Battery lifetime/cost vs. state of charge

[conachair]

Originally Posted by Pizzazz

Others, myself included, choose to be aggressive on the weight savings and install only 200AH, discharge to 20% of capacity,

Do you have any problems with low voltage when discharged so much? My fridge almost certainly would start cutting out long before that even with very healthy batteries.

[chala]

Originally Posted by sailinglegend

Let me give my 4 reasons why one big bank is better than two small banks.....

I always have been taught not to keep all my money in only one bank.
Sure if I could double the money that I have in the banks I would do that.
It is obvious that if one bank goes bust I still have money in the other banks so long that they are not interconnected.
I do not mind the complexity of using many banks.
Instead of doing a big withdrawal from one bank and deteriorate my rating, I do intelligent withdrawing from different banks. (1 & 2)
I keep wife and kids, essential and other in separate banks

Originally Posted by SV THIRD DAY

(remember 4 people living aboard so that's 4 phones and 4 laptops)

Like this, the wife and kids knows when their bank balance is low.

Originally Posted by sailinglegend

3. Doubling the service bank size also means it will be more efficient and accept more Ah more quickly from all charging sources during the boost phase up to 80%.



For “many smaller batteries in one large bank” it makes no difference each battery will take all the charge that the battery can absorb.
Doubling the weight of batteries for son dubious efficiencies is not every one cup of tea.

Originally Posted by sailinglegend

4 If you have a larger bank - or many smaller batteries in one large bank, it is easy when they start failing to just disconnect the bad ones and run on the others as long as you can until you can replace the whole bank.

Oh yes so long someone competent is present when the short circuit happen and does disconnect it in the required time without making some sparks that can ignite the explosives gas that are around.

http://www.cruisersforum.com/forums/...ot-172827.html

Originally Posted by Maine Sail

We have no good way to test state of health of our batteries so the abuses are essentially blind to the owners. Often it is not until an owner can no longer start the motor or the lights dim when the fridge kicks on or the batteries thermally run away due to an internal failure that any issues with SOH are noticed..

Originally Posted by Maine Sail

The highest rates of internal shorts/failures I see occur in flooded batteries but they can also occur AGM or GEL.

[valhalla360]

Originally Posted by Pizzazz

Let me attempt to summarize all the facts and opinions presented so far in an effort to answer my original question regarding the optimum battery bank. It would be good to remind ourselves that we use batteries to hold electricity between charging. Thus the key figure of merit for any battery system is capacity vs. cost/weight/volume/use.

- Based on manufacturers' data, going to deep discharge vs. halfway discharge reduces the battery life proportionately other things being equal. This is based on the expectation that we are going to get the same amp-hours over the lifetime of the batteries irrespective of depth of discharge. For more information see the DOD curves. No the data does not show that. The manufacturers data assumes 100% charge between cycles.

- There are other factors that could further shorten battery life, the two most important are high temperature and incomplete charging. Most people expect the combination of two or more factors to exponentially decrease battery life but we have not been able to verify this with hard data. It's well established that not fully charging a battery is bad for it. Only the degree of damage is in question.

- It is important to take Peukert's law in consideration when sizing a battery since high current draw could significantly reduce available battery capacity. Yes and could result in effectively deeper discharges magnifying the issue.

- Charging wet cell batteries above 80% is inefficient due to the rising internal resistance of the battery. This means that if charging the battery from an alternator, we need to run the engine twice as long to get from 80% to 100% vs. from 60% to 80% assuming a three stage temperature compensated charger. With a simple dumb charger it is much longer. Thus many cruisers just give up on charging the batteries above 80% unless they have excess solar, wind or generator (all options more expensive than just bulk charging and replacing batteries). This is a separate topic though. I believe with modern cruising boats and chargers the old rule about only reaching 80% is no longer a valid assumption. For many they never use the motor for charging unless they go thru several cloudy days or have unusually high electric consumption. Even if they do use a motor, an hour or two in the morning bulk charging to get to 80% and then let the solar top up the last 20%.

