• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

Switchable 12/24/48vDC battery?

Mattb4

Solar security
Joined
Jul 15, 2022
Messages
5,854
Location
Off beaten path.
On my afternoon walk I was thinking about the problems with using 12vDC LiFePO4 batteries in series to make a larger voltage battery with the separate BMS's, not allowing for cell balance and high cell voltage trips, when it struck me. Why not a select-able voltage output LiFePO4 battery? Inside the battery is the typical 16 cells of a 48vDC nom. battery and the BMS it has can control groups of 4 cells to act in series or parallel so that the output was the desired 12v, 24 or 48vDC.

Example: Battery would be rated as 50ah-48vDc/100ah-24vDC/200ah-12vDC. The user could decide what they needed.

Probably take some mad electronic tech to figure out all the details.
 
The only thing I can see is it would be expensive and not reliable, both for the same reason.

You would need a set of relays to do the switching - and they would need to be high current so more like contactors. And it would take quite a few of them to switch from 4s4p to 8s2p and to 16s... this means lots of space for the heating and the contactor area. Lots of wires.... Or it would mean a lot of MOSFET to turn on/off in a certain way to create the patterns needed.

I think the best you could hope for would be a battery where the cells were bolted so you could change the bus bar arrangement and then have a programable BMS.

Probably not enough market for either to get it done since people other than those of us on the forum have their voltage picked for them and they never think about it.

I can't see it ever being UL listed because of the user interaction bits which means you can currently buy a 16s case and do it in any configuration you want with your own BMS and bus bars.
 
The only thing I can see is it would be expensive and not reliable, both for the same reason.

You would need a set of relays to do the switching - and they would need to be high current so more like contactors. And it would take quite a few of them to switch from 4s4p to 8s2p and to 16s... this means lots of space for the heating and the contactor area. Lots of wires.... Or it would mean a lot of MOSFET to turn on/off in a certain way to create the patterns needed.

I think the best you could hope for would be a battery where the cells were bolted so you could change the bus bar arrangement and then have a programable BMS.

Probably not enough market for either to get it done since people other than those of us on the forum have their voltage picked for them and they never think about it.

I can't see it ever being UL listed because of the user interaction bits which means you can currently buy a 16s case and do it in any configuration you want with your own BMS and bus bars.
That is why you need a mad electronic tech. Since you would not be switching on the fly (under load) it should not be a big issue with current interruption.
 
That is why you need a mad electronic tech. Since you would not be switching on the fly (under load) it should not be a big issue with current interruption.
no, but whatever you switch would need to be able to handle the full current output on a contentious basis... so 100amps? 200amps? Basically the switch element would become the bus bar... in some places it would need to make a choice of 3, so now you have to bus bar to each "on" position and have it fail with things open so it is safe. If the battery goes dead how do you wake it up? How much drain is it going to take while it is just sitting there.

Latching contactors would be the ideal things... those are just a bit over $120 each --

You need 2 contactors or a bistable set at the junction of 4 & 5 and again at the junction of 12 and 13... and you would need 2 at junction 8 & 9 ...

A little chatgpt help to get the logic table

Takes - 4 bistable latching contactors - so they don't consume power except to switch them --- and the rest hurt my head to think about ... but if you study the stuff below it makes sense -

Cost would be an extra $600 to $1500 per battery + the minor logic circuits to control things....

And it would be something you might switch once in the life of the battery unless you also used it with a shunt-trip to disconnect it instead of a fuse...


AND NOW YOU HAVE MADE MY HEAD HURT ---- DONT DO THAT




Here is a detailed logic table showing the connections for each post of the 4 DPDT bistable latching contactors in all three configurations (4s4p, 8s2p, and 16s). Each contactor has six terminals: two poles (Pole A and Pole B) and four throws (two for each pole, for the two states).

Contactor Terminal Legend:​


  • C1: Contactor for junction 4&5.

  • C2-A: First DPDT contactor for junction 8&9 (parallel/series switching for cells 8 and 9).

  • C2-B: Second DPDT contactor for junction 8&9 (direction of 8&9 connection to other cells).

  • C3: Contactor for junction 12&13.

