wattmatters
Solar Wizard
All batteries have different Peukert exponents. So what?
Mixing Lead and LiFePO4
- Thread starter EmilyAndClark's Adventure
- Start date
wattmatters
Solar Wizard
You bought up the Peukert effect as if it is something to be concerned with. So it is entirely reasonable to ask you why you think it is something to be concerned with.
If you cannot provide an explanation, then why raise it?
My explanation was indicated before. . LFP has no - minimal peukert effect. It stays relatively the same in the flat part of the discharge curve no matter how much current you draw. Yes, there is *some* difference, but it is minute compared to lead. Look at the charts.
Lead acid has a large Peukert effect. The heavier the current you draw from it, the terminal voltage wants to drop faster. Look at those charts - sometimes given by a manufacturer. Take a gander at 0.5C !
If you measure only at pack-level, this will be hidden. If the voltage disparity is large enough, the LFP will be trying to recharge the lower-voltage lead-acid battery at the same time it is discharging. This is not efficient for one thing.
But again -the *best* answer that I am providing is that NO MANUFACTURER, of either chemistry condones this.
Don't test me, or I'll have to use my Simpson's Comic-Book-Guy voice.
Lead acid has a large Peukert effect. The heavier the current you draw from it, the terminal voltage wants to drop faster. Look at those charts - sometimes given by a manufacturer. Take a gander at 0.5C !
If you measure only at pack-level, this will be hidden. If the voltage disparity is large enough, the LFP will be trying to recharge the lower-voltage lead-acid battery at the same time it is discharging. This is not efficient for one thing.
But again -the *best* answer that I am providing is that NO MANUFACTURER, of either chemistry condones this.
Don't test me, or I'll have to use my Simpson's Comic-Book-Guy voice.
Here is an interesting 2016 academic article modeling the hybrid LFP and LA battery in a remote medical clinic: https://juser.fz-juelich.de/record/829025/files/PE_070_Christiane.pdf
Here is a commercial product combining chemistries: https://www.hymer.com/de/en/models/technology-innovation/smart-battery-system.
In terms of commercial products, in addition to Hymer RVs, you will find:
Here is a commercial product combining chemistries: https://www.hymer.com/de/en/models/technology-innovation/smart-battery-system.
In terms of commercial products, in addition to Hymer RVs, you will find:
- This product described as a LFP extension to LA battery, but they only do it in 12v apparently: https://www.bos-ag.com/products/le300/
- This Goal Zero/ Yeti product apparently combines the chemistries: https://www.goalzero.com/shop/yeti-accessories/yeti-expansion-tank/
Investigate whether some of those products / solutions are using mixed-chemistry by simple paralleling, or effectively isolating them by using one first, and then switching to the other.
If they are trying to use simultaneously, then that circuitry is, or should be, a LOT more intelligent than just a simple jumper-cable parallel with maybe a fuse in between.
If they are trying to use simultaneously, then that circuitry is, or should be, a LOT more intelligent than just a simple jumper-cable parallel with maybe a fuse in between.
wattmatters
Solar Wizard
I was noticing today that my Lead begins to contribute once the LiFePO4 gets to ~25% SOC. As LiFePO4 SOC falls, then gradually the lead acid battery contributes more.
I wish I could collect more data but at present I have no way to collect the data on both batteries at the same time. I can do one or the other at the moment. I'm hoping that will change in future.
But it shows when comparing the loads vs what the LiFePO4 supplies to those loads:
I know that's a lot of lines but let me explain:
ORANGE is our load power consumption (10-second intervals) from 8PM last night until 6AM this morning.
I chose that period as there is no solar PV contribution, the battery is supplying the entire load. The load varies as the fridges in each of the three buildings cycle on and off and general base load consumption of various electronics etc.
GREY is the power discharged from the LiFePO4 battery. Discharge is shown as negative value.
