate of change
Crest factor
Probably transients (millisecond, sub-millisecond) don't matter much. Iron, copper, rotating mag field rarely have any. Although, inverter type gens could.​
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Thaqnks. However, except for the 20% SMA spec, ggunn and jag already explained this.

Except for item 1. I don't know where that comes from. AC power generation principles have not changed much in the last 135 years. If you are discussing hardware store grade generators, I seriously doubt they have improved any in the last twenty years. Maybe you could define "quality power" and give some context for the type of generator you are discussing. Here are some specs that come to mind:
Frequency limits and rate of change
Voltage limits and rate of change
Crest factor
Probably transients (millisecond, sub-millisecond) don't matter much. Iron, copper, rotating mag field rarely have any. Although, inverter type gens could.​
I don't know a whole lot about the whys and wherefores, but see my post #23. With the equipment I had at my disposal eight years ago, the GT inverter recognized the genny as grid power.
 

jaggedben

Senior Member
Thaqnks. However, except for the 20% SMA spec, ggunn and jag already explained this.

Except for item 1. I don't know where that comes from. AC power generation principles have not changed much in the last 135 years. If you are discussing hardware store grade generators, I seriously doubt they have improved any in the last twenty years. Maybe you could define "quality power" and give some context for the type of generator you are discussing. Here are some specs that come to mind:
Frequency limits and rate of change
Voltage limits and rate of change
Crest factor
Probably transients (millisecond, sub-millisecond) don't matter much. Iron, copper, rotating mag field rarely have any. Although, inverter type gens could.​
The specs would come from UL 1471. Frequency is quite tight, I think between 60.1 and 59.7 for 60Hz. Voltage is +10% or -12% of nominal. Waveform must be within spec for five minutes before inverters start outputting. Those are the basics.
 
Top