  • Terminals are denoted as:

    • A1, A2: Pole A throws (e.g., connected when in state 1 or 2).

    • B1, B2: Pole B throws.

Logic Table for Each Configuration​

ContactorTerminal4s4p Connection8s2p Connection16s Connection
C1 (4&5)A1Cell 4 positive ↔ Cell 5 positiveCell 4 negative → Cell 5 positiveCell 4 negative → Cell 5 positive
A2Cell 4 negative ↔ Cell 5 negativeNot connectedNot connected
B1Not connectedNot connectedNot connected
B2Not connectedNot connectedNot connected
C2-A (8&9 Parallel/Series)A1Cell 8 positive ↔ Cell 9 positiveCell 8 positive → Cell 9 negativeCell 8 positive → Cell 9 negative
A2Cell 8 negative ↔ Cell 9 negativeNot connectedNot connected
B1Not connectedNot connectedNot connected
B2Not connectedNot connectedNot connected
C2-B (8&9 Direction)A18&9 parallel positive → Cell 1 positive8 positive → Cell 7 positive8 positive → Cell 7 positive
A28&9 parallel negative → Cell 16 negative9 negative → Cell 10 negative9 negative → Cell 10 negative
B1Not connectedNot connectedNot connected
B2Not connectedNot connectedNot connected
C3 (12&13)A1Cell 12 positive ↔ Cell 13 positiveCell 12 negative → Cell 13 positiveCell 12 negative → Cell 13 positive
A2Cell 12 negative ↔ Cell 13 negativeNot connectedNot connected
B1Not connectedNot connectedNot connected
B2Not connectedNot connectedNot connected

Detailed Notes on Connections:​


  1. C1 (4&5):

    • 4s4p: Both poles (A1 and A2) connect cell 4 positive/negative to cell 5 positive/negative for parallel operation.

    • 8s2p/16s: Only Pole A connects cell 4 negative to cell 5 positive for series operation.

  2. C2-A (8&9 Parallel/Series):

    • 4s4p: Both poles (A1 and A2) connect cell 8 positive/negative to cell 9 positive/negative for parallel operation.

    • 8s2p/16s: Only Pole A connects cell 8 positive to cell 9 negative for series operation.

  3. C2-B (8&9 Direction):

    • 4s4p: Parallel pair of cells 8&9 is connected to both cell 1 (positive) and cell 16 (negative).

    • 8s2p/16s:

      • 8s2p: Series connection established between 7 → 8 → 9 → 10.

      • 16s: Series connection extends from cell 7 → 8 → 9 → 10.

  4. C3 (12&13):

    • 4s4p: Both poles (A1 and A2) connect cell 12 positive/negative to cell 13 positive/negative for parallel operation.

    • 8s2p/16s: Only Pole A connects cell 12 negative to cell 13 positive for series operation.



And a cost estimate

To implement the battery configurations (4s4p, 8s2p, and 16s) using 4 DPDT bistable latching contactors, you can consider the following options:


  1. Albright SW88-102L:
    • Specifications: 24V coil voltage, 80A continuous current rating, DPDT configuration.
    • Pricing: Approximately $354.16 per unit. citeturn0search2
  2. TE Connectivity KUL-11D15D-24:
    • Specifications: 24V DC coil voltage, 10A current rating, DPDT configuration, dual coil latching.
    • Pricing: Approximately $108.21 per unit. citeturn0search0
  3. TE Connectivity RT424F24:
    • Specifications: 24V DC coil voltage, 8A current rating, DPDT configuration, dual coil latching.
    • Pricing: Approximately $7.02 per unit. citeturn0search0

Total Cost Estimates:


  • Using Albright SW88-102L: 4 units × $354.16/unit = $1,416.64.
  • Using TE Connectivity KUL-11D15D-24: 4 units × $108.21/unit = $432.84. *takes both this and the next*
  • Using TE Connectivity RT424F24: 4 units × $7.02/unit = $28.08.

Considerations:


  • Current Rating: Ensure the contactor's current rating meets or exceeds your system's requirements.
  • Voltage Compatibility: Match the coil voltage with your control circuit's voltage.
  • Physical Size and Mounting: Verify that the contactors fit within your design's spatial constraints.

