Adding new batts to old bank a total no-no?

Started by Powerplay, December 15, 2016, 02:04:37 PM

Previous topic - Next topic

Powerplay

I've often wondered if installing a new battery (of the same make & capacity) to an old battery bank (let's say one battery at the end of each series string) would work OK.  As long as the battery bank is still functioning at a fairly high percentage would the new battery on the end of each series act as a pusher.  The battery strings would still be balanced since each string in the bank got the same new battery on the end.

Just to spark some more battery discussion.
41 degrees N, 255 Ah 48V AGM, MagnaSine 4448 Inv, 3500W PV, Midnite Classic 150 A (12/6/15), Midnite Classic 150 B (4/14/16), WBjr, BTS, MNPV6 X 2, SPD X 2, Apps: Mini Split AC, Car Charger, Water pumps, Lighting, -> 48V string plan 500W, 94.5VmP, 112.5VoC, -15C - 40C TCVoC -.3%, TCIsC +.04%

RossW

Not sure what a "pusher" might be, but in a series string - the current will be the same through all cells.

What I've observed over a lengthy period of continuously monitoring each cell in several diverse battery banks, is that if you have a mix of older cells and newer cells, the older cells will require more watt-hours poked back into them to come up to full charge. In a series string, that means that newer cells that reach full charge before then, will be overcharged.

It gets more complicated when you're using constant voltage charging - some cells will start to rise in voltage more than others. With a constant-voltage charge across the entire pack, those cells with the higher voltage are now "starving" other cells - they're taking more volts and other cells voltages are necessarily lowered (due to the fixed charging voltage of the pack).

Left without additional balancing, you quickly destroy some cells by constant overcharging, and others by chronic undercharging.
3600W on 6 tracking arrays.
7200W on 2 fixed array.
Midnite Classic 150
Outback Flexmax FM80
16 x LiFePO4 600AH cells
16 x LiFePO4 300AH cells
Selectronics SP-PRO 481 5kW inverter
Fronius 6kW AC coupled inverter
Home-brew 4-cyl propane powered 14kVa genset
2kW wind turbine

Watt_

#2
Hello Powerplay

I don't get too wrapped up in nameplate ratings. A graph below to give a visual to some of what Ross has proven.

Here is an example of an arrangement similar to what you are inquiring about. These are not of the same nameplate capacity though and with what turns out to be good reason. Look at the cell voltage graph. The light orange lines are a 875AH traction battery and the darker lines are an older 1000AH traction battery in series. The newer, lighter line, battery has a small "balancing" load applied to it during charge to lower the voltage across those cells allowing the older battery the time for the difference in current to help bring the voltage up high enough for a full charge. We are not adding current to the weaker older cells, we are just robbing about 5A from the newer set to allow full current to the older cells. Also, although i hsve been away and not able to move the cell chargers around, i can add up to 2A per cell to even more lagging older battery cells.

http://ranges.albury.net.au/watt/index.html

I'll follow up a bit later. The monitor was down for a while and it'll show better data tomorrow

russ_drinkwater

What age difference in fla batteries do you consider to be a problem when adding to a bank?
Standalone. 20 Hyundai x 220 watts panels, 2 x classic 150's, Latronics 24 volt 3kw inverter, Whiz bang Jnr, 12 Rolls surrete  4KS 25P  batteries and WBJ.
Grid tie feed-in, 12.5 kw in 3 arrays generating 50 kws per day average. Solar river grid tie inverters

RossW

It's less about "age difference" and more about performance difference.
Internal cell resistance, SG, cycle efficiency etc.
If you can adequately monitor them *and adjust charge to those that need more (or less)* then I'm not so sure it actually makes that much difference. Clearly the lowest capacity will give up first on a deep discharge, but just because they're old isn't a reason to junk 'em IMO. *ON THE PROVISIO THAT YOU MONITOR AND ADEQUATELY CHARGE EACH CELL*.

Watt thought his batteries were junk, and that he'd ruined his new ones by using them in series in the same pack.
It was certainly headed that way, but we all started calling them "Betteries" after a while, because after some intensive monitoring and care, they were "better than they'd ever been".
3600W on 6 tracking arrays.
7200W on 2 fixed array.
Midnite Classic 150
Outback Flexmax FM80
16 x LiFePO4 600AH cells
16 x LiFePO4 300AH cells
Selectronics SP-PRO 481 5kW inverter
Fronius 6kW AC coupled inverter
Home-brew 4-cyl propane powered 14kVa genset
2kW wind turbine

russ_drinkwater

The trojans I want to add to have been babied for the 12 months of service. Well charged and watered as well as being kept under cover and clean. They have never been discharged less than 10% per day. SG's have always been 1275 plus on testing.
Standalone. 20 Hyundai x 220 watts panels, 2 x classic 150's, Latronics 24 volt 3kw inverter, Whiz bang Jnr, 12 Rolls surrete  4KS 25P  batteries and WBJ.
Grid tie feed-in, 12.5 kw in 3 arrays generating 50 kws per day average. Solar river grid tie inverters

LwK

Thanks for the discussion.  I was wondering the same thing for my system.  The batteries are similar in age, but they are different makes.  I have 12 Trojan L16E 6V 370Ah batteries hooked up for a 24V system and I just got 12 used 2V Surrette 4000 Series s-1380 1050Ah. I did some reading before I installed them and hooked the banks up though a bus bar to "distribute" the power between the different banks.  But there is a problem,... The Surrettes use a lot of water (14L after a couple of months) and the Trojans are still filled. 
Any thoughts?

dgd

#7
Looks like the Surrettes are charging and the Trojans are not.
So you have three strings of Trojans, each 4 in series to make 24v and you have one string of Surrettes 12 in series?
Then all strings are commoned to two busbars for +ve and -ve?

Its highly unlikely this arrangement will distribute charging amps across the strings as you want so that all strings will evenly charge. It only takes a slight difference in resistance to any string and it all becomes unbalanced. This will lead to some cells getting overcharged and other undercharged leading to early failures.

What charger are you using? solar controller?

http://www.smartgauge.co.uk/batt_con.html
dgd
Classic 250, 150,  20 140w, 6 250w PVs, 2Kw turbine, MN ac Clipper, Epanel/MNdc, Trace SW3024E (1997), Century 1050Ah 24V FLA (1999). Arduino power monitoring and web server.  Off grid since 4/2000
West Auckland, New Zealand

russ_drinkwater

I have the same here.
The L16's use little water and the surrettes use about 8 liters a month.
SG's are good allround and no problems. May just be a trait of the larger surrettes!
Standalone. 20 Hyundai x 220 watts panels, 2 x classic 150's, Latronics 24 volt 3kw inverter, Whiz bang Jnr, 12 Rolls surrete  4KS 25P  batteries and WBJ.
Grid tie feed-in, 12.5 kw in 3 arrays generating 50 kws per day average. Solar river grid tie inverters

LwK

Correct on the Strings of batteries, DGD. I am using a MorningStar MPPT60 for the charge controller.  I also have a small wind turbine (600 watt , actually only makes 300w, but that is a different thread :) with a MPPT charger controller working in tandem.  The Sureties sure bubble compared to the Trojans.  I was thinking about taking the trojans out of the loop for a bit and see how the Surrettes charge on their own.