Battery charging at well below C/8 rate

[PNjunction]

If you charge at less than C/12 for FLA batteries, you end up stratifying the electrolyte in a stationary battery. Kind of like never shaking a bottle of oil and vinegar dressing. Your salad will eventually sulfate.

Although you plan to hit the bank up with the genny when it reaches 75% SOC, there may not be enough charge time for it to fully destratify what you've done with the low current charging in the previous days. A one-off situation might not be so bad, but for day in day out, I think the effects will be cumulative. What you save in proper panel wattage is cancelled out by having to buy batteries a bit sooner.

Also consider that low-current charging means that the charge never goes deep into the plates, but is mostly is a surface charge, and you may not be able to sustain a large load, and when you try, the surface charge depletes and your voltage crashes fast.

[Sunking]

OK it you charge less than C/12, you risk stratification as already explained. If you charge greater than C/8 will most likely be at or above the gassing voltage and electrolyze your battery water exposing the battery plates which causes corrosion and permanent damage. The generator/charger combo should be selected to charge the batteries at C/8 to minimize generator fuel burn.

Moral to the story is the panels, charger, and batteries must be sized to work with each other. Not enough power and you harm the batteries, To much power and you harm the batteries.

So again if you use 8 Kwh per day the batteries need to be a minimum of 48 volts @ 820 AH. That means minimum panel wattage to generate C/12 is 3400 watts. Maximum at C/8 = 4900 watts. In your location to generate a minimum 8 Kwh per in December = 3700 watts. A 3700 watt panel array will generate up to 77 amps of charge current. 77 amps on a 820 AH battery is C/11 which is just about perfect and close to a perfect C/10. Go below C/12 or above C/8 and you are asking for trouble, very expensive trouble.

[kevinc_63366]

SunKing -

I understand your math, my headache comes in when I think about the fact that I'll be drawing off those batteries at the same time for my daily need.

My train of thought is as follows:

1) I need 3700 watts of panels to meet my daily need (using the batteries as storage bank to level out my load over 24 hours).
2) I would need an additional 3700 watts of power to charge at the C/11 rate.
3) Presuming a 25% discharge, I'd need that power for 11*25%=2.75 hours. Maybe less, since full power would only be applied during the Bulk charge phase, after which less power/time is applied for a longer time, right?
4) 2.75 hours is less than the 3.2 to 3.4 (depending on tilt) "full power equivalents" hours available in my area in December, so in theory a recharge could occur within a day at C/11 levels.
5) Even if full power is not available all of the "solar day", as long as it is for some decent period (say an hour or two), the stratification problem should be minimized.

QED: I should size the array at 3700+3700 watts, presuming I wanted to charge at C/11.

Now, just for fun, lets change the scenario and presume I only work 5 days a week (fiction on a farm, but perhaps not so for the workshop where by not working in it I could shed 60% of my daily demand. That would make 60% of 3700 watts, 2220 watts, available for charging, lets say weekly. Under this scenario, my train of thought goes like this:

1) I need 3700 watts of panels to meet my daily need, 5 days a week.
2) I could use an additional 1480 watts of power to charge at C/27.5 rate.
3) Presuming at 25% discharge, I'd need that power for 27.5*25%=6.9 hours
4) It will take about 2 days to recharge after a power outage, with stratification
5) On the weekends, an equalization charge could be applied to address the stratification issue

QED: An array of 3700+1480 watts would be "right sized".

Are those lines of thought reasonable?

[Naptown]

Use AGM batteries you can slam them with current without a problem.
Another advantage to this is they don't stratify, higher charging currents will allow you to quickly recharge during the day and any excess production will power the shop.

[kevinc_63366]

But AGMs come at a higher cost and have shorter lifetimes right?

I'm really NOT into instant gratification, prefer my pleasures over decades (reflected in us raising cows, not chickens!)

[kevinc_63366]

Gents?

Rather than oversizing the array around stratification concerns, how about using the virtually required generator for a weekly equalization charge? Most recommend testing a generator monthly anyhow, how about doing so weekly for equalization (which implies the batteries get a full charge from the generators if they need it first)? No crisis if you miss a week once in awhile for whatever reasons (like the generator fails...).

[Sunking]

QED: I should size the array at 3700+3700 watts, presuming I wanted to charge at C/11.

Huh? You got things backwards.