YELLOW is the LiFePO4 battery state of charge as reported by the BMS. There is 10.2 kWh of LiFePO4.
BLUE is the difference between the absolute values of Battery Discharge power and the Load power (using a rolling minute average to help avoid unnecessary aliasing noise). It is negative, meaning the battery has to discharge at a high power than the load.
It's the blue line I want to focus on.
From 8PM to about 1AM the blue line stays horizontal and sits at -110 W. That means the battery is discharging 110 W more than the load is drawing.
Some of that discrepancy in DC discharge to AC loads will be:
- inverter self consumption, which is typically 45 W
- DC to AC conversion losses, which are going to be circa 10%, so for discharge average of around 500 W, then about 50 W will be lost
So that means about 15 W is "missing". So where is it going?
Into the lead acid battery, that's where.
Effectively it is a trickle charge from the LiFePO4 keeping the lead acid battery at float. It's about 0.25 A or so. Into a 424 Ah lead acid battery. And that it right about the level of current I measure when I take my current clamp meter to check on the flows in/out of the individual battery banks when the system is in this mode of operation.
So from 8PM up until about 1AM or so the lead acid battery represents a small additional parasitic load.
But from about 1AM we can see that the blue line starts to curve upwards, meaning that the LiFePO4 battery is supplying less and less of the loads demand.
Since the load is still there, it is getting that extra energy from somewhere. IOW the lead acid battery is now supplying a small part of the demand on the batteries.
It starts to contribute a little about when the LiFePO4 batteries fall to about 25 % state of charge. The LiFePO4 contributes the majority of the load, but as they discharge further the lead acid battery picks up more of the demand.
The Lead battery actually provides a nice soft landing for the LiFePO4 batteries as they discharge down to lower levels.
Now in my system I have it set up to swap over to the grid supply if the LiFePO4 batteries drop below 10% SOC. This is because my Lead batteries are there to be a backup energy reserve should we have an outage. They are not there for daily cycling.
Coincidentally we did have a grid outage last night, for nearly 4 hours from about 10:40PM. Since we were operating from the off-grid battery system anyway it didn't affect us. However had the outage continued through to this morning then the Lead Acid battery would have continued to keep us running quite happily and eventually the solar PV would have begun to take over the supply and recharge the battery.
I wish I could collect more data but at present I have no way to collect the data on both batteries at the same time. I can do one or the other at the moment. I'm hoping that will change in future.
But it shows when comparing the loads vs what the LiFePO4 supplies to those loads:
I know that's a lot of lines but let me explain:
ORANGE is our load power consumption (10-second intervals) from 8PM last night until 6AM this morning.
I chose that period as there is no solar PV contribution, the battery is supplying the entire load. The load varies as the fridges in each of the three buildings cycle on and off and general base load consumption of various electronics etc.
GREY is the power discharged from the LiFePO4 battery. Discharge is shown as negative value.
YELLOW is the LiFePO4 battery state of charge as reported by the BMS. There is 10.2 kWh of LiFePO4.
BLUE is the difference between the absolute values of Battery Discharge power and the Load power (using a rolling minute average to help avoid unnecessary aliasing noise). It is negative, meaning the battery has to discharge at a high power than the load.
It's the blue line I want to focus on.
From 8PM to about 1AM the blue line stays horizontal and sits at -110 W. That means the battery is discharging 110 W more than the load is drawing.
Some of that discrepancy in DC discharge to AC loads will be:
- inverter self consumption, which is typically 45 W
- DC to AC conversion losses, which are going to be circa 10%, so for discharge average of around 500 W, then about 50 W will be lost
So that means about 15 W is "missing". So where is it going?
Into the lead acid battery, that's where.
Effectively it is a trickle charge from the LiFePO4 keeping the lead acid battery at float. It's about 0.25 A or so. Into a 424 Ah lead acid battery. And that it right about the level of current I measure when I take my current clamp meter to check on the flows in/out of the individual battery banks when the system is in this mode of operation.
