How do I size a solar power system?

[thaisail]

How many 1 kWh batteries are required for a house that has a single phase 15 (45) meter with consumption of around 500 units per month.

 

Can I assume that each unit is 1 kWh? I need to take into account the fact that there will only be 8 hours of electrical production per day from the solar panels.

 

I am just trying to find out what size solar plant I need to build to able to go off the grid.

 

I need to know how many solar panels to buy and how many batteries if I need to make it through the night.

[Crossy]

On 12/23/2020 at 6:14 PM, thaisail said:

I am just trying to find out what size solar plant I need to build to able to go off the grid.

 

I've pulled this out of the original thread to get better responses

 

[Crossy]

A quick and dirty response:-

 

500 units a month comes down to about 17 units a day, let's say 20 for round numbers. One "unit" is 1 kWh (kilowatt-hour).

 

Our panels make useful power for about 8 hours per day, but that's not all at full output of course.

 

Real life numbers (from our own installation) have the standard 300W-350W panels generating between 1 and 1.3 units each per day.

 

So for a simple calculation that's about 20 panels.

 

You'll need to store that energy for use at night. Let's assume you only use power at night so you need to store 20,000 Watt hours (Wh) of energy, that translates as about 420Ah of battery at 48V.

A LiFePO4 battery of that magnitude is going to cost you around 4.5 grand US$ (although prices are falling) and you'll need charge controllers and inverters too as well as your panels.

 

Still interested?

 

The above is the absolute minimum and takes no account of miserable dull days. For example we have 5.8kW of panels which can produce as much as 32kWh or as little as 6kWh in a day and of course, like buses and bananas, miserable days come in bunches.

 

If you actually have and retain a grid connection you can use that with a hybrid inverter to bridge over the dull days, otherwise you will need a lot more panels and batteries (say double) or a genset (use the genset to charge the batteries too so you don't have to run it all night).

 

It really depends what you want to do, go truly off-grid, mostly off-grid (but with a backup mains supply), or stay on-grid and offset your power consumption. The latter is what we do, the solar provides for 50-60% of our energy usage the rest coming from PEA as and when it's needed.

 

Do you intend going DIY (have a look at the Solar Car Port thread) or paying a solar contractor to do the job?

 

[BritManToo]

On 12/23/2020 at 6:14 PM, thaisail said:

How many 1 kWh batteries are required for a house that has a single phase 15 (45) meter with consumption of around 500 units per month.

 one 330/340w panel will produce 1kwh/day. So  let's say 20  panels @3,000bht/panel to cover 500 units  = 60,000bt 

assume 1/2 your consumption is at night so 8kwh or 800ah of 12v batteries (but wait if you discharge more than 50%  it kills the battery) so 1600ah of (globatt inva) battery @5,000bht per 100ah  = 80,000bht replace every 3 years.

 

You still  need hybrid inverters @6000bht/3kva, minimum 2 = 12,000bht

so you're looking at 150,000bht.

Edited December 26, 2020 by BritManToo

[abrahamzvi]

20 hours ago, Crossy said:

A quick and dirty response:-

 

500 units a month comes down to about 17 units a day, let's say 20 for round numbers. One "unit" is 1 kWh (kilowatt-hour).

 

Our panels make useful power for about 8 hours per day, but that's not all at full output of course.

 

Real life numbers (from our own installation) have the standard 300W-350W panels generating between 1 and 1.3 units each per day.

 

So for a simple calculation that's about 20 panels.

 

You'll need to store that energy for use at night. Let's assume you only use power at night so you need to store 20,000 Watt hours (Wh) of energy, that translates as about 420Ah of battery at 48V.

A LiFePO4 battery of that magnitude is going to cost you around 4.5 grand US$ (although prices are falling) and you'll need charge controllers and inverters too as well as your panels.

 

Still interested?

 

The above is the absolute minimum and takes no account of miserable dull days. For example we have 5.8kW of panels which can produce as much as 32kWh or as little as 6kWh in a day and of course, like buses and bananas, miserable days come in bunches.

 

If you actually have and retain a grid connection you can use that with a hybrid inverter to bridge over the dull days, otherwise you will need a lot more panels and batteries (say double) or a genset (use the genset to charge the batteries too so you don't have to run it all night).

