Small Off-grid Solar System: a few questions

[whazzatt]

Hi

I would like to power the following with a solar system:

A laptop @ 75W for 6 hours
Surround sound @ 70W for 4 hours
Lights @ 30W for 6 hours
Chargers @ 20W
Bass amp @ 400W 1 hour
(total 1400 W - ish!)

Using this site, I've calculated that, with an average of 5.5 hours sunlight a day, two 105Ah batteries should work for this; that a 650W inverter should do the trick; that 300 W panels should be enough to charge the batteries; that a 10 or 20 A controller/regulator is needed.

Questions:

1) Is the calculation correct?
2) 10 or 20 A controller?
3) What will I need to do in order to wire only the lights DC?
4) What do I need to consider if I would like to add a wind turbine into the mix sometime in the future?
5) (This might be asking a bit much, but...) If this were the site you were using to purchase the parts, what would you order in order to achieve a balance between efficiency, ease, and cost?

Respect,

Whazzatt (?)

[Sunking]

The answer to your questions is no, and you need to take another look using real data. One major flaw you have made is using your area average insolation figure. That can only be used on a grid tied system, not a battery system. If you did that you will go dark in winter waiting for longer days. For a battery system you have to use worse case, and in your situation is the months of December and January the jinsolation is at its lowest.

For now please read this thread and work the math. Then come back with questions. You have taken the first step by calculating your daily watt hour usage. Now you need to find your December Insolation to calculate panel wattage required and work some more on battery capacity. You need a minimum 5 day reserve plus a generator for those long cloudy spells.

[Mike90250]

... You need a minimum 5 day reserve plus a generator for those long cloudy spells.

Well, I think a home based system can get away with 3 day reserve, since you always have the option to start the genset whenever you want.

There is a huge price difference with a 3 day or 5 day battery bank. And the charger for it.

Shallow discharges allow a battery bank to last longer. Any battery bank is going to die of old age in 5-10 years, so when it's time to replace a 3 day battery bank, you will know if you need to step up to 5 days, or manage with 3.

[whazzatt]

Hi - thanks for your replies, much appreciated.

I haven't had time to redo the calculations (will get to it soon), but I have found the insolation information for the place we'll be setting up - Port Elizabeth, South Africa. Here it is - link. You'll have to scroll down a bit.

The lowest is 2.50, considerably lower than I was working with. This will mean a more expensive setup to achieve the desired power capability. I am considering still going with something close to the proposed setup and just using it for super-efficient lights and for charging a phone, camera battery, electric toothbrush, and when possible the laptop battery for occasional/rare use because...

Look at the consistency of the wind! Port Elizabeth has 2 appellations - the friendly city, and the windy city; the latter is unquestioningly true.

So I'll keep the solar system small and use it very sparingly, and then add more batteries and a few VAWTs when I manage to find the appropriate motors and make the turbines. This is something I planned to do anyway, but this forum process is helping me see that my initial power usage will have to be minuscule. That's okay though, because I'd happily go without any power rather than participate in the current system.

Any links for or recommendations on forums or sites for small hybrid systems? I've got the following so far, which, despite the title, does also integrate a solar panel: How I home-built an electricity producing Wind turbine

[Mike90250]

for wind, I always suggest Hugh Piggott's site http://scoraigwind.com/

read it, and read http://www.rc-trucks.org/home-wind-turbine.htm too.

[russ]

For wind other links are -

http://www.wind-works.org/ this is Paul Gipes site - he is one of the real experts

http://www.greenpowertalk.org/search.php?searchid=20 Rob Beckers site - a real good wind forum - honest site

A factory built small turbine is difficult to keep running - a home built turbine is almost impossible

A wind turbine has a lot in common with a boat anchor - heavy is good and light weight is useless.

Russ

[whazzatt]

Okay, so I've looked at the original proposed power needs again and tried to do the calculation using Sunking's post on the topic.

