Looking for guidance on correcting first self-taught system

[aefriot]

Location:

New York, USA
Latitude 44.6


Inventory:

12 - 150 watt mono panels
3 - 30A MPPT CCs
8 - Duracell 6V GC2
2 - DEKA GEL 375 Ah 24V (obtained used)
1 - Champion AGM 500 Ah 24V (obtained used)
1 - DEKA GEL 250 Ah 24V (obtained used)
2 - 3000 watt inverters
1 - Ryobi 2300 watt inverter generator
1 - 5k diesel brushless genset
1 - 6.5k gas Onan genset
1 - 4k gas Onan genset
2 - aluminum bus bars 1/4" x 1 1/2" x 6" surface finely sanded flat for good connection
4 - aluminum lugs double 2 ga mating surface finely sanded flat for good connection
2 - aluminum lugs single 2 ga mating surface finely sanded flat for good connection



Current Setup:

3 groups of 4 150 watt panels in series, 10 ga. 90C copper wire, 10A fused to each to a 30A MMPT CC
Each CC, 10 ga. 90C copper wire, Negative to single common aluminum bus bar, Pos is 30A fused to a single common aluminum bus bar
Power is supplied to 1 - 3000 watt inverter with 2 ga. 90C copper wire attached to positive and negative common bus bars with aluminum lugs
2 - DEKA Gel 375 Ah 24V batteries with industrial copper welding cables in parallel connection to common positive and negative aluminum bus bars
Electric paste on all connections


Background:

My wife and I purchased property which had no grid power. Instead of paying the $9000+ to get a service point we decided it was in our best interest to buy a photo-voltaic system. Upon checking the price of installed systems, I thought I might be able to assemble the system myself. While not entirely sure of what I was doing, I figured if I did some research, if I kept the proper fire extinguisher close and kept a reasonable distance plus a step or two from things that could explode, I would be okay. Well, I did some reading and upon reading suggestions to first buy cheap golf cart batteries, I decided it was where I was going to start. I then purchased the panels which came with PWM CCs and a couple inverters with one for a spare. My wife works from home. If I was to have any chance at being happy if I were to blow up an inverter, I wanted to be able to get her back up and operating in a flash because waiting a week for a new inverter to be shipped to us wasn't going to be easy on me. In the meantime, I saw on a local ad site an ad for a free electric pallet jack. I thought that would be an excellent battery backup in case I damaged the golf cart batteries. Besides, a 1000 lb battery should hold enough power for us. We just use LED lighting and no television. Well, we do have fans and can run AC as long as the sun shines if we start the AC early enough to be ahead of the heat. As it happened, I was able to obtain three more electric 24 volt pallet jacks with batteries. The price I paid, I could not pass them up. Knowing we would be off-grid, we knew we would need generators too. Out property has several outbuildings too far from each other to economically wire them together for a single power source so we have generators for each of the important structures.

I have assembled our system as in the current setup. I am using the free/cheap batteries since it's easy to lose something that didn't cost much in the first place. If I go through the free and cheap batteries, I will use the golf cart batteries. Until then, I will just keep topping the golf cart batteries up and use the industrial stuff.

The most power I have seen this system charge is about 1600 watts from about 10 am to noon in full sun. On a cloudy day I will get 400 watts or so. The inverter disconnects me at 21.3V I believe. I have hit this number a few times I'm afraid to say. I do try to use a generator (Ryobi 2300) to recharge the batteries with a 20A charger. Each CC will vary charging wattage slightly but may vary by an estimated 15% from each other. Voltage tops out at 29.1V then drops to about 26.5V mid-afternoon and stays above 25V after sundown. I have a clamp-type AC/DC ammeter and an infrared temperature meter that I use to see how things are going. I found that one battery (B) terminal was 30 degrees warmer than the others so I checked the ammeter on that battery compared to the other. It was charging 20A compared to the others' 5A. I found the reason to be a loose battery cable termination. I removed the cable, cleaned the connection, re-crimped the connection and it's much better now with zero ohms and charging amps have dropped in half. I was losing power to heat. Then I checked all other cable to terminal resistance to find that the positive terminal/cable combination on battery A has 32 Ohms resistance.

Reading posts here while awaiting to be approved for posting my topic I did read that parallel batteries should not be done. I don't see how to change the configuration and still use these batteries to utilize the 750 Ah of these batteries.

My questions are:

How do I repair Battery B cable/terminal and get the zero Ohms I believe it should have? It is a one-piece leaded terminal that will not fit a normal terminal. Can I drill and tap it to fit bolt to attach a new cable? I don't like the idea of shorting a battery with that many amps and ruining my....uh....drill.

