50% discharge on AGM batteries - old wife's tail

[SamDeleted]

This guy has an interesting take on the 50% discharge rule

AGM battery Depth of Discharge myth busted

I have been wondering for some time if the "don't discharge your batteries beyond 50%" rule really applies to AGM batteries. AGM batteries claim to be deep discharge, and 50% discharge doesn't seem very deep to me! So I did a little research...

tab-rv.vanillacommunity.com

Quote:

"I have been wondering for some time if the "don't discharge your batteries beyond 50%" rule really applies to AGM batteries. AGM batteries claim to be deep discharge, and 50% discharge doesn't seem very deep to me! So I did a little research...

If you look at the data for Odyssey AGM batteries, you find that they are good for 630 discharges to 50% but only 400 discharges to 80%. Seems obvious that the 50% depth of discharge (DoD) is better. But hang on, if you flatten your batteries to 80% DoD, you won't have to do the recharging as often, so fewer cycles. The important parameter is your lifetime capacity in Amp-hours, which you get by multiplying the battery capacity * DoD * number of discharges. For a 225 Ah battery, your lifetime capacity is 71,000 Ah if discharged to 50%, but is 72,000 Ah if discharged to 80%. Indeed, If you look at Odyssey's data for 100% DoD, you still get 72,000Ah of lifetime capacity! For Odyssey batteries, there is absolutely no reason to worry about your DoD, you're not gaining anything by starting up that generator when your battery gets to 50%.

But we don't have Odyssey batteries, we have Harris batteries for which there is a dearth of information about its discharge cycles vs. DoD. So instead I created an "Average" battery from Trojan, Concorde, Odyssey and Rolls AGM products (these were the only ones I could find with detailed DoD data)."





"I've included the 3 data points for the Harris battery, and you can see that its number of discharges under-performs the average, but has much the same overall shape. So now you can see how DoD relates to total capacity for our average AGM battery:"




"So going from 50% DoD to 80% DoD will just lose you 6% of your lifetime output from your battery. Not really worth worrying about.

Take home message: If you're using AGM batteries, don't get too fixated upon getting them recharged once you get to 50% depth of discharge. Any time up to 80% is good. More important is not to leave them discharged any longer than you have to."

 

[OzSolar]

Yep. LEU (Lifetime energy units) is really the metric you should use when sizing a battery bank and analyzing it's costs.

If I've said it once, I've said it thousand times. 99% of batteries die of unnatural causes long before cycles get them, regardless of DOD.

A client with a remote off grid property asked me about this recently. I put together this chart for him. They went with FLA.

 

[sunsurfer]

Yep. LEU (Lifetime energy units) is really the metric you should use when sizing a battery bank and analyzing it's costs.

If I've said it once, I've said it thousand times. 99% of batteries die of unnatural causes long before cycles get them, regardless of DOD.

A client with a remote off grid property asked me about this recently. I put together this chart for him. They went with FLA.

Hey Oz, Have you ran an LFP bank off-grid yet? I have used lead for many years and went to an LFP bank last year. I could never go back to lead now. The performance is just to good with LFP. You just can't beat it for daily cycling.

The issue for me isn't just cost, but how much of available sun is captured during the day for later use. With lead, so much solar gets missed when charging with its higher resistance. I still have a large 24v lead acid bank on a second system, very similar capacity to my LFP bank. The LFP blows it away in efficiency. Pair them with a fast SCC and it makes a big difference in what you can collect in a few hours. It really helps with those low/partial solar days.

Lead acid has served well for many decades but I prefer LFP now. I'm afraid lead's days are numbered.

Its like going from a flip phone to a smart phone.

 

[OzSolar]

Hey Oz, Have you ran an LFP bank off-grid yet? I have used lead for many years and went to an LFP bank last year. I could never go back to lead now. The performance is just to good with LFP. You just can't beat it for daily cycling.