So from 8PM up until about 1AM or so the lead acid battery represents a small additional parasitic load.
But from about 1AM we can see that the blue line starts to curve upwards, meaning that the LiFePO4 battery is supplying less and less of the loads demand.
Since the load is still there, it is getting that extra energy from somewhere. IOW the lead acid battery is now supplying a small part of the demand on the batteries.
It starts to contribute a little about when the LiFePO4 batteries fall to about 25 % state of charge. The LiFePO4 contributes the majority of the load, but as they discharge further the lead acid battery picks up more of the demand.
The Lead battery actually provides a nice soft landing for the LiFePO4 batteries as they discharge down to lower levels.
Now in my system I have it set up to swap over to the grid supply if the LiFePO4 batteries drop below 10% SOC. This is because my Lead batteries are there to be a backup energy reserve should we have an outage. They are not there for daily cycling.
Coincidentally we did have a grid outage last night, for nearly 4 hours from about 10:40PM. Since we were operating from the off-grid battery system anyway it didn't affect us. However had the outage continued through to this morning then the Lead Acid battery would have continued to keep us running quite happily and eventually the solar PV would have begun to take over the supply and recharge the battery.
wattmatters
Solar Wizard
So the concern you are expressing is for efficiency, not safety.
One can choose to trade an efficiency loss for other benefits if they are worth it.
I'm perfectly OK with the small inefficiency of the hybrid. It's certainly no worse than the efficiency of Lead alone.
As I have demonstrated, my Lead battery, which mostly sits at float, adds a small parasitic load of about 0.25 A to the LiFePO4 battery when it is discharging to power loads.
From my perspective this is actually a good thing, for several reasons:
i. given the parasitic load of the 424 Ah of Lead battery is just 0.25 A, it tells me my Lead battery bank is in good condition.
ii. it also means my system is pretty much always keeping my Lead battery in the 97-100% SOC range, which is right where I want it - because it's job is to be a backup source of energy. They are ex-data centre backup units, designed for exactly this purpose, and cost a fraction of new and a fraction of cost of LiFePO4 for the capacity they can supply.
iii. the Lead can provide great support for the LiFePO4, extending the duration we have power coverage for.
iv. it significantly extends the useful life of the Lead battery I already had. Given the level of parasitic load is so small, it's hardly an issue for the LiFePO4. I (mostly) have PV capacity that gets wasted anyway, so that efficiency loss is really neither here nor there.
I have expressed the safety reason before as well. This was just an additional point because here there is apparently no concern for safety.
Look you kids go have fun. Do it safely. You have been warned, not just by me, but real grownups at the manufacturing level.
wattmatters
Solar Wizard
That is not true. There is no disregard for safety.
The safety concern has been dealt with through applying appropriate battery safety measures. Battery management systems and appropriate individual bank fusing along with good system design used in an appropriate manner.
Just because a battery is all one chemistry doesn't make it safe either.
Vigo
Solar Addict
I decided to parallel lead and lifepo4 after reading a website detailing basically wattmatters’ same experience.
The lifepo4 floats and lessens the cycling demands on the lead while still operating in a fairly comfortable range of its own soc.
The only thing you need to do to parallel the 2 is not put a dumb charger to it cranked up to 14+v. If you are fine giving up the top 10-20% of lithium capacity to massively simplify your charging and wiring schemes, it works great. Trying to get as much as possible from both banks is the mentality that leads to complication. Imo.
The lifepo4 floats and lessens the cycling demands on the lead while still operating in a fairly comfortable range of its own soc.
The only thing you need to do to parallel the 2 is not put a dumb charger to it cranked up to 14+v. If you are fine giving up the top 10-20% of lithium capacity to massively simplify your charging and wiring schemes, it works great. Trying to get as much as possible from both banks is the mentality that leads to complication. Imo.