I need to test the batteries to be sure I am getting the best use of them. I don't want to continue using the parallel 375 Ah batteries if I can make better use of the single 500 Ah battery for my home. Then use the 375 Ah batteries individually in different workshops to power the lights and small tools. I know they aren't new, but still may provide me with more power than the golf cart batteries. How do I reliably test the Ah of my batteries?

I hope I have included all relevant information you need to assist me.

Thank you

[Mike90250]

Welcome.
battery terminal ? Is this the post on the battery or the cable/lug that attaches to it ? If the battery terminal is bad, they can sometimes be "re-cast" at a battery shop.
If the lug on the cable is bad, you can cut it off and crimp a new one on, or take it to a battery shop and have them crimp it. Only hydraulic crimpers (10 ton or more) can properly crimp a lug to a cable and get nearly zero ohms every time. The high pressure cold welds the copper to the lug.

The only reliable way to test batteries is with a hydrometer, using temperature compensation. (manual or automatic) once the battery is charged, you can drain it with an inverter with a known load for some hours, and let the battery rest several hours, and then measure it's voltage. if the battery cannot be charged to read 1.2 on the hydrometer, it's likely going to fail soon.

https://batteryuniversity.com/learn/article/how_to_measure_state_of_charge

[sdold]

Reading posts here while awaiting to be approved for posting my topic I did read that parallel batteries should not be done. I don't see how to change the configuration and still use these batteries to utilize the 750 Ah of these batteries.

For system design, watt hours is more useful than amp hours. You'd get the same energy with the batteries wired in series or parallel. In parallel you have 24V x 750AH = 18,000 WH, and in series you have 48V x 325 AH = 18,000 WH, and have the benefit of lower losses due to lower current required for the same power. You'd need to use charge controllers and inverters that are designed for a 48V battery.

[aefriot]

Thank you for your reply.

The batteries are industrial batteries with the cables attached/molded to the sides of the positive and negative terminals of the battery. (Battery B has previously been refurbished and had it's molded connection reworked with removable cable lugs attached) With an ohm meter, I measured 32 ohms from my positive battery terminal (Battery A) to the copper at the end of the positive battery cable. To isolate the cable from the terminal, I measured from the exposed copper at the end of the cable to a point near the battery(I penetrated the cable insulation with the probe). Again, I found 32 ohms. I conclude the cable must be faulty and should be replaced. Since the cable is molded to the side of the positive terminal, I need to remove it and attach a new one. Given some additional thought, I can attach a new cable to the original cable molded connection after removing the faulty flexible cable from the connection. By working with the molded connection to the side of the battery terminal, I don't need to penetrate the battery terminal. This is a much safer method of repairing the battery.

Thank you for suggesting that the battery be taken to a battery shop, but we live 3+ hours from the nearest battery shop. I do prefer to do my own work when possible as long as I assess I can do it with reasonable safely. I don't go head first into my projects without regard for safety. As for working on the battery, I can see five sides of the molded terminal connection and the sixth side is solid lead that I'm not going to touch. I believe that I can safely rework the original molded connection to accept a new cable connection.

As for the testing of the performance of the batteries. A hydrometer will not work with my batteries. With the exception of the golf cart batteries, they are VRLA AGM or VRLA GEL types and cannot be tested with a hydrometer (Correct?). Maybe I can just check internal resistance as suggested by Sunking in https://www.solarpaneltalk.com/forum...-float-service

Backwoods I think I gota tip you may want to consider. I know you have the skills and equipment to do this. Best of all it gives you some real data to determine battery health. You can look inside so to speak. What I am talking about is the battery Internal Resistance (Ri) using the Delta Voltage/Current measurement. Ri will tell you when your battery is toast. When Ri doubles, you are at 80% rated capacity and time for new batteries.

Really simple to do. With the batteries charged and rested (room temp), apply a light load like 1 to 5 amps and measure the voltage on the battery Temp Post. Example you measure 6.3 volts @ 2 amps. Call the voltage V1 and current I1.

Now repeat the same test except now you as much current as you can get, say 200 amps. The higher the current, the more accurate result you get. So lets say you get 200 amps @ 6.0 volts. Call the voltage V2, and current I2.

Ri = (V1 - V2) / (I2 - I1)

Ri = (6.3 - 6) / (200 - 2)

Ri = 0.3 / 198

Ri = .0015 Ohms

It works for any kind of battery, and Ri will tell you a whole lot about the battery. Example how much current can you take and keep voltage sag to 3% or less. .03 x 6.3 = .189 volts. So .189 volts / .0015 Ohms = 126 amps. I am not making these numbers up. They are your real number, look at the cut sheet. Ri = 1.5 milliohms.