The issue for me isn't just cost, but how much of available sun is captured during the day for later use. With lead, so much solar gets missed when charging with its higher resistance. I still have a large 24v lead acid bank on a second system, very similar capacity to my LFP bank. The LFP blows it away in efficiency. Pair them with a fast SCC and it makes a big difference in what you can collect in a few hours. It really helps with those low/partial solar days.

Lead acid has served well for many decades but I prefer LFP now. I'm afraid lead's days are numbered.

Its like going from a flip phone to a smart phone.

I think LFP is great! I've done about a dozen off grid systems with LFP and still can't get over how stable the voltage is under load. And I don't think anyone will miss adding water to cells. I have 20kWH of SOK server rack batteries ready to go but I still have 50kWh of FLA that do all I need them to.

In general, LFP does most things better, but its Achilles heel is operating it below 32f. In most applications that doesn't matter but there are some where it really does. Sure, you can get heated batteries but that's an added complexity and cost. (I'm a KISS sort of person)

The main point I was trying to make is that the 50% rule for lead acid is a fallacy. The other point is that if you actually study the $/LEU of LFP and high-quality FLA they are around the same. *Ignoring maintenance, weight and area needed for FLA.

What no one talks about is that lead acid batteries are very recyclable whereas LFP's are not or at least they weren't latest I knew.

 

[SamDeleted]

Also the price of used lead acid is fantastic compared to Lithium, Oz your figures at the top quoted $250/kWh

I just paid £50/kWh for lightly used, less than 6 months old gel traction batteries


5x cheaper , changes everything

 

[DIYrich]

30% federal tax credit today. Who knows what it will be when time to replace FLA? Who knows the replacement cost in x years? Need to look at the life cycle cost over the same period of time.

 

[pollenface]

Interesting info. My AGM's probably see 10-15% DOD on a daily basis. 4yrs in so far... no signs of capacity loss that I have been able to notice.

 

[Tomthumb62]

This guy has an interesting take on the 50% discharge rule

AGM battery Depth of Discharge myth busted

I have been wondering for some time if the "don't discharge your batteries beyond 50%" rule really applies to AGM batteries. AGM batteries claim to be deep discharge, and 50% discharge doesn't seem very deep to me! So I did a little research...

tab-rv.vanillacommunity.com

Quote:

"I have been wondering for some time if the "don't discharge your batteries beyond 50%" rule really applies to AGM batteries. AGM batteries claim to be deep discharge, and 50% discharge doesn't seem very deep to me! So I did a little research...

If you look at the data for Odyssey AGM batteries, you find that they are good for 630 discharges to 50% but only 400 discharges to 80%. Seems obvious that the 50% depth of discharge (DoD) is better. But hang on, if you flatten your batteries to 80% DoD, you won't have to do the recharging as often, so fewer cycles. The important parameter is your lifetime capacity in Amp-hours, which you get by multiplying the battery capacity * DoD * number of discharges. For a 225 Ah battery, your lifetime capacity is 71,000 Ah if discharged to 50%, but is 72,000 Ah if discharged to 80%. Indeed, If you look at Odyssey's data for 100% DoD, you still get 72,000Ah of lifetime capacity! For Odyssey batteries, there is absolutely no reason to worry about your DoD, you're not gaining anything by starting up that generator when your battery gets to 50%.

But we don't have Odyssey batteries, we have Harris batteries for which there is a dearth of information about its discharge cycles vs. DoD. So instead I created an "Average" battery from Trojan, Concorde, Odyssey and Rolls AGM products (these were the only ones I could find with detailed DoD data)."



View attachment 153294

"I've included the 3 data points for the Harris battery, and you can see that its number of discharges under-performs the average, but has much the same overall shape. So now you can see how DoD relates to total capacity for our average AGM battery:"


View attachment 153293

"So going from 50% DoD to 80% DoD will just lose you 6% of your lifetime output from your battery. Not really worth worrying about.

Take home message: If you're using AGM batteries, don't get too fixated upon getting them recharged once you get to 50% depth of discharge. Any time up to 80% is good. More important is not to leave them discharged any longer than you have to."

