Scalable Micro Grid Tie Solar

[BobHope]

Hi, I am looking for a sanity check for a project. I am in the UK.

I work from home with constant low draw electrical devices on ( Lights, computers, monitors, fans, fridges, freezers etc ) and I would like reduce electricity usage. I have about a fairly constant daytime use of about 200w. Everything high draw ( cooker/heating/water etc ) is on natural gas. My understanding is that on a cloudy British day with the system at 10% I would need a 2kw system in order to generate that 200w.

I would like to start very small at 330w and scale out from there.

The first system is to test performance levels at my site to see if it is worth it, and I can extrapolate from there and size out a bigger system. I would like my initial investment not to be wasted, so buying the right initial components would be good. My experience with these type of systems is that a beginner is usually wrong about many things and it is good to gain experience. As I have no battery in my system, any excess is simply given back to the grid, I don't plan on having the inspections and certifying it to local regulations ( despite it being compliant ) and selling electricity back to the grid.

My idea is to use 330w panels ( available locally in the UK for approx £190 ) and individual micro grid tie inverter(s).

There is a 350w Chinese Y&H inverter at around £55 pounds, example: https://www.ebay.co.uk/itm/403087824719 that would match this panel. These inverters seem to be popular, but are they any good ?

I have a suitable unshaded south facing 1 story outbuilding to mount the panels to. The inside could mount the inverters, keeping them out of rain/light etc and making it easy to maintain.

Price per ( potential ) KW would be about £720

The initial system setup seems almost too simple, it this right ? I simply connect the inverter directly to the home system and it floats a higher voltage into the system, which the electrical devices use - I am thinking at 330w I could just wire the ground, neutral, and live to a UK plug and plug it into a power socket. If I scale up I will use proper electrical connectors sized for the amperage.

To scale I simply add more panels, each with a micro inverter in parallel. Or, based on the maths the small initial setup creates, move directly to a 1.5/2/3/4 KW system with a single inverter from Growatt or similar. In that case I either parallel the systems or re-use the initial panel and discard the Y&H inverter.

 

[Ozark Tinkering]

Currently I have 16-320watt panels and each panel has a micro inverter attached to the backside and all 16 panels are tied in parallel to to a 240 volt 20amp circuit breaker with no neutral. Grid tie really is that simple but the downside is if you don't enter an agreement with your utility, whatever you don't use you lose. With a net metering agreement you get credit for those unused hours so nothing is ever lost...it just gets traded.
So yes you can do that but it'll be costly. There are cloudy days here where I'm lucky if my 5kw system produces 4kwh and other days when it produces 32kwh.
As you laid it out it really seems like a set of batteries and an "all-in-one" inverter would be you friend after purchasing the first panel and delete the idea of grid-tie. You can do that (if you want) with the right "all-in-one" and those aren't really that costly in the scheme of things. Are you sure you're only drawing .2kwh during the day? That seems really low for the appliances you mentioned that aren't on gas? My tele uses more than that?
Oh and then there's those days without any production or during an outage, my 5kw grid-tied system is useless. That's almost $13k just sitting there on the roof. That is what is pushing me to do what I'm working on in this thread...I want to fool my inverters into thinking the grid is still there and I describe how I think I can get'r done.
https://diysolarforum.com/threads/inverter-generator.34650/

Hi, I am looking for a sanity check for a project. I am in the UK.

I work from home with constant low draw electrical devices on ( Lights, computers, monitors, fans, fridges, freezers etc ) and I would like reduce electricity usage. I have about a fairly constant daytime use of about 200w. Everything high draw ( cooker/heating/water etc ) is on natural gas. My understanding is that on a cloudy British day with the system at 10% I would need a 2kw system in order to generate that 200w.

I would like to start very small at 330w and scale out from there.

The first system is to test performance levels at my site to see if it is worth it, and I can extrapolate from there and size out a bigger system. I would like my initial investment not to be wasted, so buying the right initial components would be good. My experience with these type of systems is that a beginner is usually wrong about many things and it is good to gain experience. As I have no battery in my system, any excess is simply given back to the grid, I don't plan on having the inspections and certifying it to local regulations ( despite it being compliant ) and selling electricity back to the grid.

My idea is to use 330w panels ( available locally in the UK for approx £190 ) and individual micro grid tie inverter(s).

................................

The initial system setup seems almost too simple, it this right ? I simply connect the inverter directly to the home system and it floats a higher voltage into the system, which the electrical devices use - I am thinking at 330w I could just wire the ground, neutral, and live to a UK plug and plug it into a power socket. If I scale up I will use proper electrical connectors sized for the amperage.

To scale I simply add more panels, each with a micro inverter in parallel. Or, based on the maths the small initial setup creates, move directly to a 1.5/2/3/4 KW system with a single inverter from Growatt or similar. In that case I either parallel the systems or re-use the initial panel and discard the Y&H inverter.

 

[Ozark Tinkering]

Where I am our insolation rate is somewhere between 4.5-5. Not the greatest sun but still worthwill to use a grid-tir system. I am comfortable with usin the 4-1 production expectation on my panels. On the average my 5kw system supplies close to 20kwh per day. Here's a insolation map of the UK so you can get an idea of what to expect.

