Could cheap panels alter the panel-battery ratio?

Eric L

A common view expressed on this forum, which I have no doubt comes from hard experience, is that a battery-based system should be sized such that the batteries can receive between 5 and 13% of the their amp-hour capacity (at the 20 hour rate) in charging amps at the charging voltage. This constraint, plus a derating-factor for the panels, typically determines the PV array size for a given battery capacity (all this is after expected loads have been established, of course). I’ve seen the 13% figure described here (by BB and perhaps others), as the ‘cost-effective maximum’.

My main question is this: could that cost-effective maximum vary if the panel-battery cost ratio changed significantly? Or is it independent of this ratio?

To see what I have in mind, suppose as a thought-experiment that:

- you knew the expected loads for an off-grid PV system you were building;
- the PV panels were free (like I said, it’s a thought-experiment), and you had the space to mount as many as you wanted.
- you had to pay for batteries for the life of the system.

What would you do in this case? My thinking is that I would build a system with a PV array large enough to deliver charging amps significantly above the 13% cost-effective maximum. I would, of course, have to find a way to make sure that the PV array could be throttled-back in full sun to avoid over-charging the batteries. Maybe this throttling-back can’t be done, but it seems like it could be. Assuming it could be done, I would then want as much charge as possible on cloudy, low-light (winter) days, with the idea being that I could thereby have a smaller and cheaper battery bank since even on low-light days the PV would be providing enough to charge the bank.

Is there a flaw with my thinking here? And could the charging amps of a very large array (relative to battery capacity) be safely ‘throttled back’ on full-sun days, as I’m assuming?

What’s motivating all these questions is that, with panels selling as low as .78/watt, batteries are now the most expensive item by far. And I’m wondering whether, if panel prices kept falling, a point could be reached where the panel price/battery price ratio made a significantly larger array relative to the battery system more cost-effective.

niel

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

what you propose is viable if you had as many pvs as you wished. there would still be a point of your not needing anymore as they would not be contributing. i do have to correct you that the 13% was not implemented due to finances, but it is a limit by which most batteries can safely be charged at. it is true that the batteries you have may safely charge at a higher rate, but there could be a tradeoff too with higher maintenance. always check with the battery manufacturer for their recommendations when it comes to high charge rates for there is a point where the batteries could also be damaged. high charge rates for the batteries will allow for a larger pv array to quickly charge up the batteries and allow for more use during non-daylight periods.

anyhow, you could use high power appliances during the daytime allowing more pvs to be actively utilized. adding that some of the mppt type controllers not only can limit the current, but also have the ability to program a max current output anywhere up to that controller maximum allowing you to stop too much current from entering your batteries. it would take some planning to make it work just right, but is doable.

icarus

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

One simple option, for an over sized array,,,load shifting. Assuming that your PV puts out twice what the batteries like under ideal sun, then draw all t he discretionary loads at full sun. Pump extra water, charge tool and lap top batteries, do your printing, in short use the excess power in any and every viable way to increase efficiency. One could even do a diversion load to preheat hot water, although that would not e very financially efficient IMHO.

Essentially, that is the way we live. We wait and use our big draw items when the sun is out and the batteries are near full, cahrge the portable batteries, pump extra water etc. It is however a great balancing act. It is best IMHO to design for normal loads and conditions and use it accordingly.

Tony

BB.

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

More or less, for a standard flooded cell battery that is not cycled too deeply, you can safely go upwards of 25% rate of charge--But you should make sure you have a remote battery temperature sensor on the battery bank. Otherwise you run the risk of thermal runaway (battery gets hot, charging voltage goes down, charger sees lower voltage and increases current, battery gets hotter and charging voltage goes down further, etc.).

A MPPT type charge control does a very nice job at limiting output current... So you could put a 10kW array on a 30 amp 12 volt charge controller, and it will output nearly 30 amps from sun up to sun down.

Remember, that even if panels are free, the mounting, wiring, and compatible charge controllers are not. In the olden days, you could plan on $1 a watt for mounts and maybe another $ per watt on installation. Which will not go away at this point.

