Planning a Low Budget, Medium Capacity Solar System for Whole House

Basspig

This is a project I want to do in stages, as I can make the money available.
The house is all electric appliances. We have a high capacity deep well pump, which is our largest surge startup (86 LR amperes @ 240V).
I have to power two small businesses I run, an electronics repair shop and a recording studio. There's also a home theater with a 20,000 watt sound system and a large commercial 4K projector.
Our minimum quiescent draw is around 4kW per hour. Then there are electric stove, oven, a large refrigerator with a defroster heating element, electric dryer, 15 computers + monitors, several 18,000BTU air conditioners and a new minisplit Mr Cool AC that I just installed this month for the studio. We currently have 200A mains service, which is just 'adequate' for our loads.
Backup power at the moment is a 13000 watt Northstar Honda V-twin generator. It can run the well pump and one AC unit and the rest of the computers and lights, but much more than that and it struggles.

My dream solar system would be a Schneider Electric inverter, 60kW capacity, but that's for later.

We get may be 2 hours of direct sun in our yard per day in summer. A bit more in winter, due to bare trees. So I'm designing the panel capacity on the assumption that we get 2 hours of full sun and the rest of the day indirect light.

The beginning test concept system will involve a Y-Solar MPPT charge controller, a PowerJack 15kW split phase inverter, 8 500 watt panels from Yangtze Solar, and around 39kWH of LifePo batteries. I may start with cheaper AGM batteries from Full River Solar, which will be the largest part of the cost by a factor of almost 8.

The idea it to start running off grid for a few hours a day and watch the battery capacity remaining and then add more solar panels until we make it through a full 24 hour period on inverter power. Below are some of the components under consideration.

Ultimately, I'll want to be able to produce 80kW of power so that we can operate like we normally do on grid, but off grid. But this small scale system will be roughly equivalent to our portable generator for short periods. I may need 50 or more large solar panels to generate enough power for a 24 hour period, given the amount of shade here.

Dave Angelini

I would concentrate my dreams on not using as much power by shifting to the offgrid mentality on consumption.
It is not easy for many to do. There is a very high failure rate when using a super large power system, when it is not required.
Shifting off of the dryer, range, and other large loads to more efficient sources.

It can be done your way and there are plenty of designers/installers that will do exactly what you want. I often get called later to fix it.

Conservation is the place to start in my opinion.

BB.

Besides measuring your loads, and conservation (converting to propane/other heating sources where practical), the 2 hours of day of direct sun is a killer... Anyway you can take down trees and/or move the array out aways and get better sun?

If you have grid and backup power (genset), why not go with Grid Tied Solar--Much cheaper, more efficient, and much less maintenance.

Guessing near Waterbury Connecticut, fixed array:
http://www.solarelectricityhandbook.com/solar-irradiance.html

Waterbury
Average Solar Insolation figures

Measured in kWh/m2/day onto a solar panel set at a 48° angle:
(For best year-round performance)

Jan	Feb	Mar	Apr	May	Jun
3.22	4.01	4.61	4.62	4.82	5.04
Jul	Aug	Sep	Oct	Nov	Dec
5.01	4.90	4.71	4.12	3.12	2.98

Not great solar, but better than 2 hours per day year round (better if adjustable tilt array)...

Figured out the long term costs of any solar power system (up front costs, maintenance, new batteries ever 5-10 years typical, new electronics every 10+ years, etc.)?

You can "make money" in many places with GT Solar power (cheaper than grid power costs). But GT Solar costs varies by utility/state/regulations... And as GT Solar becomes more prevalent, the subsidies offered by utilities are going away (no more "selling at retail", higher fixed service costs, commercial installations with more restrictions, and in some places, flat out bans or near-bans on GT solar).

I understand the desire to have a work/home place. But at some point, a more solar friendly location (or lower power costs) may be difficult to overlook.

If your utility power is reasonably reliable (say less than 1-2 weeks a year of outages from ice storms/etc.), building out a full off grid capable hybrid solar power system seems "expensive"... $1-$2 per kWH of off grid solar cost is not unusual (all in pricing just for components and upkeep). And getting close to $0.50 is hard work for off grid solar power.

