diy solar

diy solar

Hello and introduction

CavendishDave

New Member
Joined
Oct 30, 2022
Messages
3
Hello Everyone!
I am new to the forum and learning my way around. It seems like the very first step is an introduction. I’m Dave, and I‘m a retired school administrator in the US, specifically Pennsylvania. I have a little bit of diy experience with home wiring and construction. I designed and built a CNC router (including completely designing the controller) to cut wood, plastic etc. a couple years ago, and have installed and programmed a couple variable frequency drives for some three phase woodworking machines. Probably exactly enough knowledge to burn down any house or building I come across and/or fry most any piece of electronic equipment, especially if it is expensive.
I am not necessarily looking to completely install my own system, and in fact wont be doing the physical install myself, as a back injury simply makes being on a roof a foolish proposition. I do know a reputable roofer/construction company who should be able to handle it with little problem. I will also most likely not do the wiring myself as I believe I may run into homeowners insurance issues, and I know a good electrician, although I’ll have to check with him to see if he would comfortable with such an install.
Mainly I’ll be trying to do the permits, interconnection paperwork, etc, as well as the design and purchasing of the pieces myself. I’ve been working with several installers in my area, and they each seem to want to install very specific systems that don’t necessarily accomplish what I’m after, and in fact they tell me it can’t be done. In a nutshell then…
  1. We use about 14,400kw annually. Like to have some headroom to add an EV down the road, so thinking we need about 16,500-17,000 kw actual production. BUT, initially we would only use the 14.4k kw.
  2. Our roof has some partial shade at different points during the day. Roof is southern facing.
  3. System would be grid tied
  4. REALLY like to be able to have power in the house during the day in the event of a power outage, but we are willing to forgo power at night (meaning we’d like to avoid batteries or at least a lot of batteries). I *think* that is called “pass through power“, and this what the local folks are saying can’t be done. According to them, if a person wants power during an outage day or night, the system must have batteries. They cite the fact the system must be able to immediately disconnect from the grid. I understand that, as don’t want to fry a lineman.
  5. Like to have quality components
Any help, insights, or resources would be greatly appreciated!
Cheers,
Dave
 
Welcome to the forums!

We use about 14,400kw annually.
A kilowatt-hour (kWh) and a kilowatt (kW) may sound similar, but they're actually quite different. While both are interrelated units of measurement, the important distinction between kWh and kW is that a kWh reflects the total amount of electricity used, whereas a kW reflects the rate of electricity usage. What you probably mean is that you use 14,400 kWh annually or about 40 kWh per day. Not trying to be nit-picky, the stuff is complicated enough and using the wrong units off can confuse the conversation.

I'll run through some quick examples to get you started, but you can find much greater detail in the Frequently Asked Questions thread. The signature of this post will also have some links to other tools that you might find of interest.

...initially, we would only use the 14.4k kw annually...
Typically an annual figure is useless as solar will produce more energy in summer and less in winter. It's typically important to understand your monthly needs. If you were off-grid you'd need to make sure you had enough panels to cover the months with the lowest insolation (not a typo, and winter in the northern hemisphere is when it's lowest). If on-grid (e.g., electing to stay connected to the utility), it's complicated by utilities net-metering agreement (assuming there is one).

Fortunately, Pennsylvania is one of those great states that offers 1-to-1 net metering in most areas (check with your local energy provider), so we can optimize the panels for year-round production and get a pretty good guess using the annual numbers.

A program like SAM will allow you to get fairly precise, but for something quick and dirty we can use an insolation calculator. For Pittsburg with the panels aimed south at 50° tilt (for year-round optimization), the average insolation is 4.2, so with an average need of (14,400 kWh/ y / 365 d/y =) 40 kWh/d and a 90% efficient system you'd want (40 kWh/d / 4.2 h / .9) ~= 10 kW of solar panels. Although that assumes no shade.

Our roof has some partial shade at different points during the day.
... [eventually] we need about 16,500-17,000 kw actual production....
Microinverters are great for handling shade, and by far the simplest type of system to start small and expand. See this post for why microinverters are better than other types of inverters.

REALLY like to be able to have power in the house during the day in the event of a power outage...
With microinverters the battery is the hitch. Enphase has a system that works and it's what I have; I think they're a great company and I like the product. Other than Enphase, you can use a Tesla Powerwall with microinverters. There's also the previous generation of technology known as AC Coupling, but as it requires a big inverter its cost prohibitive.

