Stand alone solar power system

[g-and-h_electric]

Hey guys I need a bit of help with a newly installed stand alone solar power system, and I have NEVER done one of these before. Apologies for sounding like a DIY'er, but this got "dumped" on me, and I am lost!

Let me start off with I did NOT design, engineer, or purchase what is there, I was told it was sold as a complete system. The system was purchased from an online vendor,

There are 4 2x4' panels connected to give 24 V output ( I dont have the spec on them yet). These are connected to a charge controller, which then goes to the batteries (2x100Ah in series)...... so far so good right?

The batteries are then connected DIRECTLY to the 1000W inverter, which is running at max 300w of load (the load is a small cell transmitter, and flow controls for some sewage monitoring stuff).

The problem is that in Chicago we have had a LOT of overcast days lately, and the battery voltage is down to 22.1, the inverter went into a fault mode which I think was low voltage

My questions are this: 1) should the inverter come off the charge controller as opposed to directly off the batteries?, 2) Based upon the connected load, would a smaller, say 600W inverter be more efficient for this application, seeing as we have a dedicated and known load?

What Am I missing????


Howard

 

[GoldDigger]

The inverter must come directly off the batteries rather than off a controlled output of the charge controller (CC). Many CCs have a controlled output which can be programmed for a variety of uses, including turning it on only when the panels are not supplying power (night lighting) or only when the battery voltage is above a fixed or programmable minimum value.
But they all have in common that they are not designed to deliver full battery power to a load greater than the nominal capacity of the CC. It is true that the low voltage cutoff of the inverter itself is generally so low a voltage that lead-acid batteries will be damaged at that point. The solution is not to rely on the CC for that function or to use a relay driven by the programmable CC output or by a separate low voltage detector to control the battery supply to the inverter.
Also, it is imperative that there be appropriately sized fuses at or near the battery terminals to protect the wiring to the inverter, even if there is a fuse built into the inverter.
In general the smallest inverter with the needed maximum output capacity will be the most efficient, and will have the lowest minimum drain when the load is off or lower than maximum. There are exceptions, in that some smaller inverters may not be well designed in the first place. Also keep in mind that you will need significant extra capacity if you have motor loads. Unlike a generator, an inverter does not have inherent short term overload capability from its rotating mass. If the inverter shuts down from short term overcurrent instead of limiting the output current you simply will not be able to start the motor.

 

[gadfly56]

Hey guys I need a bit of help with a newly installed stand alone solar power system, and I have NEVER done one of these before. Apologies for sounding like a DIY'er, but this got "dumped" on me, and I am lost!

Let me start off with I did NOT design, engineer, or purchase what is there, I was told it was sold as a complete system. The system was purchased from an online vendor,

There are 4 2x4' panels connected to give 24 V output ( I dont have the spec on them yet). These are connected to a charge controller, which then goes to the batteries (2x100Ah in series)...... so far so good right?

The batteries are then connected DIRECTLY to the 1000W inverter, which is running at max 300w of load (the load is a small cell transmitter, and flow controls for some sewage monitoring stuff).

The problem is that in Chicago we have had a LOT of overcast days lately, and the battery voltage is down to 22.1, the inverter went into a fault mode which I think was low voltage

My questions are this: 1) should the inverter come off the charge controller as opposed to directly off the batteries?, 2) Based upon the connected load, would a smaller, say 600W inverter be more efficient for this application, seeing as we have a dedicated and known load?

What Am I missing????


Howard

As GoldDigger indicated, you don't want to run off the charge controller unless it's designed for it, which it probably isn't. You might gain some efficiency by moving to a smaller inverter, but again, GoldDigger has provided good advice on this.

Welcome to the perils of attempting to run a modern civilization via renewable unreliable energy. The solution is to add either more panels, more batteries, or both. Or hook it up to the grid, if possible.

One of the problems is that designs are typically based on averages, with maybe a WAG for extreme conditions. Cloudy in Chicago? Who knew?

 

[wwhitney]

One of the problems is that designs are typically based on averages, with maybe a WAG for extreme conditions.

That's just bad design. If you want your cell site to run 99.99% of the time (~1 hour outage per year) and are using solar, you need to design for the 99.99% worst case insolation.

Cheers, Wayne

 

[gadfly56]

That's just bad design. If you want your cell site to run 99.99% of the time (~1 hour outage per year) and are using solar, you need to design for the 99.99% worst case insolation.

Cheers, Wayne

This is the unacknowledged weakness of solar and wind. There are days, sometimes consecutive, when the entire mid-West USA, or all of Britain, or all of Germany or Denmark have no wind, or no sunshine, or both. You need massive batteries at a cost of trillions of dollars, or standby fossil fuel or nuclear plants. Storage of 1kW-hr of electricity is $400 for batteries alone, using Tesla as a model. And no, it doesn't get cheaper as you scale up. So 1 hours worth of storage for a 1GW plant is $400 x 1,000,000 or $400,000,000. For 24 hours, it's 24 x $400,000,000 or $9,600,000,000. Nearly 10 billion for one day's worth at one power plant. For that cost, you could build nearly 10 closed cycle natural gas plants with a 1 GW output EACH.

