Depth of Discharge Myth?

[newhaul]

Originally Posted by ramblinrod

.

I believe the concept of no reserve capacity for LFP is very bad design. If loads are higher than average, alarms may go off forcing one to get up at 0230 to start a generator. Or worse, have a disconnect and all power be off until intervention.

When one considers that one can dip a FLA bank below 50% on occasion, then there is inherrent reserve capacity in that system.

why do any of that instead a prudent mariner would plan accordingly to not get that low on power regardless of battery chemistry.
The negatives of low battery are easy to avoid run your dino juice charger for an hour in the late afternoon if there is a significant lack of adequate power available to get thru the night without doing some damage to your battery and or systems. You will also have hot water for dinner dishes.

[kmacdonald]

Unfortunately one has to design in excess capacity in LA banks since they lose capacity so quickly and below about 30% capacity things quit working due to low voltage. Of course lithium doesn't suffer from that.

[ramblinrod]

Originally Posted by newhaul

why do any of that instead a prudent mariner would plan accordingly to not get that low on power regardless of battery chemistry.

1. Because prudent mariners are human, and humans make mistakes.

2. Because not everyone else aboard is necessarily as prudent. e.g. sometimes teen agers have been known to leave electrical appliances on needlessly.

3. Because every sailors plan is written in the sand not cast in stone.

4. Because not all mariners are prudent.

5. Because the prudent marine service provider should make the system as fool proof as possible or practical.

There are likely more very good reasons if I spent but a few seconds putting my mind to it, but this is probably way more than enough to justify a reasonable amount of "reserve capacity".

[Delfin]

Originally Posted by ramblinrod

1. Because prudent mariners are human, and humans make mistakes.

2. Because not everyone else aboard is necessarily as prudent. e.g. sometimes teen agers have been known to leave electrical appliances on needlessly.

3. Because every sailors plan is written in the sand not cast in stone.

4. Because not all mariners are prudent.

5. Because the prudent marine service provider should make the system as fool proof as possible or practical.

There are likely more very good reasons if I spent but a few seconds putting my mind to it, but this is probably way more than enough to justify a reasonable amount of "reserve capacity".

And a prudent mariner has the ability to deal with dead FLA (I connect it to my LI bank) or low voltage Li batteries (I avoid that by recharging at 50% DoD by firing up the generator, or the engine, or connecting to shore power when I can get there), so reserve capacity tends to be a chimera.

The reason for having a "reserve capacity" in FLA is not because you need it, but because FLA voltage drops so much below 50% DoD with any kind of a load you fry electronics. Like the PCB on my Kabola that burned up because I (shockingly) dropped below 24 volts under load a few times with an AGM bank (see Ohm's law for details). $3,500 later, boiler problem solved, but from practical experience I tend not to think about the last 50% of an FLA battery to be of much use unless you are talking about a very, very light load.

[ramblinrod]

Originally Posted by kmacdonald

Unfortunately one has to design in excess capacity in LA banks since they lose capacity so quickly and below about 30% capacity things quit working due to low voltage. Of course lithium doesn't suffer from that.

For one with a balanced FLA charging system combining ICE, Wind, and Solar, they can use about 100% to 50% of FLA capacity normally, and down to as low as 20% on rare occasions.

There may be some loss of performance at the low end for demanding loads, but most things will work fine; some load compromises may need to be made until voltage is higher.

At that low level, the FLA bank will have its highest available acceptance rate, which will be about the maximum capacity of the properly selected ICE charging system in a balanced electrical system.

So what this means is one will normally use 50% capacity but in rare cases can use 80% capacity of an FLA system.

In contrast, one can use 80% (90%-10%) of the capacity of LFP all of the time without detriment.

BUT...

This provides no reserve capacity like the FLA system inherently has.

To add in an equal amount of occasional LFP reserve capacity the FLA system naturally has, one would need to have an LFP bank that cycled 90% -40% under average conditions, leaving 40% to 10% as the same amount of reserve for unusual circumstances.

