• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

Recommendation please

jpadie

New Member
Joined
Dec 26, 2024
Messages
7
Location
SW France
Hello all

First post here, so good to meet everyone!

Project for the first half of 2025 is a good sized solar array at my place in SW France. I'm fairly comfortable about doing all the work but I'm one of those people that loves the planning stage more than the execution....

In that spirit can I have suggestions for good books on the topic of grid tied installations please? Nothing like Delia Smith teaching readers how to boil an egg, but something that starts with an assumption that readers know the basics of electricity and PV, perhaps.

Thanks in advance.
 
Hello all

First post here, so good to meet everyone!

Project for the first half of 2025 is a good sized solar array at my place in SW France. I'm fairly comfortable about doing all the work but I'm one of those people that loves the planning stage more than the execution....

In that spirit can I have suggestions for good books on the topic of grid tied installations please? Nothing like Delia Smith teaching readers how to boil an egg, but something that starts with an assumption that readers know the basics of electricity and PV, perhaps.

Thanks in advance.

Someone will post something for you to read, I'm sure.

But a lot of people just tell us what they are looking to do, what you know about what kind of system you need, what needs you have and any preferences .. Lot of advice will flow from that point.. Q/A will help you figure out all the details and start getting a plan together..
 
Yup. I'm sure that I will want very and appreciate a lot of help. But I'd love to get the terminology and art of the possible first so that my questions are well focused and informed
 
