On Grid Solar

[savage]

Hi Guys,

I'm not sure that these numbers are right - it seems WAY high to me. My average consumption is about 3,000KWH per month. This figure will include some room for growth going forward.

Based on a calculator I found online (iPower SA), the stats basically boils down as follows:
Average consumption: +- 3,000KWH per month
Total Load Offset: 75%
Sun Hours: 8.2 (Cape Town)
Total Wattage needed on Panels: 11.89KW
With 200W Panels: 60 Panels required
Inverter Required: 12KW???

Really? Is this correct?

The basic idea is that I want to look at options for a grid-tied system. Energy will mainly come from Solar, and any shortfalls will be fetched from the Grid. If I fetch from the grid, I also presume that I don't have a choice but to send excess energy back to the grid too, correct?

I'm also unsure about something else in terms of ratings for these inverters.... Most of the inverters I've seen are very low spec systems. Is the general consensus that you just put "what is required" on Solar and leave the rest on the grid? Just boiling a kettle of water for example, the kettle will use in excess of like 2 or 3 KW of power... Way above the average rating of inverters... How do they deal with these short bursts of overload? How would they deal with long term overload?

Would the better bet also be to get a ATS (Automatic Transfer Switch), or rather opt for a grid-tied inverter?

 

[P924]

How on earth are you using 3000kWh a month? I'm not using a tenth of that. Oh, and the year average sun hours per day is closer to 5...

 

[savage]

How on earth are you using 3000kWh a month? I'm not using a tenth of that. Oh, and the year average sun hours per day is closer to 5...

My average consumption is about 3,000KWH per month. This figure will include some room for growth going forward.

Averaged over an 12 month period (winter + summer), 6 ppl, 2 geysers, and quite a few (work related) power hungry systems.

My usage is my usage, is my usage... And I sorely pay for that, every single month too. That's hardly the question here.

 

[P924]

Well the point is - you will sorely pay for that by going solar. It will be easier to get that down first. You should be able to get that down quite a good chunk by installing solar water heating, which is much cheaper than PV. Looking at your graph your usage is actually much lower than 3000? What happened in June/ July, did you get billed for both in June?

Devices are only getting more efficient, so you should not need that much headroom.

Mlt inverters make bigger inverters. And some inverters are made to be scalable, ie you can cluster loads of sunnyboy inverters together to suit you power requirements.

 

[oober]

Averaged over an 12 month period (winter + summer), 6 ppl, 2 geysers, and quite a few (work related) power hungry systems.

My usage is my usage, is my usage... And I sorely pay for that, every single month too. That's hardly the question here.

View attachment 186849

Convert all your geysers to solar first. After that get GAS for all your cooking. Then install CFL lights and LED. Doing that will save you quite a bit of money on the PV.

Then I would not recommend getting a GRID tie system. This means you have no electricity when the power goes off. Rather get a hybrid inverter that can use the grid and isolates itself from the grid when the power goes off. This way you can at least continue your life as normal.

 

[Bern]

Maybe add a small battery bank designed for the short term high load? This may reduce your overall panel requirements.

Just design it so you can add more batteries and more solar panels as needed or budget dictates.

On the inverters - if you struggle to find one big one maybe split the solar banks up?

 

[savage]

-sigh- Reasons not to ask questions on myBB 101... Is the calculations correct or not? Am I looking at 12KW?

It's all or nothing, I'm not prepared to re-wire my entire house and have bits from there and bits from that... So either the entire house is connected to the inverter, or nothing at all.

 

[oober]

Maybe add a small battery bank designed for the short term high load? This may reduce your overall panel requirements.

Just design it so you can add more batteries and more solar panels as needed or budget dictates.

On the inverters - if you struggle to find one big one maybe split the solar banks up?

Just a note of warning, if you have a small battery bank you cannot pull high loads from it for very long. One needs to pay attention to the C ratings of the batteries etc. If you do not and you pull 3/4KW/70-80Amps from the batteries for lets say 5min plus and you have a small bank, they will get damaged after a while.

My setup prefers solar over the grid but will start pulling from the grid after a few seconds if the solar is not providing enough power since I have a small battery bank.

