Series versus Parallel

[SunEagle]

I have had 7 of these panels running for 2 months (all wired in parallel) and have been on solar completely during the day. I had been using 4 batteries (400Ah total) and calculated that I should be able to run in the evening for about 2 hours (I assumed a 10% battery consumption before the inverter shut down due to low voltage). Sure enough, my calculations seemed to be inline with my observations. During the day, I was completely on solar (producing about 700 watts, consuming 500 and charging the battery with the 200). I used a 5000w PWM inverter from Harbor Freight which I got for $350. Soon as I began hooking up more motors (furnace, sumb pump etc.), the inverter gave out. This is why I am going with an industrial inverter so it can handle a surge for a longer time.

So let me get this straight. You have burned up a $350 inverter and are draining your batteries to the point the inverter shuts down due to low voltage.

I would give your (probably expensive) battery bank another month before it will fail to fully charge and then stop charging at all. Those probably cost you about $800. It must be nice to blow $1000 on a system that is not set up correctly. Good luck with your new industrial inverter.

[Ravi]

Well it is going to fail. You have done about every thing wrong so far. One huge eror is you assume you have 5 sun hours. You do not have anything close to that especially if the panels only get 8 hours of direct light a day. Only in June/July does your area even get close to 5 sun hours and that is only if you have complete unobstructed veiw of the East West South horizon, perfect solar south orientation, and optimum tilt angle where sun falls on the panel surface from sun up to sun down. . Come November-December, January and February will be 2 hours and less.

Off grid battery systems are designed for worse case. Come winter when your sun hours fall to less than 2 hours (less than 1 in your case with obstructions) your batteries will be dead and you will be in the dark all winter needing to buy all new batteries next spring. That is when you will discover you need a lot more panel wattage and much larger batteries.

Right now you do not even know how many watt hours you need in a day or Sun Hours have.

I have had 7 of these panels running for 2 months (all wired in parallel) and have been on solar completely during the day. I had been using 4 batteries (400Ah total) and calculated that I should be able to run in the evening for about 2 hours (I assumed a 10% battery consumption before the inverter shut down due to low voltage). Sure enough, my calculations seemed to be inline with my observations. During the day, I was completely on solar (producing about 700 watts, consuming 500 and charging the battery with the 200). I used a 5000w PWM inverter from Harbor Freight which I got for $350. Soon as I began hooking up more motors (furnace, sumb pump etc.), the inverter gave out. This is why I am going with an industrial inverter so it can handle a surge for a longer time.

[Ravi]

I am just trieng to plan for the future. I want to put more and more of my AC circuits on solar. I measured the current of all the circuits that I wish to slowly migrate and it came in at about 33 amps AC which gives me about 4000 watts. Some of the items are motors and have a huge surge.

[Wy_White_Wolf]

If your only using 500W why are you installing a 7000W inverter? Losses from the inverter are going to be more than your load.

WWW

[Sunking]

[ATTACH]3063[/ATTACH]I have 800w being produced from PV and typically 500 being used (I monitored using Killawatt device). This leaves me 300W going to battery. 300W * 5 hours (full sunlight -- I usually get 8 hours of full sun) = 1500wh to battery. I figured that this 1500Wh would be used at night when I have no solar and then get recharged in the day while I still have enough juice in the batteries.

Well it is going to fail. You have done about every thing wrong so far. One huge eror is you assume you have 5 sun hours. You do not have anything close to that especially if the panels only get 8 hours of direct light a day. Only in June/July does your area even get close to 5 sun hours and that is only if you have complete unobstructed veiw of the East West South horizon, perfect solar south orientation, and optimum tilt angle where sun falls on the panel surface from sun up to sun down. . Come November-December, January and February will be 2 hours and less.

Off grid battery systems are designed for worse case. Come winter when your sun hours fall to less than 2 hours (less than 1 in your case with obstructions) your batteries will be dead and you will be in the dark all winter needing to buy all new batteries next spring. That is when you will discover you need a lot more panel wattage and much larger batteries.

Right now you do not even know how many watt hours you need in a day or Sun Hours have.