He’s referencing Odyssey AGM batteries. Those are not typical AGMs and yes do have ability for deeper discharge. Most/many cheaper AGM are not as robust. I think last time I checked, a 135Ah Odyssey went for $450-500. At those prices, unless you need lead acid long periods of freezing temps, lithium is a better bang for the buck.

 

[SamDeleted]

He’s referencing Odyssey AGM batteries. Those are not typical AGMs and yes do have ability for deeper discharge. Most/many cheaper AGM are not as robust. I think last time I checked, a 135Ah Odyssey went for $450-500. At those prices, unless you need lead acid long periods of freezing temps, lithium is a better bang for the buck.

not just odyssey, also Trojan, Concorde, and Rolls AGM.

And the agm DoD vs lifespan graph he showed is very similar to other manufacturers you can find online

 

[SamDeleted]

At those prices, unless you need lead acid long periods of freezing temps, lithium is a better bang for the buck.

Looking at Oz's figures , lead acid Vs LiFePo4 per lifetime energy unit:

Lead acid $0.089

LiFiPo4 $0.071

So yea like you said better bang for buck , but not as bigger gap as some might believe....

the real problem with lithium is the huge up front costs, literally double

 

[stienman]

Yep. LEU (Lifetime energy units) is really the metric you should use when sizing a battery bank and analyzing it's costs.

This ignores the energy loss in charging, as suggested here:

With lead, so much solar gets missed when charging with its higher resistance.

The round trip efficiency of lead acid is 80%, while lifepo4 is 90%.

Figure that into the equation and it comes a lot closer.

 

[OzSolar]

This ignores the energy loss in charging, as suggested here:

The round trip efficiency of lead acid is 80%, while lifepo4 is 90%.

Figure that into the equation and it comes a lot closer.

Good point.

I thought LFP was over 95% so it might even be a bit worse for lead acid on the big picture.

Charge efficiency for lead acid also gets worse as the temps drop, but not appreciably until temps get below 45F (from a shaky memory). If we really want to get into the weeds we'd also need to keep that in mind. What happens to LFP R/T charge efficiency with temp?

My gut tells me that glass (solar panels) is much cheaper than battery kWH. So if you've got the real estate for more panels then lead acid will pencil out.

How would would the formula look?

I think like this.
Rated AH x Nominal Voltage (which is way more squirrelly with LA than LFP) x RT efficiency x cycles = LEU
Battery Cost / LUE = $/LEU

I'll work on the PV part later.

 

[HarryN]

There is a pretty wide range of values for AGM internal resistance for lower end vs higher end batteries.

The higher end ones are pretty close but the cheap ones are not even in the same category.

I mostly use the Lifeline 27Ts for a variety of reasons (including space and ability to lift them ) but they are a completely different beast than a $100 AGM battery performance wise. ( when needing to use an AGM vs LiFe)

I am interested in following this thread to see where it goes for round trip efficiency and improving how to use them.

The data sheet / user guide is amazingly detailed and they are a great resource to chat with on the phone a well.

 

[Zwy]

Justifying a lead acid battery for use in a PV system is an effort in futility.

There are a number of threads in the Battery section of the forum where lead acid batteries died the premature death. The reason is lead acid, and it does not matter if FLA, AGM, or GEL, is the absorption charge takes time due to increasing battery resistance as 100% SOC is reached. Usually the battery never hits 100% SOC because the sun tends to go below the horizon at the end of a day before the absoprtion charge is complete. The damage begins the very first time this happens and just accelerates from there.

If you want to sell someone a lead acid bank based on cost, you are misleading them. The bank will need to be over twice, almost 4 times the Ah size of comparable LFP batteries. The reasons are the depth of discharge past the 50% rule causes hard sulfation on the plates. Chemistry is still chemistry and not some fantasy. Second, the lead acid bank actually needs 2 banks within it where one bank is charging while the other is being used. The bank charging should be run fully thru absorption before being used and occasional EQ if the manufacturer specifies EQ needs to be performed. Once charging is complete, the banks are switched and the process repeats. Why should it be done this way? Because the sun still sets everyday and 100% SOC most likely will not be reached due to the slow charge rate. That charge rate is determined by the battery; large arrays won't help, multiple or high amperage chargers won't help either. It takes time which is the major limiting factor because the sun doesn't produce once it sets. For longest battery life, the 100% SOC should be reached on each charge cycle to convert normal soft sulfation back to lead and sulfuric acid. If this does not occur, eventually hard sulfation forms and the early death cycle begins.