-Bill

65DegN

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

Absolutely, in fact this is my approach to solar/wind systems. I advocate charging capacity to be far in excess of what the battery bank requires and using more power in real time as it is produced. Instead of spending that extra money on a large batterty bank that will only last a few years and become a recurring expense, I would rather buy more generation capacity which will likely not be a recurring expense and actually generate power to use and not just store it. That way cloudy or breezy days will still charge the batteries which make it really easy to keep them fully charged and excess power can be used to do other tasks, like heating water to heat your home.
Solar electric panels have an advantage of over solar thermal here because they can work at sub-zero temps which would allow space heating even if it's 30 below zero outside.

Eric L

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

Now that you mention it Tony, you're completely right that a diversion load could do the job here; in fact, some charge controllers are already set for this, I think.

Bill, you're also quite right to mention the associated costs of mounting and connecting panels; if these costs are high enough then even 'free' panels would not be worth mounting above a certain point. A couple comments here though: one is that these costs are highly variable, as you know: my DIY ground mounts (superstrut and concrete piers) came at .28/watt. Also, some of the wiring and controller costs in a larger array would benefit from economy of scale (since these things have to be bought anyway).

It would be interesting to find a way to calculate if and when it might make sense to begin significantly upsizing the array relative to battery size; i.e., how cheap would installed panels, inclusive of mounting, wiring, etc., have to be in order for this way of proceeding to become economically viable. Some obvious variables would included latitude (since above a certain latitude the array size needed for a cloudy, dark winter day would no longer make sense even if panels were free), the de-rating factor for cloudy weather, as well as the usual costs per watt and watt-hour (for batteries) averaged over the life of the system.

Someone here once linked to a very detailed cost analysis tool that incorporated details like latitude and even PV watts data, if I recall correctly, although I can't find it now. It would be interesting to play with in this respect.

BB.

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

It actually is not too difficult to make a price model for making comparisons...

(capital costs+repair costs over 20 years+permit fees+taxes) / (20 years worth of kWH USED)

You can pick any number from 7 years (the average # of years people live in their home) to 25+ years (the "typical" lifetime of solar panels, etc.).

One of the big "confounding" issue is kWH USED. With Grid Tied systems, you can "harvest" all of your kWH and either use it or push it to the grid for use a few months later (some utilities will pay for excess generated power, but usually in the $0.02 to $0.05 per kWH range--not very "profitable").

With off-grid systems, most people will probably be lucky to use around 50-75% of "available power". Only 1-3 days of useful storage means that summer excess is useless during winter (or summer deficit for Air Conditioning cannot be carried forward or backwards from spring/fall).

So, take the ~77% end to end system efficiency for Grid Tied systems vs the 52% system efficiency and the ~66% usage factor for off grid system--which multiplies out to ~34% of Panel rating (vs ~77% of panel rating for pure GT)--And you end up with very expensive off-grid power (low harvest/usage and higher initial capital and maintenance costs due to battery bank and extra hardware).

For either system, once you exceed the practical amount of power you can reasonably use, there is little need to expac the array (even if near "Free").

And for off-grid systems, much of the "power available for usage" depends on it being in the battery bank. A large array and small battery bank is not going to be much help for large over night loads and/or loads with large surge current requirements (pumps, fridges, A/C compressors, etc.).

In the end, the best bang for the buck will still be 1) conservation, 2) more conservation, 3) even more conservation, 4) know your loads, and 5) design your system to support your loads.

For seasonal/weekend cabins, many times it makes more sense to support the occasional/heavy demands with a genset. Just enough batteries/panels for lighting, presurization pump, radio/laptop/cell phone chargers). And use a genset for the vacuum cleaner, filling the cistern, running the microwave. And use propane for a frdige/heating.

For the genset, you can again work out a $/kWh price:

(price of generator+maintenance + gallons of fuel for 5 years) / (# of kWH used over 5 year life)

If you assume a backup genset was needed anyway, you can forget the capital costs and work out $/kWH for fuel costs... Say a Honda eu2000i running 400 watts (1/4 load) for 9 hours on 1.1 gallons of fuel:

1.1 gallons * $4 per gallon (getting there in CA) / (9 hours * 0.4 kWH load) = $1.22/kWH

Very similar to the costs for a well used off grid system (9 months or more occupancy and sized to load) which (for my sample calculations) usually works out to around $1-$2 per kWH (obviously, cheaper panels have brought the $/kWH costs down).

Note, if you only use a cabin 4.5 months of the year (vs my 9 month spring through fall assumption), the cost of solar power goes up by 2x (all of the hardware/batteries are there but not used). For a genset, if it is not used, there are no fuel costs and no "wear and tear" to speak of.