No reserve and lights out or immediately start ICE (or some other intervention) right now (to avoid trashing a very expensive bank), unlike the FLA bank designed with occasional reserve capacity that could be dipped into and ICE started at a later more convenient time (if even necessary at all) to recover to 100% SOC within 24 hours or so.

The thing I'm seeing here, is some are trying to claim that they can design an LFP system with no reserve capacity. I feel this is very bad design.

It is actually rare that actual daily consumption equals exactly average daily consumption.

This is similar to your home furnace.

It is generally designed to handle (100% run cycle) the coldest anticipated day, less 2.5%.

Most of the time it may only run on a 0 - 50% duty cycle.

On rare occasions, it may run 100%. In the worst condition (coldest day), you may see slightly reduced performance, perhaps the house is only 68F despite the thermostat set to 70F.

Again these are general statements and it goes without saying that YMMV based on your personal circumstances, and that is perfectly OK.

We do not need to review every single possible exception (which would be an exercise in futility).

To attempt to dictate to others that generalizations be abolished is unworkable.

The general statement, "Sailing is fun", would be declared invalid because some people don't care for it.

It would set up an impossible condition for anyone to post anything general unless it applies universally without exception.

It would be absolutely ridiculous (IMHO) to even suggest this is a requirement to post something on CF, especially something which may be of great value to the "average" CF member, or "most" members, or even just "some" members, but not necessarily "absolutely all" members.

[evm1024]

I came across another paper that looked into off grid performance of various battery types.

This paper was looking into pulsed charging and partial charging and the effects it has on battery capacity and cycle count. Pulsed charging is common to both solar where there is a varying cloud cover and wind generation. It is opposed to constant current/ constant voltage charging.

The conclusions from this paper are in line with Mainesail's testing where he shows that PSOC and deep discharge (< 50% SOC) are killers of FLA batteries.

This goes to show us that depending on occasional dips below 50% SOC in FLA as a reserve coupled with one or more days to return to 100% SOC are destructive to FLA.

And in the case of Mainesail's testing significant loss of capacity to the point of the FLA cells failing industry based capacity tests.

Here is a sample from the papers conclusions.

The lead-acid cells are also subject to severe
partial charging when charged with a variable current; this latter
effect is greater for longer pulses and reflects the poor pulse charge
acceptance of these cells. The resultant incomplete charge is
believed to accelerate sulphation and capacity fade. The LCO cells
show better overall charge acceptance and, while variable charge
results in greater partial charging, the frequency of the variability
does not affect charge acceptance. Variability does not increase
capacity fade in the LCO cells. The LCO-NMC cells degrade slower
than the LCO cells, and constant charge cells show initial signs of
faster degradation than variable charge cells, which may be the
result of incomplete charging limiting the side reactions induced by
high SOC. Degradation in the LCO-NMC cells appears to be initially
due to loss of active material, but ultimately shows an increase in
resistance as well. The LFP cells show very little degradation under
all charging protocols. No increase in resistance is seen, and the
slight reduction in capacity observed appears to primarily stem
from the loss of cyclable lithium.

See: LiFePO4 technical papers mega thread (ok mayby not so mega) - Cruisers & Sailing Forums

Basically both this paper and mainesails testing indicate that the capacity of FLA (and lead acid in general) at a SOC of less than 50% (I won't quibble the exact SOC. 50%ok, 40% ok, what is your number?).

To me this indicates that irreparable damage is done to FLA in typical cruising boat conditions (low SOC discharges and common PSOC conditions). Thus the region under 50% is an emergency reserve that you use when you are needing to get power while accepting the damage it is doing to your battery.

[ramblinrod]

Originally Posted by newhaul

sorry but I have to disagree with the both needing the same size banks.

You certainly have a right to disagree.

As do I to disagree with your disagreement.

Due to the discharge voltage curve . My refrigerator would stop when the Fla bank reaches 60% DOD...

In my professional opinion, you have an electrical system design problem if this occurs.

