Off-Grid: AC-what usable % of DC energy, now with Lithium!

Getting into the nuts and bolts of wind and solar power components, installation and performance issues, ask technical questions, answers and tips

Off-Grid: AC-what usable % of DC energy, now with Lithium!

Postby Gordon-Loomberah » Thu Feb 28, 2013 3:28 pm

In November 2010 I started a thread about how much of your DC PV ouput is available as usable AC energy,
the test was done with Lead-acid batteries.
Now I'm using LiFePO4 cells, with 800AH capacity, ~2/3 the rated capacity of the Lead-acids, although given their age, capacity was probably quite similar. The Lithium battery operates at a higher voltage, giving marginally higher (~5%) overall kWh capacity per AH.

The test does not take into account how efficient the PV panels are, nor how efficient the charge controllers are, just output from charge controller to usable AC energy.

To summarize the previous findings in Nov 2010:

Flooded Lead-acid batteries (7yo) nominal 1260AH @24V, Latronics 3kW inverter
Test Duration: 31days 1hr (Start and end SOC ~100%, but not measured precisely)
Total DC output from charge controllers: 357.3kWh
Total DC energy used by inverter: 272.3kWh (+3kW fan), at an average rate of power usage of: 369.53W
AC energy used according to the digital kWh meter: 233.5kWh, Average = 7.5kWh/day
Overall inverter efficiency: 86%
Battery charging/discharging/float etc, efficiency: 77%
Overall, about 66% of generated DC energy is available at the inverter output as 230VAC.


Feb 2013:

CALB LiFePO4 cells 800AH @ 26V, Latronics 4kW inverter
Test duration 14days 9hr 35mins (start and end @ the same State of Charge- 2.6kWh down from 100%)
Total DC output from charge controllers: 126.75kWh
Total DC energy used: 117.48kWh, at an average rate of power usage of: 339.94W (fan not logged)
AC energy used according to the digital kWh meter: 101.36kWh, Average = 7.0kWh/day
Overall inverter efficiency: 86%
Battery charging/discharging/float etc, efficiency: 93%
Overall, about 80% of generated DC energy is available at the inverter output as 230VAC

[EDIT-the charging and overall efficiencies for the Lithium cells are >1% better than stated above, due to the 3.6W fan and tracker drive motor power use not being included in DC load, but the charging has to cover it. *See March 28th for charging efficiency update]

Lithium batteries are far more efficient!

They have a number of other advantages including the ability to accept much higher charging rates than Lead-acid batteries, and the ability to accept those high charge rates up to almost 100%SOC, without the need for extended constant voltage absorb stage charging, typically about 3 hours with Lead-acid batteries of ~800AH capacity cf about 20minutes with LiFePO4 cells.
End amps for absorb charging (ie when 100%SOC is reached) is typically 1-2% of C20 capacity for Lead-acid, cf under 0.5% for LiFePO4s. The Lithiums are down to about 1% within a couple of minutes of reaching absorb voltage, the next 15 or so mins sees that drop back down to under 0.5% as the cells balance, typically to within 0.02V.
Last edited by Gordon-Loomberah on Mon Jul 07, 2014 10:15 am, edited 5 times in total.
Reason: added a note re 3.6W fan use
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5031
Joined: Thu Sep 24, 2009 10:41 pm
Location: Loomberah NSW Australia

Re: Off-Grid: AC-what usable % of DC energy, now with Lithium!

Postby jaahn » Sat Mar 02, 2013 2:21 pm

Hi :D
Thats some usefull stats there. Interesting too. :mrgreen:

Could you indicate the relative costs for the batteries and the regulators.

By my crude calcs the increase output is ~ 21%. 8-) So if you were needing extra generation the Li batteries would be equivalent to 21% extra panels. The batteries may be cheaper than the extra panels ?
Cheers jaahn
jaahn
Solar Evangelist
Solar Evangelist
 
Posts: 89
Joined: Sun Apr 25, 2010 9:42 am

Re: Off-Grid: AC-what usable % of DC energy, now with Lithium!