The Albright SW88-102L offers a higher current rating, suitable for more demanding applications, but at a higher cost. The TE Connectivity options are more cost-effective but have lower current ratings. Choose based on your specific application's electrical requirements and budget.



And for 200amps

To accommodate a continuous current of 200 amps, you'll need 4 DPDT (Double Pole Double Throw) bistable latching contactors rated for at least 200A. Here are some suitable options:


  1. Albright SW200 Series:
    • Specifications: Designed for direct current loads, including motors used in electric vehicles.
    • Pricing: Approximately $218.25 per unit.
    • Source: Cloud Electric
  2. Electric Car Parts Co. 200A DPDT Contactor:
    • Specifications: Available with coil voltages of 12V, 24V, 48V, 60V, 72V, 96V, or 120V DC; includes magnetic blowouts and UL listing.
    • Pricing: Approximately $91.00 per unit.
    • Source: Electric Car Parts Co.

Total Cost Estimates:


  • Albright SW200 Series: 4 units × $218.25/unit = $873.00
  • Electric Car Parts Co. 200A DPDT Contactor: 4 units × $91.00/unit = $364.00

Considerations:


  • Current Rating: Ensure the contactor's current rating meets or exceeds your system's requirements.
  • Voltage Compatibility: Match the coil voltage with your control circuit's voltage.
  • Physical Size and Mounting: Verify that the contactors fit within your design's spatial constraints.

The Albright SW200 Series offers a higher current rating, suitable for more demanding applications, but at a higher cost. The Electric Car Parts Co. option is more cost-effective but has a lower current rating. Choose based on your specific application's electrical requirements and budget.
 
no, but whatever you switch would need to be able to handle the full current output on a contentious basis... so 100amps? 200amps? Basically the switch element would become the bus bar... in some places it would need to make a choice of 3, so now you have to bus bar to each "on" position and have it fail with things open so it is safe. If the battery goes dead how do you wake it up? How much drain is it going to take while it is just sitting there.

Latching contactors would be the ideal things... those are just a bit over $120 each --

You need 2 contactors or a bistable set at the junction of 4 & 5 and again at the junction of 12 and 13... and you would need 2 at junction 8 & 9 ...

A little chatgpt help to get the logic table

Takes - 4 bistable latching contactors - so they don't consume power except to switch them --- and the rest hurt my head to think about ... but if you study the stuff below it makes sense -

Cost would be an extra $600 to $1500 per battery + the minor logic circuits to control things....

....
:) Yeah but once the Chinese get wind of the idea it it likely will costs only a shiny dime. Always the problem with outsourcing production.

Especially now that you did all the work for them.
 
I was imagining big U-shaped copper pins 1" thick that connect parallel or series the internal cells when you insert them into holes on the top,
depending on horizontal or vertical orientation,
with lots of surface area and a threaded lockdowns. And a see-through cover to admire them.

Probably not physically realizable. And the BMS thing.

But at least I didn't get a headache as the idea whooshed in and out. :)
 
Sounds like the most brilliant idea ever...  Not.
And then someone was crazy enough to go into a whole book about it, you guys need to get out more LOL
 
On my afternoon walk I was thinking about the problems with using 12vDC LiFePO4 batteries in series to make a larger voltage battery with the separate BMS's, not allowing for cell balance and high cell voltage trips.
This false info keeps getting repeated over and over.

Four 12v LFP batteries each with their own 4s BMS with 4 independent passive balancing bleed resistors that activate cell bleed for cell voltage above 3.4v is no different for balancing than a single 16s BMS with 16 independent passive balancing bleed resistors that also activate cell bleed for a cell voltage above 3.4v.

Only slight difference related to balancing is four separate 12v LFP batteries will not be synchronously shutting down balancing bleed momentarily when the BMS make cell voltage sampling. For the small balancing current of bleed resistors, this has little effect. Random inverter/charging current will dominate cell sampling voltage measurement variations.