With rapidly changing technology, I suspect that a book that is current will be hard to come by. As a newbie in the area, this is where I'd recommend researching to start
- your local building code requirements (this often dictates a system design more than anything) for such a photovoltaic (PV) solar system
- figure out what service you have (3-phase vs split phase or ?) Amp service level provided by utility and main load center (breaker panel) type and capacity. Any sub-panels, etc, Why? if an older house, with say only a 100A service from utility company, and a 100/125A busbar on main load center, you might be limited in how much PV you can add.. maybe... depending... ;^)
- understand the design implications of roof mount vs ground mount arrays (general rule of thumb is to use ground mount when practical possible) due to future roof maintenance implications, panel access, etc
- what are you trying to accomplish with solar electricity generation (partial vs full kW usage offset, future electrical consumption expectations (ie appliance electrification, EV, etc). See below on monitoring... first there is what you are trying to accomplish, then reality of lower PV production in winter, your specific site PV considerations (tree shading, mountains (number of direct sunlight on panels), etc), and then what makes sense financially (ie enabling to disconnect from local power grid entirely, and 100% self-generate tends to be expensive and not a good financial approach... but depends). vs simply optimizing utility bill (largely reducing, not entirely eliminating, grid power importing)
- understand the design implications of AC coupled (micro-inverter) vs DC optimized, vs DC string panel arrays . Personally, I'd say beware of anyone that one approach is always recommended (because.. not true.. it depends.. each has its own Pro's and Con's)
- local net metering rules (ie putting excess PV generated energy on local Power Company (PoCo) grid, and using that energy later. Early rules to incentivize PV installations provided very generous benefits (like using power grid as free battery), and such rules are going away.
- Based on net metering rules specific to your local power company, an energy storage system (ESS, ie whole house battery) may make financial sense (ie a positive return on investment [ROI]). Though you can decide on a battery for other reason than ROI.
- When dealing with a battery, you have to decide whether trying to
  • only have enough kWh capacity to cover peak time-of-use rates (some places {like where I live} have MUCH higher electricity rates at certain time of the the day).. This is sometimes called demand-shifting (ie using PV (and grid?) to charge battery during times of excess solar and/or law rates, then using battery during peak charge rate times
  • Whether you want backup power is grid is down (grid forming setup), vs a battery that only works when local utility power grid is working (grid following.. cheaper as a micro-grid interconnect device / MID is not required)
    • ex - for those with a high variance (difference) from off-peak to peak time-of-use rates, a grid-following battery large enough to cover peak times may provide best ROI. However, you'll be annoyed when grid power goes down during daylight hours, and you can't use your PV/battery.. side note, many simpler AC or DC coupled PV systems are grid-following (ie grid power fails, and PV system stops working). a MID (grid-forming) device of some sort required to disconnect home from grid to allow PV and/or battery to keep working (for more on MIDs and recent technical developments, see this article https://iaeimagazine.org/evolving-t...-adaptors-and-microgrid-interconnect-devices/)
  • Whether you want entire home backup vs only specific critical loads. This is often determined by cost and amount of energy required to cover entire home. For example, my peak kW usage is around 8kW (and that is with heat pump HVAC and PHEV charging (at 16A)... as such, many hybrid inverters can cover my entire house load, which is cheaper than getting either
    • smart load solution (from upper-end SPAN type breaker panel setup or whatever similar solution exists in Europe) to various hybrid inverters that can have a few limited breakers smartly controlled, vs the old-school brute-force critical loads sub-panel [ie, battery only powers sub-panels (critical loads).. and that means moving circuits to new panel, and potentially re-wiring some circuits if certain outlets need to be broken into separate circuits (breakers)] There are those with primarily electrical appliances (where as I have many natural gas appliances like oven/stove, water heater, etc) that have MUCH higher peak kW usage... and that means either a more expensive inverter, or sometimes multiple inverters (can get expensive)...
    • When planning, if you have natural gas or other non-grid powered appliances that might eventually want to cut over to PV/battery/grid-powered, then consider those loads in your peak capacity calculations
    • Bottom line, with decreasing battery costs, whole house backup is sometimes cheaper than partial/critical loads only backup, if peak load is small enough and factoring in electrician labor... depends when going the do-it-yourself (DIY) route
  • *IF* going backup route (ie a MID is present), then you need to consider system coordination for the following scenario... batteries are near full, and PV output exceeds house electrical load. In this case, the 'system' needs to signal PV system inverters to curtail production. The old-school brute-force approach is frequency shifting, but there can be more direct and granular communication .. but there are good standards for this, so check carefully. If using frequency shifting, be sure to understand implications and potential downsides.
  • And you will probably want to look at a high level on virtual power plant options in your area (may or may not be something to consider)
- General rule of thumb is that energy efficiency on the load side is usually (at least for now) cheaper than adding PV/battery capacity (ie, beware trying to off-gird power old inefficient air conditioning systems.. or certain other systems)
- The other hot-button topic is local communication vs required 'cloud' (ie internet server) communication and control systems. Along the same lines, if you want reliable network communication, use Ethernet not WiFi (when connecting various components, when practical). Numerous vendors have gone out of business, stopping certain products from working (both solar and other)... so some folks (like myself) strongly prefer a system that does NOT require Internet communication to function. I'm ok with certain reporting, etc.. for me the question is on-going operations if company goes out of business, or extended internet communication outage. Local communication also comes into play if trying to do something like Home Assistant for various automation tasks. In North America, there is hope that bi-directional EV charging standards will drive cross-vendor communication standards related to smart home energy management... only time will tell if this happens. In the mean time, a single vendor solution can sometimes be the only route today to simple reliable sophisticated smart home energy management.
- when hooking up a hybrid-inverter, beware they often have longevity of typical north american natural gas hot-water heater. As such, best practice is to have a bypass switch installed enabling you to power house directly from grid if/when hybrid inverter fails, giving you time to troubleshoot and repair/replace (which could easily take longer, to much longer, than replacing a hot water heater as they aren't as common, not as readily available, not as standardized, etc). If you can trained and capable of working with live grid-powered service entrance cables, then this may be a low concern.. for most people, if hybrid inverter/MID fails while you are away from home, your significant other (partner/family) will likely want a simple "go flip this switch, and power will be restored [from grid], and I'll deal with problem when I get home" type option. This also lets you do hybrid inverter maintenance (including outright replacement) while house still has power. If this appeals to you, be sure to plan wiring accordingly (due to extra costs involved, this is often not proposed/suggested)

