LiFePo4 addition to a SLA House battery?

[eabyrd]

Aye, there's the rub. Do not use the LFP addition as a band-aid for not keeping your lead-acid fully charged!

That's an expensive band-aid. Make sure your charging regimen is proper for your SLA, otherwise consider total replacement with LFP.

No rub, and completely disagree. There is no need to throw out good SLA and do a wholesale replacement. The two chemistries are compatible especially if your following a long life charge profile for your Li

 

[eabyrd]

Yeah, but then you need a charge profile that includes an eq cycle which means separating the SLA from the LI. Just using them lightly one in a while is plenty

Actually I don’t think you equalize SLA. That whole sealed part kinda makes bubbling undesireable

 

[Substrate]

What I'm saying is that it is important to keep the SLA charged properly, and not use the LFP add-on as an excuse to not maintiain the SLA's fully charged.

 

[gelmjw]

What I'm saying is that it is important to keep the SLA charged properly, and not use the LFP add-on as an excuse to not maintiain the SLA's fully charged.

I will be using power from the LFP to charge the AGM. I will explain why that is a good idea that needs no excuse as if I have done something wrong. A few days ago while rolling this around in my head, I wrote up how I plan to add LFP to AGM. Here it is:

Adding LFP to an AGM bank. The AGM is a healthy single 12V 100Ah battery located under a dinette seat in the trailer. I will be acquiring a 100Ah LFP bank. This will increase my Ahs from 50 for the AGM alone, to 130 by adding 80%, or 80Ah from LFP.



T he dangers are:



1. Protecting the LFP:

a. The BMS must shut the LFP off for all the over and under limits such as too hot or cold, etc.

b. The AGMs need to be the same Volts before connecting in parallel or a burst of current could cause damage.



2. Protecting the AGM:

a. AGMs over/under Volts or temps are more forgiving.

b. The LFPs need to be the same Volts before connecting in parallel or a burst of current could cause damage.





Charging devices:

1. Magnum MMS-1012 must be connected through the LFP’s BMS. The best available charge and float Volts are 14.5, followed by a float at 13.5V. The only other control is to use the remote to turn the charger on/off. This device could be a problem.



This process must be performed without the Battery Temperature Sensor connected because the charge Volts will be raised higher when the battery is cool. The 14.5V is barely lower than the allowed max of 14.6V for LFP. In cell terms that is 3.625 compared to 3.65. The LFP must be well balanced so that no cell exceeds 3.65V during charging.



2. Morningstar Corp MPPT charger must be connected through the LFP’s BMS. This is very programmable and can avoid BMS many shutdowns. Per the specs, it does require the AGM in parallel to know battery Volts(12/24) in case the BMS shuts the LFP off.



This device will be set to charge = 13.8/3.45 Volts and Float = 13.5/3.375 Volts for AGM /LFP. It will have the BTS connected and the max charge Volts, or HVD = 14.4Volts.



Operation:

After a charge each chemistry will settle to different resting Volts; AGM to 12.8 and LFP to 3.375. This happens within a few hours and both figures represent fully charged. What you want, is to begin each night with is sufficient power to last until morning.



At the end of daily charging, batteries in parallel will be at equal Volts. While supplying power, the battery with the most Volts will supply Watts to the loads until the two are equal again. Even without a load, they will try to stay equal by passing current to the lower one. The LFP will reach resting Volts first. At that time it will supply current to the AGM. The AGM needs to be healthy or it could draw excessive current. The LFP will continue to carry loads until the AGMs resting Volts are reached.

Until now, the AGM has not contributed much to the load. Now, for the rest of the night, 50% of the AGM capacity is available plus whatever is left in the LFP .

If this is true, there will be a period of time when the LFP will maintain the AGM until the AGM reaches resting Volts. Aftet that, the AGM will share the load.

Comments?

 

[LVLAaron]

I just connected my 400ah lifepo4 battery to an AGM cell I have. Charged them to 14.1

Will report if any funny business happens.

 

[LVLAaron]

here's a video he has about adding a little bit of lead to a larger lifepo4 bank.