GT solar power is probably 1/4 the cost of off grid solar power (sub $0.15 per kWH pretty easily?). No costs for batteries (and replacements), no charge controllers, smaller AC inverters needed, etc...

One difficult issue that has hit some places out in California was for Commercial power customers. Very roughly, there can be "Two Charges" for electric power--The $/kWH power cost, and the "reservation charge". More or less, the highest amount of Power Used in a 15 minute period in last month/1 year period (I am not sure of the details).

And, for a typical customer about 1/2 of the monthly charge is $/kWH and the other 1/2 is for Reservation Charges. What killed some solar customers (GT solar), was (for example) schools... They have a relatively low average power usage (9 months, non-summer, lighting and HVAC). And low usage during summer. It turned out the Reservation charges were the same for consuming power or generating power. And for GT solar, the peak power generation is ~2-3x the daily average energy usage.

So, during summer, low energy usage (no students) and high power generation (lots of sun on a GT system that pumps out 2-3x "average consumption" of the school... It ended up that Reservation charges doubled or worse because of the peak summer time generation. And GT solar installations intended to save/make money for schools ended up with the schools paying more for their power than before solar was installed.

I have no idea if any of this would apply to your place... But just be aware that there is a lot of technical and political stuff going on with solar power and GT solar power. In Nevada, in the last couple of years, the state PUC changed the billing plans and stopped all new GT installations because they no longer made economic sense.

And in California, they only promised a 10 year "grandfathering" of rate plans for GT solar--After 10 years, they reserved the right to do anything they wished (and after ~10 years, they have certainly made GT solar a bit less cost effective in California--What happens next, not sure).

Rolling in a 400 Amp service (single or 3 phase?) to your place, that some Utilities have given a hard time to Hybrid (GT+Battery) solar power systems (they are worried about you buying "cheap power at night", storing in batteries, and then selling into expensive power during afternoon peaks--They think it is "unfair" and not compatible with the idea of "green power").

Lots to research (and getting involved with utility engineers/rate folks to figure out what they will "allow" in their area, is there local distribution capable of more kVA/kW capacity, etc.).

-Bill

mike95490

Bare trees still have branches which will still throttle the PV panels.

and only 2 hours summer sun ?   And Electric range/cooktop.   you are smoking some strong stuff, and I think you have a tough road ahead. Just being blunt, and no sugar coating.  This is a big pill to swallow, and it's going to hurt your wallet deeply.

Basspig

I've looked at grid tied solar, as a friend of mine in Wilton got it installed recently. His panels generate about 4-5kWH of power and he claims he pay near zero on his electric bills. But he doesn't have a large home theater with a 20kW surround sound system.
We've made a lot of changes, as our electric bill doubled when the utility company introduced split billing (separate generation and delivery charges. We saw bills of $80-100/mo in the late 90s before the split. In 2002, the split happened and the bills were 160 from that day forward and rising. Today, we pay about $400 delivery charge and $330 for generation fees per month.
When the bill doubled, we stopped using 500W halogen lighting and 175W merc vapor lights outside all night long. I used to have a bunch of laser printers going 24/7 when I was in the publications business, and printing bulk mailings for clients (lots of ozone) and the fuser heaters wer 800W a piece.
We still have about 15 PCs in the house and numerous tablets and phones. Though we're saving money there too with upgrades and elimination of hot, power hungry mechanical drives for SSD drives that run cool and use less than a watt per drive. I got my PC power usages down from 1496W to 880W just by upgrading to modern motherboards, CPUs and SSDs.
On the lighting side we transitioned to CFL and then finally LEDs.
I'm replacing our 30 year old AC units with mini splits from Mr. Cool.
But the 20kW sound system stays. Forty years ago, I built something like the Greatful Dead's "Great Wall of Sound" in my basement. That's my raison d'etre. Loud music keeps me sane. I need to feel the earth move under my feet to get that 'high'. No drugs, just bass!
The other big load is the well pump. The well serves to supply water, but also for fire fighting. 50 gallons/minute to feed the sprinkler system that washes down the outside of the house in the event of another fire like the one in '66 which prompted me to build the house out of asbestos and put in the outdoor sprinklers. That pump has a 19,600 W startup surge.
We're surrounded by wetlands to the north and east, but we only have .5 of an acre. The house is 8' from the property line and the trees shading my yard are on a neighbor's property on all sides. I'm having 6 oaks removed on May 13, some of which block the sun to areas where I may put panels in the back yard.
Realistically, a couple of hours of direct sun per day falls on the yard, so I need enough panel capacity to store about 100kWH into batteries to cover the 4kWH steady state load. These loads are the computers, recording studio equipment, electronics shop with various scopes, analyzers and signal generators.
The idea of building a small/medium scale system is to see if we can run the base loads, and for how long. Initial outlay expected to be about $13K for eight 500W panels, the inverter and the charge controller and 39kWH of AGM batteries.
If that works out, I will look to expand it. At the very least, I can use it as back up power without running the Honda V-twin generator.
Realistically, I don't know that I'm going to put a whole lot more into this system, as our plans are to eventually expatriate to another country and rent this place out. But I hate recurring costs and electricity is getting very expensive now. If I can eliminate that cost, I'll have more money to put towards my goal of moving overseas.