- Having a larger than necessary battery bank has negatives including weight, volume and upfront replacement cost (i.e. if one battery goes down you have to replace the whole pack) while providing higher margins of safety. And smaller than needed battery bank includes frequent replacement, likely will result in higher lifecycle costs. Also there will be a greater chance of being completely without power when the small bank goes dead as it nears the end of it's life.

Based on the above, some cruisers decide to play it safe sizing their house battery bank at 400AH for a typical 120A overnight use which allows them to stay above the 50% DOD recommendation plus adds a sizeable reserve and hope to get 5+ years out of the batteries. If they want to use more than 30% of the installed capacity they invest in the three way charger or renewables and accept the inefficiencies or extra generation capacity when away from shore power. You are being overly specific on sizing. With a decent solar system, it can easily reach 50% or more of the battery bank rating when you figure during the day a percentage of the solar can go directly to powering devices while pushing the batteries up to 100% charge.

Others, myself included, choose to be aggressive on the weight savings and install only 200AH, discharge to 20% of capacity, only do bulk charging at sea and are prepared to replace the batteries every two years (assuming a limited amount of nights away from shore power, i.e. not full time cruising). The net cost difference between the two approaches is small (Calder's article mentioned 5%, other people think it could be more) and difficult to pin point as other real-life factors also affect battery life. Both approaches are valid.

Now, let's get away from shore power.

SV Pizzazz

Your analysis probably held some weight 30yrs ago when solar arrays were expensive and rare on cruising boats, so the standard was to run the motor for a couple hours per day to charge the batteries. With the cost of solar, there really isn't a good reason not to hit 100% charge most days and that knocks out a lot of the math behind your assumptions which were already a bit sketchy.

[Cadence]

Originally Posted by valhalla360

Your analysis probably held some weight 30yrs ago when solar arrays were expensive and rare on cruising boats, so the standard was to run the motor for a couple hours per day to charge the batteries. With the cost of solar, there really isn't a good reason not to hit 100% charge most days and that knocks out a lot of the math behind your assumptions which were already a bit sketchy.

Your assumption is based on what lat. as it relates to solar. It probably works great for six months a year some places and not at all for the other 6 months. No place I would live but there are those that do.

[valhalla360]

Originally Posted by Cadence

Your assumption is based on what lat. as it relates to solar. It probably works great for six months a year some places and not at all for the other 6 months. No place I would live but there are those that do.

If you cruise in Svalbard in the winter, true.

I've checked on the boat in winter in Michigan and the solar panels are still producing but no sane people are on the water in Jan. Certainly if you bulk charge in the morning with the motor, getting up to 100% most days should be doable. We been on the water in spain over he winter and they were keeping up OK.

I would think in most tropical areas cruisers tend to head to in the winter, it would do even better.

[thinwater]

Originally Posted by Cadence

Your assumption is based on what lat. as it relates to solar. It probably works great for six months a year some places and not at all for the other 6 months. No place I would live but there are those that do.

I've learned that battery consumption is lower in the winter:

• No fans. These can really eat amps in the summer.
• No fridge (leave a cooler in the cockpit--the cooler is to keep stuff from freezing!).
• Longer nights, but cabin lighting should be florescent/LED. And you should get up at dawn and go to bed early. Days are too short to waste.
• Power for heater. Very low.

You do have to learn that battery voltage and SOC changes with temperature. Also that battery temp is not always ambient.

[mvas]

Regarding "Battery lifetime/cost vs. state of charge" and referencing the Depth of Discharge chart posted in message #13 ... I wondered, what is the lowest Depth of Discharge on a consistent basis that gives me the maximum total Amp-Hours from the battery bank? Assumption: I can recharge the battery bank to 100% the next day.

Total Amp-Hours ...
80% DOD x 675 cycles = 540Ah
70% DOD x 780 cycles = 546Ah
60% DOD x 950 cycles = 570Ah
50% DOD x 1150 cycles = 575Ah
40% DOD x 1475 cycles = 590Ah
30% DOD x 2050 cycles = 615Ah
20% DOD x 3300 cycles = 660Ah <<< Maximum total amp-hours

But in the real world ...
I will never use 3,300 cycles before the battery bank dies from old age (time).
And I don't want a battery bank 5 times larger than need.
Therefore, 50% DOD and 1,150 theoretical cycles in 5 - 7 years is a better match.
With the deeper Depth-Of-Discharge, the battery bank will spend more time in PSOC,
which will reduce the actual number of cycles before the battery bank dies.