Batteries are still expensive. You might look at getting a smallish battery with a generator for now; if so, you'll want a system like Enphase, SMA, or Sol-Ark that can work with batteries and a generator. Natural Gas generators are better in my opinion than gasoline if you can get natural gas piped to your property as it doesn't go bad sitting in cans and you don't have to haul it to feed the generator. Ideally, you can resell the generator at some point when battery prices fall more.

Like to have quality components
A lot of inverters come with fairly short warranties of 5 to 10 years. But everyone is different. For example, Enphase microinverters come with a 25-year warranty and SMA is an exceptionally good brand and some inverters have a 15-year warranty. A number of vendors also offer extended warranties for an additional cost. Panels don't make too much difference quality wise beyond warranty (see Comparing Solar Panels). Breakers & fuses are another place to buy quality. Battery quality is all over the place, hate to say stick with a big name because of the cost. Will has some cells and rack-mounts he recommends - but due to the costs I bought from a quality vendor to get a good warranty.

Anyway, hope that helps!
 
Last edited:
Welcome Dave, there's enough info on here to take yourself completely off grid...
Do you have PPL where you're at in PA?
Have you looked at the list of approved inverters they allow to feed back to their grid?
That's likely why installers are telling you what you want can't be done.
I ended up not feeding back to the grid, rather dumping the extra power into batteries. You may be able to couple a 2nd inverter and some batteries to an approved inverter...

Good luck.
 
  1. We use about 14,400kw annually. Like to have some headroom to add an EV down the road, so thinking we need about 16,500-17,000 kw actual production. BUT, initially we would only use the 14.4k kw.
It's good to know that the annual production is 14.5 MWh. Now knowing the highest month is important to. Typically its summer like when I turn my two ACs on.
  1. Our roof has some partial shade at different points during the day. Roof is southern facing.
When you say partial shade, this can kill panel output. I only have several sets consisting of three panels, 100 watts each, but a shade the size of my hand across one panel reduces that series of three by 66% - 75%, of that set. The other sets are unaffected.

Need to consider unshading the area, or figure more specifically how this effects production. A tree can be cut down, but roof obstacles are unavoidable.
  1. System would be grid tied
Are you up for the intesity of a DIY install with permiting or do you plan on going through a contractor? I go through a contractor if I decide to actually get an install.
  1. REALLY like to be able to have power in the house during the day in the event of a power outage, but we are willing to forgo power at night (meaning we’d like to avoid batteries or at least a lot of batteries). I *think* that is called “pass through power“, and this what the local folks are saying can’t be done. According to them, if a person wants power during an outage day or night, the system must have batteries. They cite the fact the system must be able to immediately disconnect from the grid. I understand that, as don’t want to fry a lineman.
There's truth and regulations to this. Please keep in mind in as long as solar as been around, noone has developed a batteryless system that provides and real power, with the exception of grid tied which is like a battery anyway. There's some gimmicks out there. I can charge my cell phone, but that's not what we're talking about. There are inverters that can power an AC, but not for hours on end as clouds interrupt it.
  1. Like to have quality components
When my inexperienced self plans, I plan to spend $15k per 5 kW of panels on the roof and then at least $10k per 10 kWh of battery power, of which, I'd want three of those. For my dream system, I see $75k with 15 kW of panels and 30kWh of battery packs built to UL specs, but I won't het my dream system.
 
... Please keep in mind in as long as solar as been around, noone has developed a batteryless system that provides and real power ...
There are a couple of Vendors that provide that now. For example, Enphase has its "sunlight" system and Outback has the Skybox. They work fine if you have enough panels, but of course they're intermittent with clouds, so a battery is very useful.
 
There are a couple of Vendors that provide that now. For example, Enphase has its "sunlight" system and Outback has the Skybox. They work fine if you have enough panels, but of course they're intermittent with clouds, so a battery is very useful.
That is what I call a gimmick. If I spend $1500 on an air conditioner, I’m not hooking it up to an intermittent system.

Will has a video of him running a small AC with inverter and no batteries. It still would be intermittent and not persistent.

To me these systems that run off solar only are like drag racers. They’re good for the drag strip, but you don’t take them across country on a drive. Two different things.
 