Running remote instrumentation? Go solar. Running a civilization? Don't.

 

[wwhitney]

This is the unacknowledged weakness of solar and wind.

I think it's fully acknowledged by anyone working on grid scale solar and wind. Obviously the grid can't be 100% solar and wind without significant time-shifting capabilities, but solar certainly has a big place in the portfolio of generation. It's not an either or situation.

Cheers, Wayne

 

[synchro]

The batteries are then connected DIRECTLY to the 1000W inverter, which is running at max 300w of load (the load is a small cell transmitter, and flow controls for some sewage monitoring stuff).
... What Am I missing????

With such a power limited supply, the system design and implementation of these loads should be focused on saving energy. In portable devices like smart phones conserving energy is a design constraint at every level from the system design on down to the circuit level inside of the custom ICs within the phones.

And so in your case, who came up with the loads of "a small cell transmitter, and flow controls for some sewage monitoring stuff" ? Was that part of "a complete system ... from an online vendor" or did the customer put together this system?
If the customer did it, talk to him whether there are opportunities to shut off unnecessary equipment when it's not needed using a timer, input from a cell phone receiver, sensor, etc. Your role in this might be limited, but I think you should ask anyway because you might be able to help wire it up to make it all work. If there are no opportunities to save power, getting more power from additional panels or a grid connection like was mentioned is your only alternative.

 

[g-and-h_electric]

So here is a bit more info, since I left some of it out last night. The flow meter and data transmitter are existing. I was told that at one time there was a lift station pump there, for that a 400 Amp 3 phase 120/208 service was put in. The pump is long gone, and all the service was feeding was basically a receptacle for the flow monitor. The guys at the village asked our shop about solar as a way to in the long term save money. One of our other guys found the "solar package" online and it was ordered.

I was at the site today, and the batteries are charging albeit slowly, the inverter had not tripped out on low voltage. I took an Amp readig on the wiring to the sensor, etc, 200mA draw! .. Had a guy from the village come by, and yup basically 3 wallcubes.... ( one piece has an internal supply...)

Thinking the "static load" of the inverter ran the batteries down, and a much smaller inverter MAY be the answer..... Going to "dump this back on the guy that ordered it ,and see what info ha can come up with! (as in specs on solar array, and charge controller)


Thank you all for the help so far!


Howard

 

[jaggedben]

I don't believe that the efficiency or tare load of the inverter is necessarily related to it's max rating. I'd try to get a spec on that independently rather than making assumptions.

 

[ron]

As a side note, I built a solar car with my son's high school a while back with a similar concept minus the inverter, and due to demand of the car's accessory loads, and the problem of a LOT of overcast days, the car had to take a break for recharging after driving for a while on a regular basis

 

[electrofelon]

Running remote instrumentation? Go solar. Running a civilization? Don't.

Where did the OP talk about running a civilization?? Let's keep (your) energy politics out of it, it has nothing to do with the OP.

 

[electrofelon]

So here is a bit more info, since I left some of it out last night. The flow meter and data transmitter are existing. I was told that at one time there was a lift station pump there, for that a 400 Amp 3 phase 120/208 service was put in. The pump is long gone, and all the service was feeding was basically a receptacle for the flow monitor. The guys at the village asked our shop about solar as a way to in the long term save money. One of our other guys found the "solar package" online and it was ordered.

I was at the site today, and the batteries are charging albeit slowly, the inverter had not tripped out on low voltage. I took an Amp readig on the wiring to the sensor, etc, 200mA draw! .. Had a guy from the village come by, and yup basically 3 wallcubes.... ( one piece has an internal supply...)

Thinking the "static load" of the inverter ran the batteries down, and a much smaller inverter MAY be the answer..... Going to "dump this back on the guy that ordered it ,and see what info ha can come up with! (as in specs on solar array, and charge controller)


Thank you all for the help so far!


Howard

Nearly all inverters have a rather high no load draw/loss. There is one inverter that is an order of magnitude better than all others and that is the Morningstar suresine. That may help, I'm not saying it will completely solve the problem though.

 

[winnie]

Hmm. Figuring 4x 140W panels. Chicago in winter averaging 1.5 peak sun hours per day. This estimates average panel production at 0.84 kWh per day.

You measured 24VA of consumption, but presumably not power consumption. IMHO there is not much headroom between consumption and production. Add inverter losses and consumption will outpace production on low light days.

If the instrumentation being supplied uses low voltage DC then it might pay to ditch the AC inverter and go to separate DC/DC converters for each instrument.

But really given the already installed electrical service it might make far more sense to have the POCO change from a large three phase service to a small single phase service. For such a small load it really is a question of tariffs.

-Jon

 

[jaggedben]

The thread subject says 'stand alone' system. Is the service still there?

 

[electrofelon]

The thread subject says 'stand alone' system. Is the service still there?

Yeah confused by that, particularly the getting the stand alone solar system to save money part (if the service was already there).

 

[winnie]

My guess: there was an existing large service with corresponding large base charges. They got rid of the pumps, and now have a few instruments consuming perhaps 10kWh per month.

Someone figured that solar power would be cheaper than the monthly minimum charge for the huge service. And when you factor in base charges for having a meter this might be true for small enough loads.

Jon