Wasteful system sizing approach...

kaipo_boy · March 2016

OK, not saying I am going to do this... however, its an interesting solution to a problem I've noticed.

Bjammin in Virginia just posted (electron harvester needed) about moving into a 30ft yurt and his system needs; toward the tail end of that discussion, Dave Angelini mentioned that his system made float nearly everyday except maybe around 20 days a year... which really got me worried, as my system in winter rarely sees float. I disconnect my loads so that it can make float maybe once a week. During the summer it makes float maybe a couple times a week without help. I have grid backup, so this is relatively easy. However, it occurs to me that with the right kind of battery maybe this can be finagled a bit. I have a NiFe battery of 500ah on order and probably several months away, and NiFe is supposed to be charged around C5, so for my bank around 100A max... with my 2600w of panel I intend to run it on and a Midnite 150 cc, I think it'll max out around 84A. But then the evil side of my mind said, "dude, PV is cheap... and since NiFe is known to be fine with 30 minutes of 3x the normal charge rate, why don't you get a couple MORE charge controllers on a separate, disconnectable circuit, plus 5kw more PV panels, and run 84x3 or around 250A, which is still under the known capable overcharge rate, for 30 minutes or until electrolyte temp says you have to back off?"

Notably, this is just talk talk as its a stupendous waste of panels... normally you'd just buy batteries to connect to those other panels. In fact, that was going to be the next phase of my plan if the tax incentives hold out and if the new system showed it was a viable step to move off grid. But on a smaller scale, it occurs that it might have some merit at some point. For instance, my first system that I'm running now uses 4xGC2 Costco batteries, and I'm trying to cycle them only to 70%SOC, but have seen 50% sometimes. I think they will only last 3 years, maybe 2 as Vtmaps indicated I'm working them quite hard. I figured when they die, i'll move the current output from the Kid to my new system for around 84A + 27A = 111A into the NiFe bank, which should be fine since I don't see max production numbers like that very often. My area is apparently not like Dave Angelini's. I don't get consistently good days, there are some sunny days, mostly in summer, and helluva lot of scattered clouds that come and go. So, the alternative is to build a system massively over batteried and panelled to keep up with a tiny (relative) load... which is apparently what Dave Angelini has; or a system like the one I've posited above, capable of really heavy overcharges for short durations while carefully monitoring electrolyte temp so you know when to back off. I have talked to the weather gods and so far they haven't been able to guarantee me squat But here's the thing; batteries are the most expensive. Cc and PV, relatively cheap. So rather than massively overbattery to keep them close to float... hey, the above, and overstock on the cheaper components.

Or, rather than the cost of the above 'extra' panels and cc, the money would be much more wisely invested in a generator or course... but again, since those extra items might be eventually moved into another discrete system, they aren't actually being wasted...

Let the tirade begin!

mike95490 · March 2016

NiFe does not need to get to float, and only needs to be in absorb long enough to get through the night to the next time to fire up the genset,. It does not sulfate like lead acid does. If I had lead acid bank, I'd be burning 3x as much fuel to keep the batteries up over 80% most days.

zoneblue · March 2016

You dont need to go wasteful to create a balanced system. If your solar can charge a 0.1C, then generally youll make float every day, and thats alwasy teh best and most simple solution. Dave has it right.

oil pan 4 · March 2016

LiFePO4 is the same way.

They don't even notice they are being charged at 3C. Most are rated to be charged at 5C. They don't need to be floated or equalized, once they are full they are full and that's it.

But they are so expensive.

Dave Angelini · March 2016

Completely different system design is how I would describe my 88 clients and myself. We are large homes that run heat-pump cooling and heating, pump from deep wells and often have 4 + people. We often live at 4 to 8kwh when conserving and run well over 40khw when the energy is there. The batteries last 10 years and cost is not as big of a deal as it may be in smaller applications.

kaipo_boy · March 2016

My thinking when posting this, was to address what I perceive as a (necessary) bias in design of charge controllers and how they deal with batteries. If you get a charge controller and PV that can run your batteries up to float everyday in good weather, you have to plan on a maximum amperage output from the cc that won't burn it out prematurely. Although the better ones are protected, they seem to do this by shunting off excess current into dump load resistors internally (to produce heat as a waste byproduct), so the amount of excess current they can handle on a daily basis becomes important and you probably don't want to run them close to their design limits all the time.

But naturally, normal weather in my neck of the woods isn't blue sky everyday; its a mix and I don't see clear blue sky all day but maybe a handful of days a year; the vast majority is scattered clouds or clear at times with clouds moving in in the afternoons. So, in order to accommodate this changing weather, you have to plan on the MOST current your system can generate to set its upper handling limit.... which in my case won't be seen often but must be used as a design constraint, in order to avoid overdriving the abilities of the cc. The trouble then becomes most of the time in semi cloudy weather, you are producing far less than you COULD if you had overpanelled. So I guess what I would want is a cc that could more easily handle a deliberate overpanelling... something on the order of half again as many panels, so that on cloudy or semi cloudy days I am inputting a very hefty proportion of my allowable amps into the batteries, even on cloudy days. From my current 'test' system of 3x285w panels and 4xGC2 6v batteries in series, Midnite Kid, I see a maximum of 27A into the batteries in high sun. But, seat of the pants monitoring indicates I'm probably more near an average of around 10 or 12A during most overcast days. So, instead of the 3 panels, it would be nice to find a cc that would handle 6 panels, say, and on cloudy days still read in 20 to 25A into the batteries, but on clear days max out at the 30A the Kid is rated for. I think my Kid can take another panel, and I was going to experiment with adding another panel (2 in each string, 2 strings) to slightly overpanel it as it would still limit current to 30A and there wouldn't be much excess most of the time. In its present configuration the cc never gets uncomfortably hot.... it does get warm to the touch, but never to the point where I can't hold it in my hand, so what's that... around 150F? I've never measured. But, I opted instead to upgrade to a larger system instead... components for that system are on order. But I wanted to know if others here consistently overpanel their systems and where the line is for a stupid amount of panels? there has to be a thermal limit to how much the cc can take but I have no idea where it is or how often it can be pushed. Since the new battery bank will be NiFe, I no longer have float concerns as Mike pointed out, and the batteries can take a short duration overcharge with no problem.

thanks,

walt

Dave Angelini · March 2016

I am glad that you did not want to go down the road of "some people are wasting solar panels and they should give them to me"
Politics are really not wanted here and if they were, what would we do about the guy from Pakistan using 600kwh per day?

You have a choice, either buy another controller and get the solar that you need for average days, or, over panel the one you have and possible shorten it's life. It will current limit and protect itself. In that price range I would do that.

mike95490 · March 2016

MPPT controllers don't use a shunt or load to limit charging current, they shift the MPPT voltage to a point where the PV panel is inefficient and the power is never created.

Dave Angelini · March 2016

Actually some do also crowbar their output in combination with a shift of input voltage to protect their electronics. Cheap insurance!

Dave Angelini · March 2016

kaipo_boy · March 2016

65%!! ouch!

BUT I won't be worried after a couple years....

Wasteful system sizing approach...

Comments

Categories