My new experiment plan

BB. wrote: »

Wayne was just interested in the 180 VAC inverter. These high voltage AC inverters are not very standard around here and Wayne was trying to find out the details about the UPS.

With 3 Midnite controllers, you will be looking at 3 banks of 60 VDC battery connections (you will have to program the Midnite that this are 60 VDC battery banks and the appropriate charge parameters).

So, you are going to have 4 "taps" coming off of the battery bank. One at zero volts, a second at 60 VDC, a third at 120 VAC, and the fourth at 180 VDC.

Assuming that the battery bank is negative grounded to your building's green wire ground/safety ground system, then each tap will need a 100 amp breaker/fuse, other than the zero volt first tap.

Note, I am guessing that the battery bank is negative grounded. With a UPS, it may actually not be negative ground, and if not, then you should put a fourth breaker/fuse on the "zero volt" line too (because it is floating and not at zero volts).

Most sine wave AC inverters have isolated AC output voltage and can have the DC input ground referenced--But you need to verify that this UPS can have the negative battery bus grounded. Do not assume anything without doing the research--This is a lot of power/current/large battery bank and it can be very easy to end up with a lap full of red hot copper and battery parts+sulfuric acid embedded in the walls and inside you too.

Another note--DC current is much more difficult for a switch/fuse/circuit breaker to "open" or "turn off". AC current switches polarity 50/60 times per second. So, you will find that the DC versions of switches/breakers are much larger/heavier construction vs their AC counter part.

For example, in the US we have a line of house hold circuit breakers that are rated for 240 VAC operation (really 600 VAC) on 60 Hz AC. The same breaker is rated for a maximum of 48 VDC.

In the US/Solar RE market, the largest "standard" DC rated breakers (from Midnite) are marked to a maximum of 150 VDC. Your battery bank runs >200 VDC. The standard solar/RE breakers will not be safe for use on your high 180 VDC battery tap. You will have to find properly rated DC breakers/fuses/switches for this system.

Remember that most solar charge controllers will have their internal electronics connected directly to the "negative battery terminal" inside the charge controller. Any communication ports (like RS 232, RS 422, etc.) will be at your "negative tap" voltage (the Ethernet connection is isolated by a transformer--But you are pushing the isolation voltage ratings of these transformers).

Why are you looking at this sort of power system? Is it to save money, emergency backup power, or a full off grid power system?

There is another possibility that may end up being easier and safer, and possibly less expensive.

Many Sine Wave type AC inverters are actually bi-directional power devices. We are used to the DC Battery bank supplying power through the AC inverter to the AC loads.

However, many AC sine wave inverters can also be driven backwards. You can put the proper type of AC power source on the output of the sine wave inverter and actually push energy "backwards" through the sine wave inverter and recharge your DC battery bank.

Two things... The "proper" AC power source is a Grid Tied type AC inverter (solar panels>GT Inverter>AC sine wave inverter's AC input>DC Battery bank).

Second thing... Most "off the self" UPS and Off Grid AC Sine Wave Inverters are not designed to manage the backwards current flow. If you connect a Solar Array + GT inverter to the Sine Wave OG AC inverter, it can/will overcharge the DC battery bank. You need to control this current flow.

Two basic methods, one is to put a DC shunt load (resistor load bank/electric heater) on the battery bank that can absorb excess charging energy (once the battery bank is full). You could also put the dump load on the AC output side of the sine wave inverter too (i.e., an 8kWatt heater, turn on for 10s of seconds to minutes or more to dump excess energy. Battery becomes discharged a bit, and the dump load is turned off for a bit).

A second method is to put a "relay" in the AC output of the GT Inverter... When the battery bank is full, the battery voltage monitor simply turns off the GT inverter's AC output (through the relay, some GT inverters also have an "inhibit" input too). GT inverter will turn off and wait for 5 minutes of AC power, then turn on again.

If you are "serious" about using the 180 VDC inverter system--I would look very closely at purchasing a GT inverter (230 VAC 50 Hz for UAE?). Just one GT inverter connected to the AC OG Inverter's AC output (~10,000 watts of solar array and ~7.700 Watt worth of GT inverter maximum for your 10 kWatt system using "nominal numbers). The GT inverter's output cannot exceed the OG inverter's AC input power.