BB.

It sounds like you have been working on saving energy (conservation always being our first suggested step)...

Next, can you tell us about your power bill a bit more... Is it something like $0.10 per kWH delivery charge + $0.08 per kWH for $0.18 per kWH (made up numbers)? Do you have time of day ($0.10 per kHW midnight-noon, $0.20 per kHW noon to 9pm, $0.15 per kWH 9pm to midnight)? Do you have tiered usage ($0.15 per kWH for first 100 kWH per month, $0.20 per kWH for >100 kWH per month? Etc.

How many kWH per month are you using?

If you do not have reservation charges (in our area, peak 15 minute kWatt power usage per month--I.e., drop peak energy usage or GT feed to grid, reservation charges drop)...

In general, GT solar is something like 1/4 the $/kWH cost to generate (or even less), than anything that uses a battery bank and AC inverter (all in costs).

One of the big issues with Lead Acid batteries, is it takes many hours of charging to fill up a bank (discharge to 75% state of charge, may take 4-6 hours to fully recharge). If you have only 2 hours of "useful shade free sun" per day, then you may be "forced" to use Li Ion (LiFePO4 or similar) chemistry batteries (you can hit them with high current and they will recharge very quickly, in a couple hours is possible--If you get the "right battery" construction).

Designing a cost effective system (one that saves you money rather than costing you even more money overall)--Is as much an engineering problem as an MBA problem (spread sheeting your utility billing plans against your energy usage over the month/year).

-Bill

mcgivor

@Basspig said.  The idea of building a small/medium scale system is to see if we can run the base loads, and for how long. Initial outlay expected to be about $13K for eight 500W panels, the inverter and the charge controller and 39kWH of AGM batteries.
If that works out, I will look to expand it. 

With the limited amount of sun your proposed experimental system will not work,  AGM batteries  are lead acid which  require hours of absorption and likely fail within months due to sulfation from cronic undercharging.


The only way something like this may be possible would be to use lithium batteries, because they can accept much higher charging rates without the need for absorption. The array needed ~30 kw rated, which would output ~25 kw so 60 × 500w panels. The charge controllers needed, based on a 48V system using 600V 80A units would be 8 at least, this would allow the batteries to be charged within 2 hours provided there are no clouds. My guestimate would be the cost to do just the experimental system would cost in the neighborhood of
 $150 000 not including protection, wiring, switchgear etcetera. Imagine how much it would cost to provide for all your current useage, as most will tell you, off grid is expensive.