-Bill

Eric L

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

One of the big "confounding" issue is kWH USED. With Grid Tied systems, you can "harvest" all of your kWH and either use it or push it to the grid for use a few months later (some utilities will pay for excess generated power, but usually in the $0.02 to $0.05 per kWH range--not very "profitable").

With off-grid systems, most people will probably be lucky to use around 50-75% of "available power". Only 1-3 days of useful storage means that summer excess is useless during winter (or summer deficit for Air Conditioning cannot be carried forward or backwards from spring/fall).

Right Bill, although I think my question about whether the battery/panel ratio could change with prices, and how to compute it, can be seen as independent of this confounding issue. That is, assume we conserve as much as possible, and take our best estimate of KWH used (in either proposed system).

Another way of putting the question then is: why assume the "5-13% amp-hour rate in charging amps" formula is always the most economical, given that it seems to rely on inputs that are a function of price, and those inputs have changed, and may continue to do so?

If PV and battery prices continue to move in opposite directions, it seems like the point might eventually be reached where one is better off with 1 (or maybe 1.5) days battery 'capacity' (to some acceptable DOD) and enough panel to provide something close to the KWH used even on the darker days. So one would be working with a much smaller battery bank (savings) and hoping to offset the energy demand with a larger PV array (cost). It seems like this point could could be reached in principle, but that might be an illusion.

One reason I'm now thinking it might be an illusion is that the 'acceptable DOD' on the bank might be some value (like 20%, say) such that you get a portion of the extra days of power "for (nearly) free" in a normal system. Another reason is the one I see you intimating; a cheap backup gen. might negate the possible savings from a bigger array/battery ratio under even the most favorable PV cost assumptions.

icarus

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

It is in fact, not a diversion load per se, but load shifting. A diversion load will take excess WH when the batteries are near full and power a given load at that time. Load shifting is a strategic plan to use the power when it is available (sunny days) and not need to use it when there is little available.

For example, assume that your battery can take say 10 amps as a save charge, but your PV can put out 20 amp. You can charge at 10 amps in full sun, and use th other 10 for discretionary loads. That way, in the evening, your battery will be fully charged, the water will be pumped, the lap tops charged etc, reducing the load on the batteries making it easier to fully charge the next day, because you don't have to take as much out.

On the less sunny days, you PV might only put out the 10 amps to charge, so that day, you don't charge the lap tops or the too batteries etc, and the batteries might get fully charged as well. Load shifting is simply a way of in the net, reducing the need to draw on the batteries, and ergo the need to recharge on a daily basis, and ergo reducing the net size of the battery bank.

In the real world, my after dark loading is very small, a few hours of 3-5 CFLs, the fans, the radio and maybe the Internet connection. So while my tri metric shows an averge daily draw of ~500 WH, I might actually use 1 kw,, with only the 500 having to come out of the battery. Make sense?

Tony

niel

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

looking at it from the prospective of when the excess pv power is there along with the excessive loads to match, it works. make that imbalanced and there could be a problem. load diversion can work in some cases, but in general if there is too much pv power for the loads at the time you would need to program the controller(s) to cut back on the output current to prevent overcharging the batteries. if solar becomes variable due to clouds this could get old quickly.

same goes for the excessive loads being a problem if the sun dips behind some clouds for instance. can you run to shut off the loads as it goes in and out? this has to work more automatically and the best way is to have the batteries being able to take high currents to begin with. that leaves the concorde sunxtender batteries as the best option, but they still have a limit to them and thus the number of free pvs and the amount of loads will also have a limit. the sun disappearing will still be problematic with all of the excessive loads on, but at least the sunxtender can handle excesses to a point until the battery gets depleted or the voltage drops too low to properly operate said inverter(s) or dc loads.

as i said a great deal of planning would need to be implemented with this, but it is doable.

westbranch

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

I would like to add a slightly different scenario:

the lower PV price will allow an array to be less costly and thus built to ~ the 13% target and thus (hopefully) ensure that on the winter solstice a full charge is attainable, sun permitting. The solstice being the least 'Solar ' day of the year. The rest of the time there should be no need to sweat the small stuff, just wait for float and give 'er... lights a blazing, etc.

niel

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

westbranch wrote: »