Other things you may want to research
- residential EV charging using DC (not really a typical solution today, but coming? see Australia's Sigenergy’s Sigenstor 25kW Bi-Directional ESS/DC EV charger, Hybrid Inverter as an example of possible direction. For me, the single-phase output is too low to be practical, and some other missing features mean it isn't something I'd want... but I like the idea of limited DC EV charging to avoid the energy loss from excess AC/DC conversions (maybe). My hope.. .once local bi-directional EV charging and connection standards worked out locally, this type of solution will become more prevalent
- why some experts say trying to charge an EV from solar isn't practical (typically not cost-effective).. but how that might change with new bidirectional DC charging?? especially if not needing large amount of kWh daily for an EV
- If you are a research/data/knowledge geek, realize you are likely to get curious about your energy usage. A typical PV solution does simple energy monitoring of grid kW import/export and PV production (from which it is easy to calculate house consumption) [Typically these measurements are done using Consumption Tracking (CT) clamps]. But you are likely to want to know more about that home consumption. SPAN panels are the top-of-line type solution (offering both monitoring and individual circuit control), with something like the Emporia Vue 3 which does monitoring only (for a very small fraction of the price) .. depends on you budget

The above should be way more than enough to get started on understanding some system basics. After that, one can get into technical details of select system components, and by then you will have a high-level understanding of your system goals, constraints, etc. The art of possible is greatly expanded, but also includes options for those that are entirely off-grid (ie rural area where connecting to power grid is not feasible) that are WAY more expensive than a grid-tied solution

I hope this helps
 
Thank you @Lawrence_SoCal . I'm very grateful that you've taken so much time and care to reply here.
I'm happy to share the basic shape of the project I'm looking at and will try to do so below. I always prefer to DIY things - I have more time and competence than cash and a constant desire to learn new skills.

At the moment there are three locations on my land that I'm targeting for solar. Not all at the same time, so the solution needs to be "extendable". The locations are: 1. a terrace roof (30° slope, 4m depth, 15m length at the narrowest part. 2. the roof of the house, 30° slope, 6m depth, 15m length with a gable in the middle. Location 3, a ground array near the swimming pool. Possibly this will end up a vedette and become a pitched roof. the area is 7m x 6m.

Locations 1 and 2 are both just off full-south. 185° azimuth. tree shading is minimal through most of the year. less than 5% overall, I think. the gable on location 2 is quite significant and will shade at least 2.5m each way.

I'd start with putting panels on the terrace roof - no working at height etc. I'm a little worried about the reflections from the panels causing glare in the upstairs windows; and if it's too bad then I can always move them to the top roof.

Construction of the property is timber frame with fibre cement cladding and the typical rounded clay tiles of SW France.

I'm gathering consumption data at the moment but typically the electricity bill is 3.5k eur per annum. the big usage is a GSHP. I've throttled it to 45Hz which keeps the consumption under 2.5kW but that limits its ability to heat the house in the cold parts of winter; and cool it in summer. In time I will add ASHPs to the upstairs in the next couple of years. The pool pump potters on at about 800W for 18 hours a day between June and end September.

Location is 43.818584N, 1.7076078E.

In my mind, and this is open to all constructive input of course(!), I am looking to get reasonable panels** with micro-inverters. My thinking is that a micro-inverter system allows savings on cabling, flexibility to extend whenever I need, flexibility to leave panel or inverter faults until when I can service them (I am very often abroad for long periods of time and my tenant would not want to work at height) and avoid the problems caused by shading from the gables and dust (i'm quite rural).

In the summer, consumption is mainly day time, or can be shifted to day-time (e.g. hot water). In the winter, the core needs are through the night as the GSHP needs to fight to beat the heat loss and keep the UFH at a pleasant temperature.

in time, I may want to add battery back-up; and maybe EV charging. Again this feels like micro-inverters with a control unit (like EnVision's) might be a good route. There are typically 2-4 days per year of power cuts. Not that traumatic.

I would not get any state aid for the installation as that is available only when using certified installers. I am able to sell back in to the network though. So I am looking at a solution whereby I use what I can and sell the surplus. The network contract in France obliges the supplier to purchase although the value changes.

The meter is a "Linky" which is the French variant of a smart meter. I am not sure, yet, whether exported energy needs to be wired directly into the meter. I think not.