Postby Gordon-Loomberah » Sat Mar 02, 2013 3:42 pm

Yes, that 21% extra sure makes a difference. In fact it is amazing just how much of a difference it makes- during a couple of cloudy spells this year (in one now) I've been quite surprised at how quickly I get back to 100% State of Charge in marginal charging conditions, and not needing to run a generator (but would have had to with Lead-acid) is a big bonus :)

Whether it is equal to 21% more panels is not so clear cut- it depends somewhat on your usage patterns- day-night, how big a load, etc, and I think in some cases it may effectively be better than that. Because you can charge at a very high rate right up to over 95% SOC, you can make much better use of marginal charging conditions like brief sunny breaks between clouds as the weather clears, whereas with Lead-Acid batteries you are stuck with the slow process of absorb stage charging, which needs to run ~3 hours with this size battery for my typical usage patterns. Having a Lithium battery also allows you to better utilise a larger array of PV panels, with no real restriction on how fast you can charge (or discharge) them for typical off-grid systems.

Panels are getting relatively cheaper these days, but you dont need as large a battery if you have Lithium, so it is a bit hard to exacly compare. However, the 800AH @26V of Lithium cells (21kWh storage) and battery balancing equipment I've just bought cost a bit over $10000, including freight.

These 900AH Raylite batteries: http://www.apolloenergy.com.au/Renewabl ... 00ah-p-256
would be a bit over $6000 + fittings and freight, but if you wanted them to last a similar time to the claimed life of my Lithium cells, you could only use about half as much of their capacity, or to have the same effective kWh storage, you would need twice as many of them. Clearly, for a new system, a Lithium battery is the best way to go right now.

They do need a bit more work for the initial set up, but after that, they require less maintenance than a Lead-acid battery, and you dont have the worry about handling acid! :)

The Outback Flexmax charge controllers are able to charge the LiFePO4 cells, as they have a range of adjustments that covers what is needed for these cells, so there is no extra expense involved. The Latronics inverter also works well with these cells, the only adjustment I altered was the low battery disconnect, which I raised from 20V to 24V via internal switches. This is the primary low voltage disconnect to protect the batteries from being discharged too deeply. The Battery Control Unit also has a LVD, which is there as a backup.
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5031
Joined: Thu Sep 24, 2009 10:41 pm
Location: Loomberah NSW Australia

Re: Off-Grid: AC-what usable % of DC energy, now with Lithium!

Postby Gordon-Loomberah » Sun Mar 03, 2013 1:55 pm

In January we had a cloudy spell that lasted a few days, rather similar to the one that's lasted 3 days so far to today, so I decided to see how the Lithium batteries would go in a large discharge situation with no generator use.
Cloud was increasing on the 26th, but the batteries were at Vfloat most of the time in the afternoon when the cloud wasn't too heavy, so that's when the plots below start. The next day had a small amount of charging, and the 28th almost none with about 50mm of rain falling all day from a heavy overcast sky, but a bit of patchy sun did appear on the 29th, and the 30th was sunny.

Here is a plot of solar radiation and what I'd expect (Yellow Max Solar curve) in clear sunny weather for the duration of the test:

Jan26-30Solar-Radiation.gif
click to de-fuzz
Jan26-30Solar-Radiation.gif (8.53 KiB) Viewed 8162 times



This is the plot of gross charging and power consumption, along with gross cumulative energy production (X 0.94, see later) and usage:

Jan26-30Gross-power-energy.gif
Click to de-fuzz



Here we can see how flat the voltage curve is with increasing discharge, down to about 71%DOD
The specs say the battery should be good for discharging to this level for 3000 cycles.