That is what I call a gimmick. If I spend $1500 on an air conditioner, I’m not hooking it up to an intermittent system.
Call it what you like, it works if the sun is out and for a lot of folks that's enough while they wait for energy storage prices to drop.
In fact, those systems might be more reliable than the California grid during firing season.
 
Thanks for the feedback! A couple quick thoughts and then I should move to one of the other forums?

As far as shading goes, I’m an actual tree hugger, and even more so an air breather, so I really value trees. The trees that will be the offenders for shade though are old and/or in rough shape. When we cleared the property 12 years ago to build the house, the fellow who dropped the tRees dropped them “out” away from the house site so he could easily get at the logs, instead of “in” toward the house site so they would have been in a big pile. The felling damaged a number of the surrounding trees which are now in slow decay. They also went from being center of the forest trees to being on the edge of a yard, so they are reaching toward the open space. Reaching and decay are going to result in them coming down on their own at some point in the nearer future. Long way of saying we are good with cutting some down.

Our power company is Penelec, which is part of First Energy. I didn’t know they accept only certain inverters, so will be checking on that. That’s useful, thanks!

As far as diy, I’d still like to sort it out myself, but a long way to go on that. Mostly need to check the rules on the tax credit and see what is needed to qualify, as well as make sure my contractors and electrician are ok with doing it. Plus may need an engineer to sign off on some things for the local gov and Penelec. Not opposed to having an installer, but I really want to learn more about it overall, as the proposed plans are all over the place with brands of panels, inverters, etc.

Finally, as far as battery backup goes, in 12 years we’ve only had a handful of outages, most are 5-6hours. One has been 18hours. So ultimately not a huge problem, but I do worry about more substantial outages, blizzards, floods, cyberattack (LOL) and such. Sure would stink to have all that on the roof and still be cold and thirsty for 3-4 days…

The first proposal was for 39 Panasonic modules, Solar Edge optimizers and inverter, and a battery backup. I liked that proposal a lot, but the cost was $68k!!! That was from a guy who worked for what i’ll call a national company. It took me a while to digest, and now three weeks later, he’s gone. Emails come back undeliverable from his email, his number is no longer in service, and the National contact email bounces back too.

Vendor #2 is a more regional company but is still bigger. That proposal calls for 44 Silfab modules that would cover the back of the house, the garage, and the shop. They would equip those with IQ8 micro inverters.They literally won’t say what the other components would be. Total cost with battery bu would be $64k.

Vendor 3 thinks a plain string system would be fine as they aren’t concerned with shading. They are thinking 34 Panasonic 400w modules, a 12kw SolArk inverter and 3 Simpliphi batteries. Cost would be $58k. I know there will be shading issues, and they have made several mistakes in communication and ended up spec’ if the wrong property because of a typo in the address. Having a call with them today, so we shall see what they come up with. They are veteran owned (as is SolArk) and I’m a vet too, so that counts for something to me. At this point after the mistakes though, it’s only counting enough to keep me talking to them.

Vendor 4 is quoting Qcells, IQ7 microinverters, Enphase combiner, Enphase Smart Switch, Enphase 10.5w LiB for $57k. That vendor is very responsive and i know a couple people who have been happy with his work. Has made a couple small mistakes along the way.

Vendor 5 is small and local. He’s a little hard to get a hold of by email because he is literally on roofs every day. He’s proposing 40 Q Peak 400w modules with IQ8 inverters for $43k. He prefers generators to batteries so hasn’t quoted a battery bu. He is on top of his game and feels that shading is not being given due consideration by other vendors. I also know a couple people who he installed for and they are happy with him and the system.
 
As far as diy, I’d still like to sort it out myself, but a long way to go on that.
With help from the forums and other online sources you can become expert enough to DIY the system. That doesn't mean you have to worry about your back, you can still have the knowledge and be the "foreman" directing your handyman. Decent vendor prices for a total install are around $3/W, DIY is somewhere from $1 to $2/W.

You can use "design" tools for mounting systems (e.g., IronRidge, UniRac) to get exactly what you need for your mounts. Based on the rough calculations in #2, you can probably get a DIY system installed for under $10k. See @upnorthandpersonal 's post Building an off-grid 10kW solar system with 28kWh LiFePO4 battery, inverter, BMS and charge controllers for under €10K. Given your need for shade-free and "quality" his system might not be exactly what you want, but may give you some great ideas.

As far as shading goes, I’m an actual tree hugger, and even more so an air breather, so I really value trees.
SAM has a shade module where you can put in your trees and see what happens through the year (use the video tutorials to learn about it).

