Tuesday, January 01, 2008

My Solar Electric System and New Year's Resolution

Plans for my off-grid photovoltaic (PV) system have been pretty much the same for the past two years; enlarge the system and use more of the available power. I’ll continue to do both of these things in 2008, but I’ll also make a major change in the way I use energy from the system. In the past I’ve carefully monitored battery voltage and manually switched loads on and off. By doing this manually I not only miss opportunities to use energy from the sun, I risk damaging the batteries by over discharging them. To resolve these problems, and to relieve myself of the chore of manually switching between power sources, I plan to automate this task.

I’ll automate the switching on and off of loads using ideas from a recent post, “Getting the Most from an Off-grid System”. I’ve decided to use a transfer switch, and a circuit of my own design, to accomplish this task. Refer to the drawing below:


How it works:

The Transfer Switch is configured to use power from the inverter as the primary source of AC power for the load, only switching to grid-supplied AC power when the inverter is switched off. The inverter will be automatically switched on when the battery state of charge (SOC) is high, and switched off when the battery SOC has dropped to a predetermined value.

Thumbwheel switches, precision resistors, and a regulated reference voltage allow precise settings of the “Low Voltage Threshold” and the “High Voltage Threshold”. The thumbwheel switches and resistors create two voltage dividers. The output of the low voltage threshold voltage divider is equal to the low voltage threshold thumbwheel setting, and the output from the high voltage threshold divider is equal to the high voltage threshold thumbwheel setting. These two reference voltages are fed into two comparators.

When the battery voltage falls below the “Low Voltage Threshold” setting, comparator 1 changes state, triggering the Flip Flop, and the inverter is turned off. The Transfer Switch senses the loss of AC voltage from the inverter and switches the load to grid-supplied AC power.

When the battery voltage rises above the “High Voltage Threshold” setting, comparator 2 changes state, resetting the Flip Flop, and the inverter is switched on. The Transfer Switch senses the AC voltage from the inverter, and connects the load to the inverter.

Initially, I’ll set the low voltage threshold voltage to 12.25 volts. That voltage represents an approximate 75% state of charge (SOC). I’ll set the high voltage threshold at 14.75 volts, ensuring that the batteries are fully charged before allowing them to power the load.

It’s interesting to note that the battery voltage will not reach the high voltage threshold setting unless the sun is shining and the batteries have been fully charged. I’ll experiment with other settings in an attempt to improve system efficiency without endangering the batteries.

Benefits:

Since I’m not able to constantly monitor battery voltage, I’ve missed opportunities to use as much of the free power that my solar electric system is capable of providing. Instead, I disconnect the load when I think that the battery SOC may fall below 80%. Once I’ve implemented this plan, I’ll be able to use more of the available power from the system without the fear of damaging the batteries. Additionally, this automation will help to keep the batteries at a high enough SOC to ensure the availability of power in the event of a grid power failure.

With a Twist:

Because of a utility company plan that results in low rates at night, I’m thinking about storing energy in a larger battery bank at night when rates are low, and using that energy to power loads during the day when utility rates are high. And because my PV array is still small, I’m thinking about using a battery charger to supplement the charging that now comes from my PV array. The charger will be turned on via a timer in the early morning hours when rates are lowest, and turned off later in the morning before rates go up. When cloudy conditions limit the amount of charging my PV panels are able to provide, the battery charger will take up the slack and the batteries should be fully charged each morning. Confident that I’ll have plenty of stored energy in the morning, I’ll add more to the daytime battery load, and therefore save money on my electric bill.

With the charger switched on during the early morning hours, it’s likely that the batteries will quickly become fully charged. When that happens, the inverter will once again be used to power the load. Because the charger is still connected, and switched on via the timer, it too will provide power to the load. However, this will occur in the early morning hours when electric rates are low, and I’ll be taking advantage of the lowest rates, cutting my electric bill.

