Since the max. recommended voltage is 54.4V, my current settings under the charger section in VEConfigure are:
Absorption voltage = 55.2V
Float voltage = 54.4
Charge current = 70A (in the VRM app I limited it to 54A)
Repeated absorption time = 1h
Repeated absorption interval = 7 days
Maximum absorption time = 1h
I will say, my (Victron-certified) installer is not very knowledgeable and after some research I’m having second thoughts about these settings. I read that typical charging voltage for 3.2 lifepo4 cells is 3.5-3.65, which means absorption should be 56V? The DC input low shut-down is set to 48V.
What charge/discharge and shut-down settings do you suggest for my pack?
You need to look into the spec in more detail to decide on “absorption” voltage – which for LFP is really only needed for cell balancing. This normally starts at a voltage above the normal float voltage (54.4V in your case), and will continue up to the “maximum recommended voltage” which is usually when the SoC is reset to 100% by the BMS.
From your spec, I’d say you want to spec 57.6V as the “absorption” voltage. 48V is fine as shutdown, 40V is really low and may lead to cell damage since you’re doing top-balancing.
For my battery the “float” voltage (set by the BMS, the Quattro is externally controlled) is 54V, “absorption” voltage = 100% SoC is 57.3V, and it charges to this once a week for 2 hours, then discharges again – usually takes a few hours to drop back to 54V which is 99% SoC (calculated by integrating current).
I’m almost exclusively charging the battery pack with solar. As a result it charges and goes into absorption mode depending on the battery SoC and available solar power. Which in the summer days I presume it will be every day? I don’t know of a different way to manage this, since I’m already filling in all the fields in the Charger tab. I guess if repeated absorption interval is set to 7 days, it makes sense for the inverter to only go back into absorption mode 7 days later and not before. Is that accurate?
As for the voltages then, I will increase my absorption voltage. What is your sustain voltage? should it be like float at 54.4?
Is the dynamic cut-off useful? I’ve read it’s disabled by default since varying loads may lead to temporarily decreased voltage, leading the inverter to shut down inadvertently. The guy who set it up seemed to made up all the voltages in there, leading the inverter into absurd behavior like going into Sustain mode at like 52V (that was like 42% or so) while shutdown was set to 48 (20%). At some point I had enough and set all of them to 48V and now it’s behaving as expected.
the label I posted is ALL the info the manufacturer provided. When I asked them for a user manual or spec sheet, I was basically yelled that the battery should only be installed by authorized personnel. The thing is, the Victron authorized guy who installed the battery shows very little knowledge about these type of installations. So I have to deal with this somehow.
You could set float and absorption to the same voltage. Usually the batteries that recommend 54.4/8v do.
Second thing usually the cut off allows 10% soc reserve on the batteries that use 54.4v float.
The other observation for the same batteries is their balancing is ultra slow (which is probably the reason for the lower voltage recommend for charging to prevent the run away.
The dynamic cut off in the ess assistant overrides the low shut down on the inverter tab. You could set that to 47V for the higher C draws.
And sustain is only triggered if the battery draws under the dynamic cut off. But it would probably bee ok to drop it to 51v.
If you set the float and absorption the same then worrying about longer absorption won’t be a thing. And the battery hasn’t stated a float.
Allegedly from showroom models with “0 km” on them. They should be +98 or 99% SoH.
They’ve just published the Victron recommended setup in the product page (here).
Absorption = 56V
Float = 55.5V
Absorption time=1h
DC input low-shutdown = 45V
DC input low restart = 47V
DC input low pre-alarm=46V
Do these seem reasonable to you guys given the hardware specs? For some reason and despite being the same model, the label on my battery pack has different voltage values for SoC 20% and 80%.
Brand is called second life. If they are second life cells, it is not a terrible thing.
They are using a dally bms. So it is how they programmed it that will be interesting. They don’t have great balancing
Neat go with that. If you experience problems the can be revised down. You have to start somewhere. You will find out soon enough if the bms will allow it.
I would make this one higher if they are second life. But see how you go there.
This is normal as a new company tries to work out the best performance for cells used. Each batch will be different if they are using second life. It will depend on where they were used previously. (If they are their name throws some question marks)
They want them cycled not kept at 100% most second life companies do.
How can I tell if the cells aren’t balanced? I mean, what symptoms should I look out for?
Yeah, the DC-low settings seem very low. Based on the info I find online, 20% SoC corresponds to 3.2V, that’s 51.2V; 2.85V is below 10%, that’s 45.6V. So their settings don’t seem conservative at all to me plus they seem to conflict with their own min. recommended of 20%.
For any lifepo4 16s battery, the daily bulk charging voltage is 55.2v. The floating voltage is 54v. Universally valid for any battery of this type. When you notice differences between cells, charge to 56v, usually once a month if there are no differences or sooner if there are. Thus, the battery will be used in the safest parameters and will reach 10000+ cycles. BMS required.