Basspig

The more I think about it, the less I like AGM batteries. But in the past I've gotten 3 years out of a set in my 3000 VA UPS.
I recently bought a LifePo battery for my super bright flashlight project. It's a Dakota Lithium and it was $120 for a 12V 7Ah battery.
Research online shows the cheapest price per kWh is used Tesla automobile batteries. But the danger of those is too great to use in or near the house. That would require me building a second concrete bunker with anti-freezing mitigation, because Tesla batteries are damaged if they freeze.
My initial back of the napkin calculations called for close to 100 500W panels. That would be for a fully off grid system that could provide pretty much all the power I would ever need.
Grid tie stuff requires contractors and permits. I'm doing this the way I built my house. One piece at a time. It took me 40 years to build my house with just me doing all the work. But when I started, my electric bill at my old house was $8 in 1965. Now it's $743 in 2020. About $400 of that bill is "delivery charges".
Alibaba has the 500W panels for about $100 a piece. I could conceivably get 100 of them for about $10K. I don't know what the freight costs are as of yet though.
The Powerjack inverter is gone, but there's the Sigineer 15kW (45kW 20S surge) for about $3500, which is must more robust and also a lot larger and heavier. Saw a couple of reviews and it seems they are much heavier duty.
The really expense always comes back to batteries. If I could live somewhere where I was able to own my property outright with no rents due to any king, I would not mind investing six figures in a massive solar power system. But around here, everything is taxable and that eats the value of power generated by the system.

Basspig

So it begins... I've completed the first of two 20-panel arrays today.
This is a pilot system to test the concept of whether I can generate enough power to go off grid.
These panels are used, paid 10 cents on the dollar for them. Shipping actually cost more than the panels!

These will feed the BMS for a bank of 14kWh of LifePo4 batteries which will drive a 6kW inverter from Sun Gold Power. The batteries were bought from Treeline Power Systems. 

Shockingly, the lumber to build this (at today's prices) cost more than the solar panels! $989 spent on 2x4s and 2x6s. Two years, ago, this lumber would have been $150. I did not expect this "misc" cost to become major, but what's done is done.

I need to finish wiring up the panels. Have harnesses connected groups of eight. Each pair (top bottom) is in series, the ten pairs will be connected in parallel. I have to lay the 115' of 8awg wire to go to the battery room this week. Initially, I'm going to just connect it to the BMS and let it keep the batteries charged.

I had 6 panels connected to a 5 ohm 160W power resistor and measured about 64 volts across the resistor, meaning I'm getting just over 700W out of six panels with partial shading at 2pm on a mixed clouds and sun weather condition.

Next week, I need to grade the land on the west edge of the property and begin building the second array.

If this works out, the plan is to move to triple the batteries and buy the battery cabinet that Treeline sells, and upgrade the inverter to the 18kW version with 54kW surge capacity. That should enable us to run pretty normally.

mike95490

Just be sure to anchor your big solar sail - so it does not go on walkabout !!

Basspig

Plans to anchor it, indeed. Wind is not much of a problem here as we're in a bowl in the top of the mountain and shielded by hillsides on the south and west. At ground levels my anemometer rarely records wind velocities over 2-3 mph. We did have a F1 tornado rip through here in 2018 though. So I will anchor it for those rare situations. It's pretty stable even so, because of the way I designed the supports with lengthy overhang in the front.

This is a concept test to see if I can get enough solar output in a wooded area. I had one huge pine tree that was blocking the lower half of this array around 3-5pm, so I took the excavator out and pushed it over. If it's sunny tomorrow, I'll make a new measurement. With the shaded situation, the array made 1770 watts into the dummy load ( 3 giant 4 ohm resistors in parallel for 1.33 ohms total. 
Another thing I learned about these panels is to clean the slight film that built up on them over the years. Testing one panel at noon and then cleaning it with steel wool and testing again, revealed a 10% increase in output.

Today I dug a trench with the excavator and buried the cable and finished wiring. It's wired all the way to the battery room.
Next is to clear the land on the west side and build array #2 which will catch the morning to early afternoon sun.