Possibly you only need to trim one or two. The prime solar time is 10 to 3, so in summer you might not have any problems. In winter southern trees may cast very long shadows. I have a ton of trees south of me, but I live in a two story house (a bit misleading as there's no real bottom story ; -) so the roof is fairly high up and only a few panels are affected.

I'd definitely go microinverters, varying shade can be very difficult for string inverters and shorten the panels lifespan.

Our power company is Penelec... I didn’t know they accept only certain inverters, so will be checking on that. That’s useful, thanks!
Usually, they only care that they're UL listed. You'll probably need insurance too.

... now three weeks later, he’s gone...
So many charlatans out there... another good reason to understand it and DIY it. You'll be able to do any needed maintenance yourself.
I paid a professional to do mine, but the first two people I spoke with were charlatans. My plan was to have a small system installed "professionally", and then use that as a model to expand it. But I also did my leg-work in advance and knew exactly what needed to be done and why.

Vendor 3 thinks a plain string system would be fine as they aren’t concerned with shading. They are thinking 34 Panasonic 400w modules
Thirtyfour 400w panels at an average insolation of 4 is 34x400x4 = 54 kWh/d, or 19,856 kWh/y. That's about +15% for shading.

, a 12kw SolArk inverter and 3 Simpliphi batteries. Cost would be $58k.
Since you don't have much in the way of outages I'd say skip the batteries and go with a "sunlight" system. You can add batteries in a few years when the prices come down more or get a natural gas generator for the in-between time.
Sol-Arc is now offering microinverters, but their battery coupling solution is via AC coupling so a bit pricey.
Some systems (e.g., Enphase) include things you'll want (e.g., neutral-forming transformer, rapid shutdown, arc fault) that aren't included in other systems, which ends up jacking up their price. It'll be hard to compare apple-to-apple prices between vendors; ask a lot of questions on the forums.

...They are veteran owned (as is SolArk) and I’m a vet too, so that counts for something to me....
Have to agree, that and American made should mean a lot. Most of the innards of the devices are not manufactured in the USA, but the quality ones are still designed and tested in the U.S., and managed out of the U.S. or at least have a U.S. branch (e.g., SMA).

Vendor 4 is quoting Qcells, IQ7 microinverters, Enphase combiner, Enphase Smart Switch, Enphase 10.5w LiB for $57k. That vendor is very responsive and i know a couple people who have been happy with his work. Has made a couple small mistakes along the way.
That doesn't sound good.

Given what you've said he should be quoting IQ8s on the roof, and IQ gateway (aka Envoy). You only need the IQ System Controller (aka Enpower or smart switch) if you want sunlight backup or battery backup and it can be added later. An IQ8+ is about $170 to $190, so 34 of them (one for each panel) is about $6000 and the Envoy in the combiner box is about $900. The Enpower is about $2200.

To DIY, you'd get something like Unirac or Ironridge mounting system and bolt it to the roof (comes as a kit from your design). Connect the microinverters to the mounts, plug the panels into the microinverters and mount the panels to the rack. Run the AC down to the combiner box and do something like a line-side tap to connect it to your power system (or into your fuse panel if it's conveniently located and the backplane can handle it, see the 120% rule).

Your handyman can bolt the mounts to the roof (you'll want to watch videos on how to do this properly to seal roof penetrations and instruct them), haul the panels up, and mount the panels. An electrician can do the grid hookup (Enphase doesn't power up the microinverters until the final commissioning (you'd want to take their online courses) so the rooftop wiring is safe (although you must follow the local building codes of course for the proper gauge, conduit, and possibly roof-elevation). There are online companies that will build the plans for your county's building permits and calculate all the wire gauges needed. The power company will want you to sign a net-metering agreement and they'll install a new bidirectional meter, typically that's all for free. Once the meter is installed you activate the microinverters and you're off!

Vendor 5 is small and local. He’s a little hard to get a hold of by email because he is literally on roofs every day. He’s proposing 40 Q Peak 400w modules with IQ8 inverters for $43k. He prefers generators to batteries so hasn’t quoted a battery bu. He is on top of his game and feels that shading is not being given due consideration by other vendors. I also know a couple people who he installed for and they are happy with him and the system.
He sounds knowledgeable, and his price is $2.68/W, which is well under the $3/W for an installed system. Check to see if that includes the IQ System Controller (aka Enpower) so you're both sunlight and generator ready.

