diy solar

diy solar

Victron recommended settings for LFP 10-90% SoC

thegoogler

New Member
Joined
Apr 15, 2021
Messages
108
I've got about 2 months of usage on my DIY 280Ah LiFePO4 battery. After a 5 week trip, I decided the Renogy Rover Elite 40A wasn't sufficiently configurable, as I found myself regularly adjusting the absorption voltage in order to either speed up charging when I'd been in the shade a few days, or to cut off charging when I started the morning off at ~90%. So I'm selling it and I've installed a Victron SmartSolar MPPT 100/50 along with a SmartSense monitor.

If I just wanted to charge my LFP to 100% this would be easy. But my goal is to stop around 90% SoC, according to both the Overkill BMS and my Simarine Pico. Note that while I have 280Ah, most days when I'm in the camper I'm only using maybe 10% of the battery capacity, and when the camper is parked between trips I'm using maybe 1%. But when living in the camper I might go 4 days in the shade and be down to 50-60% capacity by the end of the trip. Also note that I have 350W of solar currently available, though I've never seen it deliver more than about 17A at peak (which is barely above 0.05C).

The Victron LiFePO4 default seems to be too aggressive for me, so I've used the LFP settings as the baseline and then I'm customizing from there.

Victron defaults
absorption: 14.2V
float: 13.5V
equalization: disabled

My initial settings:
absorption: 14.2V
float: 13.5V
equalization: disabled
re-bulk voltage offset: 0.10V
absorption duration: adaptive
max absorption time: 2 hours
tail current: 2A

I installed this yesterday. My SoC was at 100%. Upon enabling the device, the Victron went into bulk mode and was applying between 0.5A and 1.2A, depending on the amount of sun (I was in the shade). So immediately I'm confused because the Amps being applied are less than the tail current. So lots of questions for those with some experience with Victron, and I'm particularly interested in feedback from anyone who has successfully used Victron to limit SoC to <100%:
  1. Is bulk charging just "constant current @ absorption voltage"?
  2. Is there no direct way to control any of the bulk charging parameters? I assume the Victron went into bulk mode because it was able to apply constant current. But SoC should be 100%. Battery voltage was showing ~13.8 with current being applied
  3. Should I decrease the absorption voltage? With the Renogy I ended up adjusting it between 13.6V and 14.4V depending on the battery SoC each morning. With the Renogy I found if I set it to a lower voltage (i.e. 13.6V) the # of Amps was lower too (i.e. 3A) but as I increased it I could get up to ~17A @ 14.4V delivered to the battery. So if I'd been in the shade for days and was down ~100Ah I would set it to 14.2 or 14.4V when driving to ensure I got a sufficient recharge, but if I was parked in the sun and only using ~20-30Ah/day I'd leave it at 13.6V or maybe 13.8V and let it "trickle" charge.
  4. Should I change the max absorption time and/or tail current value? I feel like in my limited testing and research tail current should actually be set based on absorption voltage, not based on battery size.
I could set absorption to ~13.6V, but then if my battery has been drained it'll take ~20 hours to recharge at the lower amperage supplied. Ultimately my goals are:

1. try to bulk charge at a high voltage (ideally 14.4V) when the battery is depleted to as to get the highest # Amps possible back into the battery
2. as I get close to full (90%+) I want to decrease the target voltage (or just plain stop charging) so as not to fully cycle the battery

I'm really interested in feedback and any experience others have had accomplishing the same, especially with Victron.
 
My Victron defaulted to 14.4v for absorption, not 14.2v. Was that a typo in your post? I recently tweaked that and decreased it to 14.2v, while changing nothing else. I have 640w to 1280w available, depending on if I use the ground deployable panels.

My usage is somewhat similar to yours. Some weekends I work on the trailer and draw perhaps 80 amps of usage from loads such as the stereo, cordless battery charger, corded saw, interior fans, etc.
 
Try disabling float.
You battery should charge until full and then discharge until you hit the re-bulk trigger voltage.
Not sure what Victron might name that parameter but it is probably there.
Then you can play with the bulk/absorb and re-bulk parameters.
Like a drinking bird.
 
