short term storage to maximize time of use payback?

System

Apologies if this is covered elsewhere. I searched but with no joy.

I've just had a 2.4kW PV system installed in SF area inside PG&E territory (though as of Friday we'll be part of the Marin Clean Energy which has a more generous payback for grid-tied systems, per http://marincleanenergy.info/PDF/Net_Metering.pdf).

As so often happens (to me anyway!), now that its all in and paid for (though not operating yet as PG&E is approx 2-4 weeks behind in processing final signoffs and swapping my meter to a "smart meter" for Time of Use (TOU)), I'm busy discovering some of the subtleties.

One of the most fascinating is that maximizing power generated does not equal maximizing money. Because my system is SE facing rather than S or SW, I've calculated that a good chunk of power (almost 1/2) being generated before the peak TOU period of 1-7pm.

This naturally led me to wonder if its financially feasible to store some of that generated power for a few hours and then return it to the grid in the peak use period. Of course, if this truly made sense, it would be happening already, but I've been curious enough to run some back of the envelope numbers as follows:

Say I shifted 4 hours of maximum possible morning production to the afternoon. And, just for yucks, say it put me into the 3rd tier of peak from the 1st tier of "part peak". That would be the difference between 15c/kwH vs 47c/KwH. Call it 32c * 2kWH * 4 hrs * 26 weeks * 5 days = $332 / yr (at the most!).

Is it even feasible to build a system that could cycle 8kWh every weekday with 6 hour charge + 6hr discharge for 1/2 year at a time that would cost less than $332/yr? (Where are those flywheels!)

Oh, and is it possible to (legally) gridtie such a system?

Thoughts? Many thanks!


PS: the PV is a 10 x DelSolar 230 with an SMA3000US inverter.

BB.

Re: short term storage to maximize time of use payback?

My system faces SSE--And with TOU I just have to accept it is what it is.. (I am south of San Francisco).

I have looked at a battery based system to peak shift--and by the time you add hardware, batteries, and replacement batteries every X Years---the costs are somewhere around $0.45 or so per kWHr. And that is before you include the 10-20% loss in efficiency (charging/discharging batteries, battery charger, invertering, etc.)...

TallGirl, another poster here, has (if I recall correctly) said that it should not cost what my estimates worked out to be...

In some locations (Quebec Canada, and an eastern European country--IIRC), had rules that dis-allowed even having battery connected inverters to prevent people from "buy cheap power and selling it at a more expensive time)...

You could probably do something like this with an XW inverter from Xantrex (shift "sell voltage" based on time (currently sell voltage is a fixed amount if I recall correctly)... But, for my 2 cents, the costs of battery replacement (and shorter life because of daily cycling) makes it a difficult deal to come ahead on when costs and losses are included.

If somebody figures out how to peak shift their power--and whether they make money or not--some regulatory agency is probably going to try and outlaw it...

In the end, the rate plans are constantly evolving and I would not spend a whole bunch of capital trying to adjust for the flavor of the day rate policies... They have changed quite a bit over time--and with smart meters will change even more... And, for the most part, the changes have been detrimental to the generic Grid Tied Solar customer... One of the reasons I went earlier (~5 years ago) was because I could grandfather my installation into the old E-7 TOU plan. Very simple and peak was from noon-6pm... The new E-6 plan (E-7 is not longer available) can push peak and partial peaks way into the late evening---Making it impractical for (at least my family) to put off cooking dinner / doing home work until 9pm--long after the sun sets...

I think I will be forced to switch over to a Smart Rate Plan eventually---but I don't know how that will affect me... The plans I have seen so far probably are not as advantages as the E-7 plan (residential TOU) I have had for the last 5 years.l

-Bill

AntronX

Re: short term storage to maximize time of use payback?

Lets assume you get 6 hours of sun per day on average, PV output derated by 20% due to heat, battery charge controller is 97% efficient, wire resistance loss is 3%.

2.3 KW * 80% * 6 hrs * 97% * .97 = 10.39 KWH

Now assume you want to shift half of your generation.