Using LFP on a PV system is a no brainer. LFP likes partial SOC and will last longer if 100% SOC is not reached consistently. LFP allows full charge rate to bulk charge voltage setting. It's all about how much PV yield can be produced to flow into the battery. Charge rate is determined by what can be produced and not limited by the battery internal resistance. This allows full SOC to be reached if desired on a daily basis if there is enough yield available.

 

[SamDeleted]

Justifying a lead acid battery for use in a PV system is an effort in futility.

There are a number of threads in the Battery section of the forum where lead acid batteries died the premature death. The reason is lead acid, and it does not matter if FLA, AGM, or GEL, is the absorption charge takes time due to increasing battery resistance as 100% SOC is reached. Usually the battery never hits 100% SOC because the sun tends to go below the horizon at the end of a day before the absoprtion charge is complete. The damage begins the very first time this happens and just accelerates from there.

If you want to sell someone a lead acid bank based on cost, you are misleading them. The bank will need to be over twice, almost 4 times the Ah size of comparable LFP batteries. The reasons are the depth of discharge past the 50% rule causes hard sulfation on the plates. Chemistry is still chemistry and not some fantasy. Second, the lead acid bank actually needs 2 banks within it where one bank is charging while the other is being used. The bank charging should be run fully thru absorption before being used and occasional EQ if the manufacturer specifies EQ needs to be performed. Once charging is complete, the banks are switched and the process repeats. Why should it be done this way? Because the sun still sets everyday and 100% SOC most likely will not be reached due to the slow charge rate. That charge rate is determined by the battery; large arrays won't help, multiple or high amperage chargers won't help either. It takes time which is the major limiting factor because the sun doesn't produce once it sets. For longest battery life, the 100% SOC should be reached on each charge cycle to convert normal soft sulfation back to lead and sulfuric acid. If this does not occur, eventually hard sulfation forms and the early death cycle begins.

Using LFP on a PV system is a no brainer. LFP likes partial SOC and will last longer if 100% SOC is not reached consistently. LFP allows full charge rate to bulk charge voltage setting. It's all about how much PV yield can be produced to flow into the battery. Charge rate is determined by what can be produced and not limited by the battery internal resistance. This allows full SOC to be reached if desired on a daily basis if there is enough yield available.

Interesting, lot of good point about lead acids

The reasons are the depth of discharge past the 50% rule causes hard sulfation on the plates.

Far be it from me to argue, but seems the datasheets say different. according to rolls & several other big manufacturers, 100% cycles only degrade the battery by 14% more when compared to 50%soc cycles

 

[RCinFLA]

Flooded lead acid batteries designed for deep discharge are built with thick plates over their support grid. During discharge, acid concentration of electrolyte is depleted before lead to support grid structure is compromised, at least until they get old with a lot of lead plates eaten away, dropping to battery bottom below plates, taking some sulfur with it reducing overall electrolyte acid concentration at full charge.

This is not the case for 'marine' batteries with thin plates where deep discharge quickly starts to attack lead to grid support structure bonding.

AGM's have no support grid for plates. They use flexible lead plates compressed between the aggregated glass matt separators to support the plates. They are classified as 'electrolyte starved' design with only the glass matt separators holding the electrolyte, like a sponge.

AGM's can still suffer from deep discharging causing some of plates' areas to become thin causing high resistance and some total islanding of sections of lead plates taking the islanded section out of battery operation.

 

[HarryN]

Justifying a lead acid battery for use in a PV system is an effort in futility.

