DIY Grid tied battery backup system

[PNjunction]

.. During the bulk charge state, every watt of power produced by the panels is put into the battery. Once the battery's voltage rises to a set threshold, the charging circuit will reduce the charge current, holding charge voltage constant. After the charge current is reduced to a specific percentage, the battery is simply given a float charge.

Kind of, but in the standard CC/CV regime, the charging circuit does NOT reduce the current, it is the battery that is on the gas pedal now. More accurately, once the voltage of the CC has been reached, the rising terminal voltage of the battery itself gets closer and closer to the CC's limit - and hence the reduction in current as the difference between the CC's voltage, and the battery's own terminal voltage gets smaller and smaller.

Not a big thing, but some assume that the charger is doing the current limiting once reaching absorb.

What I would like to do is to find some good way to take the power not being used by the controller for battery charging, once the charger reaches it's absorption phase, and send this energy to a grid tied inverter.

Well, we covered the no-no of an illegal grid-tie inverter. However, don't go thinking that your panel power is somehow wasted once you reach absorb.

AGM's, in reality, need to FINISH absorb, and get this - need to float at 13.6 - 13.8 for at least 8 hours to actually finish the job - otherwise you are walking them down in capacity. That little .1% that doesn't get recharged each cycle, eventually hard-sulfates, and just keeps on growing each and every cycle is part of the reason that agm's are widely murdered before their time. Continuous floating after that initial 8-12 hours is "maintenance", but that crucial first few hours after absorb are very hard to come by with solar. Sometimes just getting *to* absorb is a chore!

What is the absorb end-current for most for agm's? About 0.05C end current canonically - or no change in current for an hour or more due to age / drying out, etc. For you that would be about 14A with a battery capacity of 280Ah.

What I'm trying to say is that with your agm's, what you might view as inconsequential panel current at the end of absorb, or going into float is not wasted! Far from it - that is only extending the life of an agm that really truly needs to be fully charged as often as possible.

[AE0MT]

One more thought... real time data monitoring / sharing is fun! I'm a big fan of PVOutput, and with their extended values fields, you can post a lot of data.

Looking at your system right now, I see your battery is at 24.87 V under a 20 W load, 41% SOC by whatever calculation is being performed. If the time stamp displayed is your local time (pacific time), and night is approaching, it looks like you've got a big gap between where you want the battery to be at the end of the day and where you actually are. Do you have a grid-powered AC charger to maintain the battery when solar isn't getting it done?

I am central time, in Minnesota.
Today was a completely overcast day, so I pulled in a grand total of 14 WH of power. Yay me! No charging today. The voltage you see is essentially resting voltage.
The batteries are likely a bit oversized for the panels I have at the moment. These batteries came from my former employer, so I was able to purchase them at a substantial discount. They are Discover AGM's.
I do not have AC grid charging capability at the moment, but I do have AC power connection ability to run the items connected to the battery. I am using a ham product called the PowerGate 40
http://www.westmountainradio.com/pro...ducts_id=pg40s to switch power from AC to Solar. I have temporarily disabled the AC charging side of this PowerGate, because that unit is designed to charge 12v batteries, and I am running the batteries in a 24v bank with a 24-14v DC-DC power supply. I do this to ensure that my loads are always seeing 13.8v, regardless of battery charge.

I don't know why it shows that it is pulling 20w of power, because there should be no load on the solar batteries at the moment. My operating theory has been to run off AC power if I didn't pull enough in from solar.

I know this is not the most efficient use of the panels, and that is what I am trying to change. I think in all likelyhood, the most efficient use of the power would be to dump every watt of solar power into the grid, and then run the radio's as sunking is suggesting, and that is exactly what the PG40S is designed to do, aside for the fact that the voltage will drop when operating on battery.

[sensij]

Currently I have 3 100w Renogy panels on my roof, connected in series to my Tracer 3215RN 30A MPPT Charge controller. This controller is working wonderfully for me, and has plenty of room for me to expand in the future. I am also using the serial data protocol to log my system's stats, and give me real time monitoring capability through a raspberry pi. (http://69.180.190.146:8081/solar/) at the moment, its pretty nifty.

I am using this system to charge a pair of 12v 140AH deep cycle AGM batteries that were previously destined to live their life in a floor scrubber. I am running my system at 24v

One more thought... real time data monitoring / sharing is fun! I'm a big fan of PVOutput, and with their extended values fields, you can post a lot of data.

Looking at your system right now, I see your battery is at 24.87 V under a 20 W load, 41% SOC by whatever calculation is being performed. If the time stamp displayed is your local time (pacific time), and night is approaching, it looks like you've got a big gap between where you want the battery to be at the end of the day and where you actually are. Do you have a grid-powered AC charger to maintain the battery when solar isn't getting it done?

[Sunking]

Stop there my fellow ham friend.

In essence what you are asking for is you want to make electricity that cost you about $1/Kwh, and sell it to my neighbor for 10 cents via the POCO lines. Why do you want to do that for. Save yourself the time and trouble and just give your neighbors cash each month. At least they will be grateful and actually benefit from it instead of using the POCO as your agent and make them a profit.

Anything you take off-grid is going to cost you 5 to 10 times more than buying power from the utility. The reason is the batteries. For each dollar you spend on the battery will only yield you 20 to 40 cents worth of over their lifetime.

As for using it in the ham shack is OK, but does not really buy you anything except something to Rag Chew about to other clueless Hams who think it is cool. You can get the exact same thing and even better just using your DC Power Supply with a set of Station Batteries sized correctly. Under normal conditions the Power Supply odes all the work and keeps the batteries charged. When and if the power fails, the batteries take over. If properly sized can last several days to a couple of weeks. They need no solar on top of the cost of the batteries.

