Battery Storage

Eoin

My solar panels are now producing a lot of energy, even in February when the sun is still low.

My next problem is how to manage that energy. The price ratio selling:buying electricity is about 1:4 and there are 2 issues I need to look at.
1. During a normal day, output is very spikey. I can be running at 30 - 40 kW and then down to 10 kW when a cloud comes over. I could do with being able to cover the troughs.
2. Night time we're running a server plus fire / intruder alarms and air con in the server room. Not a huge demand, but running all night, every night. It would be good to cover this demand as well.

Anyone done this with a 3 phase installation? The clever bit will be something that knows when the system is exporting so it sends to battery and when it's importing so it supplies from battery.

Gulfcoastguy

No but there are plenty of people on electric car forums who probably have.

Eoin

Gulfcoastguy said:

No but there are plenty of people on electric car forums who probably have.

I've seen some stuff, but generally small scale and single phase.

nolaegghead

You want a system that will run stand-alone with batteries that only exports energy to the grid when the batteries are full and demand is met. 

I don't have any experience with any system other than our residential, but there are vendors out there and the industry is pretty competitive. 

I'd spec out a good design and buy the components ala carte. 

The big expense will be batteries to tide you over for a day of minimal sun.  Every subsequent day will almost double the cost. 

The LiFePO4 battery technology is probably what I would go for.  Handles very high charge/discharge rates, great longevity and cheaper than other Lithium technologies.  I use a 24V 100 Ah battery for my boat's trolling motor.  I see people buying large numbers of these for off-grid systems.

The battery management system needs to be flexible to the battery voltage it handles to convert and invert.  There's a limit to the number of parallel batteries you can run, and it's a good option to be able to go up another in series (higher voltage).

I'm not sure an "All-in-one" inverter/charger is the way to go....if you can modularize that stuff you may have more flexibility for growth.  If that's something you want.

Gulfcoastguy

Nola has this one right.

Eoin

nolaegghead said:

You want a system that will run stand-alone with batteries that only exports energy to the grid when the batteries are full and demand is met. 

I don't have any experience with any system other than our residential, but there are vendors out there and the industry is pretty competitive. 

I'd spec out a good design and buy the components ala carte. 

The big expense will be batteries to tide you over for a day of minimal sun.  Every subsequent day will almost double the cost. 

The LiFePO4 battery technology is probably what I would go for.  Handles very high charge/discharge rates, great longevity and cheaper than other Lithium technologies.  I use a 24V 100 Ah battery for my boat's trolling motor.  I see people buying large numbers of these for off-grid systems.

The battery management system needs to be flexible to the battery voltage it handles to convert and invert.  There's a limit to the number of parallel batteries you can run, and it's a good option to be able to go up another in series (higher voltage).

I'm not sure an "All-in-one" inverter/charger is the way to go....if you can modularize that stuff you may have more flexibility for growth.  If that's something you want.

I'm looking at components now, the controller will be key. I can't cater for an overcast day, but I want to cover night use (low drain/long time) and dips during an otherwise sunny day.

nolaegghead

I'm sure there are worksheets out there, but you need to audit your usage to determine all the metrics that will help you purchase the correct equipment.  If your KWhr max generation capacity isn't grossly over your average usage, you can't generate enough to save for the overcast day anyway.  Unless you have variable rates (cheaper at night), not worth the extra batteries.

You can do a paper audit to figure some stuff out, or put an inductive current meter recorder on the feed to get real time recordings of usage.

I have a good friend who's an EE that does this exact type of work, but in the US.  I can ask him for any advice.  He owes me a favor for a drone survey.

nolaegghead

The audit will also tell you if you have spikes in usage.  You can spend a lot of money to supply intermittent high current demand from batteries, it's not worth it in most cases unless it's an essential demand for business and, say, you are on a very unreliable grid.

I think the meters used are inductive clamp recorders.  Your electric company might be able to do it using your existing meter.  Those meters are very accurate.

nolaegghead

Just some stuff I'd like to share I just learned.

In the US (don't know about UK/EU):

5% of generated electricity is lost in heat in the grid (lines, transformers).

After adjusting for inflation, major US utilities spent 2.6 cents/kWh on electricity delivery in 2010, using 2020 dollars, and spending on delivery was 4.3 cents/kWh in 2020, while utility spending on power production decreased from 6.8 cents/kWh in 2010 to 4.6 cents/kWh in 2020, the EIA said.

The amount of credit you get with selling electricity back with net meters is more a political decision.  I hope we encourage more solar conversions through good legislation to help individuals and small business to go solar with favorable net meter policy.  Looks like yours (25%) is unfavorable.  Ours is 100%, for now.  The utilities here are private but controlled by state government.  They really work for us and solar installation do take a lot of stress off our already crippled grid. 

Eoin

I have a lot of capacity, so I'm looking to utilise that for night hours and to cover dips in output. I don't want to over spend, but I think it will be worth covering those requirements. Once I have a proper useage pattern I can make the calculation.

nolaegghead

These kind of appliances have bee around for years.  They're cheap.

You can measure what you use on the longest night of the year, or extrapolate.  If you can get this meter on the computers, etc.

Lights are easy - consumption is on the bulb/fixture.

Anyway, do the math and figure out how many kw/hrs you use.  There's an efficiency factor for inverting the power back to AC from battery. 

I don't know what you're running, but chances are it's not that much electricity if it's just a couple computers.  High end servers can easily suck 800 watts or more.

The reserve for dips would probably be were you need the bigger capacity.

Tesla power wall is not cheap, but it's pretty turn-key.   You might want just run your power backup on a single phase.

https://support.solarquotes.com.au/hc/en-us/articles/115001596554-How-does-a-single-phase-inverter-on-a-3-phase-supply-affect-my-self-consumption-

Plenty of other companies copy the power wall. 

If it was me, I would go 100 or 200 Ahr LiFePO4 batteries and make my own array and enclosure (needs to vent outside and be fire safe).

nolaegghead

https://www.solarchoice.net.au/blog/solar-pv-battery-storage-sizing-payback-calculator#fullcalc