Basspig

Late this morning, my array was getting off-angle sunlight and a shadow cast across the entire array by the trunk of a large oak tree just east of it. The array was making about 500 watts at this time, so I decided to hook it to my 14kWh battery bank and top off. The batteries had about 54.2V at the time. Went out for a walk with the wife, about an hour, and came back. Battery voltage was at 67 volts! In just one hour. Amazing.
Later in the afternoon, I put the dummy load of 1.33 ohms across the wires coming from the panels and measured, 51.9V. Just over 2kW.
Still later, I was out in back of the array and noticed one of my cables hanging down, disconnected. A pair of panels was not connected to the rest of the array. The connector had broke apart (the plastic nose and hook part had separated from the rest of the connector. I repaired that. Am a bit annoyed that Renogy didn't make these junction splitters with longer wires so that they could actually REACH the four panels they are designed to lnk.
Tomorrow I have some heavy duty lugs arriving, so I am going to make a more permanent hookup to the battery bank and then switch the household power to the SunGold inverter and see how the battery voltage holds up.

Basspig

Today is the first real test of the system. At 2pm, I disconnected from grid power and have been running off the Sungold inverter and solar array #1. After 6pm, output from the array was down to a trickle, and we've been running on the 14kWh bank of Lifepo4 batteries since. I'm at 52.0 volts on the battery bank now.
Getting ready to purchase two more 14kWh battery sets from Treeline Power Systems and the steel cabinet for them.
Also about to spring for the 18,000W inverter from Sungold.
I'm wondering about hooking the three battery banks together in parallel with their three BMS'es. I'll have to charge all batteries to the exact same voltage before I connect them together, but wondering how all those BMS units will work together. I may hit up Treeline for some advise.
The bigger inverter has a MPPT solar charge controller, so I can go directly into the unit from my panels.
I have an 85 amp blocking diode in series, but during the day, that thing gets real hot. I'll heat sink it and devise a means to isolate it electrically.
Even as it stands now, this system, backed up by a generator, could get us through long term outages very nicely.

BB.

I don't understand the 85 Amp diode--Exactly where is it, and what is is supposed to do? What happens if the diode(s) "fails"--Diodes can fail open or shorted--And then what happens to the rest of your system?

One thing to do with paralleled battery strings is to have a "balanced" length/resistance for the bank wiring so that wiring resistance does not unbalance/cause strings to share current unequally... Via Smartgauge website:

http://www.smartgauge.co.uk/batt_con.html

Just don't run your battery bus voltage too close to the High and Low BMS trip points... In my humble opinion, if you have BMS shutdown (relays, MOFETS, etc.)--The BMS shutdown of your battery bus is a last ditch effort to protect your expensive battery bank from damage.

Ideally, if the rest of your system is working correctly (chargers, inverters, etc.), then they should have "soft warnings" and graceful shutdowns before the BMS kicks off. The multiple BMS should work independently of each other normally (unless they log/report battery status banks as a single "Bank Capacity).

Depending on what BMS you get--Some just monitor per cell voltage (and give you warnings/shutdown signals)... And others will actively balance high/low cells in a string too. Just do per cell/per battery voltage checks to make sure all is working within the specifications.

-Bill

Vic

In looking at the Treeline Power Systems site, for their LiFePO4 battery BMSes, it states, "The BMS will disconnect the battery in case of over discharge, over charge or high temperature".  But, see no mention about not allowing charge of cool/cold batteries, which in CT, could be a consideration   ...   perhaps, not.  [perhaps these batteries have build-in heaters] ...

FWIW, GOOD LUCK,  Vic

Basspig

I talked with the folks at Treeline Systems and they just told me it's fine to connect the battery banks in parallel. I'm going to balance them slowly through a low ohmage power resistor though, to be safe.

The diode is there to prevent current from flowing back into the panels and draining the batteries at night. I observed that my ammeter goes negative as soon as it gets dark out before I installed the diode. If it fails shorted, I just lose some charge at night due to the batts powering the panels. If it fails open, the batts simply won't charge via solar. Blocking diodes are recommended to keep shaded arrays from becoming a load on arrays that are receiving sunlight.

