LiFePo4 voltage under load

3 tons wrote:
BFL says: “IMO the real LFP figures in the OP graph should only go to the "knee" and use about 12.5v as the "end voltage". That would make the times approx”

Fair enough, but consider too that in actual practice 12.5v occurs somewhere below the chart’s stated 10% SOC - in my case this works out to around 97ish % DOD…
3 tons

I run mine down to about 12.4 on each charge cycle and just let my Redarc re-charge until it goes into float.

BFL13 wrote:

So apparently the SiO2 times include going well below 0% SOC! (confusing!) but the LFP times do not. Is this because LFP cells get damaged if you do that? BMS cut off?

As to the times themselves it seems the LFP is 60 min for 1C but the SiO2 is more like 30 min, and then at 0.5C the LFP is 120 min vs about 60 min for the SiO2. However at 0.2C it is 305min LFP vs about 240 min for SiO2?

IMO the real LFP figures in the OP graph should only go to the "knee" and use about 12.5v as the "end voltage". That would make the times approx:

1C- 40 min, 0.5C 100 min, and 0.2C 270 min

It all seems difficult to make a comparison. I don't have a comparable discharge table for AGMs or FLAs. Maybe because they can't do that at all except once or twice?

the do and they don't go below SOC,the volatge SOC is based on a resting battery not a battery under load. the under voltage sag as I call it can be hard on a battery, this is why GC batteries are better then say optima batteries, and also the difference between cheep GC batteries and expansive ones. the ability to recover from a heavy load. this is usaly due to plate thickness and manufacturing process.
I think this is where the "only use the top 50% of capacity came from so your not dipping into that low point all the time.

also you cant stop the LFP at the knee "for comparison" as this is the feature of the LFP, the rated life is based on 100% dischargeso 3000 to 3500 cycles depending on the manufacture and battery size so by making it easier for comparison you are handy caping one of its features.

pianotuna wrote:
Steve look closely at the right hand graph (stage left). It shows the loads as c=1: c= 0.5: c=0.2

ya weird I cant blow up the pictures big enought to be usefull to me it just looks like a graph of how long it would last at that rate..not the actual voltage under load, but it might be.. haha I would like to see a load no load comparason at different states of charge. and I don't have anything in my camper that would put enough load on my pack to change the voltage haha, it was purly built to make it so I could go 10 days with out sun.

BFL says: “IMO the real LFP figures in the OP graph should only go to the "knee" and use about 12.5v as the "end voltage". That would make the times approx”

Fair enough, but consider too that in actual practice 12.5v occurs somewhere below the chart’s stated 10% SOC - in my case this works out to around 97ish % DOD…
3 tons

The OP LFP graphs seem to use 10v as the "end voltage", which is 0% SOC.

Trying to compare the LFP in the OP with SiO2 discharge table (which is all I have to compare with) for how that all works, and I think I see something but not sure. Maybe it makes sense to some others who can help?



The SiO2 table seems to show how you can pick a time say 60 minutes, and then if you run ever- higher loads in amps you have to go lower in "end voltage per cell". Nothing like that in the OP info for LFP?

The thing is SiO2 resting voltage for 0% is 11.1v (1.85 per cell), but the discharge table goes down to 1.6 per cell or 9.6v

So apparently the SiO2 times include going well below 0% SOC! (confusing!) but the LFP times do not. Is this because LFP cells get damaged if you do that? BMS cut off?

As to the times themselves it seems the LFP is 60 min for 1C but the SiO2 is more like 30 min, and then at 0.5C the LFP is 120 min vs about 60 min for the SiO2. However at 0.2C it is 305min LFP vs about 240 min for SiO2?

IMO the real LFP figures in the OP graph should only go to the "knee" and use about 12.5v as the "end voltage". That would make the times approx:

1C- 40 min, 0.5C 100 min, and 0.2C 270 min

It all seems difficult to make a comparison. I don't have a comparable discharge table for AGMs or FLAs. Maybe because they can't do that at all except once or twice?

Easy to size the battery to accommodate any load. If connecting a 5000 watt inverter certainly there should be resources to drive it.

pianotuna wrote:
I'm fairly sure that a 1 C load is on the high side for any battery technology.

But with relatively huge inverters that can happen.
.

Agreed, the chart on the right is while under a load, but note too that the loads represented are continuous…While a worthy test regiment, not necessarily representative of actual user practice…

3 tons

I'm fairly sure that a 1 C load is on the high side for any battery technology.

But with relatively huge inverters that can happen.

I do know that a 5000 watt inverter in 1996 cost about $50,000.00. Now they sell for peanuts in comparison.

Steve look closely at the right hand graph (stage left). It shows the loads as c=1: c= 0.5: c=0.2

pianotuna wrote:
CA Traveler,

Do those other charts show voltage under load? All the ones I found seemed to just be resting voltage based.

the one looks like voltage at a spicif capacity, the second is just a time as a spicif discharge , nether is voltage under load as it would generaly be a pretty complacated chart and I don't know if I have even seen one for a regular battery.

LFP have lower internal resistance and close to no Pukert (speling that wrong probably but on my way to work soon) so the voltage drop under load vs resting will be close to nothing. I have heard and seen videos of people running inverters and such on 100amp banks to 10% capacity with out triggering the alarm on the inverter so that should tell you something but now you have my curiosity peaked so I'll see if I can find anything.

Steve