Michael Qulek

Rico, a bit more information from you.  What's the general location you're at?  Not pinpointed, but generalized, like East Texas, or Western Washington.  Design makes a big difference when you take into account your sunhours (sh), and what time of the year your sun is shining.

Can you come up with an itemized list of things you want to power?  What you power, and when is as important as how many watts they consume.  Items with electric motors that start under load have what's called "starting surge", or "inrush current".   The inverter you select needs to be able to handle the starting surges your stuff will generate, usually 3-5X the running amperage.

For example, my well-pump, with a head of ~250 of water above the pump, has a starting surge of ~9000W for about 500 miliseconds or so.  On the other hand, a free-spinning window fan in the bedroom only a starting surge of 1.1X.  So, my 100W fan consumes ~110W right at startup.

This makes a big difference in the choices of inverters you select to run your house.  Basically, there are two main types of inverters, low-frequency transformer based inverters, and high-frequency transformerless ones.  The LF models are big, HEAVY, and expensive, but they have massive surge capacity.  HF models are lighter, and much cheaper, but have almost zero starting surge capacity.  Ignore whatever claim they make as to how MUCH the inverter can surge to.  Pay close attention to how LONG they claim the inverter can surge to.  That is the number that will make or break an inverter in terms of starting a big-ticket item like a well-pump, or shop air-compressor.

The big LF inverters typically an surge for 5-60 seconds.  The HF models typically can surge for only 8-16 miliseconds, not long enough to start big motors.  You definately get what you pay for with inverters.  This is why the itemized list is so important.

Inverters on the market now can be divided into two styles, component, and AllinOne (AiO).  Component means you need a separate charge controller, and breaker panel system to feed into the inverter. Some of the component models are LF, but cheaper component models are HF.  The AiO models combine the charge controller in with the inverter in one box, making wiring more simple.  Almost all AiO inverters are HF.

If you go the component route, you'll also need a charge controller.  Again, they can be divided into two types, PulseWidthModulation (PWM), and MaximumPowerPointTracking (MPPT).  Just skip PWM completely!  They have no business being in a modern system.  MPPT is vastly superior in every way except price.  PWM is appropriate only for the cheapest, low budget systems.  Beware though.  Ebay has sellers marketing FAKE MPPT controllers.  A real MPPT will have a voltage limit of at least 100V, and weigh a couple of pounds.  If it says "MPPT" but only has a voltage limit of 36V, and weighs only 8oz, it's a fake.

The best solar panel deals around are local purchasing with cash and carry pickup.  You pay a steep price premium to have solar panels delivered to you via shipping.  Shop local venues like Craigslist for panels, and then drive over to pick them up.  If you worry about panel quality, bring a voltmeter with you.  Every panel has a specifications sticker on the back, listing Voc, Vmp, Isc, and Imp.  With the unconnected panel just sitting out in the sun, your voltmeter will be measuring the Voc, the open-circuit voltage.  Don't buy a panel if the measured Voc is more than 10% less than the published value.  For example, if the published Voc = 37.5V, and you measured 35.0V, that panel is OK and good to buy.  If however, you measured only 32.5V, then I'd say skip that one.  It's worn out.  A good deal with cash and carry panels is 4-5W/$.  With shipping to your house, you won't do better than 1-2W/$.

Battery choice is becoming a religious issue.  Every battery type has some kind of weakness.  Traditional lead-acid dies with chronic undercharging.  Small solar arrays kill lead-acid batteries.  On the other hand, LI batteries can not tolerate low temperatures.  They can be destroyed by placing them in an unheated shed that goes below freezing.  If you buy a goodly number of panels, and charge lead-acid at 1/8th of C, they will be happy and give you long years of service.  If you keep Li batteries inside a heated cabin that NEVER gets cold, then Li will be happy and give you long years of service.  Let your personal situation dictate what type to chose.

https://diysolarforum.com/ is a great site for beginners.  I post there on various topics, including solar water pumping.

The very first thing I would suggest is to sit down and draft a plan, before you attempt to purchase anything.  

The first part of the plan is to determine what loads you intend to power, and during what time of the day.  For big systems, I recommend identifying your single largest load, and build a system that has 2X the capacity for that.  You could go with either kWh/day, or W of your load.  Let me give you some examples.

I typically consume ~4-5 kWh of power per day, which powers the frig, TV, computer, lights, and stuff.  My location gets about 3 sunhours (sh) in December, and 6 in June.  So, if I need 5000Wh in December, I'd need 5000Wh/3sh =  1667W of solar panels.  It would be good to double that to ~3200W for catch-up after cloudy weather.

On the other hand, my 1hp well-pump is my single biggest load, consuming ~2100W of power, so following the 2X rule, I installed 4500W of panels.  They are on rotating ground mounts, so I can turn them East in the morning, and West in the afternoon.  With rotation, I can run the pump from 8am till 4pm with zero battery depletion.