Edit. Ohh another thing is you cannot add batteries later on to your current bank. Once batteries are over about 9 months old adding new batteries will deteriorate them much faster.

 

[savage]

Just design it so you can add more batteries and more solar panels as needed or budget dictates.

On the inverters - if you struggle to find one big one maybe split the solar banks up?

That's the idea yes. But for now I'm just trying to get an overall feel of what the system should look like / expand too. Hence, if 12KW is the number, then 12KW is the number - so be it.

The call on going down the route or not, will be made much further down the line once I know what it should scale to in terms of design, and what it will look like in terms of cost, obviously.

The first port of call however would be to get the numbers correct - hence the initial question, is 12KW the number, or not

 

[oober]

-sigh- Reasons not to ask questions on myBB 101... Is the calculations correct or not? Am I looking at 12KW?

It's all or nothing, I'm not prepared to re-wire my entire house and have bits from there and bits from that... So either the entire house is connected to the inverter, or nothing at all.

You can have everything wired, but you will still need to reduce the load or pay mega $$$ to supply your peak loads. Geysers do not play well with PV solar. Much cheaper to go solar.

 

[Gaz{M}]

True.. Are two smaller inverters (e.g. 2 x 3kw) cheaper than 1 x 6kw?

Maybe put your lights, TV and security on 1 inverter with a battery bank, then run the other stuff off the grid-tied inverter with no batteries. If you are at home running a business during the day you will definitely see huge savings, because you are using electricity when the sun is shining. The problem most houses will face is that they are not at home during the day to use all that solar electricity, so they need huge batteries, wiping out the cost savings.

 

[P924]

-sigh- Reasons not to ask questions on myBB 101... Is the calculations correct or not? Am I looking at 12KW?

It's all or nothing, I'm not prepared to re-wire my entire house and have bits from there and bits from that... So either the entire house is connected to the inverter, or nothing at all.

No, to generate 36000kWh/year in cape town, you need 21,1kWp of panels and inverter under ideal circumstances...

 

[savage]

No, to generate 36000kWh/year in cape town, you need 21,1kWp of panels and inverter under ideal circumstances...

Thanks - and how did you get to 360,000kWh/year from 3,000kWh/month? I presume you mean 36,000kWh/year?

Care to elaborate how you get to 21.1kWp too?

 

[P924]

Thanks - and how did you get to 360,000kWh/year from 3,000kWh/month? I presume you mean 36,000kWh/year?

Care to elaborate how you get to 21.1kWp too?

You just read it wrong. Average insolation for cape town is 6.42kWh/day/square meter. That roughly translates to 6.42kWh/day/kWp of panel if the panels are ideal.

Under real world conditions, you are looking at generating 1988kWh/kWp in capetown per year. That changes based on how you mount them and pitch and azimuth. So in the real world, you would need much more than 21kWp, and a huge roof/area for the panels.

Check out pv-calculator.ch

So it is advisable to reduce your load first.

 

[phoneJunky]

-sigh- Reasons not to ask questions on myBB 101... Is the calculations correct or not? Am I looking at 12KW?

It's all or nothing, I'm not prepared to re-wire my entire house and have bits from there and bits from that... So either the entire house is connected to the inverter, or nothing at all.

Yes it is correct.

12*8.2 = 98.4

You actually need more than 100 if you design for 3000kW per month and still make space for losses, so 15kW is probably better.

But your usage is actually around 1900kW from what I can see from your graph. So the 12kW will be sufficient.

Edit:

Better to go solar geysers first, gas ovens, then LED lighting - in that order.

And if you went of the grid, you would have to double that amount. You might also want to measure what your peak usage is per day over a few months

 

[oober]

You just read it wrong. Average insolation for cape town is 6.42kWh/day/square meter. That roughly translates to 6.42kWh/day/kWp of panel if the panels are ideal.

Under real world conditions, you are looking at generating 1988kWh/kWp in capetown per year. That changes based on how you mount them and pitch and azimuth. So in the real world, you would need much more than 21kWp, and a huge roof/area for the panels.

Check out pv-calculator.ch

So it is advisable to reduce your load first.