Now, just for fun, lets change the scenario and presume I only work 5 days a week (fiction on a farm, but perhaps not so for the workshop where by not working in it I could shed 60% of my daily demand. That would make 60% of 3700 watts, 2220 watts, available for charging, lets say weekly. Under this scenario, my train of thought goes like this:

1) I need 3700 watts of panels to meet my daily need, 5 days a week.
2) I could use an additional 1480 watts of power to charge at C/27.5 rate.
3) Presuming at 25% discharge, I'd need that power for 27.5*25%=6.9 hours
4) It will take about 2 days to recharge after a power outage, with stratification
5) On the weekends, an equalization charge could be applied to address the stratification issue

QED: An array of 3700+1480 watts would be "right sized".

Are those lines of thought reasonable?

Kevin you math is make believe and backwards. This is why you are taking a pounding.

As an example lets say you have a 48 volt 800 AH battery.

For a C/12 charge rate (66 amps) rate requires 3200 watts
For a C/10 (80 amps) requires 3840 watts.
For a C/8 rate (100 amps) requires 4000 watts.

If you did 3700 is approx the C/10 rate. If you did as you suggest 3700 + 3700 (7400 watts) would be a C/5 charge rate. As you go up in current or power, the charge rate number goes lower, not higher. Where C = Battery Capacity in Amp Hours, and X (number like 8, 10, 12) = Hours. So the formula is Amps = Amp Hours / Hours.

For most Flooded Lead Acid Batteries (FLA) you do not want to go lower than C/12, and no higher than C/8. There are some FLA batteries that can take say C/4 but those fall into the Starting and Hybrid batteries which you normally do not use in RE applications. There are cases where you must exceed C/8 like in regions where winter insolation is low. In those cases you have to use AGM batteries which can take up to 1C or higher charge rates. But that does not apply to your location.

So here is the deal, You will be using FLA batteries. That being said the charge rate must be between C/12 and C/8. Go outside that range and you are asking for big expensive trouble. You are dealing with the Laws of Physics. You cannot change or violate the Laws of Physics. If you try you suffer consequences immediately without mercy.

[Sunking]

Gents?

Rather than oversizing the array around stratification concerns, how about using the virtually required generator for a weekly equalization charge? Most recommend testing a generator monthly anyhow, how about doing so weekly for equalization (which implies the batteries get a full charge from the generators if they need it first)? No crisis if you miss a week once in awhile for whatever reasons (like the generator fails...).

Because a Equalization Charge is a OVER CHARGE CONDITION. A EQ charge causes plate damage that is permanent. Batteries should only be EQ'd when needed. EQ every week like you suggest vs when needed say once a month, means you just cut your cycle life 50 to 75%. Instead of having your expensive batteries last up to 5 years you only get 1 to 2 years. Physics my friend.

[kevinc_63366]

If you did 3700 is approx the C/10 rate. If you did as you suggest 3700 + 3700 (7400 watts) would be a C/5 charge rate. As you go up in current or power, the charge rate number goes lower, not higher. Where C = Battery Capacity in Amp Hours, and X (number like 8, 10, 12) = Hours. So the formula is Amps = Amp Hours / Hours.
.

OK - obviously I don't understand something basic. If I have a 3700 watt daily demand, those watts are not going to be available for battery charging right? Would not the inverter be functionally pulling the power from the panels via the charge controller(s)? I'm running under the assumption the battery in that situation simply provides a clamping voltage for those watts during those hours when the panel provide as least as much power as the instantaneous demand, and that only the excess power will be available for charging. With that in mind, I'm thinking the averaging over 24 hours is what is causing grief, since power is only available at instantaneous demand levels for a much shorter part of the day.

Regarding being pounded: Is this not a forum for learning? My personal learning approach is to try and understand the math behind the statements. Sucks having a BS in Applied Math, or the MS in Comp. Sci with a minor in Electrical Engineering that required me to prove that 1+0=1 using set theory and having to ignore all assumptions (btw - that took 4 of us working as a team about 3 hours to do, all straight A grad students), but it is who I am. That approach served me well for 30 years of my career, allowing me to find tens of millions of dollars of vendor mistakes and false claims - so please forgive me if I use the same approach to my personal systems.

[kevinc_63366]

Because a Equalization Charge is a OVER CHARGE CONDITION. A EQ charge causes plate damage that is permanent. Batteries should only be EQ'd when needed. EQ every week like you suggest vs when needed say once a month, means you just cut your cycle life 50 to 75%. Instead of having your expensive batteries last up to 5 years you only get 1 to 2 years. Physics my friend.

