LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

[evm1024]

Originally Posted by gscriba

For those using them, how low do you go with your voltage? SNIP %<

Sounds like you are on top of it. I know you are more interested in actual experience rather than generalizations so I'll be specific based on my experience.

For the longest time I babied my batteries and tried to keep them between 80% SOC and 40% SOC. It was a pain. Then it dawned on me that these were actually very tough batteries. So

So, I decided to use Victron's recommendation for non Victron LiFePO4 batteries. Charge to 14.2 (3.55 VPC), an hour absorb time and them "float" at 13.3 (3.32 VPC).

At 13.3 volts the typical current is at or near zero.

When I'm not going to be on the boat for any length of time I charge to 14.2 then shut off the charger and let the idle current from the BMS, Victron monitor and WiFi router use up some amps. That base current runs 3 or 4 amps. With the Victron monitor (VenusGX) I get battery state sent to Holland where I can access it from their portal (see attached image). I can also enable the charger remotely if needed.

At least one of my cells has reduced capacity and I've tried a number of differing charge/discharge cycles to see if I could recover any capacity (without success)/ It appears that there was a loss of active area in the "bad" cell. What this means is that I've been monitoring the cells capacity and have not noted any change using Victron's recommendations. Of course it needs a few years more time.

The plan if I was going to be away for months is to charge the pack to 100% SOC (14.2 volts) then I take out about 200 AH with a space heater. THis puts my cells at about 60% SOC. If they had full capacity like yours I would still take out 200 AH for a 70% SOC. And then disconnect the pack from everything. I could then switch the AGM start battery in place of the house bank (just a few switches to flip) and set the multiplus to AGM parameters and monitor the boat from the Victron Portal.

In any case if you charge fully to your normal voltage and create a discharge curve at your typical current draw you will be able to get a voltage vs SOC (AH draw) graph.

The point is that you would have a halfway good idea that at xx volts you are at (oh say) 40% and start recharging there. I think that 40% SOC is a good number.

Part of the reason for my waffaling on a specific number is that the difference in voltage of a few millivolt (5 or 10) is the difference between 80% SOC and 40% SOC and we typically do not have that kind of accuracy min measurement. But with a curve that you construct based on your own voltage measurement and SOC you can get good numbers between SOC and your measured voltage.

Right now the power is off at my boat so I've been running on battery for a few days. I've been rebuilding the icebox and using battery power over the last weekend. Attached is the discharge graph with about 166 AH taken out for a SOC of about 70%. I'll get down to the boat on Saturday and flip the shore power breaker back on. The shore power pedestal got soaked by a wave in the last storm and has started nuisance tripping every once in a while. No big deal.

Darn, After re-reading I did not state vary well. When using or working on the boat I charge to 100% SOC and then float at 13.3 volts. When the SOC gets to 40% ( of say 13.2 volts) I consider recharging. I go by SOC.

[CatNewBee]

Originally Posted by gscriba

For those using them, how low do you go with your voltage? Love my 700 ah Winstons (4 cells/700 ah each) and REC-ABMS. With a typical slow draw of anywhere between 2.5-10 amps, there must be some general cell voltage "guideline" of when to recharge that others are using? At present time, I'm recharging using my Xantrex SW 3012 with a bulk and absorption voltage of 14.6v, two stage charge, no float.
As you can see from the picture, three days following a charge (Xantrex charger cuts out at 14.6 as measured by the Xantrex, which is higher than what the REC-ABMS is measures at cut off time), the REC currently measures 13.01v for the bank, 3.257v per cell. I have attached the Winston voltage curves.
Interestingly and as a side note, the Simarine monitor is anywhere between 0.14 and 0.2v lower than the REC. So REC measures 13.01v and Simarine measures 12.86v.
So, at what point with a slow draw, should I recharge? When the REC is measuring 3.1v at the cell level? 3.2v? I am considering the REC to be the most accurate due to it's measurement of the individual cells.
Thanks in advance.

You have 296Ah drawn, SOC is as stated. Charge whenever you want to, set absorption to 14.4 or 14.5V, Float to 13.5V and you will be fine. Your Xantrex shows higher voltages while charging because of crappy wiring, connections and long distance, the higher the charge current, the bigger the voltage drop between the charger and the battery. High capacity charger usually use separate sense wires at the battery to get the real battery voltage. Not sure about the Xantrex wiring capabilities.

The voltage from REC is at the cells, all subsequent voltage measurements are behind the shunt, and will be slightly different when a current is flowing, either higher while charging or lower while discharging.

[CatNewBee]

I assume, you know all values on the display from REC, but for the others here, starting from the top

REC with active connection to the unit 1A-1134,
SOC 61%, 296Ah drawn,
Pack Voltage 13.01V, Current - 6.20A, discharging,
Bus number 2, CMAX 3.257V (highest cell, cell 2)
Bus number 2, CMIN 3.250V, (lowest cell, cell 3)
Bus number 2, TEMP 14.0°C, SENSOR 2,
Time to empty est. 66h 03min
Current power draw - 0.08kW
SOH 99%

Then you have overview, cell voltage details, cell resistance details, cell temperature details, BMS temperatur details, settings and error status window.

[gscriba]

Good idea to summarize.
Thanks.