My Victron defaulted to 14.4v for absorption, not 14.2v. Was that a typo in your post? I recently tweaked that and decreased it to 14.2v, while changing nothing else. I have 640w to 1280w available, depending on if I use the ground deployable panels.

My usage is somewhat similar to yours. Some weekends I work on the trailer and draw perhaps 80 amps of usage from loads such as the stereo, cordless battery charger, corded saw, interior fans, etc.
No typo, that's what it shows for position #7 (LiFePO4). Maybe it's slightly different between Victron models (I'm using the SmartSolar 100/50) or maybe it's a difference in firmware?

For the moment I may just set 13.8V absorption and 13.4V float for now. SoC will show 100%, but at least it shouldn't (drastically) overcharge. And I can play with it when I'm back in the trailer in a couple weeks.

Side note: I have 350W on my roof and an extra 100W panel that I can plug into the side. I've considered switching everything to 24V input but (a) I'd need another flexible panel for the ground/mobile use to make that 24V and (b) I already can get shading of one roof panel depending on where I'm parked at times so I'm not sure going to a 24V setup will (always) help me. If I was always in the sun, perhaps. The 100/30 would've been sufficient for my current needs, but I wanted the ability to add panels in the future without yet another SCC swap
 
Try disabling float.
You battery should charge until full and then discharge until you hit the re-bulk trigger voltage.
Not sure what Victron might name that parameter but it is probably there.
Then you can play with the bulk/absorb and re-bulk parameters.
Like a drinking bird.

I've considered that, but two things stopped me:
  1. This lowers the number of cycles, but it still charges to 100% if I use any absorption voltage above 13.4 (or maybe 13.5). And if I set it to 13.5V then my panels charge at like 2-3A. But when it's set to 14.4V the same panels will charge at 16A or more.
  2. There is a benefit to floating at low voltage when I have a small load on the camper. The 12V devices will basically run off the solar (i.e. 1A in, 1A out). If I disable float then I'm basically charging to 90% and immediately draining (even during the day when it's sunny) but not charging until my voltage drops sufficiently (which might not happen until night time, in which case I'm starting the evening off at perhaps 70% SoC).
I think what I want to is configure the tail voltage in order to stop charging. The thing I can't yet figure out is if there's a way to bulk charge at higher currents (and voltages) but cut back to float as soon as the bulk charge is done.

Thinking about it a bit more what I feel like I want to do is set:

absorption: 14.4V <- encourages "rapid" charge during bulk phase
float: 13.2V (or maybe even 13.0V) <- enough to power devices but always lower than the LFP battery so it never charges. If I expect to sit at 85-90% SoC most of the time when unattended and not under load then I generally want this to be whatever battery voltage is at perhaps 80% SoC?
equalization: disabled
re-bulk voltage offset: 0.10V <- I think re-bulk only kicks in when Vbat < (Vfloat - Vrebulk) which means the battery would need to hit 13.1V before another charge cycle starts, which is basically never. But after 1 hour the charge algorithm apparently restarts anyway (as per the manual).
absorption duration: adaptive
max absorption time: 0 hours <- skip absorption step
tail current: 11.2A <- This is 4% of my capacity... just a swag but 1% or a bit more would probably be sufficient.

I'll take advice on the above and what I'm setting wrong...
 
I've considered that, but two things stopped me:
  1. This lowers the number of cycles, but it still charges to 100% if I use any absorption voltage above 13.4 (or maybe 13.5). And if I set it to 13.5V then my panels charge at like 2-3A. But when it's set to 14.4V the same panels will charge at 16A or more.
  2. There is a benefit to floating at low voltage when I have a small load on the camper. The 12V devices will basically run off the solar (i.e. 1A in, 1A out). If I disable float then I'm basically charging to 90% and immediately draining (even during the day when it's sunny) but not charging until my voltage drops sufficiently (which might not happen until night time, in which case I'm starting the evening off at perhaps 70% SoC).
I think what I want to is configure the tail voltage in order to stop charging. The thing I can't yet figure out is if there's a way to bulk charge at higher currents (and voltages) but cut back to float as soon as the bulk charge is done.