Last edited:
wattmatters
Solar Wizard
It's not they are not welcome, it's they need explaining.
Dumping a comment saying "X is bad" without providing reasonable rationale for making that statement is not welcome.
Making the same statement along with a supporting rationale is welcome.
Just dumping the statement "because Peukert differences" is not helpful and why I asked for further clarification of why that's a problem.
I'm confused as to why anyone would want to in the first place. I've watched several videos, and checked out the "BBMS" Whether or not it is safe, I don't see any advantages, and you are tossing out and ruining the advantages you get from LFP. Namely, high capacity with a light weight, no Peukert (which makes accurate monitoring of SOC possible), and no battery maintenance(watering, periodic equalization, etc.). Why have Lithium in the mix at all if you lose the advantages from it?
You are much better off tossing the lead and having a smaller Lithium battery than a larger hybrid imho. Or if you need the larger battery and can't afford to do it all in Lithium, do it all in lead.
You are much better off tossing the lead and having a smaller Lithium battery than a larger hybrid imho. Or if you need the larger battery and can't afford to do it all in Lithium, do it all in lead.
Now that is a reasonable statement that you don't see any advantages and you perceive some disadvantages, namely a loss of high capacity, continued weight of lead, watering and maintenance of lead, and Puekert effect concerns. In my experience, first, as I perhaps did not explain well enough previously, my off-grid cabin had 330Ah @ 24v, of FLA. It was less than two years old, and while I can afford whatever I want, I am not inclined to toss it out when it has so much life in it. Others may feel differently. Rather, I was wishing to add capacity and also to get the advantages of Lifepo4, without ruining a perfectly good battery. So, I added 200Ah in parallel with the FLA, with a device (BBMS) that enables them to be joined the vast majority of the time but disconnected when the Lifepo4 is full and the controller charging, which allows the charging voltage to increase when it drops off so as to top off my FLA per normal parameters. When done charging the FLA, the voltages come back together and it again joins them. Since the FLA is only 165Ah in practical terms (limited to 50 percent discharge), this more than doubles my capacity. You expressed concern that somehow this loses the advantage of "high capacity" found with LifePo4. I am not sure how you would even come to anticipate that, but I am open to clarification if you wish. For me it is just the opposite in that I have higher capacity. With regard to the dreaded ol' Puekert effect, unlike when using solely FLA there is little voltage sag because the Lifepo4 in the mix discharges more rapidly most all of the time. Only if I were to have a very, very high amperage draw, or if I were to take my Lifepo4 very, very low would the FLA come to predominate and result in some significant FLA-style Puekert effect voltage sag you expressed concern about. To be clear, I don't experience that in regular use; rather, voltage remains quite stable. So, I get both higher capacity and no Puekert concerns, and while I could see that as a hypothetical concern, mixing them doesn't ruin those LiFePo4 characteristics at all in my experience. Additionally, the FLA stays nearly fully charged most of the time, and the trickle from the Lifepo4 overnight into the FLA is very minimal. No big energy loss due to that fear some express. Also I anticipate that trickle will only benefit the FLA in the long run. Addressing your other expressed concerns, it is true I still have to monitor the water in my FLA. That is a valid point based on my experience. However, since this is a stationary installation, the weight concern you expressed is of no import, but I understand the issue for mobile applications which is why in my vehicle based system I use solely LifePo4. In summary, of your worries about the LifePo4 advantages you fear would be lost, I can validate only the remaining higher weight and having to water the FLA occasionally. With regard to a loss or ruination of the LiFePo4 advantages high capacity, and of the small Puekert effect, these have simply not been born out by my experience. Now, in my special circumstance, I might add that I have additional advantages in that I am on the central plains and have a very good small wind turbine that works very well with my FLA. Such wind generators are in essence traditional alternators, which you may know cause havoc for Lifepo4 (at least without the additional circuitry of newer smarter alternators, I am told). In my circumstance, leaving lead in the circuit allows that lead to do all that it should vis a vis that turbine, dampening the effect of a variable voltage source. In fact it runs even better now because wind gusts do not push the voltage up as drastically with the lower resistance of the Lifepo4 (again, good riddance to significant Puekert effect). It essentially allows more room for that rapid charge of higher winds and results in much less braking because voltage doesn't spike.
