Does This Old Dual-Inverter Charging Setup Make Sense?

Leomania · June 2022

Hello! It's been six years since my last post, and this one is related to the previous, but focused on what I think is an odd charging setup on my now 22-yr old solar installation. Not quite stone age, but yeah, it ain't new.

I've got two Trace (Xantrex) SW4048 inverters that share one set of four 12V deep cycle batteries. Both inverters are set up with charging profiles, so both are therefore charging, and both are monitoring the battery voltage to determine when to switch from bulk to float. Aren't they going to be fighting with one another?

What I've seen is that when there's a discharge event (backup power kicks in), one of the inverters will switch to float after a few hours but the other stays on bulk. While it's there, I don't see any DC current from the PV cells flowing per the display on the three C40 DC charge controllers (I have three panel groups). I tricked this inverter into switching to float mode by lowering the bulk voltage to 57.0V and reducing the absorption time to 0, then setting things back where they were. At that point, the charge controllers began to show some DC current flow.

One thing to note is that the bulk voltage had been set to 58.0V based upon the battery spec sheet. I did this six years ago when replacing the batteries, up from 57.4V the installers recommended. While the inverter had been in bulk mode I measured the battery voltage and it was only at 55.2V. Checking the four batteries individually was interesting; they were:

B1: 13.52V

B2: 14.38V

B3: 13.49V

B4: 13.73V

These had been driven down quite a bit overnight due to an outage (transformer blew out nearby). These are VMAXTanks SLR-125 AGM batteries. At six years they're getting pretty old, but not thinking to replace them as the system is so darned old. I'm considering putting the system in bypass and shutting down everything to charge the other three batteries separately to see if they can be re-balanced.

My primary concern is the charging setup; I can't see how having two charging/monitoring systems running at the same time can work properly. Any opinions/insights?

Thanks,

- Leo

Dave Angelini · June 2022

Time to modernize with a networked (all devices) power system that has graphicly based monitoring.

Leomania · June 2022

Dave Angelini said:

Time to modernize with a networked (all devices) power system that has graphicly based monitoring.

Yeah, I know folks who keep their cars for a few years then sell them when the first issue crops up. Me, I drive mine to 200K or more. This system is aging but functional and as I'm probably moving in the next few years, replacing it isn't likely to be the optimal financial move.

My query has more to do with this implementation issue that dates back to the time the system was installed; I'd like to determine if it makes sense or, as I surmise, the two charging systems may not play nice together. I may not do anything about it.

- Leo

BB. · June 2022

The following is probably common knowledge for you--But just to make sure the bases are "covered":

As long as you follow good wiring practices and keep the overall length/resistance per battery string "equal":

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

Then having several charging sources not "in sync" with each other is not the end of the world. The charging source with the "highest voltage" set-points will "win".

So--With identical chargers (say two solar arrays with two solar chargers) setting them to the same float/absorb voltages would be the normal setup. They will not always be in sync--One will probably transition from absorb to float before the other. And that is not a "bad thing".

In some cases, that will be normal... Say you have a South West and a South East facing array ("virtual tracking" to get "more hours of sun per day--Lead Acid batteries "like more hours on charge") that would be normal.

In other cases, you would set different set-points. If you have mains charging and solar array for backup power for your home. You could set the mains charger voltages lower than the solar chargers and "favor" the solar charging for your battery bank (use solar more, and utility power less).

For systems that are not tracking correctly--You can always tweak one a bit higher voltages if it tends to "charge less" than its mate (internal "voltmeters" have some offset from each other.

I am guessing that B1/B2/B3/B4 are four batteries in series with each other (not 4 different parallel strings)? Assuming that these are the voltages of 4x 12 volt batteries in series for 48 volts... What else can you tell us?

Are these flooded cell lead acid batteries? What are the specific gravity readings per cell?

What is the bank AH capacity and the float charging current and voltage of the bank (since you bank spends most of its time is float)?
the
More or less, float charging current should be way less than 1% for a floating lead acid battery bank. As batteries age, then float current starts to rise. When it rises to 2% or over, then the batteries are near the end of life (or your float voltage is "too high").

Note for flooded cell lead acid batteries... You want the plates covered with electrolyte at all times. But do not (for example) fill the cells to the "top" before charging/EQ. If you fill to the top and then start hard charging/EQ, the batteries will get hot (electrolyte expands) and gas bubbles will form and push electrolyte fluid out the top of the batteries (loss of electrolyte, and a big mess). Top off the cells at the end of the charging cycles.

When charging, the 12 volt batteries should be (ideal) within 0.180 volts of each other with float voltage around 13.6 to 13.8 volts (or whatever the Mfg. says) or 54.4 to 55.2 volts overall.

When measuring the per battery voltage during charging/loads/floating, if you have a >0.180 volts between "high and low batteries", then your bank needs Equalize charging (or when SG range is >0.030 SG units between "high and low cells").
EQ is done for several reasons--But only "when needed" as it is "hard" on batteries (erosion of plate material from gassing, raising temperature of batteries, etc.).

