Small Off-Grid Solar Setup Questions

Cariboocoot wrote: »

Oh boy; equipment bundles. "Let us sell you what we make the most profit on" is the real description there. Four 100 Watt panels would be just over 20 Amps so you're good with a 30 Amp controller then. But four 100 Watt panels are going to cost you more per Watt than buying larger panels. In fact at 400 Watts you're at the 'tipping point' between using standard panels & PWM controller and switching up to GT style panels & MPPT controller: the $1 per Watt or so difference in panel price comes to around $400 which when added to the cost of the PWM controller makes MPPT feasible. This has become even more so with the introduction of MidNite's Kid 30 Amp MPPT at <$300. Expanding a system is rarely easy or cheap.

The problem with putting 5 Amps to 420 Amp hours of battery is that it will be fairly ineffectual no matter how long you leave it on. It's about maintenance level; it will keep the batteries charged but it won't actually bring them up from discharged state. Unfortunately battery charging isn't as simple as replaced used Amp hours by apply 'X' Amps for 'Y' hours. With four of the panels you'd be applying perhaps 22 Amps (with PWM charger the exact panel specs will matter) which is the minimum 5% rate so over time that could recharge 420 Amp hours.

A couple of things: first, the batteries aren't being charged now which means they are slowly dying as they wait. Second, if you use power from them while they are charging once the panels are installed that rate falls below minimum and becomes less effective. It's a good thing they are used batteries you haven't got much in to.

The problem with using MSW inverters with anything is that the inverters' outputs aren't all alike. Some are just square wave, some are pretty close to sine. You can pretty much bet that low cost = low quality waveform. Then there's the other unknown: the power supply of what you are trying to run. Motors won't like it, that's a guarantee. They will draw more power and run hotter, slower. In some cases so much so as to be ineffectual. Other types of power supplies are a gamble. Maybe it will work, maybe it will burn up. There's no way to know in advance. At the very least expect noise in any audio/visual equipment and be very pleased if there isn't any.

Wire sizing depends on how much current is expected and how much Voltage drop can be tolerated. You could pull 100 Amps from a 420 Amp hour battery bank for a short time. You could pull 5 Amps for a long time. Depending on which is your intent you would use either 1 AWG or 14 AWG, each fused according to its capacity. Where Voltage sag is a concern you use larger size to reduce that (longer length and/or greater Amps = larger Voltage drop requiring bigger wire).

Take one 400 Watt inverter. It will probably draw about 40 Amps maximum at minimum input Voltage. The instructions that came with it should have recommended wire and fuse sizes, and the wire size may be different according to length. It will say something like "Use 8 AWG wire up to five feet. Use 6 AWG wire up to eight feet." If you skip that and go to standard 4 AWG auto battery wiring it should handle 50 Amps up to 15 feet with minimal Voltage drop for 400 Watts on 12 Volts.

Cariboocoot wrote: »

420 Amp hours @ 12 Volts is a respectable amount of power. There's a couple of things you have to understand about "how much inverter it can support". Inverter size is in maximum Watts (actually Volt Amps) which corresponds to the maximum current draw from a battery bank at a given moment. Normally for flooded cell batteries this should not exceed 25% of the Amp hour rating. In this case that's 105 Amps or 1260 Watts DC. The inverter will take some power to run itself and there is some conversion loss in converting to AC power. So the battery bank should be good for a maximum of 1000 Watts AC. That's a pretty standard size inverter for a 12 Volt system. You may even run a bit over that without too many difficulties. But, and this is important, it won't support it for long.

Here's the other thing to understand: Watt hours. It's Amp hours with Volts. That same 420 Amp hours at its maximum safe DOD multiplied by nominal system Voltage amounts to 2520 Watt hours. See the difference? One is the biggest load it can handle at a moment without 'going flat' the other is the quantity of power it can supply over time.

And then they intermix. In theory that 2520 Watt hours could be 5 Watts for 500 hours or 500 Watts for 5 hours. In practice the real capacity diminishes as the load demand at the moment increases. So the maximum 1 kW load would not actually last for 2.5 hours. Complicated, isn't it?

Usually you only have to worry about maximum Watts for something like getting a motor started and the demand doesn't last long enough to make a difference. The average load over time (Watt hour capacity) becomes the more important aspect.



Nope. That is a problem with solar: every install is unique to the particular location. You may be able to copy basic specs and equipment, but most of the unique aspect is in the wiring. Two systems may have the same PV, controller, batteries, et cetera but because one's array if 20 feet further off it will have different wiring.

This is why you need to understand the reasoning behind the choices made, instead of just using a given wire. There are two main factors is choosing wire size: current handling and Voltage drop. The first is a minimum that must be met: if it can't handle the expected current it will fry. The second is to reduce power loss through Voltage sag which may be present depending on Voltage, current, and distance. The largest problem here occurs between panels and controller due to long distances usually involved. The second most likely trouble spot is between batteries and inverter due to the high currents usually involved.