I'd suggest sizing your inverter to 2-3X the size of your biggest load.  So, for my 2100W pump, I purchased a 6800W Schneider inverter (XW+6848).

Lastly, size your batteries to what what your biggest load is.  Traditional lead-acid likes to be drained at no more than 1/10 of C, while Li might tolerate 1/4th of C.  C being the capacity listed in amphours at the 20 hour rate.  Taking the example of my pump above, the inverter will draw 2100W/50V=42A, so I purchased 400Ah Trojan L-16 batteries.

I would suggest that manually disconnecting your controller to stop charging is in general a bad idea.  Either you will forget to disconnect it in time and overcharge, or you will forget to reconnect it and let it drain to minimum.

I think the best course of action is to replace whatever controller you have and purchase a controller specifically programmed to charge Li.  Then just leave it on auto.  You didn't mention anything about what kind/wattage solar panels you have?  Getting a good quality MPPT controller, and wiring your panels in series to raise the incoming voltage is usually the best strategy.

If you can document the Vmp/Voc and Imp/Isc of your panels, and your controller specifications, we can recommend the best wiring combination.  Knowing your extreme winter lows is important, because the Voc of you panels goes up as the winter temperatures go down.  Please get back to us with those numbers.

Aaron Yarbrough wrote:We installed

After looking at the module available now I'm considering just replacing my 320 watt modules with 450 watt bifacial ones. That way I could use my existing racking system and my generation would be 5.4 - 6.5 kW.

In the picture above you can see that the lowest row of modules is still partially shaded at 10:30 am. So, I also was thinking about ditching my inverter, charge controller, and the combiner box for an inverter with a built in charge controller and dual mppt so the performance of the partially shaded panels doesn't drag down the rest of the system. Seems like I could get all of this for $4000 not including the federal tax credit if I install it before the end of the year. Then I could use my still good quality existing components on a project down the road.

Any downsides with taking this route (other than probably convincing myself that I need a new a battery bank as well)?

I would suggest neither replacing you panels, or your inverter/charge controller is a good idea.  You can simply upgrade your system by adding additional strings in another location.

First, before diving in, please explain in detail your inverter and charge controller settings, what the voltage/amperage limits of each?  Those numbers are critical to proper design.  What voltages and amperages are you running through your wiring right now?

I could suggest a couple of things, based on my own personal experience with my own system upgrades.  It looks like you have your panels wired 4S3P, is that correct?  I guestimate that your strings are running at ~150VDC?  Is that correct?  What is the Voc of those panels?  How low are your winter lows?  I would guess the Voc of your strings might exceed 200V around freezing or so?

One easy way to add capacity is through what is referred to as "virtual tracking", that is positioning additional strings of panels in directions other than South.  For my own system, my primary arrays also face South, but I have additional arrays designed to face East, and West.  I say "designed" because they are all on single-pole rotating ground mounts.  You could do exactly the same.  You could place a ground mount facing due East to catch morning sun, or a West-facing mount for the late afternoon.  Besides my 4500W of South-facing panels (but can be rotated), I have 1000W facing East, and another 2000W facing West.  I'm planning on adding another 1000W on to the West-facing array (3000W total) because I typically need a lot more at 5pm, instead of 8am.

It's important that string voltage be within 5% of each other, but amperage may vary from string to string.  That means you do NOT have to have exactly the same panels, just match the voltages.  Any string putting out between 143-157V is going to work just fine.

Another option is you simply add a second charge controller to the system.  You will need to match up the charging parameters so they are exactly the same, and it will work best if you use a brand that is designed to be paralleled and can communicate with each other.  But, with the second controller, you have total freedom as to your string voltages and amperages, independent of your existing strings.

With ground mounts, and string voltages in the 150V range, you can run at least 150' from the charge controller with zero voltage drop.  Right now, I'm running 120VDC for ~130' through 10 gauge copper, and I can't measure any voltage drop at all.

I've seen this myself, shopping at HomeDepot, and placing my hand on a few of their bags of compost.  I could feel the heat in them.  So yes, it works.

Notice though the great size of the the piles, and the diesel-powered equipment used to construct it?  If you live next to a sawmill, you might have access to enough raw material to make it happen?  Homesteaders like me would be hard-pressed to find the cubic yards of material needed to build this?

Sardine salad sandwiches.  Just make Sardine salad just like you would Tuna salad, with mayo, mustard, salt and pepper.  Then spread it onto you favorate bread just like you would Tuna salad.

I checked my own well water with a confirmed LST test for fecal coliforms and E. coli.  The water was completely sterile by that method.

You can also try Lemi Shine.  I add it both to the dishwasher when I run it, and also add a spoon-full to the canner when I'm processing some jars.  One thing that works extremely well on rusted cookware is to sprinkle some Lemi Shine on a steel-wool pad, and clean off the rust with that.

In your case, you might try a sprinkle on a wet green scrubbing pad, and scouring off the film with that?