solar grid-tie inverter when running on generator

[TampaSolar]

Looking to build and install a 10kW solar grid-tie system. Normal operation will be with local utility grid attached. When utility is down ....

Already have a 24kw diesel generator on-site and want to be able to get the solar to sync with generator. Need to find the inverter designed to limit panel generation so as to NOT backfeed excess current into generator windings. In other words, an inverter that can sense line frequency, not just to force a disconnect of the inverter, but sense line frequency and realize the generator seeing reduced load and in turn reduce the transfer of panel energy to the load.

This seems to be possible if an inverter is designed smartly. Any feedback would be great.

 

[davidbeach]

I'm aware that Sunny Boy has something along those lines as I've seen it come through on interconnection requests.

 

[TampaSolar]

I just talked with an oriental sounding engineer at SMA about Sunny Boy. He said that grid-tie devices have the ability to burn out a generator and that there is no current limiting in the 8,9, or 10 kW units. If the line and freq stays within specs, the inverter will max out the current it delivers.

According to the engineer the Sunny Boy will not ramp down the current even if the frequency or voltage drifts higher. Nor can it be configured to do so. It is an all or nothing device.

Disappointing. Still hoping some engineer has thought this through and provides a solution for those of us in the hurricane states.

 

[davidbeach]

Ask them about the "Sunny Backup"

 

[wayne440]

Is your total load greater than 24kw? If not, it seems you are going to a lot of trouble that will save a little fuel, while doing much to promote wet stacking your genset. Consider simply dropping the solar generation when you have a utility outage, and let the diesel do what it was made to do.

 

[TampaSolar]

Proposed Project Purpose:

to utilize solar energy during a utility power outage

The Situation:

isolate house, run on off-grid inverter or generator, tie with solar inverter and load managed variable dummy load.

solar inverter being and on or off device can pass more energy at any given moment than the house can consume. Grid-tie diverts that energy out to the utility system.

Off-Grid usage must use a 240 volt ac electric heater strip (5 or 7.5 kw) to discard unused solar panel energy.

logic:

Heater to be virtually off when line voltage is 240 ish or lower

When line voltage rises to and beyond 245 volts, use scr or triac to pass current to heater element and draw current up to 100% in order to attempt to regulate voltage to 245 or less. This should allow solar inverter to remain on, thinking it is grid-tied.

YOUR THOUGHTS PLEASE? Have you seen anything already made that will do this?

 

[waross]

I remember seeing an installation where a number of design decisions ranged from poor to incompetent self serving. The installation was a complete failure and was abandoned in less than a year (At the taxpayers expense!) One of the questionable decisions was to use load bank voltage control.
Now load bank control may work for some installations but in your case I find it a little scary.
Fuel consumption; At low loads most of the fuel a diesel uses is for overhead. Just keeping the engine turning and up to temperature.
After overhead, you can expect a little over 13 KWHr per gallon of regular diesel fuel. Since the solar system will be taking the top off the load on the gen set, 13 KWHr per gallon is a pretty accurate rule of thumb for the savings from solar. Some sets will achieve 14 KWHr per gallon after overhead.
You can run light loaded if you want but I have seen a gen set glaze the cylinder walls and pump all the lube oil out the exhaust in a couple of hours.
Wait, I lied, I have seen two sets do that.
You can expect a frequency droop of 3% with most diesel sets. The frequency will drop 3% (1.8 Hz.) From 61.8 Hz. at no load to 60 Hz at full load.
I understand that customer pressure may drive poor decisions and then we must do our best.
I would suggest keeping the base load on the set at 25% or more of rated load. If there is any sign of oil pumping this may be remedied by running at 75% or more load on a load bank for several days to seat the rings. I have had to run for 5 days to set the rings to the suppliers satisfaction to avoid having the warranty voided on a 275 KW set.
Remember, if you have to do a load bank run for several days you will blow enough savings out the stack to offset 5 or 10 years or more of power failure, solar contribution.
Bring lots of water. I boiled off several hundred gallons of water a day load banking the 275 KW set.
I would look into modulating the DC into the inverter rather than trying to make the inverter do something it doesn't want to do.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[TampaSolar]

Really looking to migrate to a 'no generator' plan.

Thinking about a couple of kw 240v inverter from a battery bank. The last post is the way to manage quick load changes. Another variation is a current control on a battery charger to the batt bank where the charger current is regulated to prevent excessive line voltage, i.e. 240-245 range.

My initial system is stuck at grid-tie inverter from solar panels to meet utility contract, or I lose rebate incentive.

 

[LionelHutz]

You will lose any potential grid-failure savings once you build and maintain a large battery bank. In my opinion, batteries are by far the worst part of an off-grid system.

You're also no being very green when you pile a big bank of lead acid in your basement once you consider the enviromental cost to build and recycle them.

From your handle are you installing this in Florida? Do you really have enough power outages in Forida to need to worry about a few $$ more fuel being used? After all, if you mess this up you will be replacing a 24kW generator or a 10kW solar inverter.

 

[TampaSolar]

It would really be a shame if Florida got hit with another big hurricane and we had days (or weeks) without power and had to look at perfectly good panels, glistening in the sun after the storm, with no output whatsoever because the grid was down.

I don't want to offend anyone, but I was hoping for some electronic skills feedback rather than political opinion.