You can also use the same concept to measure the whole plant resistance including cables. If the batteries are 1.5 x 8 batteries = 12 milliohms. Cable connection resistance should be 25 micro-ohms of less and there would be 20 connection x 25 micro-ohms = 500 micro-Ohms = 12 = 12.5 milli-ohms. total resistance. Makes it real easy to know if you have a problem and how to find it.

I just need to find that data on my batteries to compare the current condition to that of when they were new. Finding Ri would be better than putting the batteries through a cycle to determine their condition and losing a cycle just to satisfy my own curiosity.

[aefriot]

For system design, watt hours is more useful than amp hours. You'd get the same energy with the batteries wired in series or parallel. In parallel you have 24V x 750AH = 18,000 WH, and in series you have 48V x 325 AH = 18,000 WH, and have the benefit of lower losses due to lower current required for the same power. You'd need to use charge controllers and inverters that are designed for a 48V battery.

Thank you for your time and reply.

Yes, I would prefer to have the 48V system, but I already had the equipment before finding my batteries. Honestly, I have found such great purchases by price, my system has been ridiculously cheap. I have been quite fortunate that I have been at the right place at the right time when gathering equipment for our experience with solar. I could lose our entire system and not be out more than a few hundred dollars. It has been a great learning experience!

I really wanted to develop my home for 120V (for example) DC operation. We heat and cook with wood, have no appliances and do not watch television. The only use of power is for lighting, laptops, fans and washing machine. I could eliminate the expensive inverter and use a few DC to DC converters for the few electric devices we have. I could use cheap conventional solid two conductor 12-16ga wiring, but I think I would still prefer the three wire ground-fault for safety sake in case of cabinet/case shorts. Motors in devices could be more compact or more powerful and easily regulated for speed (PWM vs. Frequency). I still have 9 months before building my house to decide. Maybe someone else could chime in about their ideas and experiences. It's nice to learn from others' experiences. I don't want to stick with convention just because it's what's always been done.

Thank you!

Edit: My charge controllers are okay for 24V or 48V. My inverters are for 24V only so I cannot go to 48V system. If a good buy comes up for a 48V inverter I might think about increasing system voltage to 48V.

[sdold]

I really wanted to develop my home for 120V (for example) DC operation.

I can't think of any devices that use 120V DC, do they exist?

The only use of power is for lighting, laptops, fans and washing machine.

How do you plan to run a washing machine without an inverter? Is there such a s thing as a DC washing machine?

I think if anything you should go with 48V and a couple of inverters, a smaller one to run your low-power devices, and a big one to run the washing machine. Actually I think your best bet would be to sell everything and pay the $9000 to have the grid connected

[aefriot]

I can't think of any devices that use 120V DC, do they exist?

I don't know, really. We don't use appliances other than a clothes washer so I haven't looked for them.

How do you plan to run a washing machine without an inverter? Is there such a s thing as a DC washing machine?

120V DC motors are readily available. All it would take is a bit of design, engineering and programming to make a controller to make that 120V DC motor operate and the washing machine do everything a 120VAC washing machine do. Sounds like a challenge...if only I had the time.

I think if anything you should go with 48V and a couple of inverters, a smaller one to run your low-power devices, and a big one to run the washing machine. Actually I think your best bet would be to sell everything and pay the $9000 to have the grid connected.

LOL! That's good one! Since we have assembled this system we have had zero electric expenses (Except for some cable, terminal and lug purchases). Even at the minimal bill (not using any power), our photo voltaic system had a 42 month break even point. It will have paid for itself in another two years or so without paying for the power too! Right now, it's all gravy! It doesn't need to work hard and we don't have any electric expenses. It's a member of our family now. We could never sell it out to BIG CORPORATE POWER! We don't have to look at ugly aerial power lines, poles or the on-ground transformers. We have all of the power we need from the very thing that warms our faces, grows our vegetables and lights our way (day and night). Why would we sell it and connect to the grid?

You do give me an idea though! I do have an extra 3000 watt 24V inverter that I can connect in series to be able to jump up to the 48V battery bank I would like to have our system evolve into. Doubling the voltage halves the amperage demands of the system. I don't need the extra power, but it's an economical way to get to the 48V that would get me away from the parallel charging that I have right now. It would also free up a charge controller for use as a spare or I could install a couple 300 watt panels I have in the barn I forgot I had. Out of sight, out of mind! I wasn't using them because I didn't have a charge controller for them. But, with a 48V system, I can rearrange the configuration and add in the two extra panels. That would bring me up to 2400 watts which would be easier to keep the battery topped off and ready for winter when the clouds make the sun's visitation less frequent.