Thinking about it a bit more what I feel like I want to do is set:

absorption: 14.4V <- encourages "rapid" charge during bulk phase
float: 13.2V (or maybe even 13.0V) <- enough to power devices but always lower than the LFP battery so it never charges. If I expect to sit at 85-90% SoC most of the time when unattended and not under load then I generally want this to be whatever battery voltage is at perhaps 80% SoC?
equalization: disabled
re-bulk voltage offset: 0.10V <- I think re-bulk only kicks in when Vbat < (Vfloat - Vrebulk) which means the battery would need to hit 13.1V before another charge cycle starts, which is basically never. But after 1 hour the charge algorithm apparently restarts anyway (as per the manual).
absorption duration: adaptive
max absorption time: 0 hours <- skip absorption step
tail current: 11.2A <- This is 4% of my capacity... just a swag but 1% or a bit more would probably be sufficient.

I'll take advice on the above and what I'm setting wrong...
I'm going to address the 10-90% SOC as that was the title of your thread.
I think there is more going on here but one thing at a time...

I suggested that you disable float and play with bulk/absorb voltage and re-bulk voltage to cycle your cells.
LFP cells like to be cycled.
The don't like to sit full and they don't like to be floated.
Holding a battery above its full resting voltage causes voltage stress which sequesters lithium which becomes permanently unavailable to the battery chemistry.

Charging to 90%
I suggest charging at a relatively high charge voltage so that no or minimal absorption is required.
The means staying in constant current if possible until you get ~90SOC.
BTW charge voltage and battery voltage will be slightly different.
It takes potential difference to make current flow.
Try 14.4 with 0 absorption time and see what your shunt says about SOC at the termination of charge.
Notice the voltage drops down on termination of charge and even more if your battery starts to service a load.

Discharging to 10%
Instead of discharging to 10% I suggest you discharge at a current representative of your use case until your weakest cell starts to nosedive.
Note the voltage where the cell starts to diverge.
That is a good point to start charging your battery.

Voltage and state of charge best correlate at the extremes.
That is why shunts are so important to our usage model.
 
I'm going to address the 10-90% SOC as that was the title of your thread.
I think there is more going on here but one thing at a time...

I suggested that you disable float and play with bulk/absorb voltage and re-bulk voltage to cycle your cells.
LFP cells like to be cycled.
The don't like to sit full and they don't like to be floated.
Holding a battery above its full resting voltage causes voltage stress which sequesters lithium which becomes permanently unavailable to the battery chemistry.

Charging to 90%
I suggest charging at a relatively high charge voltage so that no or minimal absorption is required.
The means staying in constant current if possible until you get ~90SOC.
BTW charge voltage and battery voltage will be slightly different.
It takes potential difference to make current flow.
Try 14.4 with 0 absorption time and see what your shunt says about SOC at the termination of charge.
Notice the voltage drops down on termination of charge and even more if your battery starts to service a load.

Discharging to 10%
Instead of discharging to 10% I suggest you discharge at a current representative of your use case until your weakest cell starts to nosedive.
Note the voltage where the cell starts to diverge.
That is a good point to start charging your battery.

Voltage and state of charge best correlate at the extremes.
That is why shunts are so important to our usage model.
Thanks, that makes sense. I'll give the zero-absorption config a shot. Might go over this evening and turn on some higher load fans just to bleed off some capacity and see how it does. I'm not sure I can disable floating but I am sure I can at least turn it down to a point where it's (hopefully) not doing anything as it's well below the resting battery voltage.

I likely won't discharge to 10%, at least not as currently built out. Right now I don't have an inverter in the trailer, so it takes a lot of fans, small inverters (like for the laptop), lights, and furnace running to make a real dent in the SoC. With the family I can't really camp more than 4 or 5 days tops before we *have* to move so we can dump our waste tanks and refill water. I've seen SoC drop 10-15% per day if I'm mostly shaded based on this usage, but thus far I haven't gone much below 68% I think. However future plans entail an inverter and a 120V fridge so that may change. But in short, yes if I get down anywhere near that low I will watch to see where the lower knee really is and configure my alarms and cut-offs for that level.
 