I'm working on 2.5 hours insolation. I've dropped the bass amp from the list, leaving an absolute max total power need of 1000W. With the 'fudge factor', that comes to 1600W using a MPPT controller.

The array therefore needs to be 640W. The batteries at 12V on a 3 day reserve (I'm happy with that) needs to be 250 Ah.

Does this look more like it?

I'm a bit confused about the charge controller. I can get 3*210W panels, making a near enough 630W array, and according to Sunking's numbers, this means 630/12 = 52.5 A controller. Is this correct? I'm sure I read elsewhere that there is a slightly different equation to take into consideration for several panels, lowering the size of the charge controller?

Thanks again.

Whazzatt(?)

[Sunking]

I'm working on 2.5 hours insolation. I've dropped the bass amp from the list, leaving an absolute max total power need of 1000W. With the 'fudge factor', that comes to 1600W using a MPPT controller.

Do you mean watt hours. Watts and watt hours are not the same thing.

Watts = Power which is measured at a specific moment in time. Example: a 100 watt light bulb uses 100 watts
Watt Hours = Energy = Watts x Hours. Example a 100 watt bulb uses how much energy in 10 hours? 100 watts x 10 Hours = 1000 watt hours

If you mean watt hours then yes you will need 640 watts. But be aware going only with a 3 day reserve time in batteries with no generator is setting you up to be dark a few months during a calendar year. Just one cloudy day and you have to shut down and wait for the sun to return and have one full day of sun to recharge before you come back on. If you are not going with a generator you really need to up the wattage quite a bit so you can fully recover in a full day during winter.

I'm a bit confused about the charge controller. I can get 3*210W panels, making a near enough 630W array, and according to Sunking's numbers, this means 630/12 = 52.5 A controller. Is this correct? I'm sure I read elsewhere that there is a slightly different equation to take into consideration for several panels, lowering the size of the charge controller?

What are you confused about? Have you read this thread explaining the differences between MPPT and PWM controllers? For MPPT controllers Ohm's Law is used. Amps = Watts / Volts So with 630 watts of panels on a 12 volt battery requires 630/12 = 52.5 amps. You will need a 65 amp Charge controller.

You have also built yourself a trap. Since you decided to only go with 3 day battery reserve, and have poor winter insolation, you will have to use expensive AGM batteries. You will not be able to use less expensive flooded lead acid batteries. The maximum charge rate you can apply to a FLA battery is C/8 where C = the 20 AH rated capacity. So if you use 250 AH FLA battery the maximum charge current is 250/8 = 31.25 amps. You panels will deliver 52 amps and they batteries cannot be charged that fast. They would overheat and boil over destroying them in a few days.

AGM batteries on the other hand can take up to a C/4 charge rate so 250/4 = 62.5 amps maximum charge current so you are OK with 52 amps from the panels. But here is the catch, AGM batteries cost 2 to 3 times more than FLA.

So if I were you I would do some pricing. Me thinks it would be less expensive to use a 5 day reserve using FLA vs 3 day AGM? GOTTCHA!

One last word on batteries. Do not buy several 12 volt batteries to make the AMP Hour Capacity. Buy lower voltage batteries with the required AH capacity so you only have 1 single string. DO NOT use parallel strings if you can avoid it, and you can avoid it.

[billvon]

So I'll keep the solar system small and use it very sparingly, and then add more batteries and a few VAWTs

VAWT's are nice, but in all cases, swept area is what you care about. If a VAWT has a small swept area it will produce little power.

Also, the #1 issue in wind turbine mounting is altitude. If you can put up a tower so that the lower edge of the turbine blade is SIGNIFICANTLY higher than any nearby obstacles wind may work for you. If not you will likely be disappointed with your production.

I'm a bit confused about the charge controller. I can get 3*210W panels, making a near enough 630W array, and according to Sunking's numbers, this means 630/12 = 52.5 A controller. Is this correct? I'm sure I read elsewhere that there is a slightly different equation to take into consideration for several panels, lowering the size of the charge controller?