 

It really depends what you want to do, go truly off-grid, mostly off-grid (but with a backup mains supply), or stay on-grid and offset your power consumption. The latter is what we do, the solar provides for 50-60% of our energy usage the rest coming from PEA as and when it's needed.

 

Do you intend going DIY (have a look at the Solar Car Port thread) or paying a solar contractor to do the job?

 

What is the approximate costs to have  a system that would safely provide cover for a house of app.120 Sq. M. , including all equipment necessary?

[Crossy]

19 minutes ago, abrahamzvi said:

What is the approximate costs to have  a system that would safely provide cover for a house of app.120 Sq. M. , including all equipment necessary?

 

That's really like asking "how long is a piece of string?". With no indication of actual usage (like our OP gave) it's nigh on impossible. Start off by letting us know your monthly power usage (from your bill).

 

Do you want totally off-grid, off-grid with mains backup (both will need batteries) or on-grid to supplement your power usage (no batteries so cheaper).

 

Do you intend going DIY or paying a contractor?

 

For an idea, Global house will do a 5kW on-grid system for just under 200,000 Baht.

https://www.globalhouse.co.th/product/detail/9229052020104

 

[abrahamzvi]

3 minutes ago, Crossy said:

 

That's really like asking "how long is a piece of string?". With no indication of actual usage (like our OP gave) it's nigh on impossible. Start off by letting us know your monthly power usage (from your bill).

 

Do you want totally off-grid, off-grid with mains backup (both will need batteries) or on-grid to supplement your power usage (no batteries so cheaper).

 

Do you intend going DIY or paying a contractor?

Many thanks, Crossy. The answers are: 1) usage - app. 150 units per month

                                                        2) Totally off-grid

                                                        3) I would use a reliable contractor

 

All I am looking for are approximate costs.

 

Thanks again.

[thaisail]

21 hours ago, Crossy said:

A quick and dirty response:-

 

500 units a month comes down to about 17 units a day, let's say 20 for round numbers. One "unit" is 1 kWh (kilowatt-hour).

 

Our panels make useful power for about 8 hours per day, but that's not all at full output of course.

 

Real life numbers (from our own installation) have the standard 300W-350W panels generating between 1 and 1.3 units each per day.

 

So for a simple calculation that's about 20 panels.

 

You'll need to store that energy for use at night. Let's assume you only use power at night so you need to store 20,000 Watt hours (Wh) of energy, that translates as about 420Ah of battery at 48V.

A LiFePO4 battery of that magnitude is going to cost you around 4.5 grand US$ (although prices are falling) and you'll need charge controllers and inverters too as well as your panels.

 

Still interested?

 

The above is the absolute minimum and takes no account of miserable dull days. For example we have 5.8kW of panels which can produce as much as 32kWh or as little as 6kWh in a day and of course, like buses and bananas, miserable days come in bunches.

 

If you actually have and retain a grid connection you can use that with a hybrid inverter to bridge over the dull days, otherwise you will need a lot more panels and batteries (say double) or a genset (use the genset to charge the batteries too so you don't have to run it all night).

 

It really depends what you want to do, go truly off-grid, mostly off-grid (but with a backup mains supply), or stay on-grid and offset your power consumption. The latter is what we do, the solar provides for 50-60% of our energy usage the rest coming from PEA as and when it's needed.

 

Do you intend going DIY (have a look at the Solar Car Port thread) or paying a solar contractor to do the job?

 

Thanks for your calculations. I have a steel roof with the perfect sun angle that is 16 meter long and 5 meters wide. The roof is already strong enough to handle the weight of the panels. How many panels could I realisictally fit on that roof? 

[Crossy]

1 minute ago, abrahamzvi said:

Many thanks, Crossy. The answers are: 1) usage - app. 150 units per month

                                                        2) Totally off-grid

                                                        3) I would use a reliable contractor

 

All I am looking for are approximate costs.

 

OK, 150 units per month equates to about 5 units a day.

 

For our experience 5 panels would make that on average, but since you want to be totally off-grid I'd go with double that to get you over dull days (you should still have some sort of backup system too - a baby genset). Panels say 4,000 Baht each.

 

Batteries to hold 2 days production - 10 units = 200Ah @ 48V, LiFoPO would cost about 60,000 Baht.