There are a number of threads in the Battery section of the forum where lead acid batteries died the premature death. The reason is lead acid, and it does not matter if FLA, AGM, or GEL, is the absorption charge takes time due to increasing battery resistance as 100% SOC is reached. Usually the battery never hits 100% SOC because the sun tends to go below the horizon at the end of a day before the absoprtion charge is complete. The damage begins the very first time this happens and just accelerates from there.

If you want to sell someone a lead acid bank based on cost, you are misleading them. The bank will need to be over twice, almost 4 times the Ah size of comparable LFP batteries. The reasons are the depth of discharge past the 50% rule causes hard sulfation on the plates. Chemistry is still chemistry and not some fantasy. Second, the lead acid bank actually needs 2 banks within it where one bank is charging while the other is being used. The bank charging should be run fully thru absorption before being used and occasional EQ if the manufacturer specifies EQ needs to be performed. Once charging is complete, the banks are switched and the process repeats. Why should it be done this way? Because the sun still sets everyday and 100% SOC most likely will not be reached due to the slow charge rate. That charge rate is determined by the battery; large arrays won't help, multiple or high amperage chargers won't help either. It takes time which is the major limiting factor because the sun doesn't produce once it sets. For longest battery life, the 100% SOC should be reached on each charge cycle to convert normal soft sulfation back to lead and sulfuric acid. If this does not occur, eventually hard sulfation forms and the early death cycle begins.

Using LFP on a PV system is a no brainer. LFP likes partial SOC and will last longer if 100% SOC is not reached consistently. LFP allows full charge rate to bulk charge voltage setting. It's all about how much PV yield can be produced to flow into the battery. Charge rate is determined by what can be produced and not limited by the battery internal resistance. This allows full SOC to be reached if desired on a daily basis if there is enough yield available.

I mostly work on mobile applications so the requirements are what drive the AGM vs LiFe decision for me, not cost.

Similarly, the panel orientation is completely different than on a typical home install.

Usually it is the low and high temperature range requirements of the project that drive it vs cost. You are right that LiFe has some distinct advantages and interestingly for smaller battery banks like these ( 2 - 10 kW-hrs) AGMs sometimes have advantages as well. I am agnostic about the matter - I can make either one work just fine.

One of the key differences for mobile vs home is that solar charging starts as early in the day as possible with a goal of achieving full charge by early afternoon on a typical day. It isn't uncommon for RVs and vans to have solar panels that capture the sunlight mounted both horizontally and vertically.

In addition, the batteries are in fact routinely being charged / discharged during the process with one battery bank. The charge settings used are not necessarily identical to what you are used to.

Adding to the fun, it is common for their to be multiple chargers going in parallel from various energy sources.

I understand what you mean though. Personally I am not willing to give up garage space for a battery pack so it has to work mounted outside no matter what.

 

[wattmatters]

The biggest problem with those lifetime throughput charts is they represent an idealised/optimised scenario. Real life is not like that.

One needs to consider the discharge and recharge protocols those lifetime throughput values are based on, and what is the impact of any deviation from those protocols.

e.g. such charts will more than likely assume a full recharge using the manufacturer's optimal charging protocol occurs immediately after discharge. They will probably use a standard and non-variable discharge rate as well.

The real world is more often not going to follow the manufacturer's optimal discharge/charge protocols. We charge when we can and at variable rates based on available solar PV. Some might kick in supplemental charging with a generator or grid supply as needed. Discharge can occur at highly variable rates.

And it's all these deviations from the discharge and charge specifications where additional battery damage and accelerated degradation can occur with lead acid.

It will also occur with LiFePO₄ however LiFePO₄ is far more tolerant of such protocol charge and discharge deviations than is lead acid.

The round trip efficiency of lead acid is 80%, while lifepo4 is 90%.

Figure that into the equation and it comes a lot closer.

LiFePO₄ is 99%. Any losses on top of that will be in DC-AC-DC conversions (same for any battery).

I have a hybrid battery with both LiFePO₄ and SLA. The SLA are ex-data centre backup batteries.

They serve different purposes.

