Saturday, March 15, 2008

Carpe diem

Carpe diem, or “seize the moment”, is a good way to describe my latest effort to get the most from my off-grid solar electric system. Because electric rates are lowest during the early morning hours, I’ve implemented a plan to shift my electricity usage to that period of time. And no, my plan doesn’t involve staying up all night to do laundry or to cook meals.

My Variable Electric Rates

Although my actual rate varies from day to day, the chart below is typical of my rates for any given day. Notice that I pay about 2 ½ cents per kwh for electricity during the first five hours of the day, and much more than that at any other time of the day.

Building on the system I already have, I’ve added a sophisticated battery charger and a timer. The premise is simple: Charge batteries at night, when rates are low, and use the stored energy in the early morning hours when electric rates are higher. Shortly after that, the loads are powered by the sun. The diagram below is of my system, including the timer and battery charger.

Here is a picture of the charger and timer. Actually, I had to replace the timer. This one wasn’t “heavy-duty” enough.

Theory of Operation

The AC loads are powered by the batteries when battery SOC (state of charge) is high, and by the utility grid when battery SOC is low. That part of the system automation was explained in a previous article, and so I’ll not repeat it here.

The timer is set to apply AC to the battery charger between the hours of 1:00am and 5:00am every day. The battery charger is a three-stage charger, meaning that it will not overcharge. It switches to “float” mode when it senses full batteries, and it does not represent a discharge path when it is switched off.

Due to the evening loads, battery voltage is low just before 1:00am. When the timer applies AC to the battery charger, battery voltage rises rapidly. Once the batteries reach a predetermined SOC, the loads are switched from grid-supplied AC to AC from the system. At that point the charger continues to charge the batteries, and it also provides the power necessary to run the AC loads. At 5:00am the timer disconnects AC from the charger, and the loads are powered with the energy stored in the batteries. Soon after that, the sun comes up. On a sunny day, the solar panels keep the batteries charged and provide power to the loads. When the sun sets, the loads are once again powered by the energy stored in the batteries, and battery voltage declines. When battery SOC drops to a preset level, the loads are transferred to grid-supplied power. This cycle repeats every day.

Initial Test

The data logger was set to take battery voltage readings every 30 minutes. The sharp voltage increases (look just to the right of the vertical grid lines) show that AC voltage was applied to the charger. The vertical grid lines indicate midnight. Battery voltage stabilizes somewhat, and then sharply falls off when AC to the charter is removed. The batteries continue to power the load for awhile after that, resulting in a gradual decline of battery voltage. Then, the sun shines on the panels and battery voltage increases once again.

Except for the first day of the test, batteries were not fully charged. This is disturbing, since chronically undercharging batteries can shorten their life. For the duration of this test the load consisted of a refrigerator, a freezer, a TV, and a cable box. To correct the problem, I’ve reduced the load by removing the TV and cable box.

I initially set the timer to apply power to the charger for three hours each night. I determined that this was not enough, and increased the charger on-time to four hours.

The jagged lines on the graph are the result of devices switching on and off.


It looks as though I’ve managed to cut my use of utility-provided electricity just before the sun comes up, accomplishing my goal. But wildly fluctuating electric rates, and the inefficiencies of storing and retrieving utility-provided electricity, make it difficult to estimate any savings I might realize as a result of this plan. This has been an interesting experiment, but I don’t think I’ll continue it. I’ll be better off with a bigger battery bank, eliminating the need to use this timer/charger arrangement. Still, I can think of two benefits of this plan:

1. Batteries will tend to be kept at a higher SOC, extending their life.

2. This process is similar to the experience of those who actually live off-grid. Those who live off-grid usually have a secondary source of electricity, a generator perhaps. When lack of sunshine necessitates the use of the generator, using it for brief periods to charge batteries is the best strategy. In my case, I’ve substituted grid-supplied power for generator-derived power. Living off-grid is my ultimate goal, and this experience helps me learn more about that.



d said...

Interesting experiment: load shifting is a useful thing to do. At home we try to avoid the UK winter peak demand period 4pm to 8pm for the big appliances like the dishwasher and washing machine, but I also try to have my Web servers draw less juice at their local peak times. I hadn't particularly considered just moving (say) the fridge/freezer onto battery power overnight.

BTW, my flat rate (non-TOD) is about 3 times your peak from what I can see!

Also, congrats as you turned up in my Google Alerts even before I went hunting for your blog of my own accord this morning. Fame, maybe? B^>



Monitor de LCD said...

Hello. This post is likeable, and your blog is very interesting, congratulations :-). I will add in my blogroll =). If possible gives a last there on my blog, it is about the Monitor de LCD, I hope you enjoy. The address is A hug.

Câmera Digital said...

Hello. This post is likeable, and your blog is very interesting, congratulations :-). I will add in my blogroll =). If possible gives a last there on my blog, it is about the Câmera Digital, I hope you enjoy. The address is A hug.

Anonymous said...

Complex Post. This record helped me in my university assignment. Thanks Alot