UPDATE: I meant no tail current either.
This is the way I charge.
 
BTW I'm starting to think my overkill bms's current accounting if not very accurate.
I have a rasbery pi quering my bms and controlling my charger.
I had it cycle by soc and the end voltage steadily increased cycle after cycle.
For the last week or so I charging/discharging based on voltage and the soc at charge termination is steadily decreasing.
I suspect its because I'm only discharging to ~50%.
 
BTW I'm starting to think my overkill bms's current accounting if not very accurate.
I have a rasbery pi quering my bms and controlling my charger.
I had it cycle by soc and the end voltage steadily increased cycle after cycle.
For the last week or so I charging/discharging based on voltage and the soc at charge termination is steadily decreasing.
I suspect its because I'm only discharging to ~50%.
Yeah I really don't use the BMS SoC (mine is the knockoff Overkill version) as I've noticed it seems slightly off sometimes vs what I'm seeing in my Simarine panel in my trailer. It's generally close, but I think the overkill BMS is sampling current draw every second or so and calculating based on that, or there's some other minimal load (like the BMS itself) that is throwing it off. My Simarine shunt seems to be more accurate. I think.
 
Update: I set my absorption to 13.6V for now with 0 absorption time, and float to 13.2V.

Result: The SCC will bulk charge at 1-2A for 6-8 hours each day before kicking over to float (which does not apply a charge). Once the bulk charge is done, the battery eventually drops down to ~13.3V. According to my BMS and my Simarine SoC monitor I only use *maybe* 2A per day, which is <1% of the battery. However the SCC records that it's producing 90-100W each day before going into float mode. (Note: I have a SmartSense attached to the battery and I *assume* that's the input for the above data).

Questions: I'm a bit confused here.
  1. How does the SCC generate 8A into a battery that purports to only have used 2A? I realize the SCC is powering other appliances but literally I'm consuming maybe 100mA/hour.
  2. I'm trying to cut off charging when the battery hits ~90%, but each day I'm starting off at 99% and then the SCC is pushing me to 100%+. Why does it bother to restart the bulk charge cycle?
  3. Do I need to go through a discharge cycle each day in order for the SCC to work as I intend it? At the moment the trailer is parked and so the battery is more or less idle and fully charged until I use it again, probably over Labor Day.
I splurged on the $$$ for the Victron but at the moment it's not buying me anything over the cheap Renogy I had...
 
Update: I set my absorption to 13.6V for now with 0 absorption time, and float to 13.2V.

Result: The SCC will bulk charge at 1-2A for 6-8 hours each day before kicking over to float (which does not apply a charge). Once the bulk charge is done, the battery eventually drops down to ~13.3V. According to my BMS and my Simarine SoC monitor I only use *maybe* 2A per day, which is <1% of the battery. However the SCC records that it's producing 90-100W each day before going into float mode. (Note: I have a SmartSense attached to the battery and I *assume* that's the input for the above data).

Questions: I'm a bit confused here.
  1. How does the SCC generate 8A into a battery that purports to only have used 2A? I realize the SCC is powering other appliances but literally I'm consuming maybe 100mA/hour.
  2. I'm trying to cut off charging when the battery hits ~90%, but each day I'm starting off at 99% and then the SCC is pushing me to 100%+. Why does it bother to restart the bulk charge cycle?
  3. Do I need to go through a discharge cycle each day in order for the SCC to work as I intend it? At the moment the trailer is parked and so the battery is more or less idle and fully charged until I use it again, probably over Labor Day.
I splurged on the $$$ for the Victron but at the moment it's not buying me anything over the cheap Renogy I had...
Bump
 
Victron Smart Battery Sense measures only temperature.

The Victron solar charger enters bulk each day because the settled volts on the battery, 13.3, is lower that the charge target of 13.6.

If you want to keep the battery at a 60% to 70% state of charge set the charge volts to less than 13.4. When you ane not using the RV use a charge volts of 13.3 and a float of 13.2, set this as a user profile and store. When the RV is in use have a charge of 13.8 vo,to and a float of 13.4 volts.

