Alan Dennis's posts

I realized I never replied to this.  Lead does not operate below freezing.  Lithium can operate at diminished power down to -20C (or rather, whatever the electrolyte variant is rated to, but -20C is typical).

I made a couple additions to my last post regarding safety.  LiFeYPO4 is the safest form of lithium.  You may be familiar with LiPoly, which are commonly used in laptops and RC airplanes.  They have twice the energy density, but are prone to exploding and catching fire if over charged or punctured.  If you use LiPoly in an off-grid array, expect your house to burn down shortly thereafter.  That being said, LiFeYPO4 is safer than lead acid (especially non-sealed cells).

Lithium makes more sense now than ever.
Lead specs are on the left / Lithium specs are on the right

Type: Universal brand AGM / Thundersky brand LiFeYPO4
Voltage: 12.8V / 3.2V
Rated Ah: 200Ah / 200Ah
Rated Capacity: 2,560 Wh / 640 Wh  (Ha, just wait!)
Weight: 132 Lbs / 16 Lbs
Volume: 1,666 cubic inches / 307 cubic inches
Toxicity: Lead and acid / inert
Fire Danger: can vent hydrogen / none
Orientation: Upright only / Upright or on its side


Hour rating: 20 hours  /  1 hour
Recommended Depth of Discharge: 30% / 80%
Cycle Life: 550 cycles / 2,500 cycles
Shelf Life: 6 years / 10 years
Self discharge: 3% per month / 3% per month
Peukert's Constant: 1.20 / 1.08
Usable watt-hours (at rated discharge): 768Wh / 512Wh
Price: $375 / $250


Cell capacity at various loads.  Peukert's Constant, as well as the hour rating is the key here.
C/100 =275.9 Ah / 251.8 Ah
C/20=200.0 Ah / 232.3 Ah
C/5=151.57 Ah / 216.8 Ah
C/2=126.2 Ah / 207.1 Ah
1C =109.9 Ah / 200.0 Ah
2C=95.6 Ah / 193.2 Ah
3C=88.2 Ah / 189.3 Ah
Lithium has a much flatter capacity curve than lead.

Dollars per watt-hour (lower is better).
Lead on left / Lithium on right
C/100: $0.353 / $0.388
C/20: $0.488 / $0.420
C/5: $0.645 / $0.450
1C: $0.885 / $0.488
Even at C/20, Lithium still wins on a $/Wh basis.


Let's consider a battery pack that we want to get 10KWh (Usable) at a C/20 rate:
Again, lead is on the left / Lithium is on the right
Cell count: 13 / 16
Usable Capacity (Actual capacity under load and to DoD) : 9.98KWh / 9.52KWh
Rated Capacity (The actual marking on the cells):  33.28KWh / 10.24KWh
Price: $4,875 / $4,000 - Lithium wins
Weight: 1,716 lbs / 256 lbs - Lithium wins (by a lot)
Volume:  21,658 cu in / 4,912 cu in - Lithium wins (by a lot)


Lithium wins on price, weight, volume, cycle life, shelf life, layout flexibility, and safety.  Convinced?  Any additional information I should provide?
By the way, my lithium scooter is still going strong.

Where are my numbers coming from?

For my solar project I bought six "100AH" (Rated 92AH @ 1/20C) 12V deep-cycle AGM lead acid batteries from the local Batteries Plus.  These were model #WKDC12-100P.  They cost me $200 each.  I then made my own 2/0 gauge interconnect cables from parts purchased at a local Marine Store and Home Depot.  These were additional costs, but I'm not including these here.

Lead Acid AGM
$200 / (92AH * 12V * 30%) = $0.604 per watt-hour
(92AH * 12V * 30%) / 63 Lbs = 5.26 watt-hours per pound

----------------------------------

For my electric moped, an XM-3500li that I'm adding extra batteries to, I purchased four 40AH 3.2V ThunderSky cells from Evolve Electrics.  They were $58 each, 5 day shipping included.  The interconnects came with the batteries, and they were actually already wired up in series, saving me extra time and cost.

Prismatic LiFePO4
$58 / (40AH * 3.2V * 80%) = $0.566 per watt-hour
(40AH * 3.2V * 80%) / 3.3 Lbs = 31.03 watt-hours per pound

----------------------------------

The operating temperatures come from the manufacturers' data sheets.

Sorry about the sloppy looking original post.  I had the epiphany, but I didn't have much time to explain myself.  The main point to take away is this:

If you buy a 100AH lead acid battery and want it to last, you'll only use 30AH (30% Dod).  That will get you around 3000 charge-discharge cycles.

However, with a 100AH LiFePO4 (lithium iron phosphate) battery you can use 80AH (80% DoD) of its capacity and still be left with 3000 cycles.  In case you ran it to 70% DoD you would get 5000 cycles.

So, even though the lithium batteries are more expensive for the same rated AH, you actually get a lot more usable capacity with LiFePO4.

Now, add the fact that lead acid batteries can only reach their rated AH capacity if you discharge them at no more than a 1/20 C rate and you watch your real world capacity diminish further.  LiFePO4 generally are rated between 2.5C and 3C, giving them much greater overhead and efficiency.  Take this example:

A 12V, 100AH lead acid battery has an ideal usable capacity (30% DoD) of 360 watt hours, at a discharge rate of 60 watts (1/20C).

a 12.8V, 100AH lithium iron phosphate battery has an ideal usable capacity (80% DoD) of 1,024 watt hours, at a discharge rate of 3,840 watts (3C).