 

[LVLAaron]

My two batteries were connected and charged to 14.1 - Charger disconnected.

I saw ~1.5 amps going to the lead battery for maybe an hour... and then .5 amps until the lifepo4 pack reached 13.3. No power movement at this time. I will let it sit for 24 hours and monitor the BMS to see how much paracsitic draw the lead has on the lifepo4 pack.

 

[Horsefly]

I'm still mostly convinced that putting lead acid directly in parallel with LFP is a really bad idea.

1. My AGM batteries should to be charged (bulk / absorption) at 14.6V @ 25°C. That's a full 3.65V per cell of LFP. It is virtually certain that the BMS will cut out due to high voltage on at least one cell every time. My desire is to charge my LFP to only about 3.5V per cell.
2. AGM charging has to be temperature compensated, so if the temperature is below 25°C, the bulk / absorption voltage will be higher than 14.6V (i.e., more than 3.65V per cell).
3. AGM batteries like to stay in absorption for a long time, until the tail current gets down to 1%-2% of the 20hr C rating. Per #1 above, the LFP BMS will have cut out with a HVD several times during that period.
4. If you choose to not charge as high as the 14.6V, your AGM battery will sulfate over time and will be a doorstop. A person who sells thousands of AGM batteries every year has told me that the #1 cause of AGMs dying an early death is that people don't fully charge them by staying at absorption voltage until the tail current drops to 1% of C.
5. The lead acid discharge curve is somewhat steep, while the LFP discharge curve is very flat. So as current is pulled from both batteries, eventually the LFP is mostly sending current over to the AGM. That's lots of energy moving around not doing anything useful for you.

Rule A: The best way to get a reasonable life out of an AGM battery is to treat it right: Never discharge below 50% SoC, fully charge to the correct (temperature compensated) bulk / absorption voltage, and don't drop into float until the tail current drops to 1%-2% C.

Rule B: The best way to get a good life out of LFP is to treat it right: Stay between the knees. Charge to only 3.45V-3.5V per cell bulk / absorption, and the tail current will drop to near 0 very quickly. Discharge only to 3.0V. If the BMS cuts out, you've got something set wrong. Don't do any temperature compensation in charging.

You cannot do both rule A and rule B, ergo if you put lead acid in parallel with LFP you are choosing to sacrifice one, the other, or both.

 

[LVLAaron]

Good info, and there's two scenarios I see people doing here.

1. Someone who has a bunch of lead that's still got some life left, but wants to use lithium. This is the camp I'm in. I have a camper with 2x 12v batteries, but they're not enough. I also don't want to throw them away or deal with trying to sell them for a few bucks.

2. Someone who's charging from an alternator, and wants to keep the lead in place to prevent damaging the alt. In that case, a cheap crap motorcycle battery is adequate.

 

[Horsefly]

Good info, and there's two scenarios I see people doing here.

1. Someone who has a bunch of lead that's still got some life left, but wants to use lithium. This is the camp I'm in. I have a camper with 2x 12v batteries, but they're not enough. I also don't want to throw them away or deal with trying to sell them for a few bucks.

2. Someone who's charging from an alternator, and wants to keep the lead in place to prevent damaging the alt. In that case, a cheap crap motorcycle battery is adequate.

Yeah, I get that. I'm against putting LFP in parallel with LA, but I am also sympathetic.

I'm currently working a 42' sailboat transition to LFP for the house battery. We've decided to keep AGMs for a separate bank on the bow thruster and windlass, with a DC-to-DC charger between the two. The DC-to-DC charger acts as a load in the event that the LFP BMS shuts down, so the alternator (connected to the LFP) will not freewheel without a load. So there's a good use of LA in conjunction with LFP: A separate bank!

As for your camp, I think that is probably OK. Just be aware and watch your BMS carefully. It would be good to watch (via clamp meter) the current between the LFP and LA, but that may be overkill.

 

[shadowsteve]

So this is all about using up the last bit of life in an existing LA battery bank?