 

3.5kW of solar would need a 80-100A MPPT charge controller (or several smaller ones) - 12,000 Baht

 

Inverter, say 5,000W to cover surge loads 20-30,000.

 

Say 150,000 for the important hardware.

Add the same again for installation hardware and the installation itself.

 

Very (very, very) rough ball-park say 300,000 Baht budgetary estimate.

 

You may want to add more batteries, much depends just how much energy you use during the day (so it doesn't need storing).

 

It would be interesting to hear what your contractors are suggesting, both as a system and as $$$.

 

[BritManToo]

33 minutes ago, abrahamzvi said:

Many thanks, Crossy. The answers are: 1) usage - app. 150 units per month

                                                        2) Totally off-grid

                                                        3) I would use a reliable contractor

 

All I am looking for are approximate costs.

 

Thanks again.

 

150units/month = 5/day ........... say 6x 340w panels 3kbht each  = 18kbht.

6x  150ah 12v globatt inva @8k bht = 48kbht (assume  these need replacing every 3  years)

3x 3kva hybrid inverters@6kbht = 18kbht

 

If you were in Chiang Mai, i'd install that  for 50kbht including  wires/welding/steel (I don't climb  on roofs)  but Thais seem  to want a  much higher profit margin, so assume  100kbht plus.

 

So around 90kbht DIY, 140Kbht ME, 190kbht THAI. (same guarantee = none)

cheaper than Crossy, but he uses expensive components that seem to fail more often than my cheap stuff.

 

Edited December 27, 2020 by BritManToo

[Crossy]

11 minutes ago, thaisail said:

Thanks for your calculations. I have a steel roof with the perfect sun angle that is 16 meter long and 5 meters wide. The roof is already strong enough to handle the weight of the panels. How many panels could I realisictally fit on that roof? 

 

Standard panels are about 2m x 1m so you should be able to get two rows of 15 (30 panels) on there without problems. 

 

Max cramming factor 5 rows of 8 panels (panels in landscape orientation) would squeeze on 40 panels, but they would be much more difficult to maintain / replace should one go bad.

 

Panels weigh about 25kg a piece so you're looking at nearly a ton of static loading on your roof.

[Crossy]

10 minutes ago, BritManToo said:

cheaper than Crossy, but he uses expensive components that seem to fail more often than my cheap stuff.

 

Always best to over estimate and get a nice surprise, than to under estimate and get a less pleasant surprise.

 

A concept which seems to be constantly lost on my contractors when estimating timescales. Job will take 2.5 weeks, they tell me 2 weeks and are 3 days late. If they told me 3 weeks and came in early they would make me much happier.

[BritManToo]

7 minutes ago, Crossy said:

 

Always best to over estimate and get a nice surprise, than to under estimate and get a less pleasant surprise.

 

A concept which seems to be constantly lost on my contractors when estimating timescales. Job will take 2.5 weeks, they tell me 2 weeks and are 3 days late. If they told me 3 weeks and came in early they would make me much happier.

 

6-7 days for me to install  6 panels, I take my time welding, tire easily and go home at lunch time.

Backup generator is always a good idea for the monsoon, add 10kbht.

Edited December 27, 2020 by BritManToo

[Muhendis]

I have 7.2 kw of solar panels, 750 Ah batteries and an 8 Kw inverter which is ample for my needs. I won't go into detail about my consumption but first you need to establish how much power you are going to need at any given time. That will tell you what size inverter you need. For example. I you may be having a shower (6kW) and your wife wants to boil the kettle (2kW) at the same time, then your inverter will need to be able to deliver 8kW but there may be a fridge running and the TV may be on and a few lights and so on.......Your panels and batteries can probably be quite small but the inverter needs to be able to drive big loads for short times. You need to work out the worst case scenario and plan around that. 

Edited December 27, 2020 by Muhendis
Amp hour decimal point wrong place

[davemos]

Dont forget the cost of replacing panels, batteries over a long term period. Like all batteries they have a use by date  .

[Crossy]

14 minutes ago, davemos said:

Dont forget the cost of replacing panels, batteries over a long term period. Like all batteries they have a use by date  .

 

Indeed. 

 

But panels should be good for 25+ years barring accidental damage. 

If you treat them nicely you should get 3000+ cycles out of a set of LiFePO batteries.

 

[Muhendis]

36 minutes ago, Crossy said:

 

Indeed. 