I understand that equalization using an overcharging voltage cause the H2O in the acid solution to electrolyze (gas out). I was unaware that doing so caused plate damage. That assumption was based in no small part by my reading of the Xantrax XW MMPT 80 600 user manual, which states:

"Boost charging allows for better utilization of flooded lead acid batteries under
moderate cycling in off grid applications. Boost charging encourages a short
duration charging voltage—above the gassing voltage—at the beginning of the
absorption charge state. Testing has shown that boost charging improves battery
performance by providing a regular mixing of the liquid electrolyte. Boost
charging specifically discourages capacity-robbing acid stratification and plate
sulfation."

and goes on to say:

"3. Boost charging may result in higher than normal water consumption.
However, the benefits of boost charging are likely to be greater than the extra
watering effort. Check battery water levels at least once per month."

Basically, their Boost charging is a very short equalization period.

Do you have references I can read that would highlight the plate damage vs. reduced sulfation such equalization causes?

Kevin

[inetdog]

OK - obviously I don't understand something basic. If I have a 3700 watt daily demand, those watts are not going to be available for battery charging right? Would not the inverter be functionally pulling the power from the panels via the charge controller(s)? I'm running under the assumption the battery in that situation simply provides a clamping voltage for those watts during those hours when the panel provide as least as much power as the instantaneous demand, and that only the excess power will be available for charging. With that in mind, I'm thinking the averaging over 24 hours is what is causing grief, since power is only available at instantaneous demand levels for a much shorter part of the day.

Regarding being pounded: Is this not a forum for learning? My personal learning approach is to try and understand the math behind the statements. Sucks having a BS in Applied Math, or the MS in Comp. Sci with a minor in Electrical Engineering that required me to prove that 1+0=1 using set theory and having to ignore all assumptions (btw - that took 4 of us working as a team about 3 hours to do, all straight A grad students), but it is who I am. That approach served me well for 30 years of my career, allowing me to find tens of millions of dollars of vendor mistakes and false claims - so please forgive me if I use the same approach to my personal systems.

Most "daily" demands are not continuous. So when the A/C, freezer, pump, etc. is not running you have the full PV watts available for charging. If the charger can handle that current and the batteries are low enough that the CC stays in bulk mode, then you will be alternating between charging at the high rate and charging at a low rate or discharging, depending on the exact balance.
Sporadic high rate charging once you get up to the Absorb voltage will probably be good enough for mixing and counteracting stratification where charging at the lower average rate may not be.

Learning is good. Some regulars on the forum are just getting burned out by having to explain the same things every day or two.
Search for the answer (using Google search with "site:solarpaneltalk.com" instead of using the built-in forum search.

PS: Using the formalism that "1" is defined as the set of all sets which have 1 member? Or using existence of a 1-1 mapping? And defining addition how?
Set theory based math is such fun!

[kevinc_63366]

Most "daily" demands are not continuous. So when the A/C, freezer, pump, etc. is not running you have the full PV watts available for charging. If the charger can handle that current and the batteries are low enough that the CC stays in bulk mode, then you will be alternating between charging at the high rate and charging at a low rate or discharging, depending on the exact balance.
Sporadic high rate charging once you get up to the Absorb voltage will probably be good enough for mixing and counteracting stratification where charging at the lower average rate may not be.

Learning is good. Some regulars on the forum are just getting burned out by having to explain the same things every day or two.
Search for the answer (using Google search with "site:solarpaneltalk.com" instead of using the built-in forum search.

PS: Using the formalism that "1" is defined as the set of all sets which have 1 member? Or using existence of a 1-1 mapping? And defining addition how?
Set theory based math is such fun!

Thank you for the kind response. Indeed much of my most critical load is sporadic (freezers in particular), I will review the rest to see what their profiles look like.

Most helpful.

You know, I did that proof in 1986... if I recall, it was to prove the identity theorem or some such (e.g. "Prove that a value exists such that the summation of the value with a 2nd value does not change the 2nd value"). I do recall it was heavy in set theory and may well have required a side proof that addition was valid on the set.

Update: Extended my load worksheet to add a "continuous" draw column. Defining that as "continuous draw likely during charging hours" which excludes things like freezers but includes the computers and other office equipment. Ends up its pretty small, with the current list at least, at about 425 watts. That will be most useful, thanks again.

[Sunking]

Kevein the problem is you ar trying to go outside proven engineering concepts. The first thing to do is determine your daily watt hours in a 24 hour period. That determines battery size right off the bat. Next i spanel wattage and that is determined by daily 24 hour use, location, and time of year use. The panels only generate their rated power for a few precious minutes around solar noon. During the course of a day only works out to a few hours of charging. Say 4 hours and in that 4 hours your panels not only have to supply what ever load is demanding at the time, plus enough for the remaining 20 hours of the day.

If you cut cornersd and go outside of the Laws of Physics and accepted design practices, only failure can be expected. It was learned by trial and error, a lot of error. So learn from past mistakes or you are doomed to repeat it.