I doubt very highly that this system follows the general FLA guidelines that I posted previously (300 A-hrs for every 100 A-hrs of average daily consumption).

If it does, then rather than going to the expense of changing batteries, I may recommend reducing compressor RPM (takes about 30 seconds for a BD-35) and longer duty cycle, thereby drawing less instantaneous current.

On the reserve capacity it has been well discussed and established in this thread that a regular discharge of Fla to 80% DOD will do harm to the life expectancy of said bank however the same can't be said to a significant degree with Lfp.

OK this is where you are missing the point.

If one regularly goes to 20% SOC, this is not unusual dip into reserve capacity; it is usual energy demand.

Again a sign of a simple system design problem.

The bank capacity is too light for common demand.

It is actually well discussed and established, that if one goes "occasionally" lower than 50% SOC with a FLA bank and gets back up to 100% fairly soon, life expectancy impact is minimal if not negligible.

Another established point is that Lfp will recover faster than Fla to a full state regardless of charging means.

A contrary established point is that this is not necessarily relevant.

As long as the bank recovers so as to provide the energy as needed and avoid permanent damage, this may have little to no bearing on actual system performance.

Again, if equal size LFP and FLA banks are drawn down to 20% DOD, and then on a sunny day the solar charging system gets the LFP bank to 100% by noon, and the FLA bank to 100% by 7 pm, this may have little to no impact on the user; the drinks are equally cold; the ice equally hard.

From a balanced electrical system design standpoint, the faster acceptance rate of the LFP was not needed, and characteristic or feature wasted, having little to no benefit realized by the system and end user.

It is hard for some to justify any additional cost for a "feature" that result in no actual "benefit".

It was established early on this thread by Bruce Schwabb, and I agree completely, "If solar and/or wind charging meets your charging capacity requirements, stick with FLA" (or something to that effect).

If one doesn't have solar or wind charging capability, they may wish to consider it before switching from FLA to LFP. It may be easier, less costly, and more enjoyable.

(I acknowledge there may be individual cases where this may not be, but that doesn't change that in many, it may.)

[evm1024]

Back in my power grid engineering days the general rule was that we only have plans for single contingency faults. The multiple contingency faults were too unpredictable to develop detailed mitigation plans.

What would a multiple contingency fault look like? Well how about hitting a submerged container (that greats significant flooding) and a fire breaks out.

You could cope with any one of those (good luck with the fire but you get my point). Plug the hole or put out the fire.

But both at the same time and all bets are off. the are off big time. You get the fire out only to sink, You plug the hole and burn to the waterline.

The real action you would likely be taking is to get the ditch bag and abandon ship.

What does this have to do with this thread? (glad you asked!)

The point I'm making is that I do not need to have a 100% failsafe system. If I run my batteries down (single fault) I can mitigate that by:

1) reducing the load (turn off the fridge)
2) get up in the morning and start the generator or main engine
3) see what solar is doing and shut the ICE off it it is doing fine.

But to say that it is a total disaster is to invoke more than one fault (total battery failure)

Just not needed to prove a point or true

[Dockhead]

Originally Posted by evm1024

I came across another paper that looked into off grid performance of various battery types.

This paper was looking into pulsed charging and partial charging and the effects it has on battery capacity and cycle count. Pulsed charging is common to both solar where there is a varying cloud cover and wind generation. It is opposed to constant current/ constant voltage charging.

The conclusions from this paper are in line with Mainesail's testing where he shows that PSOC and deep discharge (< 50% SOC) are killers of FLA batteries.

This goes to show us that depending on occasional dips below 50% SOC in FLA as a reserve coupled with one or more days to return to 100% SOC are destructive to FLA.

And in the case of Mainesail's testing significant loss of capacity to the point of the FLA cells failing industry based capacity tests.

Here is a sample from the papers conclusions.