I hope that explains how, in reality, it is true that a few advantages of an all LifePo4 battery are diminished with a hybrid system. Weight remains with lead in the mix, as does watering. But the other expressed concerns of losing capacity and keeping some effects of Puekert are either non existent or not a significant concern. And there are some advantages related to alternator charging, and keeping the lead you have in service until it's eventual death. I anticipate adding a smaller amount of lead when my FLA does eventually die for the reasons stated.
I hope this clears up the confusion for you. Of course, this setup may not be for all or even many, but for some significant lot of us it is ideal. In any such setup, the banks have to have proper over current protection devices (separately fused) to prevent malfunction of one causing any damage to the other.
Last edited:
My capacity statement was in regards to buying a bank that is all lfp, vs one that is physically the same size, but hybrid. Many installations have a space constraint. I do, and the switch to lfp tripled my usable capacity in the same space.
My issue with Puekert isn't what you discuss. Its that because of Puerkert it is practically impossible to know the state of charge of a LA battery. You _think_ you are using 50%(or whatever you think you use) of it. But without a whole lot of work and very sophisticated monitoring and regular capacity measurements and calibration, i promise you, you are wrong. With LFP knowing the SOC is easy, until you parallel a LA and screw that up. This is really the biggest issue for me . Its very real you might think you have 50% total capacity left, then all of the sudden your inverter turns off I would rather have a smaller battery and be certain what's left.
Regarding maintenance, you don't appreciate that yet. But i switched from FLA to LFP. I 100% ignore it now, except for an occasional glace at my SOC. Its been a year since i bothered with anything. I'll never go back. FLA is agonizing to tend to. But if you are used to it, you don't appreciate yet what you're missing.
wattmatters
Solar Wizard
I agree it's not a suitable solution in most circumstances.
But just because you may not have thought of one doesn't mean such a scenario doesn't exist. Perhaps you haven't considered all the possibilities. I know I haven't.
So let me explain my individual situation.
We need outage backup. We average a dozen longer outages a year (I have years of stats on this). We had a petrol generator and it works well however it was no good if I am not here, as my wife and my elderly frail mother (who lives in a home we built for her on my property), are just not capable of moving and operating a petrol generator.
Hence a desire to build an automated, self-fueling system, i.e. battery + solar PV.
I installed a small 4 kW AIO inverter and used the same transfer switch we already had installed for the petrol generator we used to use as backup so that our home's circuits can operate from backup when needed, excluding some non-essential and/or high power draw appliances.
I also already had my Lead batteries. Their sole purpose is to be a reserve of energy storage for grid outage coverage. They are not used for daily cycling. The Lead batteries I have are designed precisely for this purpose - they came from a telco/data centre (who swap these backup batteries out on a regular schedule). High quality but cheap. Enersys SBS units. 20 kWh of them.
To keep the batteries topped up and to extend the capacity of the system to operate during long outages, I installed a small 2.22 kW off-grid PV array on my garage. I gave a second life to some used 375 W panels I picked up, along with some used solar PV racking.
So now I had a fully functioning PV + battery system, ready to supply power when we needed it during outages. And use it we have.
.
But because the PV array had more capacity than was needed to keep the batteries at float then it represented an opportunity to run some loads from it, at least during the daytime. So I started with the pool pump. Moved it over to operate from the off-grid system during the day. Perfect, the back up battery bank is not affected and the pool pump now runs off-grid.
But the off-grid PV array was still under-utilised while my system was quite capable of doing more daily work. But the loads I was most interested in moving off-grid were overnight loads when I was importing power from the grid.