10.39 KWh / 2 = 5.2 KWh

For this estimate, I will use Lithium Phosphate batteries, you can cycle them down to 70% depth of discharge and get 3000 cycles, and get 98% charge - discharge efficiency. Lead-Acid battery is very inefficient for this project. Your inverter will be 95% efficient. First, how much will this battery cost. Then how much energy it will store:

5.2 * 1.3 = 6.76 KWh / 3.2V = 2113 Ah * $1.1 = $2324
5.2 * 3000 * 95% * 98% = 14,524 KWh

So, this system will shift 14,524 KWh over 3000 battery cycles, or 8.21 years, assuming 1 day = 1 cycle.
Next let's calculate cost per KWh of battery only:

$2324 / 14,524 KWh = $0.16 per KWh

Here is your "profit":

$0.32 - $0.16 = $0.16 * 14,524 KWh = $2,324
$2,324 - $2,324 = $0

Congrats! You broke even. Just ignore inverter and charge controller cost. And cost to rewire your array. And cost of other minor parts to make this project work. On another note, this is very crude estimate, and I could have missed something. If you fine-tune this with your real numbers, maybe you will profit something?

AntronX

Re: short term storage to maximize time of use payback?

Forgot to write this: Battery will not fail at 3000 cycles. It may go on to much more cycles, but it's capacity will fade. So, if you keep doing this you may even pay back for other costs.

I think this exercise shows that excess energy shifting by using Li-phosphate battery is economically feasible, only as long as you are paid for performing such service. This is beneficial to the grid for smoothing demand peaks and absorbing cheap excess energy. Wind power comes to mind.

russ

Re: short term storage to maximize time of use payback?

İt seems that if this type of thing worked utilities would pickup on it.

The fact that they don't use it probably tells us something.

BB.

Re: short term storage to maximize time of use payback?

At this point, with rare exceptions (for example, perhaps hydro)--the cost of storing the energy is more than the costs of generating it... A power plant + fuel costs less than power+storage system costs today...

Knowing your loads (presumably a heavy industry running 24x7x52) is a more "utility friendly load" than others (retail, residential) where time of day, outside air temperature, etc. are more cyclic throughout the day and have high peaks at certain times of the year.

-Bill

AntronX

Re: short term storage to maximize time of use payback?

russ wrote: »

İt seems that if this type of thing worked utilities would pickup on it.
The fact that they don't use it probably tells us something.

It does not tell us much. Utilities have no use for this technology at this time. Some are testing lithium battery systems, not for bulk energy storage but for frequency regulation. That is a niche where good profit can be had. A123 Systems makes 2 MW shipping container sized unit for this purpose. Only when utilities are mandated to switch to solar, wind power generation, they will require this technology.

BB.

Re: short term storage to maximize time of use payback?

Yes, there are a few threads here--but I cannot find one with details in it at the moment...

Typically, when I have done generic calculations with charger/gt-hybrid inverter/lead acid batteries... It cost, just to shift the power, around $0.45 per kWhr (assuming 20 year system life, and battery replacement every XX years because of wear-out from cycling).

Another person here did the calculation with the new Lithium Iron Phosphate batteries--that the battery replacement cost alone could be as low as $0.15 per kWhr.

So, figure out how much your peak shifting system would cost to install. How often and how much you would need to replace the battery bank every XX years. Divide those costs by the kWHrs to be shifted over 20 years.

In the end, if you had a Time Of Use billing plan, you may be able to break even after all costs are taken into account. (remember that batteries are 80-90% efficient charging, plus charger+inverter losses--so you have to buy another 25% power to waste during load shifting).

Usually, a pure GT inverter + solar panel system and a nice 1 year net metering agreement with the utility makes the best economic sense.

If you have a need for emergency power (ice storms, hurricanes, etc.)--then a Hybrid System (Grid Tied + Off Grid backup), while more expensive--may be worth its weight in gasoline for your generator if you have several week of no grid power once or more per year.

-Bill