Also the whole exercise is futile since getting so much solar panel won't help you jack if your loads are not during those 5/6 hours and you only have grid tie...there is no place for all this energy to go and that 12KW+ array will be a BIG waste of money in a grid tie only setup.

Grid tie inverters also do not charge batteries, hence the hybrid grid tie inverter suggestion.

 

[savage]

So it is advisable to reduce your load first.

Not going to happen. Large household, business from home (which also actually counts in my favour), large amounts of systems... I'm not interested in changing my usage, or lowering it. I pay for what I use, and will continue to do so too. It's really as simple as that.

This is about lowering the amount of energy that is used from the grid, by generating (some) of the energy myself. It will start small, and it will gradually scale upwards.

Yes it is correct.

12*8.2 = 98.4

You actually need more than 100 if you design for 3000kW per month and still make space for losses, so 15kW is probably better.

But your usage is actually around 1900kW from what I can see from your graph. So the 12kW will be sufficient.

Yes, I've increased my average monthly usage to already make room for more power usage (not that I'd think there would be anything significant), and therefore the inverter is actually already over spec'ed as you've stated. Obviously getting the RIGHT inverter and getting it grid-tied would be a different story all together, but that is something that I will look at when I'm ready to cross that bridge.

Better to go solar geysers first, gas ovens, then LED lighting - in that order.

Not interested in Solar Geysers, all my geysers are already on timers and on for +- 2 hours per day. Gas Stove may be something to look at, although there's not much cooking going on, and already have some LED lights too... Hardly makes a dent.

Please, no offence, but I didn't ask for advice on how to save energy. I run a business from home, I have MONSTER systems here (actually, 2 cabinets full of servers), and monster networking gear. Just the network switch sits with 2 x 3000W power supplies. This isn't an average household that we are talking about, and I have never said that it was too. There are reasons why my usage is high, it will stay high, regardless of what I do.

 

[P924]

Also the whole exercise is futile since getting so much solar panel won't help you jack if your loads are not during those 5/6 hours and you only have grid tie...there is no place for all this energy to go and that 12KW+ array will be a BIG waste of money in a grid tie only setup.

Grid tie inverters also do not charge batteries, hence the hybrid grid tie inverter suggestion.

Agreed. I was working under the grid tie assumption just to illustrate the scope of trying to generate so much...

If his energy needs are so high, 12kW+ of grid tie won't be a waste, the larger the setup, the lower the overheads in "admin" charges?

Sure he would still be off when load shedding occurs, but even to go hybrid would require a huge battery bank to carry his load average and last a reasonable time.

 

[oober]

Agreed. I was working under the grid tie assumption just to illustrate the scope of trying to generate so much...

If his energy needs are so high, 12kW+ of grid tie won't be a waste, the larger the setup, the lower the overheads in "admin" charges?

Sure he would still be off when load shedding occurs, but even to go hybrid would require a huge battery bank to carry his load average and last a reasonable time.

If his house pulls 12KW+ loads constantly then it won't be a waste yes but will it have such constant loads?

 

[savage]

Also the whole exercise is futile since getting so much solar panel won't help you jack if your loads are not during those 5/6 hours and you only have grid tie...there is no place for all this energy to go and that 12KW+ array will be a BIG waste of money in a grid tie only setup.

Grid tie inverters also do not charge batteries, hence the hybrid grid tie inverter suggestion.

But my load *IS* during those times! That's the whole freaking point! This isn't a household that we're talking about, this is a energy hungry business from home! My load are 24x7 actually.... When IDLE (i.e. all appliances off, except business related services), I sit here currently pulling 5A @ 220VAC, that's 1.1kW of energy...

There's two things that I want to address here,
1) I want to supplement feed by offloading energy from the Grid to Solar. Yes, even if that's just during those few hours during the day
2) I want to be able to run for those 2 or 3 hours during load shedding (we are shed from 4PM till 6:30PM), and I know there may not be enough sun left at those times to do what I want - especially not during winter.
3) I want to gradually scale this up (i.e. more panels / batteries) over time as budget allows.

I don't think it's unrealistic, and I don't think it is unreasonable too. Frankly, I actually believe very much that it is do-able...