Mike
 
Victron Smart Battery Sense measures only temperature.

That is not true. Screenshot right from https://www.victronenergy.com/accessories/smart-battery-sense:

1629981410666.png

The Victron solar charger enters bulk each day because the settled volts on the battery, 13.3, is lower that the charge target of 13.6.

If you want to keep the battery at a 60% to 70% state of charge set the charge volts to less than 13.4. When you ane not using the RV use a charge volts of 13.3 and a float of 13.2, set this as a user profile and store. When the RV is in use have a charge of 13.8 vo,to and a float of 13.4 volts.

Mike

I would expect the Victron to enter bulk, but then to exit bulk fairly quickly (within a few minutes) as if the battery is already >90% SOC then the SCC should switch from constant current to constant voltage automatically.

I could set the charging voltage very low, or just turn the SCC off as with a 1Ah parasitic drain per day the trailer can sit for ~9 months before the battery runs out, and then change the profile every time I use the trailer. But I'd prefer to come up with a "set it and forget it" configuration rather than needing to remember to change it back and forth, as I have enough things to remember each time I park or pick up the trailer that I'm likely to forget to do it occasionally and then find I'm starting off a trip with less battery capacity available than expected.
 
YES please look at my response in that thread, https://diysolarforum.com/threads/float-lifepo4-or-not.27417/post-326734

I DISAGREE with SmoothJoey's assertions regarding the use of Float. It is boldly explianed in the thread noted.

There are some general misconceptions with regards to the use of Float & what is the 100% of your battery.
Float in the Context of Lithium is used as Constant Voltage, Variable Current, which allows the cells to saturate at the "set" voltage to balance out / level off and reduce in Amps taken as a result. An immediate consequence is when Charge Voltage stops, the normal settling that occurs will be much more level & smooth and not as deep.

Max LFP Voltage Range is 2.500-3.650.
The Nominal Voltage is 3.200 - 50% SOC
The Working Voltage Range is 3.000-3.400 which is for all intents & purposes 92-95%
Below 2.900-2.800 is the Voltage Cliff Drop
Above 3.500 is the Voltage Cliff Climb *!* This is also where runner cells will take off & hit 3.65+ Hi Volt cutoff.

Additionally, EndAmps / TailCurrent is 0.05C-Rate. for 280AHx0.05=14A
If there are multiple packs in a Battery Bank, the highest AH battery pack is used for the EndAmp calculation.
 
Looking at the manuals there is a Tail current setting for the SmartSolar charge controller under Advanced settings:

And the SmartShunt can be configured to use tail current to indicate when the battery is considered at "100%":

The problem with tail current is that the scc doesn't know tail current from the batteries perspective.
The solar charge controller can't tell if its powering a load, charging a battery or some combination of both.
 
If you couple the Victron SmartSolar with the Victron SmartShunt, I bet it can tell the tail current. At least with my BMV-712 shunt, it does communicate with the Victron 100/50 SmartSolar.
 
The problem with tail current is that the scc doesn't know tail current from the batteries perspective.
The solar charge controller can't tell if its powering a load, charging a battery or some combination of both.
Thankfully the tail current doesn't take effect until the charger switches to absorption phase. So at least for LFP, the "penalty" is only at the top of the curve once the controller has switched from CC to CV. At least that's my understanding. So if your goal is to stop at around 90% SOC (as in my original question) then this effect is probably fine. But if you're trying to get to 100% then it might stop you

For the SmartShunt, would increasing the "charged detection time" help in that case, at least with "bursty" loads (like running a microwave or hair dryer off an inverter, or a fan or pump kicking on for a few minutes)? The SCC manual indicates the tail current is determined over the period of one minute, but I don't see a way to adjust that directly (and I don't have a SmartShunt) :-/

For the SCC's, I would think re-bulk voltage offset might help trigger the SCC to kick back into bulk mode if you're significantly undercharged, though there'd still be a delay for the battery to settle, as well as the risk of microcycling under normal usage since you have to go into bulk mode before you can drop back into absorption.
 
Back
Top