Jan26-30Li-discharge.gif
Click to de-fuzz


Interestingly the usual 0.93 factor I apply to the cumulative SOC curve didn't quite fit in this case, because there is no float time represented in the graph- only charge and discharge, so the battery efficiency worked out to be a bit better, at 94%
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5031
Joined: Thu Sep 24, 2009 10:41 pm
Location: Loomberah NSW Australia

Re: Off-Grid: AC-what usable % of DC energy, now with Lithium!

Postby Cherokee Solar » Sun Mar 03, 2013 6:42 pm

Hi Gordon,

The thing that stands out to me is that the voltage is rock solid. Do you then measure the SOC through a comparison of kW in to the battery versus the kW load? Do you see this on the regulator, inverter or the data logger?

The other thing that stands out is how quickly the batteries can be recharged. The amps they can absorb is quite amazing. Do they get hot during this process? I think the batteries here are to be recharged at no more than 25% of their total capacity (300A!), well up until they get to about 85% SOC at which point it tails off considerably.

The sunny breaks you get over winter will be quite telling and it is good to see that the batteries will be able to absorb them all. I hope you keep us all informed.

Gordon-Loomberah wrote:Here we can see how flat the voltage curve is with increasing discharge, down to about 71%DOD
The specs say the battery should be good for discharging to this level for 3000 cycles.


They should give quite a long service life as I can't imagine you'd get down that low every day?

What affect does ambient temperature have on the batteries? Are they as sensitive to temperature changes as lead acid?

I read on the spec sheets that if my batteries are operated at above 35 degrees, they will have a reduced service life (probably due to loss of electrolyte?). Fortunately they haven't been beyond 29 degrees that I'm aware of.

I'm thinking of changing the battery room door here by changing the fire rated roller shutter to a crimsafe stainless steel mesh shutter so that I can leave the room open on cool nights over summer and let more fresh air in to cool the batteries and inverter down. The hot overnight weather this summer has been unprecedented.

PS: Last Wednesday (27/2) I never got to Float here. It was the first day not to get to Float since August last year...

Regards

Chris
Off grid solar + hot water. Heavily insulated + owner built flamezone house BAL-FZ. 300 mixed fruit trees + herbs + flowers + vegetables. Bees + heritage chickens. High up in the mountains north of Melbourne. http://ferngladefarm.blogspot.com.au/
Cherokee Solar
Solar Crusader
Solar Crusader
 
Posts: 822
Joined: Tue Jan 05, 2010 10:11 pm
Location: Cherokee, Victoria

Re: Off-Grid: AC-what usable % of DC energy, now with Lithium!

Postby Gordon-Loomberah » Sun Mar 03, 2013 8:23 pm

Cherokee Solar wrote:The thing that stands out to me is that the voltage is rock solid. Do you then measure the SOC through a comparison of kW in to the battery versus the kW load? Do you see this on the regulator, inverter or the data logger?


Yes, the voltage only dropping down about half a volt by the end of that big a discharge from its normal ~26.5V overnight is very nice :)
I calculate and plot that SOC curve by multiplying cumulative energy output (V,A sampled at 5 sec intervals) by the 3 charge controllers by 0.94 (in this case with no float, normally 0.93), less the cumulative energy used by the inverter. The actual efficiency is slightly better though, as the (3.6W) load of the Clivus Multrum fan isn't being logged at the moment. I just need to add another shunt for it, which will also be used for the 2nd inverter to run a 2nd circuit in the house, mainly for the RCACs, plus a few extra GPOs for when I want to run multiple high power devices.

The amps they can absorb is quite amazing. Do they get hot during this process?


I've seen up to 150A going in a few times, including on 29/1, but they dont get hot at all at that charge rate ~19%.

I usually only go down to 15-20% DOD each night

These cells don't need temperature compensation for charging, their temperature seems to follow the room temperature fairly closely. I'll whack in a thermocouple one of these days to monitor their temperature on the data logger.