 

But panels should be good for 25+ years barring accidental damage. 

If you treat them nicely you should get 3000+ cycles out of a set of LiFePO batteries.

 

Well said.

The life expectancy of the panels is generally defined as 20 years at 25 C and they will have degraded to 80% of output when new, so they are still giving a useful output. I have heard that some new innovations can improve on that but I don't know if China has copied them yet. 

Additional to Crossy's 3,000+ cycles from LiFePO4 batteries. They are more expensive than the various Lead acid chemistries but they have a better performance so you may not need so many.

[Crossy]

36 minutes ago, Muhendis said:

Additional to Crossy's 3,000+ cycles from LiFePO4 batteries. They are more expensive than the various Lead acid chemistries but they have a better performance so you may not need so many.

 

What he said ^^^

 

Anyone building a "serious" system either off-grid or hybrid would be well advised to go for (or at least carefully consider) LiFePO4, the prices for equivalent capacity are approaching parity and the "fit and forget" maintenance of the Lithium technologies is very attractive.

 

[BritManToo]

18 minutes ago, Crossy said:

 

What he said ^^^

 

Anyone building a "serious" system either off-grid or hybrid would be well advised to go for (or at least carefully consider) LiFePO4, the prices for equivalent capacity are approaching parity and the "fit and forget" maintenance of the Lithium technologies is very attractive.

 

Amazon want $600 for a 12v 100ah LiFePo.

https://www.amazon.com/LiFePO4-Phosphate-Rechargeable-Overland-Applications/dp/B08M8YWC6N/ref=psdc_15719921_t1_B07X4Z8FML

 

I  can get a wet deep cycle 12v 100ah for 4,200bht ($140).

https://shopee.co.th/แบตเตอรี่-Globatt-Inva-100Ah-12V-Deep-Cycle-i.229263273.6217937411

 

That doesn't seem like parity to me.

Edited December 27, 2020 by BritManToo

[Crossy]

14 minutes ago, BritManToo said:

Amazon want $600 for a 12v 100ah LiFePo.

I  can get a wet deep cycle 12v 100ah for 4,000bht.

That doesn't seem like parity to me.

 

How much is the lead-acid battery on Amazon? (don't forget you will need about 200Ah for equivalence).

 

I (and you) can get a 100Ah x 12V LiFePO4 (equivalent to 200Ah Lead acid @ 50% discharge) for for 7,200 Baht on Lazada/Shopee. Not quite there, but when you consider the life of the LiFePO vs the Pb not a million miles off.

 

And no mucking about with distilled water keeping the beast happy.

 

[Muhendis]

1 hour ago, Crossy said:

I (and you) can get a 100Ah x 12V LiFePO4 (equivalent to 200Ah Lead acid @ 50% discharge) for for 7,200 Baht on Lazada/Shopee. Not quite there, but when you consider the life of the LiFePO vs the Pb not a million miles off.

Two years ago when I got my Lead acid Gel's (250 Ah) from Yangtze Solar they were just over 7,000 Baht each and Lithium were nearly double plus I would have needed a bit more electrical stuff. At 30% discharge Gels are good for 2,400 cycles which was good for the price but now I would probably go for lithium. Price against performance, lithium are good value in the long term.

[xtrnuno41]

IF you have some more money left, the n maybe this battery:

 

[BritManToo]

2 hours ago, Crossy said:

How much is the lead-acid battery on Amazon? (don't forget you will need about 200Ah for equivalence).

 

I (and you) can get a 100Ah x 12V LiFePO4 (equivalent to 200Ah Lead acid @ 50% discharge) for for 7,200 Baht on Lazada/Shopee. Not quite there, but when you consider the life of the LiFePO vs the Pb not a million miles off.

Not sure you're right about that, one person who bought the $600 battery claimed it gave him  70ah (70%).

 

As for the 7,200bht LiFePo ........ the reviews ain't that great (operated a 335w washing machine for 5 mins then dead when it  should have powered it for 2+ hours).

 

Prices are falling so fast, I'm not sure I want to buy an expensive untried product first.

In 3 years time, when my wet batteries are dead, I'll probably be able to buy the new technology for 1/2 the current price while the current LiFePo will be outdated.

 

You buy one first, and tell me how it goes!