[Ravi]

I am just starting out and from everything I read, I gathered that I needed the batteries and so I purchased them. Currently, my total cost is less than $6000 since I did all the work myself. I sized the number of batteries as follows:
I have 800w being produced from PV and typically 500 being used (I monitored using Killawatt device). This leaves me 300W going to battery. 300W * 5 hours (full sunlight -- I usually get 8 hours of full sun) = 1500wh to battery. I figured that this 1500Wh would be used at night when I have no solar and then get recharged in the day while I still have enough juice in the batteries.

16 batteries * 105ah per battery * 12 volt battery = 20160 Wh (fully discharged). I dont want the batteries to suffer so I planned on 10% daily discharge = 2016 watt hours being drawn from batteries (thus keeping the inverter going without shutting down due to low battery voltage). This along with routine de-sulfination (the MPPT controller has this built in) should make the batteries last 15-20 years (i am told). I also have a portable 12 volt desulfator that I can use if necessary to revive dead batteries. My return on ivestment should be about 7 years.

Going with the 16 batteries also provides me that surge of power when things like my refrigerator, microwave, sumb pump, furnace kick in.

If all goes well, I can forget the batteries in the future, change my inverter to grid tie (cant do that in my city yet) and continue onward.

I have enclosed the modified drawing of my system. I realize theoretical is not necessarily practical so i certainly expect my design to have flaws and welcome all input as I have no practical experience in this area. This is my first system and I am doing it for my home in southern illinois close to st.louis, missouri.

[Ravi]

SolarSetup.pdfI am just starting out and from everything I read, I gathered that I needed the batteries and so I purchased them. Currently, my total cost is less than $6000 since I did all the work myself. I sized the number of batteries as follows:
I have 800w being produced from PV and typically 500 being used (I monitored using Killawatt device). This leaves me 300W going to battery. 300W * 5 hours (full sunlight -- I usually get 8 hours of full sun) = 1500wh to battery. I figured that this 1500Wh would be used at night when I have no solar and then get recharged in the day while I still have enough juice in the batteries.

16 batteries * 105ah per battery * 12 volt battery = 20160 Wh (fully discharged). I dont want the batteries to suffer so I planned on 10% daily discharge = 2016 watt hours being drawn from batteries (thus keeping the inverter going without shutting down due to low battery voltage). This along with routine de-sulfination (the MPPT controller has this built in) should make the batteries last 15-20 years (i am told). I also have a portable 12 volt desulfator that I can use if necessary to revive dead batteries. My return on ivestment should be about 7 years.

Going with the 16 batteries also provides me that surge of power when things like my refrigerator, microwave, sumb pump, furnace kick in.

If all goes well, I can forget the batteries in the future, change my inverter to grid tie (cant do that in my city yet) and continue onward.

I have enclosed the modified drawing of my system. I realize theoretical is not necessarily practical so i certainly expect my design to have flaws and welcome all input as I have no practical experience in this area. This is my first system and I am doing it for my home in southern illinois close to st.louis, missouri.

[Mike90250]

...

Finally, no, there is no generator in the system because if there is not enough juice in the battery, the system will automatically switch over to grid power via my automatic transfer switches...

You have calculated out (haven't you) that the battery replacement cost from using all their cycles, is way more than grid cost. Have you considered grid as primary power, and the batteries as backup, and use a hybrid inverter to backfeed the grid and unwind the meter ? (if that's permitted in your area)

[Ravi]

I see I made a mistake in my drawing. Since the inverter can have a 14000W surge and my battery would be 48v (14000/48 = 292Amps DC); 292 amps/4 parallel strings of batteries = 72 amps per 48v battery set.
A 0 gauge wire is rated for upto 150A.

So I will use 150A breakers not the 300A breakers as shown to go between battery and bus bar for the batteries. This way up to 2 parallel strings can fail and I still have plenty of power coming from the batteries without any hazards.

I will also change the 0 guage wire from the battery bus bar to 300 amp breaker and to inverter with a A THHN 350 MCM wire since it is rated for upto 350 amps.