Yes, Lead-acid batteries do have reduced life if they are hot, but reduced capacity if cold. I dont know if the Lithium batteries will show any reduced capacity at colder temps, but haven't read anything to that effect.
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5031
Joined: Thu Sep 24, 2009 10:41 pm
Location: Loomberah NSW Australia

Re: Off-Grid: AC-what usable % of DC energy, now with Lithium!

Postby Cherokee Solar » Sun Mar 03, 2013 8:47 pm

Gordon-Loomberah wrote:3 charge controllers by 0.94 (in this case with no float, normally 0.93), less the cumulative energy used by the inverter. The actual efficiency is slightly better though, as the (3.6W) load of the Clivus Multrum fan isn't being logged at the moment.


Interesting stuff.

How did you come up with the 0.94 multiplier?

Do your controllers use any noticeable power especially under high amp DC to DC conversions, or do you take this into account in the multiplier as an average over the day?

It is great to hear how good they are. I've wondered whether an off grid system would be better with smaller capacity batteries and a much larger capacity for generation. These batteries seem ideal for this scenario and would ultimately cut the price down significantly for an off grid system.

Chris
Off grid solar + hot water. Heavily insulated + owner built flamezone house BAL-FZ. 300 mixed fruit trees + herbs + flowers + vegetables. Bees + heritage chickens. High up in the mountains north of Melbourne. http://ferngladefarm.blogspot.com.au/
Cherokee Solar
Solar Crusader
Solar Crusader
 
Posts: 822
Joined: Tue Jan 05, 2010 10:11 pm
Location: Cherokee, Victoria

Re: Off-Grid: AC-what usable % of DC energy, now with Lithium!

Postby jimbo » Sun Mar 03, 2013 9:10 pm

What size lead acid bank would be fairly similar to your lithiums? How many cycles will you get at 50% dod?
jimbo
Solar Crusader
Solar Crusader
 
Posts: 550
Joined: Mon Jul 16, 2012 4:48 pm
Location: Ballarat

Re: Off-Grid: AC-what usable % of DC energy, now with Lithium!

Postby Gordon-Loomberah » Sun Mar 03, 2013 10:41 pm

Cherokee Solar wrote:How did you come up with the 0.94 multiplier?


Best fit to the data! I know when the batteries are at 100% and I know when they get back to 100% after being discharged, multiplying the charging energy by a factor of 0.94 brings the SOC charging curve back to zero when the battery reaches a SOC of 100%. As I mentioned, the actual efficiency is a bit more, probably closer to 95%, as I wasn't logging the energy used by the fan. I'll get that logging sorted out sometime in the next month or 2.

Do your controllers use any noticeable power especially under high amp DC to DC conversions, or do you take this into account in the multiplier as an average over the day?


They do use a little, but are quite efficient- in the high 90s%, but this is not part of the calculations, which only measure from the output of the CCs.
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5031
Joined: Thu Sep 24, 2009 10:41 pm
Location: Loomberah NSW Australia

Re: Off-Grid: AC-what usable % of DC energy, now with Lithium!

Postby Gordon-Loomberah » Sun Mar 03, 2013 11:35 pm

Jimbo, they specify: "Cycle life: 2000+ to 80% DOD, 3000+ to 70% DOD, measured, not just claimed",
so I would estimate something in the 4000-5000 range of cycles to 50% DOD

Claimed number of cycles to a certain DOD for Lead-acid batteries varies quite a bit between manufacturers, so I couldn't say specifically what size Lead acid battery would last as long as the Lithiums, other than perhaps one of about twice the rated capacity, for an equal useable amount of energy, and similar life.
http://gunagulla.com Loomberah weather and astronomy including live solar radiation intensity and UV + Gunagulla aquaponics, organic eggs and cherries
User avatar
Gordon-Loomberah
Community Moderator
 
Posts: 5031
Joined: Thu Sep 24, 2009 10:41 pm
Location: Loomberah NSW Australia

Next

Return to Components, Installation And Performance

Who is online

Users browsing this forum: Google [Bot], qed and 1 guest

new solar power specials