The meter is 200m away from the property.
In the property there is a main cut off and fuse which then feeds the consumer unit/distribution board.

the consumer unit is in the basement. I don't want to put battery back up there as the basement is somewhat prone to flooding, particularly when there is a power cut as then the evacuation pumps are not active (a good use for solar stored energy however there are local regs that require there is no energy injection to the network during a power cut - so this would need to be via cut-over).

That's as far as I have got in my thinking at the moment. And I'm sure I am exposing more ignorance than wisdom, hence wanting to read lots before taking the plunge!

** ref the panels - there seems to be a cost difference between a 20% efficient panel and a 23.x% efficient panel. Of those that I have looked at, a good quality bifacial P-doped panel is £90 and has an efficiency at 20% with a curve suggesting 0.55% losses per annum. a panel of N-doped cells with 23% efficiency and back connections seems to be £310 per panel and a little better on the loss curve. I'm not sure that the maths work to justify the extra costs - it feels more logical to install £100 panels now and replace sections in ten years when technology has brought the cost down or improved things otherwise.

sorry for the long input and I'm very happy to receive all and any feedback.
 
The general rule of thumb regarding panels is to pay a price premium for higher Watt panels / higher efficiency panels, when you are limited in panel space (typically roof mount). When you are not panel space limited, then not paying for the top-most panels tends to provide a higher return on investment.
Are you limited? that depends on your kW usage model and other details. On this site, there are folks that have referenced web sites that can take a look at a satellite photo of your location, and with your input, make estimates. Only then will you know if you you energy production needs and available roof space means you do, or don't, need to pay for premium higher output panels.

I did what is known (locally at least in the roofing industry) as a "Lift and Lay", which is pulling up all of my concrete Spanish S style roof tiles, replacing ALL of the underlying felt (with double layer, 40 year felt), repairing any wood damage (I'm near the ocean, so some wood rot is common due to frequent marine layer) and then putting the tiles back. I then installed the solar system a couple of years later. What you don't want to do is put a PV system that should last 25+ years on a roof that will require panel and mounting system removal well before then.
Also, be sure to understand your local regulations on roof panel setback requirements (for me, that means 18 inches/46CM from top center line as well as 36in/1M from one side needs to be free of panels ... ie, allow fireman to move around on that area of roof)

I have an Enphase based micro-inverter based system. In the DIY world, it is common to see straight DC-string installs (and on this site you will come across those that very strongly advocate for such at all times). As I have a chimney in the bottom middle of my south facing roof, and local regulations on rapid shutdown, etc... a micro-inverter based setup made sense for me. Some folks have indicated that for smaller grid-tied systems, AC-coupled/micro-inverter (MI)based systems can make a lot of sense. However, when getting into larger systems, DC-coupled has benefits (and own associated risks from high voltage DC, limitations on panel orientations, etc). Micro-inverter is much simpler in that regards, though of downside of a piece of equipment (the inverter) up the roof that could fail. Each has its own Pro's and Con's. When getting a whole house battery, being able to direct DC charge sounds theoretically beneficial. But how much actual difference it makes... ???
For my AC-couple MI system (425W panels, 384W MI), each 10-12 panels gets its own 20A breaker. What is common, it to run all of the solar panel wiring/breakers into its own electrical combiner box (sub-panel). This can be sophisticated device, or a very simple small sub-panel. This make shutting off all PV to house a matter of a single breaker. It also makes measuring all PV input easier this way.

Not putting solar onto grid when there is a power outage is a normal safety consideration. Do some research as I mentioned on grid-following vs grid-forming setups... In my setup, if the grid goes down (grid power outage), my solar system stops producing, even if mid-day bright sun... this is a grid-following setup.

That meter is 200M away is probably a non-issue assuming single 'service entrance wires' from meter to 'consumer unit/distribution board' {what you will read about here described as 'main load center' or '{circuit} breaker panel'. A sub-panel simply means another circuit breaker panel/ load center fed from (connected to) main load center

There are some batteries (like Tesla's PowerWall 3) [and others?] that advertise being able to be partially submerged... but obviously not a great idea. I'm guessing flooding doesn't get to high to short out your consumer unit/distribution board {breaker panel}? At that point, I'd be considering re-locating the breaker panel.... but if at most all that can happen is a little water on the floor (couple of inches)? and the electricity is safely higher? is such a situation, maybe having the house batteries wall-mounted up high would work? there are a number of possible recommendations for battery placement.