Note--Many devices are rated at very high power levels, but when the actual demand (or back driving of power) at rated power--They can overheat/fail. You might want to limit power used/generated to 80% of name plate until you do some experimenting to make sure they do not overheat/fail. Many devices if operated at 100% rated power for 5 hours (such as recharging the battery bank through the sine wave inverter during the day) will overheat.

"AC Coupling" has been done on some OG AC Inverters and is known to work--And may be easier/safer than using the three Midnite charge controllers on the DC side. You need to do your own research and make sure that your installation is safe (and you have battery charging limit hardware reliably installed and maintained). In the US electrical code, that would mean there would be two independent methods of controlling the charging current (such as two battery charge monitors and two AC relays to turn off the GT inverter).

Shunt controller type battery charging is less than ideal (series charge controllers such as the Midnite on the DC side is better). But shunt control of battery charging is done all the time.

-Bill

Cariboocoot wrote: »

In addition to what Mike said about making sure you have all the specs on the AC side of things ...



Each of those controller outputs needs at least a fuse or breaker on the positive side for safety. As Bill mentioned if it is 'floating' then a double poll breaker disconnecting both positive and negative is safest.

Will you actually be using two parallel strings of batteries? I thought you indicated before that they were 200 Amp hour 12 Volt units all in one series string.

Cariboocoot wrote: »

Not necessarily wrong, but is it needed to have the 200 Amp hours?
In other words what are the specifications on the individual batteries?

Cariboocoot wrote: »

Yes, but what batteries are you actually proposing to use?

For example you could use these 12 Volt 115 Amp hour batteries http://www.solar-electric.com/repoba12vo11.html and need two parallel strings to get to 230 Amp hours capacity.

Or you could use these 12 Volt 530 Amp hour batteries http://www.solar-electric.com/crinba6.html and have more than double your desired capacity on one string.

There are many variations between those two examples. Including 24 Volt industrial, 6 Volt cells (220 Amp hour common GC2's), and 2 Volt cells. It is best to have only one string of batteries if possible: fewer connections to go wrong and fewer cells to check.

Cariboocoot wrote: »

I wasn't suggesting you buy those batteries; merely using them as examples to explain the different possibilities. Shipping on batteries is a huge cost so it is nearly always best to buy them locally.

If you've got 200 Amp hour 12 Volt batteries and only need 200 Amp hours then you only need one string of them. At 180 Volts that is a whopping 18kW hours of stored power. Your proposed 6kW hour load need is well below that.

You're quite right to distrust someone selling you AC rated breakers for DC as they won't work. At least not at the same Voltage level. You actually need breakers like this for the controller outputs: http://www.solar-electric.com/misomn150vdc.html Notice the smallest of these is 80 Amps so the wire should be sized accordingly. Note too that they are not double pole breakers.

For the inverter input you would need something like this: http://www.solar-electric.com/miso30dccibr.html

solar_dave wrote: »

How about taking a different tack, and run the desalination only in daylight hours, should cut you battery requirements quite a lot.

solar_dave wrote: »

Exactly, the water is durable and perhaps doing most of it just off panels is much more effective. The battery bank you propose is pretty huge. If the location is truly off grid you would only need enough battery to buffer the daytime solar variability.

solar_dave wrote: »

So what is more cost effective? Small battery with lots of solar X 2 desalination plants running part of the day? Or huge battery with all the issues to get 7X24 operation. Do these desalination plants need down time for maintenance?

pleppik wrote: »

Where are you getting the power consumption number? Ten thousand gallons of water per day is enough for 20 houses (I assume this is for irrigation or something similar), and 6kWh doesn't seem like even remotely enough power to desalinate using conventional techniques like reverse osmosis.

I did find this article, though, about a demonstration solar desalination plant in California. It looks like it's basically a giant solar still.

If that's what you want to attempt, then the plant will only desalinate water while the sun in shining anyway, so there's no point in pumping at night. The energy to actually desalinate the water is coming from the sunlight, and I would assume the 6kWh/day is what's required to operate pumps and fans to keep the whole thing running.

Edited to add: According to the company's website, they use thermal storage to allow 24/7 operation, so you can still pump at night. Buying batteries to allow 24/7 operation off-grid will be more expensive, though, so with a major capital purchase like this it probably warrants a more careful cost analysis.