The graph below shows the expected results over a 24 hour period. As a result of charging from the PV panels during the day, and charging from the battery charger at night, the load will be powered by the batteries most of the time.

John

6 comments:

Unknown said...

Hi SJ,

I found an upper threshold voltage of somewhat under 14V, ~13.8V, worked better for my equivalent of your scheme.

Rgds

Damon

Anonymous said...

John,

How many PVs can be built before we run out of gallium and indium?

How many batteries can be made before we run out of nickel or lithium or whatever else they're using to manufacture them?

I admire your optimism, but there is no way in hell that we can replace the fossil fuels we're using now without drastically reducing the amount of energy we're currently using. This means living without air conditioning, electric dryers and all of the gadgets we've come to expect as our birthright.

Furthermore, do you honestly expect that everyone will switch over to using solar panels to charge their Chevy Volts?

Ultimately, we need to re-scale and re-localize. We need to rebuild our national passenger rail system. We need to get people out walking and cycling, especially for trips less than 3 miles. And don't give me the usual nonsense about inclement weather or disabled people as the logic goes "human power might not always make sense so therefore we should drive everywhere all of the time."

I seriously hope to be proven wrong by this, but it is mammothly irresponsible to assume that we'll simply switch over to another technology. We need to start making other plans, including a back-up that doesn't rely on things that might not even have a future.

Sean

John said...

We don't have to replace all of the fossil fuels we're currently burning with other energy sources, but we do need to conserve and use available energy sources more efficiently. Anyone who powers their home with PV will tell you that the first step is to reduce consumption. Having done that you can get by with a smaller system, and therefore use less of the limited resources you speak of. Other sources of energy, such as wind and hydro, will also limit the need for PV panels and the resources they require.

If you're buying groceries for a family of four, and you only want to go to the store once a week, you're going to have trouble getting them home if you're on foot or on a bicycle.

I'll keep adding to my PV system, and I'll buy a PHEV when they become available. I also supplement my home heating by burning corn. I don't understand your statement "...it is mammothly irresponsible to assume that we'll simply switch over to another technology." I've already done that!

I don't expect a mass exodus from grid-supplied electricity to PV panels anytime soon, since it's still cheaper to buy electricity from a coal-fired power plant. With low nighttime rates in some places, it's likely that most PHEV owners will charge their vehicles with power from the grid. For that reason, those who want to use clean alternatives should have no trouble doing so, since supply and demand issues shouldn't be dramatic.

Adding PHEV charging to the already strained electrical grid will be a headache, but it can be fixed. My utility company makes huge profits, and will be making even more when PHEV's are added to the load. It's time to use some of that money for improvements. It's also important to note that most PHEV's will be charged at night, when electricity demands are low.

I'm all for mass-transit, walking, and biking when it makes sense to do so. Perhaps those who are most responsible for consuming resources irresponsibly and for polluting the environment should have to pay accordingly. Oil and coal should be phased out gradually, giving us time to adjust. I believe we'll start to see a change if we get a president who truly cares about the environment, but I encourage ordinary people to take the lead instead of waiting for politicians.

sj

Unknown said...

Dear SJ,

I found your postings very educational. I would like to invite you to checkout my new free public service page, for 'Predictive performance diagnostics', at:
www.pvperformance.com. I believe your input will be valuable in helping us to evolve it into a truly useful tool for many smaller PV system owners. It is put up in the hope to reduce waste due to un-detected under-performance in PVs. Thanks,
Steve scyang(at)wattminder.com

John said...

Update:

I've installed the transfer switch over the weekend. Now I'm ready to move on to the battery control circuit I described in the original post.

sj

Unknown said...

Ah, interesting! Keep us posted...

I worked out that my contribution to the UK's electricity will be about 2x10^-9 when I've installed my next PV panels, which is a little daunting, but we all have to try IMHO! But it's good to know that I already eliminated more than 10x that waste with hardware and software upgrades to my systems (and I've just been writing up my 'low-power software' thoughts today)...

Rgds

Damon