If starting with solar PV panels only to start, and wanting to be able to expand over time, then I'd take a look at what a "full" system with whole house battery, grid-forming hybrid inverter, etc would look like. And set up your wiring such that adding all of those components later is easy. What you don't want to is to do simple AC-coupled panels into main main load center... and then want to re-wire (with insufficient spare wire length) to hybrid inverter later .. this is a case of some simple planning ahead can save you a lot of time and effort later. It also means adding in bypass switches and the like initially, to make adding parts later less involved (especially anything involving the 'hot/powered' power feed lines from power company to your distribution unit... unless there is a cut-off switch already out by the meter)
 
thank you again.

There's a lot of technical terms being used. the normal electrical ones I'm familiar with (although there seems to be a US centric usage ; e.g. breaker panel rather than distribution board/consumer unit etc). Easy enough to understand. The terms of art that have developed in the PV industry seem slightly more challenging! A new language to learn.

electrically (and in theory) it's not that hard to envision a system that uses panel inverters and supports grid-tie, battery and network isolation in the context of a black out etc. there would be increased losses as the AC from the micro-inverters would need to be reconverted to DC for battery charging. I don't know whether single solution units exist on the market - they'd need to be a sort of "controller" but not electronically or electrically complex at all. Are these what you are calling "hybrid" inverters perhaps?

Another couple of starter questions, if I may?

1. In my current BOM I've got 23 440W bi-facial panels coupled with EnPhase iQ8ACs. which are rated at peak 380W. This coupling was "chosen" by the software I'm using as a design assistant. Is derating the output via the inverter normal? There are no shading considerations worth factoring in.

2. how are the systems earthed typically? I'm almost positive my house has an earth rod. If this is the case (the earth rod) I assume I connect the panels and inverters directly to the earth bus-bar and don't need to worry about isolation of earth leakage from the network in a black-out event? France uses a TT system so there is a tie between neutral and earth at the local transformer as well. But unlike the UK there's no tie more locally as I understand it.
 
A typical Hybrid inverters can be AC or DC coupled PV array, DC-connected to battery, and performs MID (microgrid interconnect device function... ie, isolate local load from grid and allow grid-down operation)

With micro-inverter, then you'd be right in thinking you don't necessarily need a Hybrid Inverter... unless you decide to get a DC-coupled battery.. it depends. With no battery, and a simple grid-tied PV array (no power when grid down), I do NOT have a hybrid inverter. As soon as you want a battery to provide power when grid power is off/down, then you need a grid-forming device to disconnect grid, isolate house, etc. A hybrid inverter is one device that can do that isolation, but not the only device type
1. In my current BOM I've got 23 440W bi-facial panels coupled with EnPhase iQ8ACs. which are rated at peak 380W. This coupling was "chosen" by the software I'm using as a design assistant. Is derating the output via the inverter normal? There are no shading considerations worth factoring in.
Yea, I have 425W panels with the IQ7HS 384W micro-inverters. The panel rating is theoretical/lab condition (basically a marketing number, not an ever expected actual output value). So a slightly lower inverter rating is normal

2. how are the systems earthed typically? I'm almost positive my house has an earth rod. If this is the case (the earth rod) I assume I connect the panels and inverters directly to the earth bus-bar and don't need to worry about isolation of earth leakage from the network in a black-out event? France uses a TT system so there is a tie between neutral and earth at the local transformer as well. But unlike the UK there's no tie more locally as I understand it.
I'll let someone else cover this, as I can explain what my installer did here is USA southwest... but your codes/design connections?

in my install, the AC circuits from the roof micro-inverters (2 circuits at 20A each) ran into own sub-panel (acting as combiner box) with ground and neutral separate. Then 4 wires from sub-panel to main load center (service entrance circuit breaker panel) where neutral and ground are combined, busbar connected to earth copper rod(s)
 

diy solar

diy solar
Back
Top