All it needed was some storage capacity more suited to daily cycling.
So I added some LiFePO4. It works out to be a perfect solution. The 10 kWh of LiFePO4 now does the daily cycling, while the 20 kWh Lead acid bank just sits there, always topped up ready to rumble if we have an extended outage.
It's working really well. The house now runs through the off-grid system. During the day it runs in Utility pass through mode and that power is supplied by my grid-tied PV system, while at night it runs from the off-grid LiFePO4 battery system, which gets recharged by the off-grid array during the day. Meanwhile the Lead acid bank just sits there, ready to be called upon if ever we need it.
I can cycle the LiFePO4 as much as I like and have zero concern about whether I have enough capacity to cover us in an outage. Keeping sufficient LiFePO4 in reserve would be a monumental waste of money and resources when I already have the Lead acid bank sitting there.
.
So why not a grid-tied battery system?
Because of the way things work here, that would cost nearly 4 times what the off-grid storage I have does. Grid tied storage here makes zero economic sense but using the off-grid system does make sense.
wattmatters
Solar Wizard
For the recent 2.5 month period I have detailed data, our grid imports this year compared with the same period for the previous three years have dropped by an average of 8.8 kWh/day. That's largely due to just adding some LiFePO4 to our existing backup system, originally built from a low cost AIO inverter, some used panels, racking and batteries. It's also due to diversion of PV energy to our hot water system.
Appreciate what you're saying regarding capacity and space to store batteries. In my case, I already had the lead in my cabin's garage and some space for a bit of Lifepo4 was not hard to come by. I suppose if I were trying to increase capacity all within the same space it would not fit, adding an additional Lifepo4. Still, my capacity more than doubled and it just took a little more space in my garage. Had I replaced it all with LifePo4 then it would all have fit in my FLA space. Space requirements may be an issue in a particular case, just not in mine. Regarding SOC concerns, I don't have any. My Lifepo4 BMS shows it's SOC accurately, and I know the voltage of the system and where the lead is in relation to that until Lifepo4 is nearing depletion. The flatness of the Lifepo4 is beneficial of course in that my LA is largely preserved throughout use of the Lifepo4. So the sudden unexpected realization of LA depletion scenario is not a concern for me and hasn't occurred. I never depended on a shunt to know capacity remaining of my LA before anyway, so it's no different. Others may need that in their setup, granted. Forgetting battery watering sounds great. I do, in fact, experience that in my mobile system, just not at my cabin, so I appreciate what you're saying about giving up that task. I guess it's the price I pay for the overall benefits of a mixed system.
I've had FLA only (we probably all have), Lifepo4 only (my mobile application), and mixed (at my off grid cabin). The alarming worries and fears expressed about the latter scenario are, in my experience, overblown, and they don't seem to come from actual experience. Of course, this particular system architecture might not suit your needs because of space or weight limits, a desire for very low maintenance requirements, etc. It functions very well in some applications, however. As always, proper over current protection of each battery bank is critical.
Not overblown - like playing chess, think one move beyond the immediate:
If your boat burns down, and the inspectors see a hybrid-bank, even though it wasn't the cause, blame it and deny your coverage? Does the ABYC or any other worldwide marine entity condone this type of use?
Use at your own risk. It's a risk you are comfortable with, and not my job to sway you. I'm just very uncomfortable with it, because far too many of my friends have said "overblown" before an incident with either their wallet or the hospital.
If your boat burns down, and the inspectors see a hybrid-bank, even though it wasn't the cause, blame it and deny your coverage? Does the ABYC or any other worldwide marine entity condone this type of use?
Use at your own risk. It's a risk you are comfortable with, and not my job to sway you. I'm just very uncomfortable with it, because far too many of my friends have said "overblown" before an incident with either their wallet or the hospital.
wattmatters
Solar Wizard
Similar threads