 

Edited December 27, 2020 by BritManToo

[Crossy]

6 hours ago, BritManToo said:

In 3 years time, when my wet batteries are dead, I'll probably be able to buy the new technology for 1/2 the current price while the current LiFePo will be outdated.

 

I agree, I certainly wouldn't swap out perfectly good batteries until they actually die.

 

Prices are faling and technology improving all the time, and it's definitely a case of caveat emptor when buying this kind of thing from inline sellers (if it looks too good to be true, it probably is).

 

Rest assured that when we actually have to go hybrid and get batteries the whole story will go on here, warts and all.

[Caspersfriend]

10 hours ago, BritManToo said:

Not sure you're right about that, one person who bought the $600 battery claimed it gave him  70ah (70%).

 

 

From personal experience, many of the LiFePO batteries for sale on Lazada/Shopee are pre-used.

 

They don't [of course] admit that in the listing, but when I have asked them on CHAT they usually admit it.

 

In fact: most of the lower-priced Lithium-based single cell batteries are either pre-used or of inferior production/quality; with inflated capacity claims 

[BritManToo]

Some deeper thinking on battery power (while out walking this morning).

 

We've not considered the recommended discharge rates on batteries.

For a 12v wet deep cycle battery, recommended discharge is over a 20hr period (no idea about LiFePo).

So my 100ah battery should only provide 5a for 20hrs (10hrs for 50%) to keep within it's specifications (= 60w)

Storage capacity of the battery drops significantly at higher discharge rates.

 

My entertainment system runs at 350w (TV + computer/console + sound bar + modem +NAS + HDs)

So I'm driving my battery at 6x it's recommended discharge rate reducing it's storage to around 50ah, then adding the 50% consideration, so only 25ah or 'real world' storage, around 1hr of backup power.

.

To keep within spec. I need 4x 150ah GloBatt INVA batteries costing 30,000bht.

150ah/20hr =7.5a, 7.5a x 12 = 90w, 90w x 4 = 360w

(and only discharge to 50% if I want them to last 2,000 charges, so 10hrs 'real world' storage)

 

Please let me know if you think I've made any serious errors in my thinking.

Edited December 28, 2020 by BritManToo

[maxpower]

The EVE-280Ah type Lifepo4 cells are becoming popular in the DIY storage world but not much info around for long term tests. One of our engineers has a 48V 280Ah pack on his hybrid solar inverter and a 12v pack inside his UPS. Good results after almost ten months in service.
  
A 280Ah 12V pack with 100A BMS can be built for around $400 plus component shipping.

 

Results from builders average around 270Ah.

 

 

 

China produces some excellent cells at all grades but sadly it also churns out a lot of junk and false claims.

Always do lots of research before purchasing lithium cells for DIY projects.

[Crossy]

44 minutes ago, BritManToo said:

Please let me know if you think I've made any serious errors in my thinking.

 

I don't think you're far wide of the mark.

 

One thing about lead-acid under high discharge rates is that they "recover" much of the lost capacity if they get a rest, so if your heavy load comes in shortish bursts with a period of lighter loading in between you wouldn't actually lose as much as you fear.

 

LiFePO4 are much more tolerant of big loads, theoretically losing little to no capacity. I'll do some real world testing on one of my (confirmed good) 6Ah cells once it's charged.

 

[Crossy]

12 minutes ago, maxpower said:

One of our engineers has a 48V 280Ah pack on his hybrid solar inverter and a 12v pack inside his UPS.

 

I've seen Eve cells mentioned on some of the solar power sites, generally good reviews and the price is good.

 

Does your man have a link to a "known good" seller of these chaps?

 

[maxpower]

1 hour ago, Crossy said:

 

I've seen Eve cells mentioned on some of the solar power sites, generally good reviews and the price is good.

 

Does your man have a link to a "known good" seller of these chaps?

 

 

Our engineers buy their DIY stuff from the local electronics market which really is like a fruit and veg market where physical appearance and source changes daily. 

 

There's a lot of surplus production in China purchased on the spot market. Not all the cheap stuff is junk.

 

Most of the DIY guys outside of China are buying and negotiating at Alibaba. Unlike places like Lazada Alibaba has lots of bulk shifters not in the game to knowingly cheat.

 

I guess poke around the tube and forums might help inspire confidence in a particular seller.