At those powers you are getting into the range of higher voltages (to reduce required wiring sizes.) You may also want to consider going to a 24 or 48V battery system; everything (cabling, fusing, losses) gets better/easier/cheaper when you go up in voltage.

For example, at 630 watts you're going to need a 60 amp charge controller at 12 volts. From Blue Sky Energy a 60 amp MPPT controller is around $500, and the wires will be expensive. For a bit over 5% losses, and panels 50 feet from the battery bank, you are looking at 2 gauge wire - so that's another $100 or so. Your inverter cables will probably be 4/0 and that's some _very_ expensive wire.

Now, go to 48 volts and you'll only need a 15 amp charge controller, and your wiring costs will be around 1/4.

[Sunking]

Now, go to 48 volts and you'll only need a 15 amp charge controller, and your wiring costs will be around 1/4.

I agree with everything you summed up.

As a general rule if panel wattage is above 500 watts I look at moving into 24 volt battery. However with that said there are a lot of variables that come into play.

If the cable distances can be kept short between panels and CC, use high wattage panels most of the better MPPT CC's operate most efficiently with standard solar panels made for batteries up to twice the nominal battery voltages. For example Outback FM 80 optimum efficiency for:

12 volt is 17 volts @ 96%, and 24 volts @ 94.5 %. Go above 24 volt panels on a 12 volt system and converter efficiency starts to drop off sharply

For me I fortunately have a program that does that for me to keep the total wire and converter losses to 7% or less. Typically 2% on the wire and 5% in the CC.

All I am getting at is you have to work the numbers. Fortunately for the DIY type Outback has pretty good wiring tables to make the job easier. FWIW shoot to keep wiring losses to 2% or less. and overall 8% or less from panels to output of CC.

[whazzatt]

You guys are great

I think I may be looking into candle technology soon!

My initial response is to do the following:

So 1000 Wh requirement* 3 days reserve / 48 Volts = 62.5 or closest Ah battery, 15 A controller

[billvon]

BUT: those numbers are based on 12V panels; do I now need to look for different voltage panels to suit a 48 V battery and controller system? Or is a 24V system the only real option for the system-size I hope to be dealing with?

2 12 volt panels in series will give you 24 volts; 4 12 volt panels will give you 48 volts. As long as you work in multiples of 2 (or 4 for 48 volts) then you can use the same panels for any system voltage.

[Sunking]

BUT: those numbers are based on 12V panels; do I now need to look for different voltage panels to suit a 48 V battery and controller system? Or is a 24V system the only real option for the system-size I hope to be dealing with?

Yes you will need higher voltage to go to 48, or more of what you already have. If you already have the 3 panels you only have two choices right now. Use all 3 in parallel with a 12 volt system, or only use 2 of them for 24 volts. 3rd is queer guy out. You would need a 4th to go to 48 of the exact same make and model.

But that is the beauty of using MPPT controllers because your are not stuck with having to use panels made for battery systems. Panels made for battery systems are 36 cell panels that produce 16 to 18 volt @ max power. It also means you can use much higher wattage panels because the ones made for grid tied system have more than 36 cells.

Do not get to existed about the smaller AH battery as it is not any smaller than 12 volt. I will give you an example let's say the requirement is for 400 AH at 12 volts, or 100 AH at 48 volts. Your supplier carries 100 AH 12 volt batteries. Either way you still use the same 4 batteries, only difference is how they are configured. The 12 volt verion is with all 4 batteries in parallel, the 48 volt version is all 4 in series. Where you wil save a few bucks is less material wirng the 4 batteries in series.

OK so with the at this low of a wattage is to find either 2 or 3 panels so that when wired in series the Vmp voltage is above 62 volts, but less than the maximum voltage the controller can handle.