LiFePO₄ does the daily cycling while the lead acid is there as reserve capacity. The lead acts as a nice ballast in the system and gives the LiFePO₄ a "soft landing" when the LiFePO₄ SOC is getting low. I don't typically discharge lower than the nominal capacity of the LiFePO₄ unless we have an outage and I need to use the SLA reserve capacity. Meanwhile the LiFePO₄ helps keep the SLA at a nice float level most of the time (and given their purpose, this suits the SLA nicely).

As a result the combined round trip efficiency (DC) of my hybrid battery is 96% as that accounts for the SLA getting a little recharge each day plus the ongoing float trickle charge.

Lead acid has the problem of slower and longer charge time, lower round trip efficiency and they don't like not being recharged fairly soon after discharge - they are not a chemistry you want to leave at a partial SOC for long. LiFePO₄ couldn't care less (although best not leave them at a very low SOC for long but you can leave them in the 20-80% range for months, if not years).

 

[Zwy]

I mostly work on mobile applications so the requirements are what drive the AGM vs LiFe decision for me, not cost.

Similarly, the panel orientation is completely different than on a typical home install.

Usually it is the low and high temperature range requirements of the project that drive it vs cost. You are right that LiFe has some distinct advantages and interestingly for smaller battery banks like these ( 2 - 10 kW-hrs) AGMs sometimes have advantages as well. I am agnostic about the matter - I can make either one work just fine.

One of the key differences for mobile vs home is that solar charging starts as early in the day as possible with a goal of achieving full charge by early afternoon on a typical day. It isn't uncommon for RVs and vans to have solar panels that capture the sunlight mounted both horizontally and vertically.

In addition, the batteries are in fact routinely being charged / discharged during the process with one battery bank. The charge settings used are not necessarily identical to what you are used to.

Adding to the fun, it is common for their to be multiple chargers going in parallel from various energy sources.

I understand what you mean though. Personally I am not willing to give up garage space for a battery pack so it has to work mounted outside no matter what.

I have LFP in my truck camper, it was easy enough to install a heating pad under it with a thermostat to heat the battery for charging below freezing.

I have both horizontal and adjustable tilt panels on it also.

RV with limited roof space for PV, LFP is a no brainer. With lead acid, it is difficult to achieve full SOC with full absorption with limited charging capacity, even with a DC to DC (which requires burning more fuel) or a generator. LFP is the way to go, it doesn't care if 100% SOC is achieved. Plus the added discharge capacity within the same amount of space is a bonus.

For my house system, the battery bank is in my basement. No temp extremes there. If there wasn't the room in a garage or other place, I'd do one of 2 things. Build a heated insulated structure for the battery bank or build a lean to on to a heated garage.

 

[Zwy]

Interesting, lot of good point about lead acids

Thank you, the days of lead acid on a PV system are gone. It makes no sense really.

Far be it from me to argue, but seems the datasheets say different. according to rolls & several other big manufacturers, 100% cycles only degrade the battery by 14% more when compared to 50%soc cycles

Anyone can claim something but in the manufacturer tests, the batteries are recharged immediately reducing hard sulfation on the plates. That is an important distinction. Also, many will require EQ on a regular basis in order to hold battery capacity.

I've recovered severely sulfated batteries using repeated charging/discharging using EQ voltages for charging after absorb. I've also used a DC welder to blow sulfation off plates. In both cases, the battery will suffer capacity loss for 2 reasons. One, lead is removed from the plates and settles to the bottom. The second is that lead is sulfated and the acid that was in the electrolyte is now in that sulfation permanently and can't be recovered.

The best I ever recovered any deep cycle with various methods returns 90% capacity, many just break 80%. Starting batteries will still output high CCA but the capacity is reduced, the reduction in plate material and acid concentration is the cause. Basically, it just becomes a smaller battery with the same CCA.

What is interesting the 14% you stated mirrors the expected recovery (80 to 90%) I've seen with recovering sulfated batteries. So that tells me right there absorption and EQ is important. With PV charging, it is hard to achieve unless a method is adopted whereby 100% SOC is reached with regular EQ.