One reason is if some cells/batteries are charged more than others (unequal cell SG/voltages). Using EQ or "controlled" over voltage charging (depends on battery brands/models) of 60-62+ volts to get 2.5% to 5% rate of charge (i.e., 100 AH bank or 2.5 to 5.0 Amps EQ charging current) to bring up the "low cells" during charging--This can take hours to do. On a fully charged bank, EQ charge and measure the SG of each cell and when all cells stop rising, that is the "new" full charge SG level for each cell (log the temperature adjusted SG for each cell/battery).

And there is EQ charging that is usually recommended for "tall cell batteries". Once a month EQ for an hour to get the batteries to "fizz" (light bubbling, not a "rolling boil"). This is to mix the electrolyte in the cells (electrolyte can stratify with heavy SG layer forming at the bottom of the cell, and light SG at at the tome). The bubbling helps mix the electrolyte.

If you have AGM (or GEL) cells--Then elevated EQ is not a good thing (you don't want the batteries to vent H2 and O2 gasses and acid mist). Instead for AGM cells, one policy to to EQ the batteries for ~8 hours once every 6 months--In this case, EQ is just the normal "absorb" voltage set-point held for 8 hours. You don't want the batteries to get hot/gas/etc.

You can also double check the batteries under load. Run your normal loads from the battery bank and monitor the per battery voltages (discharging to 75% State of Charge as an example) and make sure the batteries support the your loads and the bank overall meets your needs.

If you have one battery that has much higher or lower voltage than the other two--Then it is possible that EQ may "fix" the problem, or you may have a cell/battery going bad.

-Bill

BB. · June 2022

I should add--If you want to charge/EQ each battery separately, that is fine too.

-Bill

Dave Angelini · June 2022

He did say 6 year old AGM's. Some can be EQ'd but at that age doing nothing may be the best plan.
Old batteries in poor condition, can have failure modes that I would not want in my home. Just saying....

BB. · June 2022

Thank you Dave... I missed that.

Yes, 6+ years on AGM is getting towards the end of life for many batteries.

And "old batteries" start to take more charging current just to "float"... Watch the battery temperatures and listen for hissing (gassing) and "boiling" sounds--If so, that is too high of charging current and/or a shorted cell or other battery problem elsewhere.

One source I found years ago said that batteries floating >2% rate of charge are just a fire waiting to happen.

And AGMs should always float at much less than 1% (probably 0.1% or even 0.01% or less?). High float current on AGMs is and indication of end of life.

-Bill

Leomania · June 2022

Thanks for the helpful replies, gents. My responses below.

BB. said:

So--With identical chargers (say two solar arrays with two solar chargers) setting them to the same float/absorb voltages would be the normal setup. They will not always be in sync--One will probably transition from absorb to float before the other. And that is not a "bad thing".

Okay, that's good to know, it sets my mind somewhat at ease.

What is the bank AH capacity and the float charging current and voltage of the bank (since you bank spends most of its time is float)? More or less, float charging current should be way less than 1% for a floating lead acid battery bank. As batteries age, then float current starts to rise. When it rises to 2% or over, then the batteries are near the end of life (or your float voltage is "too high").

I have three panel groups, 16x100W Siemens panels. At this age, the best output I've seen on a panel group is 18A @54V => about 1KW. One of the groups is on a low angle, north-facing roof and it hits 8A for a narrow band of time each day. It gets way more shading than it did when installed due to tree growth. So at summer peak I can do 1KW + 1KW + 0.4KW => 2.4KW which is half the original max capacity (which even in 2001 I could not hit due to the south + east + north facing panel groups). This is in part why I went with replacement batteries that were half the size of the originals, also cost was a factor. We don't lose power all that often, but that 12-hour outage last week was enough to drain them with mainly fridge, lights, and misc. vampires as loads. So I'd agree with Dave, the batteries are degraded. They've had very little cycling.

I'm not sure whether my charger settings are ideal, these old inverters don't have a lot of tweaks. I had things set as follows:

Set Bulk Volts DC 58.0

Set Absorption Time H:M 02:00 (default)

Set Float Volts DC 53.6 (default)

Set Equalize Volts DC 57.6 (default)

Set Equalize Time H:M 00:00

Set Max Charge Amps AC 6

Set Temp Comp LeadAcid

The installers had recommended 10A AC charge rate but that was for the 2x larger batteries, so I dropped it down to 6 for these. But using AC amps has always seemed odd to me, not sure how to equate that to DC charge rate other than a basic AC -> DC calculation.

These AGM batteries have specs of 14.4-14.9V for charging voltage, so 58.0V seemed safe. The 00:00 equalize time was recommended by the installers. The float voltage is spec'd at 13.5-13.8V so the 53.6V in retrospect is a bit low.

At this point, anything you'd suggest changing? Besides the batteries, I mean.

A buddy just had a system installed that has panels with 3x the output. I could get rid of that inefficient panel group and replace the 32 panels with just 16 and have an even better system than I had before with optimal placement on those south- and east-facing groups. But I'll need to get an estimate to decide if all of that would be worth it; my annual electricity true-up cost is running about $1200 currently.