After taking into account the DoD of these two chemistries, lead acid comes in at around $0.60 per watt-hour, and LiFePO4 comes in at around $0.57 per watt-hour.

Other advantages of LiFePO4:
-Operating temperature of up to 85ºC, rather than lead's 60ºC
-Truly a sealed, maintenance-free, non-toxic, and non-spillable paste
-No worry about plate thickness or corrosion

The only thing we would need are solar charge controllers and inverters designed for lithium iron phosphate.  That is, a charger with a goal of 3.7V per cell (3.2V nominal), and an inverter with a low voltage disconnect of 2.7V per cell.

Let met start by saying that I built my own off-grid 400 watt solar PV system with 6 of 100AH @ 12V lead acid batteries.  My new project is with an electric moped that uses LiFePO4 cells.  I put together a table that shows their relative merits.  Keep in mind these are the chemistry that isn't susceptible to thermal runaway like LiPo or Lion.  Is it time to make the switch for PV installations?

(Sorry for the layout, I can't post a proper table without using HTML tags)

*Lead Acid
Volts:           12
AH @ Standard C: 92
Cost:            $200.00
Lbs:            63
Operating Temp:    -40ºC to 60ºC
Standard C Rate: 0.05
Cycles @ Standard C: 3100
Power Density:   55.2
DoD:            30%
Usable Whr:    331.2
$ / Whr:     $0.60
Lbs / Whr:    0.190217391


*LiFePO4
Volts:            12.8
AH @ Standard C: 40
Cost:            $232.00
Lbs:        13.2
Operating Temp:   -45ºC to 85ºC
Standard C Rate: 3
Cycles @ Standard C: 3000
Power Density:    1536
DoD:            80%
Usable Whr:    409.6
$ / Whr:    $0.57
Lbs / Whr:    0.032226563

I'm thinking of making a battery box for 300AH of 12V AGM batteries from a wheeled Coleman 50 quart ice chest (the kind that you'd take car camping).  I was also thinking of cutting out some holes in the casing so I could mount a 1500W inverter and charge controller inside while having the plugs and displays accessible from the outside.

My reasoning:
*Portable for when I move out of my rental house.
*Have the whole thing self-contained and lockable.
*AGM batteries will be kept warm, and the inverter will even act as a battery heater.

My question:
Aside from the inverter, is there any reason to install cooling fans?  Is there any reasonable chance of explosion from the hydrogen gas and a spark from the inverter without a fan?

I've decided to resell the 5-watt panels I bought earlier and now I'm ordering 16 x 15-watt amorphous solar panels   But while their Voc is 21-23, the voltage under load is only 13.5-18.

I'm thinking that if I get a bottom-of-the-barrel panel at only 13.5 volts that they won't be able to charge a 12V battery bank.  So my plan is to string them together to create 27-36 volts and then use a 24->12V MPPT charge controller to charge the 12v batteries.  Is this a good idea, or can I save some money and safely stick purely with 12V?

I narrowed it down to the Morningstar Sunsaver (15A) and the Blue Sky SOLAR BOOST 3024i (30A).  I'm leaning more heavily towards the Morningstar since it's cheaper and can handle 240 watts (24V * 10A) with 50% overcurrent capacity.

I believe I found an answer from the "How to size a battery bank" page at the university library link.

Daily use in watt-hours: 500
Depth of discharge: 50%
Ambient temperature: 60 degrees (1.11 modifier)
Battery voltage: 12

By those calculations I'd need at least a 92.5 Amp-Hour battery.  But like you said, Michael, by getting a bigger battery I could lessen the DoD and prolong the battery life.  With a 105AH battery like the MK-8A31 I'd get a 44% DoD which might last 6-7 years.

Thanks for the information Michael.


I'm curious how you came up with that number.  Did you plug in numbers for a 6 or 12 volt battery?

350 watts / 12 volts = 29 amps
80 amp hours / 29 amps = 2.75 hours

From my math, if I ran 350 watt-hours from a 12V 100AH battery I would have approximately 71AH left which is well within the capacity limits of the battery.

However, my real question is if a 12V 100AH AGM battery could sustain a 29 amp discharge rate for any length of time, or is it limited by the plate surface, battery chemistry, heat, etc?  I don't want to ruin a battery by discharging it too quickly.  Obviously you can't discharge a battery in 1 minute or it's likely to overheat and die. I just haven't found any literature that lays out discharging guidelines.

Hi everyone,

This is my first post here.  I just picked up 24 5W solar modules (.35A x 15V) at about $3.33/watt. I'm looking to put together a simple off-grid system that will usually be called upon to run a floor lamp (45 watts) or to charge a laptop (90 watts).  Occasionally, however it would be nice to charge some power tool battery packs with it at about 350 watts (3A @ 120V for 1 hour).  I wouldn't try drawing over 400 watts for any length of time. 

I need help choosing an AGM battery that can support these intermittent loads.  Do I need a 100AH battery to support the 350WH load, or do I need something approaching 200AH?

Additionally, I'm looking at a charge controller (Solar Boost 2000E 25A, 12V) and a pure sine inverter (Samlex 600W, 12V).  Are these overkill for my project? I'm probably not going to add more panels in the near future (Not until I invest in a full 3-5KW grid-tied system).