I have the same battery, installed two days ago, and I’m having the same thoughs about the multiplus configuration values. Did you come to any conclusions regarding the recommended values by the vendor?
I emailed them inquiring about the discrepancy in settings published and recommended on the label. Their response was basically that the website settings are slightly more aggressive for those users wanting to get more out of their packs.
I believe I kept mine the same (meaning more conservative), I have it on my notes somewhere, have to dig them up, but will let you know in case you’re still interested.
I think I’ll do the same (conservative settings), but it might be interesting to have a set of values for winter (I live in northern Spain, and the PV produces little during the season).
Just checked my notes and went with 55.2V absorption and 54.4V floating. I am running the same settings I listed in my first post. 48V as shutdown.
However, with the mentioned settings here are a few things you should note:
VRM app SoC% reading slightly differs from the battery pack screen SoC reading. In my case, the app is always slightly higher, by 2-5%, I don’t think it’s concerning and is likely due to the fact that the ESS goes by voltage and can’t directly read the pack SoC, hence the drift.
With the above in mind, and considering that LiFePo4 cells have a flatter voltage discharge curve, when my app SoC reads 10% my battery pack screen is maybe at 5-7% and the voltage reading is about 50.8V. This leads me to believe that the set voltage levels in the BMS (or microcontroller just to show a user-friendly display) are higher than the 48V listed in the specs? Or that the SoC shown in the battery screen is actually showing the “useable” range based off of safer harcoded limits set by them. Maybe that’s to prevent any damage and protect it from user errors? My reasoning is that if at displayed 5% SoC the voltage level is around 50.5V, it seems unlikely I can get as low as 48V without going negative SoC?
The variable most useful I found for “tuning” based on seasons is the charge current. I allow 40A (0.2c) for summer time, because there is plenty of sunlight and it is only a matter of time that the battery gets fully charged, therefore why push the battery to harness all of it early? In winter I set it to 60A (0.3c). We have more overcast days (I’m in the North East), but often there are short windows where the sun actually shines very bright and if I only allowed 40A most of the energy during those peaks is wasted and dumped to feed in. To minimze that waste, I set it higher, although still quite conservative at 0.3c. Note that these values work for my 5kWp and 5kVA inverter, yours may vary but the reasoning may still hold.
Given that at 10% SoC my pack voltage sits at ~50.7, I set in the ESS the min battery SoC to 10% and that’s how I use it. In home assistant I integrate the charge/discharge curves and it kinda makes sense with the rated capacity. So I don’t think I’m leaving that much on the table, despite the higher minimum voltage.
Their customer service is, to put it mildly, not very technical inclined so I gave up on trying to figure out why the heck less than 10% SoC sits at 50V yet the published (safe and) usable voltage range that goes down to 48V. If you ever find this out, or yours behave drastically differently, please let me know! I hope that helped!
I’m also using 5kWp and 5kVA so my installation is very similar. Your experience will be so useful for me.
Regarding the battery itself, are you monitoring something using BMS software?. It seems that the daly BMS is using a custom firmware and for example, in the screens that read data from UART port the only confident value is the voltage (i’m my case SoC % only matches victron MP2 SOC at 100%, if i discharge a bit from the battery the victron always shows and accurate SOC but the screens on the battery easily shows 0%). In my opinion, that custom firmware on the BMS is prepared to manage the battery but not for end user consumition.
I tried to access the BMS a long time ago, just to check and monitor values, but I couldn’t figure it out. Did you manage to do it?
For sure their nmc battery packs come with a Daly BT dongle, but even in those I couldn’t manage to pair the device and access it. I assume the LiFePo ones must’ve used something along the line.
I think that LiFePo4 models comes with 100 Balance Daly model. This BMS comes with builtin bluetooth connectivity. The external screen (that shows Volt and SoC) is attached to UART port to show the values.
In my case I installed from Play Store Daly BMS app. Also i installed the 100 balance app downloading apk from vendor site.
It’s true that in a first moment, no BMS device appeared when i opened the android application. I tried some times and no BMS appears.
After turning off an on the battery with lateral DC breakers (i turned off to connect anderson and power on the Multiplus), the two BMS begins to appears on the apps and finally, It’s possible to connect and for example, verify every cell and take a look for unbalance situations.
Damn! That’s actually very interesting, I wish I’d known this before. I never turned off the breaker, that’s probably why I never got to the see devices show up.
I’ll have to schedule a shut down since I don’t think I can just switch the breakers one at a time, and the MP does need the battery on to keep operating.
Were you able to see all the cells info in both apps or only in one of the two? Do you need to switch the breakers every time for the devices to show up? Or once set up you can monitor the cells from the app without messing with the dc breakers.
Both applications work. No nees to turn off dc switches every time, only stay on bluetooth coverage. 100 balance app maybe is discontinued, it seems that some time ago was published on play store but now, it’s gone and you need the apk download.
Brand is called second life. If they are second life cells, it is not a terrible thing.