I chose to pull out my 4x 6v AGMs in my 5th wheel and replace them with LFP as they were 6 years old and down to about 80% capacity. I sold them and recouped about 20% of my initial cost on them. Saved about 170# of weight too

 

[wattmatters]

My AGM batteries should to be charged (bulk / absorption) at 14.6V @ 25°C

My "48V" AGMs bulk charge voltage is 56.4V (2.35V/cell). That's a lithium cell voltage of 3.525V/cell.

Float voltage is 54.0V (2.25V/cell) = lithium equivalent of 3.375V/cell.

This sounds like a match made in heaven.

I have 380Ah of 48V SLA. Adding 100Ah of LiFePO4 would be pretty neat. I won't be doing it unless 100Ah of LiFePO4 was available for SFA but a neat trick to keep up one's sleeve.

 

[Horsefly]

My "48V" AGMs bulk charge voltage is 56.4V (2.35V/cell). That's a lithium cell voltage of 3.525V/cell.

Float voltage is 54.0V (2.25V/cell) = lithium equivalent of 3.375V/cell.

This sounds like a match made in heaven.

I have 380Ah of 48V SLA. Adding 100Ah of LiFePO4 would be pretty neat. I won't be doing it unless 100Ah of LiFePO4 was available for SFA but a neat trick to keep up one's sleeve.

If that's really what the AGM manufacturer recommends, then maybe you are right. Mine said 29.2V for my 24V bank, so 3.65V per cell LFP. 56.4V is only 14.1V @ 12V, which sounds closer to float for an AGM than bulk / absorption. Trojan says 57.6V (3.6V/cell LFP equiv), which is about as low as I can recall seeing. What brand are your AGMs?

You still have to deal with the flat LFP vs steep AGM discharge curve. As long as you know what you are getting in to, go for it. It is your choice.

 

[wattmatters]

What brand are your AGMs?

Enersys Powersafe SBS190F

 

[wattmatters]

You still have to deal with the flat LFP vs steep AGM discharge curve. As long as you know what you are getting in to, go for it. It is your choice.

Well I'm not doing anything but the way the voltages work it would seem to be quite viable option to consider at some stage.

 

[gelmjw]

Yeah, but then you need a charge profile that includes an eq cycle which means separating the SLA from the LI. Just using them lightly one in a while is plenty

I was suggesting that the BMS could use High Voltage Disconnect(HVD) during the eq cycle of the lead. No extra charge profile needed.

 

[12VoltInstalls]

may be advantageous to charge each chemistry with it's own charging infrastructure

My thoughts exactly.

 

[gelmjw]

Actually I don’t think you equalize SLA. That whole sealed part kinda makes bubbling undesireable

Are you saying your SLA is a Sealed Flooded Lead Acid. A 'sloshy'?

I am certain that acid stratification is a real thing. How do you stir the acid solution?

The other definition for equalization is when you believe the cells of your lead have become unbalanced. This is when you rely on a longer timed eq period at extra high Voltage. Some cells are deliberately overcharged in order to restore the weaker cells.

 

[Horsefly]

Enersys Powersafe SBS190F

Well the manufacturer recommends the charging voltage of 2.40V per cell, or 57.6V for your 48V bank (at 20°C). That would be 3.6V / cell for LFP.

I've read about those batteries before, and got the impression they were really intended for backup (float) applications, rather than off-grid (cyclical) applications. However, I see they do claim it can be used in cyclical applications. However, they emphasize that it has to get back to full charge between discharges. If you are only charging to 2.35V per cell, you (strictly speaking) are not getting the batteries to full charge.

 

[Horsefly]

I was suggesting that the BMS could use High Voltage Disconnect(HVD) during the eq cycle of the lead. No extra charge profile needed.

First off, if you are talking about equalization you must be talking about flooded lead acid, not AGM. You do not want to equalize AGM batteries.

Second, I would never recommend planning for an HVD or LVD from a BMS as a normal course of using the LFP bank. The LVD and HVD are meant as a last line of defense for your LFP cells. But, to each his own.