I would like to add a slightly different scenario:

the lower PV price will allow an array to be less costly and thus built to ~ the 13% target and thus (hopefully) ensure that on the winter solstice a full charge is attainable, sun permitting. The solstice being the least 'Solar ' day of the year. The rest of the time there should be no need to sweat the small stuff, just wait for float and give 'er... lights a blazing, etc.

westbranch,
it isn't quite that simple for having it at 13% at winter solstice could pump many times more than most batteries can take in the bulk stage during the warm months. this could be held back with current limits on some controllers limiting the charge to the batteries to that 13%, but then you won't take advantage of that excess pv power as the controller won't deliver it even if the extra loads are presented to take up that extra power. that means putting the controller higher up to allow that excess power through. remember that this could involve many controllers as you are feeding from extra pv arrays that would need their own controllers when power levels are too high.

in the case of using a sunxtender battery, they are rated for several c as a max charge rate. you then don't have to worry about limiting the high pv charge so much when the sun would go in and out from clouds with a high charge rate allowed. this also allows more flexibility on the extra loads going off causing a spike to the battery charge rate too. of course i am looking at this from a bit of an extreme as one could apply it to batteries that would say be limited to maybe 20%. add 7% more charge and 7% more load straight up without inhibiting the controller output. this would present more power to the batteries when conditions are worse as there would be more pvs collectively outputting a smaller amount of power for the conditions at the time. this 20% example would not normally present itself as able to supply 13% at winter solstice for it would probably be far lower unless boosted say with snow or water reflections.

65DegN

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

BB, I installed a 48V test system using Trojan T105 batteries. The owner typically runs his generator less that 10 hours per summer and I ran into him two days ago, he said he has 2 hours on his gas generator this winter. He doesn't have huge loads but runs a wood working shop and has a standard size off the shelf fridge which has no problem at all starting.
He lives in a recent burn area and has very good wind that blows relentlessy, almost no trees and unobstructed solar access. This in effect increases the useful size of his generation capacity. He has 630 watts of solar (three REC210's that produce ~ 1,000KWH annually) and 1KW of wind.
According to Trojan the T-105 is rated at 75A for 115 minutes. Thats 3600 watts @ 48V so I'm sure it would surge well over that to start loads and is a nice fit for the VFX3648 he is running.
And when it comes time to replace his battery bank it will be ~$1,400 instead of $3,000. Then if he wants to he can take that extra $1,600 and buy more solar panels.
Since the batteries are rarely cycled very deeply, due to the near constant charging, they will likely last longer than what is typical.

I have 14 Arco 16-2000 solar panels that are about 25 years old and on a FX80 they put out at least 450 watts, which is their rated power.
"...an opportunity to measure performance of modules (AS16-2000)that had been in operation for 25 years showed up...Nineteen of the twenty modules obtained average peak power values which were less than 2.0 % lower than data obtained in the beginning of the modules’ operation..."
http://energimyndigheten.se/Global/Forskning/Kraft/Performance%20of%20old%20PV%20modules.pdf

I suspect solar modules will be functional for 50 to 75 years. Thats a good investment.

65DegN

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

niel, the pv controllers all have (static) current limiting. So there is no need to worry about hitting them with too much current unless there are other additional charging sources like wind.
I have been talking to wind turbine and charge controller manufacturers trying to get them to work on dynamic current limiting. So that the controller looks at the total current entering the batteries and adjusts accordingly. The industry is still in it's infancy so I am confident that at some point it will come, it is inevitable.

niel

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

i guess i'm having trouble trying to get my points across here. the whole point was to send extra power with many extra pvs into a smaller battery bank and with the option of having large extra daytime loads on them to eat up the extra power that technically was to be too much for the batteries. if you limit the current via the controller then it won't be available to the extra loads. don't limit the current via the controller and if something happens to the loads the batteries would get smacked with high currents they weren't meant to handle. could you be there when that happens all of the time or foresee that fuse popping in advance dumping that excess current to the batteries? i think not. remember the system we are talking about here has more pvs and controllers than the battery bank would normally be able to take. by then it's too late so i don't recommend pv systems larger than the battery banks can take as any imbalance could prove detrimental one way or another.

for instance take a 200ah battery bank and feed it 200a from pvs. the extra controllers will be set to allow all of that power through so you can power the extra loads you have awaiting to bring that charge current down to a safe level for the battery bank. if that safe level is 30% then that's 60a the bank can safely take and you must dissipate the extra 140a into those loads. the only way to stop the extra current flow when the loads get disconnected is to manually set the charge output of the controller for those that can have a settable current limit.