The lead-acid cells are also subject to severe
partial charging when charged with a variable current; this latter
effect is greater for longer pulses and reflects the poor pulse charge
acceptance of these cells. The resultant incomplete charge is
believed to accelerate sulphation and capacity fade. The LCO cells
show better overall charge acceptance and, while variable charge
results in greater partial charging, the frequency of the variability
does not affect charge acceptance. Variability does not increase
capacity fade in the LCO cells. The LCO-NMC cells degrade slower
than the LCO cells, and constant charge cells show initial signs of
faster degradation than variable charge cells, which may be the
result of incomplete charging limiting the side reactions induced by
high SOC. Degradation in the LCO-NMC cells appears to be initially
due to loss of active material, but ultimately shows an increase in
resistance as well. The LFP cells show very little degradation under
all charging protocols. No increase in resistance is seen, and the
slight reduction in capacity observed appears to primarily stem
from the loss of cyclable lithium.

See: LiFePO4 technical papers mega thread (ok mayby not so mega) - Cruisers & Sailing Forums

Basically both this paper and mainesails testing indicate that the capacity of FLA (and lead acid in general) at a SOC of less than 50% (I won't quibble the exact SOC. 50%ok, 40% ok, what is your number?).

To me this indicates that irreparable damage is done to FLA in typical cruising boat conditions (low SOC discharges and common PSOC conditions). Thus the region under 50% is an emergency reserve that you use when you are needing to get power while accepting the damage it is doing to your battery.

I'm glad someone besides me is still interested in the original topic.


I think that it is pretty clear that we don't understand everything that goes on inside lead-acid batteries, despite all attempts at simple explanations. The authors of this paper don't seem to understand, either -- "is believed to accelerate sulphation."


Different responses to different charging regimes and currents show that degradation of lead acid batteries is not a linear function of time x amount of lead sulphate available for dissolving/recrystallization.



At least the question of "whether" has been answered, if not the question of "why".

[evm1024]

Originally Posted by ramblinrod

SNIP

Again a sign of a simple system design problem.

The bank capacity is too light for common demand.

It is actually well discussed and established, that if one goes "occasionally" lower than 50% SOC with a FLA bank and gets back up to 100% fairly soon, life expectancy impact is minimal if not negligible.



A contrary established point is that this is not necessarily relevant.

As long as the bank recovers so as to provide the energy as needed and avoid permanent damage, this may have little to no bearing on actual system performance.

SNIP

Super, I have that thread where I am listing papers and testing results for batteries.

I would like to include the paper/test-results that show where permanent damage begins in FLA PSOC and DOD regimes.

Shoot me the link and I will included it that thread. The thread title says LiFePO4 but FLA can be included as well.

http://www.cruisersforum.com/forums/...ga-206796.html

[ramblinrod]

Originally Posted by evm1024

I came across another paper that looked into off grid performance of various battery types...

To what detrimental impact?

If the normal FLA bank daily discharge cycle is 100% to 50%, and once every 2 weeks it is dipped to 30% by sunrise, but back up to 100% by sunset, what is the actual impact on bank life expectancy?

I would suspect that EEE, it may reduce life expectancy by 10% tops.

This would mean that rather than lasting 5 years, the bank would last 4 years 6 months.

No problem.

At the current cost of LFP transition, it would simply take too long to break even and realize any return on investment to make the transition economically justifiable, especially considering the risks involved.

The PSOC problem arises when people leave batteries at a low PSOC.

If I woke up in the morning and saw my FLA bank at 20% SOC, I would have the engine on in a heartbeat, and it would be back to at least 50% PDQ and 100% by the end of the day (topped off nice and slow by solar).

Conversely if one had a LFP bank with half the capacity, they would have hit 10% SOC much earlier, and either be woken up at 2 am to start the generator, or the LVD tripped and possibly the running lights extinguished, resulting in their vessel being caught under the bow of one of those many 20 knot ocean freighters they may be dancing around, when the AIS and radar screen goes ____________________________________.


(This is meant to be funny, I know the person on watch shouldn't be asleep all night under way, but I hope you get my point).

[kmacdonald]

Not including the full blow infomercials pushing solar panels, this has been a great thread. Lots of great information from various perspectives. It certainly points out that one size doesn't fit everyone.