I'm running the battery banks in the basement. It never goes below 62 or above 70 degrees F, so freezing is not a problem. Whether they are charging or supplying power, I've never felt them get even slightly warm, which is interesting.

I'm going to pull down a big oak tree which is casting a shadow across the entire array from 11am to 2:30pm, this weekend, if my crossbow arrives, which is needed to shoot the leader line over the limb so I can pull the big rope up there. Was hoping I could keep the tree as it's my antenna tree which supports a 90' vertical HAM antenna, but it's killing probably 50% of the array's output with the 36" trunk casting a huge shadow.

I need to fell another double oak on the west lot which is casted shadows on the array at 4pm onward, before I level the ground there and build array #2. Much work to do, but these $800 (soon to break $1000 with Eversource rate increase) are huge motivators.

mike95490

Nearly all Solar Charge Controllers built since the 80's, have the "Backfeed" diode funnction built into them.

I hope you are using a charge controller, as a battery's BMS is only a last ditch " save the battery" system, not really a charge regulator.

mike95490

Blocking diodes are recommended to keep shaded arrays from becoming a load on arrays that are receiving sunlight.

Only recommended by diode sellers.   I'd love to see that in print somewhere.

 A shaded array, will still produce Voc, but just not with any measurable current.  Well, if the shaded array is in a cave with no light, it will suck a tiny bit of power, but who would put an array in a cave ?

 Hooking arrays up in parallel is a perfectly accepted way of creating a large Virtual array, one array faces east, one faces west.  The array in the shade does not detract from the illuminated array.

Photowhit

Lith batteries have very little internal resistance, hence no warming.

The charge controller will prevent backfeeding the array... The charge controller it's self will use a tiny bit of energy.

Basspig

The larger inverter I'm about to order has an MPPT charge controller. Presently, I do not have a charge controller with the 6kW Sungold inverter. That's when I noticed the back flow of about 100mA at sunset.
I did some research and apparently the junction boxes on the back of panels contains some blocking diodes. It's quite surprising that while testing a single panel last month, if I shade most of ONE cell, the panel output power drops by about 50%.
My setup will use two arrays, one facing east for the morning to noon sun, the other facing southwest for the afternoon sun.
I would expect these batteries to have very low internal resistance, but curiously, when I connect the solar array to the system, the battery voltage increases by several volts immediately. Hmmm.
Today, I dropped the big oak. The tree was 32" at the trunk where I cut it. It was leaning at 30 degrees off vertical. I used a crossbow to shoot a leader cord over the bough about 40' up and pulled up a big rope, which I attached to an excavator, the purpose being to prevent the tree from falling too close to the array when it comes down. I cut my notch and did a stepped plunge cut to reduce the possibility of barberchairing when it comes down. Put the rope under tension and make the back cut. The tree came down exactly where I wanted it to.
Unfortunately, later on, while cutting branches with the chainsaw, I lost my footing while stepping on a pile of branches and gouged my left knee on jagged piece of oak branch on the ground. Went to the bone and bled like a waterfall. All bandaged up now, but it's gonna leave a scar. I'm taking the rest of the week off to recover.

BB.

Wow. Sorry about the injury. Not even a chainsaw injury.  

Blocking diodes were used (as I understand) decades ago when you had paralleled strings of panels... They were used instead of fuses/circuit breakers to block backflow of current from the rest of the array into a shorted solar panel/string. Now use Fuse/circuit breaker per string for protection.

And, yes, before solar charge controllers were used for charging battery banks, a blocking diode could be used to prevent backflow at night from (typically 24 volt or higher voltage) battery bank. Today, any modern charge controller will prevent (significant) current backflow).

Today, solar panels generally have "bypass" diodes instead... These are needed for systems >~12 volts (array voltage) and using series connected panels (or "higher voltage" panels).

The idea is that solar cells are really just giant diodes. Each "diode" (cell) generates around 0.5 volts (silicon panels). The problem is that solar cells go "high resistance" in the dark. And the solar cell "diode" can withstand about 12 volts max of "reverse voltage" before the cell/diode fails.