to feed say 100a into a controller that may output 60a is not very wise as there is waste when solar intensities are higher and that extra power won't be available to extra loads being it is held back by the limiting of the controller. the solution is to have a battery that can be smacked with the high current without damage like the sunxtender for they can be given 200a into a 200ah battery.

now it can be done on a smaller scale with other batteries if designed right, but the extra loads and the extra pvs will need to be appropriately sized so as to not allow the batteries in question to ever receive more than they are capable of receiving. the limit of the controller does not play a part here because that must be open to allow that extra current to go to the extra loads unimpeded. the 200ah battery and 60a charge at 30% rate is a good example as you could use a max of 60a in pvs after losses to the batteries. loads then shutting down for whatever reason will not kill the batteries if the system is held to stay under the maximum of the batteries by design rather than regulation. if that system has 40a of load on at the time it will still send 20a to the batteries at a 10% rate. better yet a 50a load will still see 10a going to the batteries for a 5% charge rate. once any of the systems get into a float charge or the lower points in absorb (which it will do in a fast rate of time being a high charge rate) this frees up even more power to be used by loads and this one holds true for any system that has charged batteries and is still receiving pv power.

simply, part of the power is being used by the loads to reduce the impact to the batteries, but if the current is too high to be handled by the batteries at any time then i don't recommend doing it as stuff happens. some generic batteries may go in the 15%-20% range for a short time, but if you aren't there it could mean exposing plates or even warping them.

vtmaps

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

I am glad to see this discussion here. I hope it will get the industry to come up with a solution. My situation is slightly different, but the problem is the same. My PV array (940 watt) and Outback fm60 charge four L16 batteries (in series). My inverter charger is an Outback vfx3524, which will allow for future expansion. My system is well designed for my domestic needs, but I did not design it to handle my workshop. I use my generator for large power tools, and when the generator is operating the vfx3524 is set to limit the AC for charging to 9 amps (about 40 amps DC to the batteries).
Sometimes I need to start the generator on a sunny morning during bulk charging mode. Unfortunately, the two charging sources do not talk to each other during bulk charging phase, and this can lead to 70 amps DC into the batteries. That is more than is desirable. I don't know of any solution other than manually turning off one of the chargers.
I am surprised that the industry has not come up with a solution. This thread is evidence that the problem will become more acute as PV arrays become cheaper. I guess it just goes to show that the solar industry is really just in its infancy. My message to the industry: Grow up already!
--vtMaps

edit: Further down this thread stephendv points out that Sunny Island inverters have the capability to dynamically limit their current during bulk charging.

niel

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

vtmaps,
i think the pvs are most of the way grown up, but are moderating in their pricing. the electronics in the industry is still growing, but the big sticking point right now is batteries. little progress is going on here and prices are going through the roof. i do believe you understand my points quite well with the overcharge potential to present day batteries. the best of the lead acid class is the agm in taking high currents and that varies considerably with the sunxtender line being rated to take the most abuse in that area to my knowledge. most i believe would want some more battery capacity rather than feed a 200ah battery 200a, or at least i would, but it is good to know that i can smack my batteries with a high charge should i ever need to. i doubt i'll ever have the ability to approach the limit on my batteries max charge rate.

westbranch

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

I remembered reading this a while back, from a 2001 document by Christopher John, of GNB, about a 2.5 Mw Absolyte battery bank at Vernon Battery...

Float Life Calculation
based on a 20-year Life = Hours of Design Life (175,320 hrs)
* Normal Float Current (2.16 A) = 378,691 Design Ah of Overcharge
Actual Overcharge (136,520 Ah) ÷ Design Life (378,691 Ah)
= 36% of Total Life Consumed
= 16.6 Years of Total Projected Life

Actual Years of Operation (6 years) ÷ % of Total Life Consumed (36%)

Cycle Life Calculation based on 1200 cycles at 80% Depth of Discharge
n Cycles (1200) * Depth of Disch (0.80) * C/6 Capacity (4500 Ah)
= Design Ah Throughput (4.32 MAh)

n Actual Throughput (769,823 Ah) ÷ Design Throughput (4.32 MAh) = 18% of Total Life Consumed
n Actual Years of Operation (6 years) ÷ % of Total Life Consumed (18%)
= 33.7 Years Total Projected Life