The battery bank would not be large by environmental standards. Just 4 walmart Series 27 marine batts. 4x 105ah. Enough to run off-grid 240vac inverter through occasional cloudiness. The planned load-managing charger enough to sink excessive solar current during load shifts.

I'd like to be smart and plan for survival as well and a time when there may not be diesel fuel around or generator breaks.

Is there any other application where a batt charger is regulated in this fashion?

Thanks for everyone's patience.

 

[waross]

I worked on a solar/battery/diesel generator system in the tropics.
Battery replacement ran over 20% annually.
A big part of the problem was the customer overriding engineering decisions with hunches, guesses and misconceptions. When the sun was shining the solar panels would supply about 60 Amps to the batteries. When the generator was running a large battery charger and an over sized automotive alternator supplied an extra 120 Amps.
We needed it, because the customer ran the system in a very foolish manner for the sake of convenience.
Back in those days I would valiently struggle through to ultimate failure. Now I've been there, done that. If I hear too many Ya-buts I leave.
But, there is one last possible solution. Transform and rectify the output of your generator and feed it into the input of the inverter. Better, drive a DC generator with the diesel. That still leaves wet stacking issues that may still blow your savings up the stack.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

What is the true expected load of the house?

It's not very smart to run 10kW of inverter with 7.5kW of power being wasted trying to power a base load somewhere around say 1kW or 2kW.

OK, try this.

Enphase Energy builds a solar inverter for a single panel. There are likely other companies that do this but Enphase Energy is the only one I'm familiar with. The inverter is ruggedized for outdoor use with weatherproof connectors. The idea is you install an inverter on the rack under each panel and they are quick and easy to install.

Take the above inverters and create small PV banks consisting of 1 to a few panels and inverters. You can then switch these banks on and off according to the current building load. You'll need to build a control panel with some type of power meter and a programmable controller.

Make sure you keep enough base load on the genset you don't kill it.

 

[LionelHutz]

I didn't comment again on the batteries in the last post. However, it may help knowing why I suggested the Enphase Energy inverters instead of a battery system.

If you want a battery-backup based system for off-grid and off-generator use then you need to have the solar and generator produce battery charging DC and then use inverters to power the building.

You would use a charge controller between the panels and the batteries and one or more inverters on the output of the batteries to power the house. As already suggested, you'd also have to convert the genset to do battery charging instead of power the AC systems in the house. You need to buy equipment that will do a correct job of maintaining the battery bank, as opposed to doing something like trying to use a transformer and rectifier on the output of the genset and "hoping" it will produce about 14.3V to properly charge the batteries. This adds more equipment and complexity compared to the genset direct to the building connection.

I seriously doubt your plan of using 4 x Walmart deep cycle batteries will succeed. You should be looking at batteries specifically for this purpose, and likely more capacity as well. You need to build/buy the proper storage box and venting system for these batteries. You also need to maintain the electrolyte level by checking them and adding distilled water every few weeks or buying the special caps that reclaim the water as it evaporates.

In the end, I personally think that using a battery based system add too much maintenance and causes such an efficiency loss that it hurts your normal on-grid operation to the point the payback or break even point will never be reached. You will need to switch the solar system from the battery backup system to direct grid-connected inverters for normal operation, since that will be the normal operating mode for >99% of the time. Still, the battery system needs to be float charged and periodically tested for operation so you know it's there for the few times when you need to use it.

The above is why I suggested you consider using the Enphase Energy inverters. You can then use the solar system to partially offload the genset during the day without moving to a battery based system.

 

[Sparweb]

When considering a grid-tied system, with battery back-up, then I would expect the batteries to be kept at float and fully charged all the time, normally. In the event of a grid outage, the batteries remain available for all or partial needs.

If they're full, then they won't be of much use as a load - particularly 4 golf-cart batts under a 10kW solar array! Those batteries, feeding an inverter, probably won't be capable of starting a 3/4hp motor. Been there, done that. Fat wires or not, the voltage drop across the cell terminals will probably cause the inverter to drop out on a low DC voltage.

How about other loads that can be given higher priority? For example, electric water heat can absorb a lot of energy, and may be a suitable diversion load. You will need a dedicated regulator for this. Now you need to manage the priority between the electric water heat and the grid, or running the generator when absolutely necessary. But at least it will improve the customer's stated goal of capturing more of the solar PV energy, even when the grid quits.

How often, and for how long, does the grid quit? Seconds/minutes/hours at a time? These factors have an impact on your choices.



Steven Fahey, CET

 

[TampaSolar]

It has been pointed out to me that grid-tie inverters are 'all or done' i.e. on at 95% efficient or off.

I ponder that if less output is desired why the mppt optimization can't be severely skewed to accomplish such. We know that the logic is to shift the dc loading or the panels such that the current times the voltage is maximized. Why can't a design be made where the user or management logic allows for a higher voltage and thus a lower current when less output is desired. It seems a no brainer and quite possibly the case in off-grid applications. It does not seem this should manifest anymore heat or be much less efficient.

Could it be the utility industry and/or the NEC or UL codes prohibit such?

I am still in search of a way to use a stock grid-tie inverter when there is a utility outage.

Thoughts?

 

[LionelHutz]

Grid-tied inverters and stand-alone inverters have completely different control requirements. I would expect that the UL listing for grid-tied would not allow the control scheme to be changed to stand-alone since that could lead to the inverter feeding power onto the grid during an outage. I've never looked at the UL file for the grid-tied inverters though.