Could cheap panels alter the panel-battery ratio?

65DegN

Re: Could cheap panels alter the panel-battery ratio?

westbranch, The sunny island is basically a stand alone inverter that charges batteries intelligently, unlike most (or all?) other AC coupled inverter systems. It can do this because it monitors battery current through a shunt and adjusts charge current accordingly. It is also usable as a grid tie inverter or a grid tie inverter with battery back up. Very versitile.
The sunny boy is a MPPT solar grid tie inverter. Used together the sunny boy applies solar power to the microgrid created by the sunny island.
Big setback is the cost. I priced a sunny boy/sunny island system last year and (from memory) it was ~$10K. Also I don't see any option for using excess power. It has two relays that are suggested to be programmed as gen start and load shedding designed to save batteries.
Also the PV array size can not exceed the rated AC input current for the sunny island. Adding a KW of sunny island for every KW of solar would add a lot to the cost of PV's.

westbranch

Re: Could cheap panels alter the panel-battery ratio?

For a while I had thoughts that a Midnite Clipper, see diagram on page 1 http://www.midnitesolar.com/pdfs/ClassicClipper_operation.pdf , might be able to fill the gap but it is Voltage controlled too...

westbranch

Re: Could cheap panels alter the panel-battery ratio?

Here is a bit of info http://photovoltaics.sandia.gov/docs/PDF/batpapsteve.pdf
that should have an impact on the worries expressed in this thread re "overcharging"
particularly the closing statement,

Charge efficiencies at 90% SOC and greater were measured at less than 50% for the battery tested here,
requiring a PV array that supplies more than twice the energy that the load consumes for a full recovery charge.

stephendv

Re: Could cheap panels alter the panel-battery ratio?

westbranch wrote: »

Charge efficiencies at 90% SOC and greater were measured at less than 50% for the battery tested here,
requiring a PV array that supplies more than twice the energy that the load consumes for a full recovery charge.

That's not too bad, since it only applies to the last 10% of charge - and remember that they're talking about energy, not power. From 90% onwards the battery would already be in Absorb, so the power from the panels would already be throttled back.

Eric L

Re: Could cheap panels alter the panel-battery ratio?

Here’s a follow-up for those who might be interested. I’ve tried a version of what I suggested in the first post above at home over the last few months. Before anyone reads all this, I should say that I don’t have anything profound to report – many of the things I’ve encountered were mentioned by posters above –- but I offer a few observations and attempts at cost-comparisons below.

The set up is not as ambitious as what I suggested in the opening post, but I do have enough panel (4.1 KW STC) to deliver an 80+ amp charge to my 48 volt bank. The batteries are 360 ah, so this works out to over a C/5 charge rate, although I have the controllers set up so that after loads, they see a max of about 55 amps.

The system is: 8 Evergreen ES-A 215s and 12 Evergreen ES-E 200s for 4120 watts nominal. Two Classic 200 CCs, Magnum MS-4448 PAE inverter and Magnum mini-panel. 8 360 AH U. S. Battery L-16 batteries. It is located about 10 miles in from the Gulf Coast near the AL/MS border. One advantage to my latitude and location is that according to PV watts, and also my past experience with solar, we have pretty steady monthly irradiance year-round.

I wound up with two panel types (and thus needed two controllers) because I ordered the panels at two different times and couldn’t get the same type on the second order. That delay added about $770 to the system cost for the second controller, and extra panel shipping costs (two pallets).

My system is not a true off-grid system, but is grid-supported. The system switches to grid power if the batteries fall below about 75% SOC. Still, I think most of what I’ve found could be reasonably extrapolated to a true off-grid set up. For instance, if you imagine replacing the grid in my case with a generator plus AGS system it would give roughly similar results. Note though that in my system all charging is done by PV; the charger function of my inverter is off. Had I had a generator running at 50% instead of being on-grid, there likely would have been some extra charging time for the batteries available.

In operation, this system has been operating nearly all of our 120v olt circuits nicely on a daily basis. After attempting to factor-out opportunity loads, it has been producing an average of about 7.6KWh/day for the last two months. Of course most days it could produce more if loads were greater or the batteries discharged further. . That average includes very cloudy days; the lowest output was 1.7 KHh on a very overcast rainy day. Next-lowest was 3.9 KWh. (Opportunity loads include a well pump and washing machine, and those have raised the actual daily average to over 8 KWh.) With no opportunity loads, the system needs about 8.1 KWh/day to complete or mostly complete a 3 hour absorption cycle with loads attached. It has delivered at least this much for 50 of the last 59 days, or 85% of the time. So most days, there is enough pv power for loads and to complete a full charge. Of the 9 days it didn’t complete, it went to grid power 7 times, usually for a few hours in the early morning of the following day. For a three-day very overcast period that included the 1.7 KWh production day, our house was probably on grid about half the time.

I originally ran this as a 2.4 KW system for two months before adding the extra 8 ES-A panels. In this configuration, it was averaging 5.8 KWh/day of production, but that was earlier in the year (Jan-March), and included some overcast stretches. So it’s not a perfect comparison, unfortunately.


Anyway, some observations:


- The inverter does not track time on grid, but I would estimate that my system has been on grid at least 60 hours over the last 59 days. Weather was mixed for this time of year, and included that cloudy, low-production spell.
- This system seems ‘low maintenance’ because I know it’s regularly completing a full charge regardless of usage patterns or weather, which is nice.
- Whether having a high panel:battery ratio makes sense may depend a lot on the type of cloudy days the system will see. We get a lot of overcast days here that are nonetheless fairly bright; shadows are present but soft. I tried to get some sky in the picture below so that you can see what I have in mind. Sometimes this is fog, and sometimes high clouds that break into partly or mostly cloudy skies.



- I have found that production under these conditions is pretty good. Here you can see 1.5 KW of output under these conditions; about 35% of STC rating. This picture was taken right after the one above. The system completes a full absorption charge before the end of the day under these conditions. On the other hand, dark overcast days are pretty hopeless, even with the extra panels. So if you get a lot of days like that it would probably make more sense to get a bigger bank to maximize harvest when it's sunny.




Cost:
As we all know, every installation is unique. But for comparison purposes, this system has cost me $2.79/watt; that includes shipping, wiring, batteries, and everything else I can think of except for tools and my labor. Remember though that this price includes a small battery bank given the panel size. The extra 8 panels (1.72KW) cost $2341 inclusive of everything (freight, wires, a second charge controller, mounting, fusing), or $1.36/watt.

Would that $2341 have been better spent on a mix of panels and more battery? It’s a bit hard to say since in my case adding a larger battery bank would have considerably increased the per-amp-hour battery cost (the type I bought were the most cost-effective by far), unless I exactly doubled the bank, in which case I would not have had sufficient panel without spending more. Also, as folks noted above, the batteries are a more frequent recurring expense relative to solar panels; so increasing the battery storage would have increased the net cost over time.

Perhaps these cost comparisons help. As I said above, better timing/planning on my part would have saved a few hundred by eliminating the need for a second CC. Overall it's working out nicely for us though.

offgrid me

Re: Could cheap panels alter the panel-battery ratio?

Thanks for the update Eric
I have a similar set up. 4.2k of panels and a 450ah battery. It has worked out very well for us with only 12hrs of generator run time during the long dark winter here in Maine. I really love that i can get a full charge into the batteries almost every day even during the almost two weeks of rain we just had. I have come to believe that one should buy as much panel as they can afford ever if it means going with a smaller battery bank. I rarely need to worry about how many days of backup my battery can supply because I get some charge every day.