So picking the right float Amperage can extend VRLA life span... ?? Looks like it

ADD:From the Absolyte manual

Temperature corrected 77°F (25°C) Percent
Float voltage per cell Reduction
Minimum Maximum in Battery Life
2.23 2.27 0%
2.28 2.32 50%
2.33 2.37 75%

65DegN

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

neil, a lot of assumptions here that are not necessarily the only way to approach this but it's great that you bring these things up. I am talking about something that is fundamentally different from the usual system configuration, consequently normal system set up will likely not work. Things need to be done differently in order to progress from where we are now to where we need to be, IMO.
Until some equipment manufacturer puts a dynamic durrent limiting controller on the market or some electronics design engineer puts an open source schematic on the net some other method for limiting the current will be needed to simulate dynamic current regulation. For instance current activated relays that progressively reroute entire arrays to do other work in real time when max current to the batteries is achieved. It is a static adaptation to dynamic current limiting. This would prevent blown fuses and the need for human intervention with changing conditions.
Then on a cloudy day in the dead of winter all the array's can be putting a full charge on the batteries without the need for a generator. Over time the additional arrays will pay for themselves in fuel savings and generator repairs or replacement plus a whole lot of work will be able to be done with the massive amount of energy produced on sunny days. I could heat my cabin and save on firewood, heat my greenhouse, any number of things.
If we can eliminate or substantially reduce the need for fuel to run a generator and space heating, while decreasing the cost of a replacement battery bank by ~60% it would take us that much closer to being energy self sufficient.

Rick, WL7BDK

niel

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

are you talking an overcurrent connect? opposite an lvd as this would be a oic. it would detect current over a programmable set point and activate another load either via a standard or solid state relay.
are we on the same page? i'm sure this could be done if you know enough of electronics. i'm guessing a standard shunt would set up a voltage differential and the higher reading would be from higher current flows, but still too low to be useful. with maybe an op amp to amplify the difference and a v comparator this could trigger a transistor to activate said relay.

thoughts?

vtmaps

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

niel wrote: »

i'm sure this could be done if you know enough of electronics. i'm guessing a standard shunt would set up a voltage differential and the higher reading would be from higher current flows, but still too low to be useful. with maybe an op amp to amplify the difference and a v comparator this could trigger a transistor to activate said relay.

You might be interested in a thread over at the outback forum. This link will start you at the twelfth post in that thread:
http://www.outbackpower.com/forum/viewtopic.php?f=17&t=5906&start=11
forum members EW Zuber and tallgirl had some interesting things to say about this subject.
--vtMaps

niel

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

i only got about halfway through it and it strongly parallels this thread. i kind of hit the nail on the head didn't i?

stephendv

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

Interesting subject! In my case, I wouldn't mind throwing away extra peak production from the panels, say limiting them to 60A when they could produce 80A, because the worst solar days are overcast winter days. Currently with the right type of high clouds the panels can produce 300W from 2.8kW installed. Which is really quite impressive - what would be even better is that when the dense fog rolls in and the panels only produce 50W or so, that could be boosted to 200W. This would allow them to more or less keep up with the loads during the day and help get the batteries through to the next sunny day.

I've heard some recommendations to use amorphous panels or micro-amorph for better low light or diffuse light applications - might be a good mix (on their own charge controller) to throw in with my existing poli's ?

Regarding current limiting for multiple charge sources, the sunny island inverter solves this for both DC and AC coupled systems. For DC coupled systems using any charge controller - you have to install a shunt which is then read by the sunny island, so it always knows how much current is going to the battery from the external charger. It will then limit how much it sends to the battery if you start the generator at the same time.
I guess, any AC coupled system would also solve this, since there will just be 1 central charger (or stacked set of chargers).

65DegN

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

vtmaps, thats me (EW Zuber) over at the Outback forum under my real name.