Now here is the problem. The larger 40, 60, and 80 amp MPPT CC can handle up to 150 volts. One brand can even take up to 600 volts. But when you get into the small stuff like 35 and below, the voltage goes way down like 70 to 120 volts. So if you buy 1 that can only handle say 70 volts max, your window is very small from 62 minimum to 70 volt max.

Understand?

[Sunking]

Ok to charge a 12 volt battery it requires a minimum of 16 to 18 volts maximum power from the panel. So to charge a 48 volt battery stack requires a minimum of 4 x 16 volts = 64 volts. With me so far?

If you look at panels made for 12 volt battery systems they consist of 36 cells and you will note the Vmp spec is 16 to 18 volts. See the connection? So if you were to use 12 volt panels for a 48 volt battery system you would need at least 4 panel wired in series. Still with me?

However today with MPPT controllers your hands are not tied to using standard 12 volt panels. MPPT charge controllers can operate at much higher voltages. With higher voltages means less current for a given power level. Thus that in return means you can use smaller less expensive wire between the panels and charge controllers.

Standard 12 volt panels have another limitation. Since they are made with 36 cells means the highest wattage they can be is limited to around 160 watts. So if you need to make 2000 watts is a lot of panels and wiring.

With MPPT controllers you can use panels made for grid tied application which have more cells, thus higher voltage and higher power levels. You can by a 70 cell 300 watt panel that has a Vmp of 35 volts. Put two in series and you have 600 watts with a 70 volt Vmp. Perfect for a 48 volt battery system.

Now here is the magic of a MPPT controller. If you used the two 300 watt panels I just referenced too at the input of the controller you apply 70 volts @ 8.6 amps (600 watts), and at the battery output 48 volts @ 12.5 amps (600 watts). Note what happens the voltage was down converted, and the current was up converted.

Try that same thing with a PWM controller with the same panels and ouch. You would input 70 volts @ 8.6 amps (600 watts), and output to the battery 48 volts @ 8.6 amps ( 413 watts). Note what happened; Input Current = Output Current. You got screwed out of 33% of you power.

So what I am telling you is to match your panel voltages up to the controller. You can use standard 12 volt panels if you wish. If you do that and operate at 48 volt battery you will need 4 panels. With panels made for grid tied you have some more options.

However there is a catch which I said earlier. When you work with small MPPT controllers less than 40 amps, th einput voltage is rather limited on the low side, and you cannot exceed it.

One last comment on MPPT controllers. They allow you to grow in terms of battery voltages and panel wattage. For example if you buy say a 40 amp MPPT controller it has the following power input limitations vs battery voltages:

12 volt @ 500 watts
24 volt @ 1000 watts
48 volt @ 2000 watts.

[whazzatt]

Let's see how far I've understood things so far.

To do this, I'm choosing to use 2*210W panels with a Vmp of 22.7 volts each. Based on what I now know, I'll work backwards to calculate the watt hours per day these panels can give me:

watts = watt hours / sun hours
420 = Wh / 2.5
2.5*420 = 1050
1050 divided by fudge factor of 1.6 = 656.25 watts hours

12 V Battery: 5 days * 656.25 / 12 volt = 273.44 Ah
24 V Battery: 5 days * 656.25 / 24 volt = 163.72 Ah

Controller for 12 V system: 420 / 12 = 35 A (so 40A controller)
Controller for 24 V system: 420 / 24 = 17.5 (so 20A controller)

The 2 * 210 W panels have Vmp of 22.7, therefore 22.7 * 2 (series) = 45.4 Vmp.

For 12 V system:
Input at CC = 420 / 45.4 = 9.25 Amps;
Output at CC = 420 / 12 = 35 Amps

For 24 V system:
Input at CC: 9.25 Amps
Output at CC: 420 / 24 = 17.5 Amps

Panel voltages and controller matched up either way.

Am I right in thinking that I can indeed use a 12 or 24 volt battery setup?

Hope I haven't missed something obvious.

Thanks in advance,
whazzatt.