After the wire sizing is worked out the fuse or breaker can be selected according to the current expected. These are the things figured out last, once the basic equipment and locations have been worked out. And if you are planning on future expansion it helps to wire for it at the start, rather than have to change it all out later to accommodate greater current.

Cariboocoot wrote: »

Okay, so the current would be 4*Isc for sizing wire. That's probably going to be around 24 Amps. The individual panel fuses would be sized for the maximum series fuse rating, or if that's not given then Isc * 1.56 or probably 10 Amps.

Cariboocoot wrote: »

Automotive style fuse holders are generally not good enough quality for RE use. They tend to burn up even when not stressed. Look for a more robust unit; I think Blue Sea makes a module capable of multiple fuses. This will have to be weather-proofed because it will likely be outside. The Blue Sea equipment is for marine use and they may have something suitable as-is. (This to avoid going with the more expensive RE equipment such as MidNite combiner boxes.)



There should definitely be a fuse between the charge controller output and the battery. You may even want to use a circuit breaker here. It appears 30 Amps would be right providing no further expansion is planned.

Cariboocoot wrote: »

You don't really need any form of over-current protection between the array and the controller as PV is self-limiting on current. People will put a circuit breaker on that line but it functions only as a disconnect.

BB. wrote: »

ore
I am not quite sure I understand your connections. The traditional wiring method would be to wire two panels in series, and put a fuse (or breaker) in that string. Then tank the 2 series panels + breaker (or fuse), and then put the X strings together in parallel.

You want the fuse/breaker to protect a single shorted string from the rest of the panels (i.e., 2 or more parallel strings feeding a single shorted string).

The "nice" way to do this would be to get a "combiner box" with breakers. The breakers can be turned off (servicing/debugging a failing string, etc.). The wires are all combined together and then fed on a pair of cables to the PV input of the controller. Normally the combined current of the panels (Isc*1.56 if you follow NEC) should exceed the cabling back to the controller--And you do not need a second breaker to protect that wiring (the array can never exceed the wire current rating).

BB. wrote: »

Back to basics--What exactly is the charge controller (PWM or MPPT) and its current rating. And what is the exact rating of the array.

If the charge controller is a 30 amp MPPT charge controller, then the wiring between the controller and the battery should be:

30 amps * 1.25 NEC derating = 37.5 amps ~ 40 amp rated breaker/fuse minimum

The 1.25 derating is used because fuses and breakers if they carry more than 80% (1/1.25) of their rated current, they can trip.

BB. wrote: »

In general, that is the correct answer. When you have 3 or more parallel strings of panels, the total output current of two of the parallel panels in the array exceeds the "series fuse rating" of the third string of panels (if they are shorted).



More or less, the battery cables need to be rated for 1.25x the maximum expected continuous current you want to send through the battery string(s) (feeding loads and/or charging, which ever is worse case).

It usally makes mores sense to do the actual calculations for your design. So, now that you have a handle on your equipment power power requirements... Can you give us a list:

X solar panels (Vmp/Imp rating)
Brand/Model of ABC charge controller
Brand/Model of Battery bank (number of batteries, AH/voltage of batteries/cells, battery bank voltage)
Brand/Model of AC inverter (or inverter/charger)
Brand/Mode of AC battery charger (if used)

We can then step through the process and help with the calculations/section of wiring/breakers and double check hardware (controllers, battery bank, etc.).

We try for a "balanced system" that will meet your needs. It does get a bit iterative as you make a selection of a piece of equipment and see how that changes the parts needed around it.

-Bill

• (4) 100 watt Renogy Monocrystalline Solar Panels RNG-100D (http://renogy.com/wp-content/uploads/2013/05/RNG-100D.pdf)
• Renogy 30 Watt PWM Charge Controller PWM30CC (http://renogy.com/wp-content/uploads/2014/05/PWM-30A-Controller-Manual.jpg)
  -This may change to something like a Tracer-2210RN 20 Amp MPPT Charge Controller (http://www.epsolarpv.com/en/index.php/Product/pro_content/id/157/am_id/136) which would allow me to wire my (4) 100 Watt panels in Series-Parallel (Would this eliminate the need for fusing between panels?)
• (4) Trojan T-605 6V 210AH Batteries in Series-Parallel for a total of 12V 420AH
• Xantrex PROwatt SW 1000 Inverter

Cariboocoot wrote: »

Yes, crimping your own can be an adventure in loose wiring. #3 is kind of an odd size. Might be worthwhile to just get some pre-made #4 cable for the interconnections. It's not like there will be a lot of them, right?

Cariboocoot wrote: »

No.
Connecting two panels in parallel with no fuse and then connecting the third with a fuse offers no protection: you still have three panels in parallel. EACH panel must have a fuse or breaker when connecting more than two in parallel. There's no shortcut for this.

Cariboocoot wrote: »

If you're using a 30 Amp controller and running current under 30 Amps using at least 10 AWG wire that will work fine.