Cheers

- Leo

BB. · June 2022

Leo,

At best, usually the "best harvest" from a solar array (assuming your batteries/GT Inverters/etc. are "accepting" 100% of the array's output) is around 75-75% of array rating (as in a few minutes to an hour a day on cool/clear days typically spring/fall near solar noon a few times a year). Or:

1,600 Watt array * 0.77 panel+controller derating = 1,232 Watts "typical best case harvest a few times a year)

If your panel tilt/off solar noon angle is less than ideal (90 degrees to sun at "peak" solar energy), then you can further derate by the Sine (angle) to sun. I.e., if 75 degrees to sun:

Sine (75 degrees) = 0.97 further derate
1,600 Watts * 0.77 * 0.97 = 1,195 best case watts in this example.

So, seeing a 1,000 Watts from the array is not far from what could be expected.

For AC inverters, yes, the "charging current" can be defined it two different ways (and typically is). There is the maximum charging current (like 10% to 20% of bank 20 hour AH capacity). So, for example, if you have a 400 AH @ 48 volt battery bank:

400 AH * 0.15 rate of charge = 60 Amps

And then there is defining the DC Charging Current (really power) by the limitation on the AC input. This can be both to "set" charging current to the battery bank by the Inverter, but also limit AC input draw from the Inverter's AC power source (say you have a "smaller" AC genset for backup).

400 AH * 58 VDC charging * 10% rate of charge = 2,320 Watts into battery bank
2,320 Watts into battery * 1/0.85 typical AC inverter eff * 1/120 VAC input = 22.7 Amps AC input "battery charging" current
22.7 Amps AC charging current / 2 inverter/chargings = 11.4 Amps per inverter/charger

(just examples, fill in your own numbers). You have to check your inverter's specifications/configuration. I am pretty sure they have the AC current limit... They may or may not also have the DC charging current limit too.

In general, 14.4 volts (57.6 volts @ 48 volts) is typical for AGM batteries--Higher voltage charging can create more gassing which causes the catalysts per cell to "wear out" faster (and the catalysts+batteries to "run hotter", also less than ideal).

You might set manual AGM to 57.6 volts and 8 hours for "EQ" time--Done once every six months. Light EQ charging.

And, did you have a chance to measure the "per battery" voltages? Ideally, they should be within 0.18 volts of each other (resting, charging, discharging). If you have one (or more batteries that are higher or lower voltage than the rest) batteries in the bank. EQ charging may help bring "weak cells/battery/batteries" into better balance... But very high or very low voltages may indicated failing batteries (or poor wiring/connections).

If you have a current clamp DMM--You can measure the current per parallel string (if you have parallel strings) and make sure they are properly sharing current.

https://www.amazon.com/gp/product/B00O1Q2HOQ (inexpensive meter)
Klein Tools CL800 Digital Clamp Meter (mid priced meter on Amazon)

If you find a couple "weak batteries"--Those could be taking down the rest of the string(s)... You can try taking the remaining "good batteries" and make that your new battery bank for now.

-Bill

m151 · June 2022

If you do decide to upgrade your system, your working sw4048's will have a very high resale value and will help offset the cost of new equipment. Don't throw or give them away!

Leomania · June 2022

BB. said:

Leo,

At best, usually the "best harvest" from a solar array (assuming your batteries/GT Inverters/etc. are "accepting" 100% of the array's output) is around 75-75% of array rating (as in a few minutes to an hour a day on cool/clear days typically spring/fall near solar noon a few times a year). Or:
1,600 Watt array * 0.77 panel+controller derating = 1,232 Watts "typical best case harvest a few times a year)
If your panel tilt/off solar noon angle is less than ideal (90 degrees to sun at "peak" solar energy), then you can further derate by the Sine (angle) to sun. I.e., if 75 degrees to sun:
Sine (75 degrees) = 0.97 further derate
1,600 Watts * 0.77 * 0.97 = 1,195 best case watts in this example.
So, seeing a 1,000 Watts from the array is not far from what could be expected.

Hi Bill,

Yeah, at this age it is what it is, and I'll ponder an upgrade after we get through our "should I stay, or should I go?" thought process.

You might set manual AGM to 57.6 volts and 8 hours for "EQ" time--Done once every six months. Light EQ charging.

And, did you have a chance to measure the "per battery" voltages? Ideally, they should be within 0.18 volts of each other (resting, charging, discharging). If you have one (or more batteries that are higher or lower voltage than the rest) batteries in the bank. EQ charging may help bring "weak cells/battery/batteries" into better balance... But very high or very low voltages may indicated failing batteries (or poor wiring/connections).

If you have a current clamp DMM--You can measure the current per parallel string (if you have parallel strings) and make sure they are properly sharing current.

I gave the EQ a shot; that one higher-voltage battery persists, and I assume it's degraded as the voltage across its terminals when charging jumps around a lot (14.2-14.6V) as opposed to the ~13.5V on the other three. I think I may replace it and hope I get a couple more years out of these batteries.

Thanks for the time and thought you put into the responses, it is greatly appreciated.

- Leo

Does This Old Dual-Inverter Charging Setup Make Sense?

Comments

Categories