65DegN

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

neil, yes thats the basic idea. I am sure that there are a multitude of makeshift ways to do this and I suspect some rather crude approach is the only way it will get done until someone with the talent needed to pull it off with sophisticated electronics tackles the job. Ideally I am thinking of some kind of master dynamic current controller that will allow a preprogrammed level of current to pass to the batteries. All extra current will be sent to do work elsewhere with a number of output ports assigned to different loads and all programmable. A thought crossed my mind about using the voltage from a shunt into tricking a standard MPPT controller (one in a group of controllers) into thinking that the battery was over absorb voltage and cause it to connect power to work loads as one way to get crude current limiting. This way the systems that are already in place in the Outback or Midnite controllers to do this could be used.
Have a few general thoughts on how to do that but nothing I am satisfied with yet. Switching (for instance a small supercap in series)between the battery and the controller is out of the question since it would cause an unloaded controller for a fraction of a second.
Just thinking out loud here.

65DegN

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

stephendv wrote: »

Regarding current limiting for multiple charge sources, the sunny island inverter solves this for both DC and AC coupled systems. For DC coupled systems using any charge controller - you have to install a shunt which is then read by the sunny island, so it always knows how much current is going to the battery from the external charger. It will then limit how much it sends to the battery if you start the generator at the same time.
I guess, any AC coupled system would also solve this, since there will just be 1 central charger (or stacked set of chargers).

Thats very interesting. Will the sunny island also divert extra power?

On AC coupled systems, the problem with those as I understand it is that with most AC coupled inverters they have no idea what the battery state of charge is.

65DegN

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

stephendv, it's starting to come back to me now. I had a long discussion with a technical sales guy about a year ago about the sunny boy and sunny island combo. At the time I didn't see the full potential this system offers. One could use a sunny boy solar grid tie inverter to install a large solar array on the microgrid that has been created by a sunny island. The sunny island would use power off of the micro grid to keep the batteries charged.
As I recall the sunny island talks with the sunny boy via the AC frequency. The sunny Island will slightly shift the AC frequency to let the sunny boy know how much power to put onto the micro grid.
Digging into the sunny boy manual now and I see that it has load shedding. Need to read more. This might do what I want.

westbranch

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

65N, am I understanding this correctly: what that Sunny system is , is in reality, 2 controllers in series?

jagec

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

westbranch wrote: »

65N, am I understanding this correctly: what that Sunny system is , is in reality, 2 controllers in series?

Yes it is. The Sunny Boy is a standard grid-tied inverter, the Sunny Island is a fairly clever DC inverter/charger that monitors both the AC and DC (battery) sides to ensure that the AC microgrid remains stable, and the batteries will either be charged or discharged as needed to keep the microgrid up and absorb excess generation or loads. If the batteries are fully charged and the Sunny Boy is still pumping more power into the microgrid, the Island will actually shift the frequency slightly to cause the SB to throttle its output. It works with wind, too, I believe. Pity it costs so darn much...

It looks like this:
Panels->Sunny Boy-->(microgrid, AC loads etc)<--Sunny Island<---Batteries

BB.

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

Just to clarify (as I understand) the Sunny Boy is "almost" a standard Grid Tied inverter... However, it has a mode (locally programmable setup?) that allows it to run in a system where it varies its output based on the "local" frequency.

I don't know the real numbers, but probably something along the line of 59 Hz (or 49 Hz for Europe), inverter outputs 100% of available solar array power and ~50% of rated power at 60 (50) Hz. At 61 Hz (or 51 Hz), the inverter outputs near 0% of the available array power. Proportional output based on line frequency.

This is different than what the Xantrax/Schneider hardware currently does (again, as I understand, much of this is probably software programmable and may be covered by patents too)--GT inverter outputs 100 % of power at 60 Hz, and outputs 0% of power if >61 Hz or <59 Hz (on or off).

The proportional output really seems neat and allows a entire local grid to support power sharing/throttling without requiring an external signal path.

Neither setup is ideal if you rely on 50/60Hz for clocks and accurate timers than run off of line power (but very few off grid inverters have an "accurate" timebase anyway--So that "knock" is more academic at this point than functional).

-Bill

stephendv

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

BB. wrote: »

Neither setup is ideal if you rely on 50/60Hz for clocks and accurate timers than run off of line power (but very few off grid inverters have an "accurate" timebase anyway--So that "knock" is more academic at this point than functional).

Would you believe that the SMA engineers thought of that too The sunny island keeps track of how long it runs over frequency (e.g. 61Hz). Then when it's no longer necessary to throttle, instead of just switching back to 60Hz, it switches back to 59Hz for the same amount of time it spent at 61Hz. Still the clocks could be out of sync a few hours a day, but probably not more than that.