73's to you

KF5LJW out.

[sensij]

Thank you for your suggestion, but $3,500 is simply not in the cards. I have 300w of panel at the moment, and I doubt I will be going more than 800w in the near future. This seems to be significantly overkill for my needs, and if I understand correctly, would remove everything existing in my system except the panels.

I don't know where you are located, but around here in sunny San Diego, 800 W of panels would produce something like 1400 kWh annually. That is probably worth around $250-$300 annually. How much of that energy potential do you think will be excess and wasted by the battery system in absorb and float? Relays that connect to "opportunity" loads like water heating when battery voltage is being held above a trigger level are a conventional way to increase self-consumption, although what you spend to implement it will take some time to recover.

As I think you are seeing, the most cost-effective way to use panels (if you have net metering or a healthy feed-in tariff) is to make the system exclusively grid-tied.... your avoided energy costs may pay for that system in a few years. As Logan005 is suggesting, if the grid goes down, a manual transfer to a battery system that has otherwise been sitting in grid-powered float is reasonable, and the investment you have to make in pure off-grid equipment (that will otherwise be sitting unused) to recharge the batteries in that extended outage situation is not overwhelming if you can keep your critical grid-down energy requirements low.

[Logan005]

In my case, I have a Bus Panel, with both sources of power. 13.8 power supply and battery I use Australian plugs for my DC and can move loads easily from power supply to battery. you can also use a load diverter but is unnecessary if your home and need the power. You can also create a panel with DPDT switches that allow even easier transfer of sources.

[SunEagle]

How do I use the excess during the day? I agree, that if I can run things on grid, I should, but I want to maximize the amount of power generated from the panels.

The only other thought is to dump power straight to the grid, and charge batteries as needed from an AC charger, but that seems somewhat counter intuitive.

You might be able to attach additional small DC loads to your batteries and use the excess power when the sun is shining and the batteries are full and happy.

Otherwise you need a hybrid inverter (as mentioned by Logan005) that can charge your batteries and send power to the grid. Those are expensive and still require a contract with your POCO to send power to the grid.

If you attempt to use a cheap (and illegal) "grid tie inverter" to send power to the grid, you will get caught or worse may cause a fire due to those not inverters conforming to UL listing standards

[AE0MT]

How do I use the excess during the day? I agree, that if I can run things on grid, I should, but I want to maximize the amount of power generated from the panels.

The only other thought is to dump power straight to the grid, and charge batteries as needed from an AC charger, but that seems somewhat counter intuitive.

[SunEagle]

Thank you for your suggestion, but $3,500 is simply not in the cards. I have 300w of panel at the moment, and I doubt I will be going more than 800w in the near future. This seems to be significantly overkill for my needs, and if I understand correctly, would remove everything existing in my system except the panels.

Unfortunately there isn't any type of low cost, safe & legal device that would allow you to connect to your grid power. There is also the problem of sending power back onto the grid without permission which can get you in trouble with your POCO. Or in most cases actually cost you because your power meter can't "subtract" the power you send to the grid but will calculate it as if you purchased it.

[Logan005]

If you have grid avail, run your equipment from a 13.8 volt power supply, use the solar to maintain your batteries w/o cycling them, you can use the excess during the day and only switch to battery during power fail. Daily cycled batteries do not last very long. backup batteries last much longer. and less maintenance.

[AE0MT]

Thank you for your suggestion, but $3,500 is simply not in the cards. I have 300w of panel at the moment, and I doubt I will be going more than 800w in the near future. This seems to be significantly overkill for my needs, and if I understand correctly, would remove everything existing in my system except the panels.

[Logan005]

I like the OutBack Flex Power One. it is very expensive, but it brings the battery bank up to float and sells power to the grid. It also does not cycle your batteries unless the grid lapses, therefore greatly extending the life of your battery bank. It does not make any sense to cycle batteries if you have grid available.

[AE0MT]

Hello everyone.

This may have been covered before, but I am scratching my head a little to try and figure out the best way to manage my system.

Currently I have 3 100w Renogy panels on my roof, connected in series to my Tracer 3215RN 30A MPPT Charge controller. This controller is working wonderfully for me, and has plenty of room for me to expand in the future. I am also using the serial data protocol to log my system's stats, and give me real time monitoring capability through a raspberry pi. (http://69.180.190.146:8081/solar/) at the moment, its pretty nifty.

I am using this system to charge a pair of 12v 140AH deep cycle AGM batteries that were previously destined to live their life in a floor scrubber. I am running my system at 24v

This energy is currently used to power my Ham Radio station, my raspberry pi's, my cable modem/router etc. In the winter, this balance works out well, but I am planning on expanding the system in the future, and I am optimistic that I will be producing a much greater amount of power this summer, so I would like to find a good way to send the excess to the grid.

As most of you are well aware, charging batteries happens in multiple stages. During the bulk charge state, every watt of power produced by the panels is put into the battery. Once the battery's voltage rises to a set threshold, the charging circuit will reduce the charge current, holding charge voltage constant. After the charge current is reduced to a specific percentage, the battery is simply given a float charge.

What I would like to do is to find some good way to take the power not being used by the controller for battery charging, once the charger reaches it's absorption phase, and send this energy to a grid tied inverter.
The basic idea being to properly charge the batteries, and keep them topped off every day, and what energy isnt used by the charger is sent to the grid.

Anyone have any idea how I can accomplish this goal, preferably without replacing my MPPT charge controller?


Thanks for the help.