[newhaul]

Ok guys for those that are actually using Lfp and a shunt type battery monitor I need input.
I'm thinking of adding one but most are to expensive to be considered by me
So I need opinions on this monitor
https://www.aliexpress.com/item/3294...7764499cqd2JU0

[john61ct]

If it doesn't have inputs to take Peukert, CEF etc into account, I wouldn't call it a true bank monitor.

There are lots of Ah meter, coulometer, coulomb counters available in that low price range.

For reliability / longevity, this one apparently does well, if you don't need higher ampacity

https://powerwerx.com/watt-meter-ana...e-dc-powerpole

although you may find it cheaper elsewhere, they are a trusted vendor.

If you like the "panel meter look" of yours, I'd go for the eBay seller for a much better / longer guarantee, probably wouldn't even need to send it back if you filed a claim

https://www.ebay.com/i/223649048058

[john61ct]

And just for comparison, Alan @Bay Marine's listing for a real BMV-712, might be a little less off his website direct, excellent warranty and customer support.

https://www.ebay.com/itm/Victron-BMV...g/173004820450

Just cook at home for a few days maybe a week to pay for a device a full order of magnitude higher quality.

Way I'd go anyway. . .

[newhaul]

Originally Posted by john61ct

If it doesn't have inputs to take Peukert, CEF etc into account, I wouldn't call it a true bank monitor.

There are lots of Ah meter, coulometer, coulomb counters available in that low price range.

For reliability / longevity, this one apparently does well, if you don't need higher ampacity

https://powerwerx.com/watt-meter-ana...e-dc-powerpole

although you may find it cheaper elsewhere, they are a trusted vendor.

If you like the "panel meter look" of yours, I'd go for the eBay seller for a much better / longer guarantee, probably wouldn't even need to send it back if you filed a claim

https://www.ebay.com/i/223649048058

the same meter and yes I would get from eBay just posted aliexpress page due to easier to read . All data is an easy scroll without extra advertising .

Hence why I asked for input from actual users of the various monitors.

[newhaul]

Originally Posted by john61ct

And just for comparison, Alan @Bay Marine's listing for a real BMV-712, might be a little less off his website direct, excellent warranty and customer support.

https://www.ebay.com/itm/Victron-BMV...g/173004820450

Just cook at home for a few days maybe a week to pay for a device a full order of magnitude higher quality.

Way I'd go anyway. . .

way to expensive for me . That bvm 712 costs more than a years fuel for my home/ boat.

[john61ct]

Originally Posted by newhaul

Hence why I asked for input from actual users of the various monitors.

That particular one only showed up a few months ago, so you might need to be the first if you like the look of it.

I've seen some other cheap-chinese types in the under $20 range if that really makes a difference for you.

The ampacity you need makes the biggest difference, the smaller rated shunts give a better accuracy.

[evm1024]

For me accuracy is the most important thing. That and consistent results.

I recall that you have a LiFePO4 house bank so the various corrections (Peukert, CEF) are of less importance than with a lead Acid bank. This is of course true because LiFePO4 Peukert and CEF are both close to 100%. Still a WAG but a "better" WAG.

I just upgraded to a BMV-702 from a BMV602s for networking compatibility issues.

Both the BMV602s and the BMV712 (lower current draw than the BMV702) are great kit. And again for me the ability to count electrons accurately is the most important thing. Voltage accuracy is important too but I always check my monitors vs my Fluke. YMMV

John's recommendation of the Victron BMV712 is good

Originally Posted by john61ct

Way I'd go anyway. . .

And I presume he uses that on his Beneteau 36.

[newhaul]

Originally Posted by evm1024

For me accuracy is the most important thing. That and consistent results.

I recall that you have a LiFePO4 house bank so the various corrections (Peukert, CEF) are of less importance than with a lead Acid bank. This is of course true because LiFePO4 Peukert and CEF are both close to 100%. Still a WAG but a "better" WAG.

I just upgraded to a BMV-702 from a BMV602s for networking compatibility issues.

Both the BMV602s and the BMV712 (lower current draw than the BMV702) are great kit. And again for me the ability to count electrons accurately is the most important thing. Voltage accuracy is important too but I always check my monitors vs my Fluke. YMMV

John's recommendation of the Victron BMV712 is good



And I presume he uses that on his Beneteau 36.

yes I am Lfp soon to be 240ah up from 100 i have now 100% is not needed but within a couple % is good . I don't need one my solar controller has voltage but for a weeks fuel I figure why not. Shouldn't take long to get my mental chart figured out.

I don't need the network connectivity hence a cheaper option is available.

[john61ct]

Well respected vendor here in the hobby community

https://www.amazon.com/GT-Power-Anal.../dp/B00C1BZSYO

Similar, rated for 150A
https://www.amazon.com/Rcharlance-An.../dp/B07DL3MY21

[john61ct]

Originally Posted by newhaul

yes I am Lfp soon to be 240ah up from 100 i have now 100% is not needed but within a couple % is good.

Even those costing hundreds are often off by 6-8%.

Of course, begs the question, "how do you know?"

Stick to just one, so you don't need to fret about which one's closer 8-D

[newhaul]

Originally Posted by john61ct

Well respected vendor here in the hobby community

https://www.amazon.com/GT-Power-Anal.../dp/B00C1BZSYO

Similar, rated for 150A
https://www.amazon.com/Rcharlance-An.../dp/B07DL3MY21

those in line units don't look good when installed where they are visible hence the round inset bezel units:-):-)