Thank you for your input, sdold! Guess you never really know where a person will get their inspiration.

[sdold]

You do give me an idea though! I do have an extra 3000 watt 24V inverter that I can connect in series to be able to jump up to the 48V battery bank I would like to have our system evolve into.

Are you thinking of connecting two 24 inverter inputs in series so that they could be connected across a 48V battery? Did I read that correctly? If so, that's a terrible idea because they will not share the battery voltage equally unless they are presenting precisely the same load to the battery, which will never happen.

[jflorey2]

I really wanted to develop my home for 120V (for example) DC operation. We heat and cook with wood, have no appliances and do not watch television. The only use of power is for lighting, laptops, fans and washing machine. I could eliminate the expensive inverter and use a few DC to DC converters for the few electric devices we have.

For those loads 12V is probably sufficient. For the washing machine you could turn the inverter on only when you need it.

120 volts DC isn't really an option. 12V is common and lots of RV equipment works at that voltage. 370/400 volts DC is becoming common for data centers and some buildings in China but there's almost no 'normal appliances' available for it.

48 volts isn't a bad option but you are always going to be stuck with 1) custom connectors for anything you want to plug in and 2) DC/DC converters to get 48 volts down to 12.

[aefriot]

It's a good thing I am not someone who acts fast on something I think of. I tend to let things soak in so I can reflect on the consequences.

My inverters, while of the pure sine wave variety, have no easy (manufactured in) method to ensure both are phase-synchronized. They could not be connected to the same circuit. And, as you say...

that's a terrible idea because they will not share the battery voltage equally unless they are presenting precisely the same load to the battery, which will never happen.

While I don't like to be told I have terrible ideas, it's most likely true. I wanted to find out how terrible it would be so I searched online for experiences with just such a terrible idea. It's hard to find people who have even asked such a question, "Can you have two electronic loads in the same current path?", or similar. Short answer is, "No". Semi-short answer, and more semi-informative answer to boot, is "No, you cannot have more than one electronic device powered by the same circuit trying to control the current on that circuit."

My case scenario: (From what I read only, I did not try this)

Identical inverters, A and B,each is 1/2 of DC battery voltage, in series on the same DC power circuit. I already know the output of my inverters cannot be connected to the same AC circuit since they do not have the capability of parallel output, they most likely will be out of phase, and I don't want that. So, they have their own output circuits, Ao and Bo.

Let's assume they both are simple, on or off, with no self health check to mess with current flows. A is turned on...nothing!. No lights, no fan, no "Hello" beep. Why? Well, B is not turned on, of course, and therefore, no current is flowing through their shared circuit. Oh! Okay!

A finger to the switch of B. I don't see a flash of light and a bit of electronic smoke, A and B appear to be able to use an identical amount of current! A and B are idling along only powering themselves. I'm amazed and very happy at this point. Maybe this is going to work! I am eager to see if I can power something simple like a light bulb.

I plug a light bulb into the outlet of A. One of the following may have happened.

1) Nothing! B may be limiting the current to it's own needs at idle since it knows nothing of what's been asked of A. Heck, it doesn't even know of A, so it does not allow more current to flow in the curcuit. A doesn't have enough current to make the light bulb operate.

2) A shuts off since it was asked to supply current to the light bulb. When it tried to get current on the circuit, voltage dropped below its minimum operating current so it shut itself off. As it does so, B is left without a circuit to remain running and is without power.

3) B burns up as A attempts to flow current through a circuit that will is not designed for the load. A is also left without a circuit and is without power.

4) A and B go up in a puff of smoke! Oh, man! I'm bummed!


Honestly, I don't know everything that could happen, but these ideas are enough to keep me from trying to connect two 24V inverters in series to connect them to 48V. I give these examples only to provide reasons not to try this. I like to know why when I am advised one avenue over another, not just yes or no. Perhaps,I'm an infinite toddler. Since I found so little about this online, I wanted others to be able to find some reasons not to do this. I'm sure there are many more, as well.

I'm okay to keep my system safe, as it is now. The lights are on and batteries are full. My wife and I are happy.

[bcroe]

Nothing wrong with lots of ideas, but then research them to see if they are terrible.

Several knowledgeable people have already tried to explain why it won't work, better listen up.
Reminds me of some shielded room testing of a phone system early in my career. I accidentally
pulled the wrong power lead and miss connected several refrigerator sized DC power supplies.
One blew up and we had to send it back to HP for repair. When it became obsolete decades
later, it ended up as a spare in my own lab, to remind me of my past transgressions. Bruce Roe