[StuM]

Originally Posted by mvas

With the deeper Depth-Of-Discharge, the battery bank will spend more time in PSOC, which will reduce the actual number of cycles before the battery bank dies.

And I suspect that fact will be far outweigh the extra Ah based on published DOD cycle life.

[mvas]

Originally Posted by StuM

And I suspect that fact will be far outweigh the extra Ah based on published DOD cycle life.

With all of the 50% DOD talk, I was actually expecting to see the peak for Total Amp-Hours to be somewhere around 50% DOD. I did not expect to see the Total Amp-Hours available to keep growing and growing with each shallower depth-of-discharge calculation. Unfortunately for us, no manufacturer (except Firefly) will provide us with any information regarding the bad side effects caused by "the amount of time spent in the PSOC state". These DOD charts do not include any time in the PSOC state - ie their battery is recharged to 100% immediately (best case scenario for their test results). If we were to add in the bad side effects of leaving the battery bank in a PSOC state, then we will get even fewer cycles, and fewer Total Amp-Hours, at the 80% DOD (deep discharge) end of the chart. The shallower discharges, like 20% to 30% usage, will be minimally affected by PSOC because we can and should achieve 100% recharge very quickly. With real world "time spent in PSOC", then the Total Amp-Hours available at the 80% DOD (deep discharge end of the chart) must be even less (ie worse) than I previously calculated. All evidence seems to point to ... many shallower cycles (within the battery's expected lifetime) and less time spent in the PSOC state is what provides us with maximum Total Amp-Hours.

[kmacdonald]

Originally Posted by kmacdonald

From Trojan Battery:
20% DOD 4000 cycles
50% DOD 1600 cycles
80% DOD 1000 cycles
If you cycle to 50% DOD instead of 20% DOD you loose 2400 cycles of battery life to gain 30% capacity.
If you cycle to 80% DOD instead of 50% DOD you loose 600 cycles of battery life to gain 30% capacity.
From an energy delivered standpoint it looks the same for 20, 50, and 80% DOD:
If x is the total energy stored in the battery then at 20% DOD the battery will deliver .2(4000)x=800x over its life
At 50% DOD: .5(1600)x=800x over its life
At 80% DOD: .8(1000)x=800x over its life

This tells me to cycle deeper and save some weight.

Tanglewood said the same thing as I just showed above.

The above assumes you are recharging to 100% SOC.

The best available information is given above. Anything else is speculation.
Clearly the total energy delivered during the battery life is not a function of DOD.....it's a constant.

[Maine Sail]

Originally Posted by kmacdonald

The best available information is given above. Anything else is speculation.

What is given above is essentially meaningless laboratory data that is useless to try and derive real world data from. When Trojan starts publishing PSOC data then we can more accurately predict the effects of DOD on cycles.

Originally Posted by kmacdonald

Clearly the total energy delivered during the battery life is not a function of DOD.....it's a constant.

In the lab, where they treat the batteries with kid gloves, so they can mislead us with marketing mumbo-jumbo, this formula might work. We must also remember that these data sets are "formulas" not actual testing to 80% or 50% DOD etc. but "projected cycles" using a formula.

It does not look like this once the batteries are removed from the white coat/white glove lab where the batteries are treated like precious little cupcakes. We also need to remember that lab data can not at all predict PSOC use and battery quality and how they handle PSOC is widely varied even among brands with similar "deep cycle" claims. FWIW we were denied any data sets we asked for, regarding PSOC use, by all the manufacturers we asked. Either they don't have it, or refuse to share it.

In the Practical Sailor PSOC testing all the AGM batteries were treated equally and still there were massive variances in permanent capacity loss over just 30 cycles of deep-cycle PSOC use where the batteries were discharged to 11.7V at the 20 hour rate then recharged for exactly 1 hour at a .46C charge rate. A LifePO4 battery was used as a control as they do not sulfate.