 

[Ozark Tinkering]

Comparing to an insolation map of the US...

 

[400bird]

I think your proof of concept system sounds like a great start. I can't speak to the quality of the £55 Chinese ebay micro inverter. But it seems too cheap to be reliable. Probably still good enough for a proof of concept.

But, in the long run when you decide to build out the larger system, you can leave this one panel connected as a parallel, separate system, so this investment isn't wasted.

 

[Ampster]

The other advantage of using Micros is that you can use different orientations without worrying about string sizing or separating onto different MPPT inputs.

 

[Ozark Tinkering]

The other advantage of using Micros is that you can use different orientations without worrying about string sizing or separating onto different MPPT inputs.

String sizing, depending on what you mean exactly might have limits like my Enphase system has. It might be a NEC thing so it wouldn't necessarily apply to the UK, but my Enphase system limits me to 5kw per string (20 amp circuit) and the 240vac string is required to be on a 20amp 2 pole breaker. If I want to add 16 more panels the of same wattage I must feed a second string from another 20amp breaker. My end plan for my grid-tie system if I chose to offset nearly 100% of what I take from the grid, I would need 3 full strings.
I guess I could go down and really dirty and run #6s from 1-60 amp 2 pole breaker and let'r rip, but for some reason that seems like a bad idear?

 

[Ampster]

I meant string sizing not from a Amperage perspective but because of the different outputs which different orientations would have and how they would interact with the MPPT controller. I know it can be done with some thought and calculation with string inverters but my point was it is much simpler with micros.

 

[Ozark Tinkering]

I meant string sizing not from a Amperage perspective but because of the different outputs which different orientations would have and how they would interact with the MPPT controller. I know it can be done with some thought and calculation with string inverters but my point was it is much simpler with micros.

I knew there were 2 ways to take what you said once we start mixing micro inverter systems with dc fed from a series/parallel circuit, so I qualified my comment. I could go on the roof and cover panels 2, 4 and 6 with a towel and the other 13 panels wouldn't skip a beat. I can have 8 panels partially shaded and 8 panels in full sun and it doesn't matter. Each panel stands on its own.

 

[Ampster]

Each panel stands on its own.

Of course with micros you could actually see the output of each panel which might be useful when trying out different orientations.

 

[Ozark Tinkering]

...I started to post up a screenshot of my production but it gave up just a little more than I wanted, but yeah it's easy to track from my phone

 

[shavermcspud]

Hi Bob, those panel prices look high (£170 for a 330w) for the UK, unless that includes the micro inverter? , drop me a PM if you want to source Enphase micro inverters and decent panels for cheaper.

 

[12VoltInstalls]

the quality of the £55 Chinese ebay micro inverter. But it seems too cheap to be reliable.

I am running a 600W version with 200W of panels going into it. Seems to work just fine.

 

[shavermcspud]

Of course with micros you could actually see the output of each panel which might be useful when trying out different orientations.

Those cheap ebay micro inverters have no monitoring function at all so you'll never know their output, The next step up is the Wifi monitored Chinese units at £85 each with 10 minute logging , but if you're going to pay that you might as well get the enphase iq7 at £98 with realtime local panel monitoring

 

[Ozark Tinkering]

Those cheap ebay micro inverters have no monitoring function at all so you'll never know their output, The next step up is the Wifi monitored Chinese units at £85 each with 10 minute logging , but if you're going to pay that you might as well get the enphase iq7 at £98 with realtime local panel monitoring

I wonder if you could just stick a watt meter on it on the line side?

 

[Ampster]

I wonder if you could just stick a watt meter on it on the line side?

On each micro?

 

[Ozark Tinkering]

On each micro?

Yeah...just a cheapy. Then you could walk around all day with a clipboard writing down readings and add them up. Oh wait one more watt meter on the entire string solves the math problem.

 

[BobHope]

I wonder if you could just stick a watt meter on it on the line side?

I have a smart meter with a readout and data for the past few months ..I was guessing I could monitor the improvement using that but it is a bit finger in the air.

 

[BobHope]

Those cheap ebay micro inverters have no monitoring function at all so you'll never know their output, The next step up is the Wifi monitored Chinese units at £85 each with 10 minute logging , but if you're going to pay that you might as well get the enphase iq7 at £98 with realtime local panel monitoring

I have just been looking into those, only snag I can see is docs state :
"
EVERY Enphase microinverter system requires an Envoy device to commission the system AND it needs to be left on site permanently in order to comply with G98/G99 regulations. Instructions for registering a system can be found here, whether you are a homeowner, first time installer or self-installer.
Please note you may receive a mixture of INT and FR version of this micro inverter. They will need a gird profile applied on install, this is completed using the Envoy."

 

[Ozark Tinkering]

I wonder if you could just stick a watt meter on it on the line side?

On each micro?

Yeah...just a cheapy. Then you could walk around all day with a clipboard writing down readings and add them up. Oh wait one more watt meter on the entire string solves the math problem.

I have a smart meter with a readout and data for the past few months ..I was guessing I could monitor the improvement using that but it is a bit finger in the air.

I thought I gave you a pencil and clipboard?