Also, the reason i went with the morningstar mppt60 is because i am a programmer familiar with modbus and can write my own monitoring app if I dont like the MSview app that I can use to monitor everything over the web or msview app.

Finally, no, there is no generator in the system because if there is not enough juice in the battery, the system will automatically switch over to grid power via my automatic transfer switches.

Aside from the breakers, any concerns?

[Mike90250]

Breakers protect wires. If you use a 300A breaker, you need wire good for 300A. O wire is not enough for for 300A if I recall. Smaller breakers or larger wire.
Be sure its a DC rated breaker for 75VDC.

Charge controller. I have a morningstar MPPT60, and like it. But I like the Midnight Classic better, and it comes with meter built in. (MS sells the optional meter, it ships with a blank faceplate)

And PDF seemed to hard to read, but I have a 10 yr old laptop too.

Inverter, does it have a built in charger, so your generator can efficiently charge your batteries when the sun fails (clouds) ? I don't see any generator in the system

[Ravi]

Thank you all sooooo much for your input. I have taken all your advice and created a design that I am enclosing in PDF form. I have gone with 48v battery bank and 48v strings for PV and have the circuit breakers as you suggested. Does everything look ok now? I built in plenty of future expandability. I hope I did the math right.

[Sunking]

I am done here. Any more help from me cost money.

[Sunking]

Thats the website where i got the idea of keeping all the battery cables to the same lengths. I was proposing going with Method 3 in which each battery (in my case of a 48 volt system, each battery set is made of 4 twelve volt batteries in series) is connected to a single point (1 point for positive and 1 point for negative) as shown in the picture.

Since my batteries are 12 volt and not 48 volt batteries, I would connect them in sets of 4 in series. Then each series set would go to a common bus bar (1 bar for positive and 1 bar for negative) just like method 3 shows.

I figured going with 48v battery system would allow me to add more panels in the future until I reach 3200w from the solar array going into charging the 48 volt battery system.

I got a great deal on 12v 115ah deep cycle marine battiers ($105 each)

Remember plan for the future. If you see going 48 volts, do it now. All it takes is a different inverter. Everything remains the same. It also cuts the parallel battery strings in half. That means less wire and connectors.

Marine batteries are not really a deal, they are not deep cycle, they are hybird a cross between cranking and deep cycle and will only last about half as long as a deep cycle.

[Ravi]

Thats the website where i got the idea of keeping all the battery cables to the same lengths. I was proposing going with Method 3 in which each battery (in my case of a 48 volt system, each battery set is made of 4 twelve volt batteries in series) is connected to a single point (1 point for positive and 1 point for negative) as shown in the picture.

Since my batteries are 12 volt and not 48 volt batteries, I would connect them in sets of 4 in series. Then each series set would go to a common bus bar (1 bar for positive and 1 bar for negative) just like method 3 shows.

I figured going with 48v battery system would allow me to add more panels in the future until I reach 3200w from the solar array going into charging the 48 volt battery system.

I got a great deal on 12v 115ah deep cycle marine battiers ($105 each). Strangely enough, I work for an electric company and have access to a hydraulic crimper, so thats no problem. These guys here however work in Megawatts and their controllers and inverters are for power plants and huge city wide installations. So I have little help for my little startup installation for home except academic questions

[Mike90250]

I have purchased 16 - 12 volt deep cycle batteries (105Am each) and can configure them anyway you suggest, 24 or 48 volts. It sounds like I ought to configure my 16 batteries as 4 strings of 48 volt batteries which would then be in parallel. I would like to start out small but be able to expand in the near future without wasting money and equipment. It also seems that I ought to go with a 48 volt inverter. Did I get that right?

Pretty close. Read about connecting parallel batteries at the smartguage site

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

even with 48V, 4 parallel banks will be a bit of a problem.

With all those interconnects, you should buy a hydraulic crimper in the +6 ton range,
and make cables yourself. Will save a bunch of cash.

And with 4 banks, you will need 4 fuses or breakers - one on each bank, before they combine to the inverter & charge source.