[evm1024]

Originally Posted by Dockhead

I'm glad someone besides me is still interested in the original topic.


I think that it is pretty clear that we don't understand everything that goes on inside lead-acid batteries, despite all attempts at simple explanations. The authors of this paper don't seem to understand, either -- "is believed to accelerate sulphation."


Different responses to different charging regimes and currents show that degradation of lead acid batteries is not a linear function of time x amount of lead sulphate available for dissolving/recrystallization.



At least the question of "whether" has been answered, if not the question of "why".

My day job is in a combined Computer Science and Electrical Engineering department of a University.

I fully understand the wording that they used. Their wording gets them off the hook for describing the mechanics of capacity loss.

Right now a bunch of our students are defending their PHd thesis. If they claim outside of their thesis we may just grill them for hours on those claims. In some cases we have had them under the "lights" for 6 hours.

Advice to students, stay on topic, do not make any claim you are not able to defend verbally in your defense.

[newhaul]

Originally Posted by ramblinrod

Again, if equal size LFP and FLA banks are drawn down to 20% DOD, and then on a sunny day the solar charging system gets the LFP bank to 100% by noon, and the FLA bank to 100% by 7 pm, this may have little to no impact on the user; the drinks are equally cold; the ice equally hard.

here you just said it yourself and don't understand it .
Lfp full charged by noon vs Fla by 1900.
Extra 7 hours of solar output to run the holding plate refrigerator and the watermaker off of without using any battery power.
Which for the refrigerator means that I won't need to run it as long on the battery thereby saving battery power . If it reduces my Lfp use by 20% that means even less time the next day to charge so now I have 9 hours to do other stuff with my solar .

Is this starting to make sense to you yet?

[Dockhead]

Originally Posted by evm1024

Back in my power grid engineering days the general rule was that we only have plans for single contingency faults. The multiple contingency faults were too unpredictable to develop detailed mitigation plans.

What would a multiple contingency fault look like? Well how about hitting a submerged container (that greats significant flooding) and a fire breaks out.

You could cope with any one of those (good luck with the fire but you get my point). Plug the hole or put out the fire.

But both at the same time and all bets are off. the are off big time. You get the fire out only to sink, You plug the hole and burn to the waterline.

The real action you would likely be taking is to get the ditch bag and abandon ship.

What does this have to do with this thread? (glad you asked!)

The point I'm making is that I do not need to have a 100% failsafe system. If I run my batteries down (single fault) I can mitigate that by:

1) reducing the load (turn off the fridge)
2) get up in the morning and start the generator or main engine
3) see what solar is doing and shut the ICE off it it is doing fine.

But to say that it is a total disaster is to invoke more than one fault (total battery failure)

Just not needed to prove a point or true

Well, having reserve capacity in a given bank only delays the inevitable, if there is a fault in all your charging sources and you find yourself unable to produce power. That's not a solution.


The solution is to have redundancy in power production. I suppose there is some cruising boat somewhere which has only solar as a charging source, but I don't think I've ever seen one. If you start to run out of power, you simply crank up the generator or main engine and start charging -- that's the standard solution.


If you have only one alternator on one main engine and no other charging source, then of course you are vulnerable to a fault in that single point of failure. But keeping some "reserve capacity" in the batteries is not a solution to that, unless you are simply never too far away from plugging in somewhere that this "reserve capacity" can actually get you home. But even in that case it's a backwards way to engineer the system and a poor substitute for backup power production capacity. A cruising boat which needs electrical power for important or mission-critical purposes (like navigation systems) should be able to rely on being able to PRODUCE power, not just having some power stored.


On my boat, this is not an issue -- I can produce 6.5kW with the generator and 2.5kW from the second alternator on the main. Besides that, there are two more alternators charging two dedicated separated start batteries. An extreme case perhaps, but not all that unusual. So the lower 50% capacity of a lead-acid bank is just "dead lead" (as MaineSail calls it), for me as for everyone with a reasonably well designed power system, of practically no value whatsoever.