CALB LifeP04 - Lost 0% of Ah Capacity in 30 PSOC Cycles
**Firefly AGM - Lost 0% of Ah Capacity in 30 PSOC Cycles
Odyssey TPPL AGM - Lost 6.5% of Ah Capacity in 30 PSOC Cycles
**Northstar TPPL AGM - Lost 8.5% of Ah Capacity in 30 PSOC Cycles
Lifeline AGM - Lost 11.6% of Ah Capacity in 30 PSOC Cycles
Deka / East Penn AGM #1 - Lost 29.7% of Ah Capacity in 30 PSOC Cycles
*Deka East Penn AGM #2 - Lost 27.5% of Ah Capacity in 30 PSOC Cycles

*A second battery was tested because initial results were so poor.

**Marketed as sulfation resistant.

Capacity losses over 30 cycles were not recoverable, beyond what is shown, in any of the batteries that lost capacity.

Trying to predict cycle life based on lab data does not work in the marine environment until the manufacturers start giving us PSOC data..

[valhalla360]

Originally Posted by kmacdonald

The best available information is given above. Anything else is speculation.
Clearly the total energy delivered during the battery life is not a function of DOD.....it's a constant.

Nope, you have plenty of speculation and your results are not consistent with the graphs put out by the manufacturers. (see post 112 which shows how it produces more overall amp-hrs). The big issue is all of these charts assume 100% charge between cycles. Also, they don't say but they most likely accelerate the testing by starting the charge as soon as the DOD is reached so you aren't likely seeing comparable results where there are typically at least a few hours early morning where the bank sits partially discharged.

Basically, the 50% DOD is a compromise between maximizing total amp-hrs over the life of the batteries versus keeping the battery bank size reasonable. It also gives you a nice reserve capacity for use in emergencies.

With 80% discharge you give up some amp-hrs over the life of each battery and if anything goes wrong, you are far more likely to run them dead. Let's say your fridge thermostat sticks on in the early evening. With 50% discharge, odds are you find out in the morning when your food is frozen and the batteries are still at 20-40% charge. If you plan on 80% discharge, you likely have 0% charge (and damaged batteries) by morning.

[kmacdonald]

The capacity lost due to PSOC happens regardless of the DOD. If you recharge to full after each use PSOC degradation doesn't happen.

[mvas]

Originally Posted by kmacdonald

The best available information is given above. Anything else is speculation.
Clearly the total energy delivered during the battery life is not a function of DOD.....it's a constant.

Why is your chart from Trojan, for their T-105RE, which includes CARBON (like the FireFly), the best chart, but charts from other manufacturer's are only speculation?
Are you saying that the majority of the batteries in use, now include carbon?

[Maine Sail]

Originally Posted by valhalla360

The big issue is all of these charts assume 100% charge between cycles.

Yes most every industry cycle life test cycles the battery to 100% SOC before discharging again.

Originally Posted by valhalla360

Also, they don't say but they most likely accelerate the testing by starting the charge as soon as the DOD is reached

This is most often exactly how it is done and the batteries do not stay in a static sulfated state.

They are also in a controlled temp environment, temp compensated and the discharge rate is rarely the 20 hour rate. Why not a 20 hour discharge rate? Number one is time and number two is that it would result in lower overall DOD on each cycle thus reducing the number of "cycles" they can claim on their glossy marketing sheets.

Much of the cycle testing done on deep cycle batteries, such as golf cart batteries, is done at a 75A rate to the desired cut-off voltage. A 75A discharge rate results in a shallower & kinder discharge, because of Peukert, than an 11.25A discharge rate (20 hour rate) on a 225Ah GC-2 battery. Cycle life data is then "extrapolated" in an attempt to predict varying cut off voltages for cycle life. It is usually far too expensive to tie up the lab testing machines conducting 20 hour cycle life testing so they do it at a higher rate, to the same cut off voltage, a discharge rate that you or I would never use.

Lab cycle life data misleading? Yes very, very often it is...