You must ensure the 12v LFP battery BMS is capable of breakdown voltage of the total four series stack maximum voltage. The battery spec should state they are series stackable to four 12v LFP batteries. This basically means the BMS's are built with 80v or 100v breakdown MOSFET's and not 40v breakdown MOSFET's in a BMS that cheaper 12v LFP batteries may use.
 
This false info keeps getting repeated over and over.

Four 12v LFP batteries each with their own 4s BMS with 4 independent passive balancing bleed resistors that activate cell bleed for cell voltage above 3.4v is no different for balancing than a single 16s BMS with 16 independent passive balancing bleed resistors that also activate cell bleed for a cell voltage above 3.4v.

...
My apologies then. I just notice a lot of folks coming to the Forum that have problems when taking 12vDC LiFePO4 batteries and connecting them in series for 48vDC.
 
My apologies then. I just notice a lot of folks coming to the Forum that have problems when taking 12vDC LiFePO4 batteries and connecting them in series for 48vDC.

Primarily because they didn't prep them properly, i.e., fully charge each to true 100%. Many just string them together when they are likely at different states of charge, which guarantees one or more will trigger OVP.
 
You must ensure the 12v LFP battery BMS is capable of breakdown voltage of the total four series stack maximum voltage.
How many of the el-cheapo 12V batteries meet this requirement? I'm guessing none. If they don't say, they can't do it.

I'm also guessing that if the 12V battery could handle 48V, they would be advertised as such, and cost more for the feature.
 
How many of the el-cheapo 12V batteries meet this requirement? I'm guessing none. If they don't say, they can't do it.

I'm also guessing that if the 12V battery could handle 48V, they would be advertised as such, and cost more for the feature.
Excellent point. 40 Mousefeets or 80 is not something I see on the ads. Pretty gruesome what mad electronic techs do to make one of their devices work out.
 
I just notice a lot of folks coming to the Forum that have problems when taking 12vDC LiFePO4 batteries and connecting them in series for 48vDC.
My wild speculation based on personal experience is that they are following voltage based 90% protocols parroted across the webz and the balance algorithm is never triggered (or not for long enough).
I'm now floating my 2s4p 12v at 29v for 3-5 hours per day. SOH issue fixed.
Damn the torpedoes and full speed ahead!
 
...

You must ensure the 12v LFP battery BMS is capable of breakdown voltage of the total four series stack maximum voltage. The battery spec should state they are series stackable to four 12v LFP batteries. This basically means the BMS's are built with 80v or 100v breakdown MOSFET's and not 40v breakdown MOSFET's in a BMS that cheaper 12v LFP batteries may use.
Exploring this a bit more. Since the majority of low end 12vDC LiFePO4 batteries do not usually come with a 5 page booklet with full specifications, so are likely to have the lower voltage MOSFET's, what is the significance of having them in series? Is this guaranteeing an early BMS failure?

This could be a very important bit of information for the folks going the series battery route so it would be nice to clear this up. Thanks.
 
Has been discussed on earlier posts. Just look for a statement that the battery can be series stacked to four 12v batteries.

Examples looking at 12v LFP's on Amazon. Most describe it as 4P4S capable.

Power Queen -
"Optimized Expandability and Adaptability: Scale your power needs effortlessly with expandability up to 4P4S for a 48V 400Ah system"

Watt Cycle-
-Grade A+ Cells: The EV Grade A+ cells in our 100Ah LiFePO4 battery ensure a lifespan of up to 10 years, withstand up to 15,000 cycles, and support a 4S4P configuration with a capacity of up to 20.48kWh.

DC House-
  • Wide Application: LiFePo4 lithium battery is the most stable battery and it is perfect for RV, Boat, Trolling Motor, Fish Finders, Ice Fishing, Camping, Solar System, and Home Alarm Systems. Please charge the battery with a dedicated LITHIUM BATTERY CHARGER, and no more than 4 batteries in series or parallel.
VATRER POWER-
Vatrer Power Lithium Iron Phosphate battery 12V 460Ah LiFePO4 batteries could be expanded to max. 4P4S as a 51.2V 1840Ah battery system to get 51.2kW output power and 94.2kWh usable energy.

I don't see anything on LiTime batteries on the Amazon verbiage for 4S connections..
 

diy solar

diy solar
Back
Top