The 2 or 3 bypass diodes are used on a "12 volt" (Vmp~18 volt) solar panel. If one of the cells in a series string is blocked (leaf, shade, etc.), then the current from the rest of the cells goes through the bypass diode to prevent "over reverse voltage" on the "dark cell".

That all works well and good... But you can see if you have two diodes in an 18 volt array, that means that dark cell loses the ~9 volts generated by the bypassed cells (plus another volt or 2 bypass diode drop). For lower voltage parallel stings--Losing 8 volts "worth of cell" in a parallel string pretty much takes that sting out of commision.

For higher voltage single and parallel string arrays (such as 400 Vmp GT inverter arrays)--Losing 8 volts on a 400 volt array will not cost much energy:

• 400 VDC * 8 amps = 3,200 Watts
• 392 VDC * 8 amps = 3,156 Watts

And for a 400 Vmp array... One string with 400 Vmp and a second string with 392 Vmp (blocked cell)--That is not a big difference. You can have a 5% (20 volt) or upwards of 10% (40 volt) difference between Vmp strings--And they will won't lose much energy.

The down side, solar panels are not great radiators of heat... A 8 amp * 2 volt drop diode--That is 16 Watts inside a (typically) plastic j-box with a glass top (panel)--And those diodes can get pretty hot (possible future thermal failure).

Here is a nice write up:

https://www.electricaltechnology.org/2019/10/blocking-bypass-diode-solar-panel-junction-box.html

Blocking diodes also have the issue of voltage drop... A typical diode may run a 1.0 to 2.0 volt drop. Put 8 amps through a blocking diode:

• 1 volt drop * 8 amps = 8 Watts
• 2 volt drop * 8 amps = 16 Watts

Compared to your 100 mA power loss:

• 0.1 amps * 24 volts = 2.4 Watt night loss

Adding blocking diodes are (usually) just an overall system power loss.

-Bill

Basspig

I'm healing little by little each day. Gonna be two weeks before I can really bend this knee without the scab cracking and a month til it's fully back to normal. Thank you for your concern!

Helpful information in this post. I did happen across that page about the bypass and blocking diodes when I was researching this myself last month.

The 85 amp Schottky diodes I'm using have a 0.4V drop as measured on a Fluke 887. I'm still getting plenty of power to charge batteries during the day. But I will try with and without the diodes when I upgrade to the 18kW inverter.

I have an oral surgery coming up this Friday and I have to see how much it's going to take out of my solar budget as it's not covered by insurance. Once that's paid for, I'll figure out if I can do the rest of the batteries and inverter upgrade. About $12K there.

Basspig

I have been busy this past month. Got the 40 panels assembled.
Just got in the additional batteries and cabinet and have started wiring up the first of three banks with BMS. 
The house has been running on one bank of 16 cells since mid April, during daylight hours. We've already seen a 42% reduction in energy usage according to our recent electric bill.
Once the additional batteries come online, we should be able to run 24/7 off grid. I am considering buying four new 450W Renogy panels and placing them on the flat roof for additional capacity, as the roof gets sun for more  hours and the new panels should put out almost 2kW. With all of that, we should be pretty well on the way to energy independence.

Basspig

Just completed wiring my battery cabinet. 42kWh of capacity now.

The 6kW inverter is temporary. An 18kW unit is on order.

Basspig

Here's the one year update on my solar power system. Total expenditure down to the last copper wire, nut and bolt = $25K.

PV capacity is not at 21kW. Actual realized power into the battery at 12:30pm, 18.2kW.

The new mono crystalline panels do quite well under cloudy conditions. I've been able to maintain battery charge through a week of cloudy/rainy weather back in the end of April. Nearly 2 weeks of overcast in the 40s. I have been able to avoid touching grid power. The two new arrays have been a game-changer.

I've made a video tour which is up to date. An exhaust fan has made all the difference in indoor temperature!

https://youtu.be/iM_7cYUSgwU