From what you said in the OP, your long-term power needs are roughly 12.3 kW of panels with net-metering to average your bill with no shading. 40x400W panels is 16kW, so about a +33% buffer.

Keep in mind you don't have to go 100% coverage annually. For example, if you went with 10 kW of panels now it would cost a lot less to install. Even better, most utilities charge a low rate for less than 1MW/month usage and a surcharge for over that. If you have enough panels installed you'll always be avoiding the surcharge. You can run for that a year, adjust tree trimming/moving in your yard as desired, and then add more panels later (easy to do with Enphase microinverters as nothing needs to match). That has to be balanced against the state and federal incentives, most are 1-time (e.g., no credit for expanding the system later).
 
Last edited:
With help from the forums and other online sources you can become expert enough to DIY the system.
That's exactly how I built the CNC, and it worked out great. Took about 18 mos. of research, but built a $25,000 machine for $6,000.
You'll probably need insurance too.
As in additional homeowners?
Since you don't have much in the way of outages I'd say skip the batteries and go with a "sunlight" system.
I think it's going to be difficult to talk any of them into it. Maybe when they see I know what is going on (when I get to that point), they will change up a bit.
So many charlatans out there...
I don't think any of the remaining vendors are charlatans per se as most of them have someone recommending them, but I'm starting to feel like they install what they like to install and are used to installing, and also possibly what they get the highest margin on. I just want to make sure I'm the right system at a fair price. And I certainly don't mind them making money on it, but also don't want to ridiculously overpay.
you can still have the knowledge and be the "foreman" directing your handyman
I've actually been "auditing" a couple solar design courses online from an online system. Not sure if the rules prevent me from saying the name or not? It has been helpful, and covered a couple rack systems, although not in substantial detail. Enough though I think to know that it isn't rocket science as I had originally thought it was. I always get nervous about roof penetrations, but the things they showed were basically a piece of aluminum flashing with a "bracket" premounted and sealed on it. A lag bolt with a shoulder and a rubber gasket fed through a rubber "cork" and went through the bracket into the rafter. Seemed pretty straightforward.
Enphase doesn't power up the microinverters until the final commissioning
I was wondering how that would work, as my thought was the panels would start producing electricity immediately.

Question: in the event of a grid outage, if there is no backup, where does the electricity generated go? Fed into earth ground?
Keep in mind you don't have to go with 100% coverage immediately.
I hoping to make this one and done! I doubt very much the loml will agree to it without being done the first time. She's totally onboard, but construction disruption is REALLY not her thing.

Thanks a ton to everyone who has helped so far, btw
 
As in additional homeowners?
Yes. My utility required insurance, but when I called for my system they added it at no additional cost.

... I always get nervous about roof penetrations, but the things they showed were basically a piece of aluminum flashing with a "bracket" premounted and sealed on it. A lag bolt with a shoulder and a rubber gasket fed through a rubber "cork" and went through the bracket into the rafter. Seemed pretty straightforward.
It is simple, once you know how. I suspect most failures are from not following the instructions or getting the right type for the roof.

I was wondering how that would work, as my thought was the panels would start producing electricity immediately.
Different systems work in different ways, Enphase microinverters won't turn on until commissioned. Given microinverters should have anti-islanding, they should never activate until they detect the grid.

Question: in the event of a grid outage, if there is no backup, where does the electricity generated go? Fed into earth ground?
TL;DR: It doesn't go anywhere, but test to be sure.

Long Answer
Electrical energy always goes from high to low voltage. If you break the circuit (e.g., a light switch), resistance becomes very high and so any voltage difference becomes very small and so no/little current will flow.

No Backup
When the grid goes down, microinverters detect this and enter what is known as anti-islanding mode. This is like turning a light switch, the power is just off and doesn't go anywhere. It's a requirement in most (all?) countries to not back-feed the grid.

Backup
When the grid goes down, a switch (ATS) inbetween your system and the grid disconnects the grid and activates the backup. In the case of Enphase the ATS is one of the components in the IQ System Controller (aka Enpower). The backup system must have some sort of brain to control the microinverters otherwise the system could overcharge the batteries or make the voltage dangerously high.

Some AC Coupled systems "control" microinverters by forcing them into anti-islanding. All inverters approved for the U.S. can also have their power controlled by varying the frequency (this is how the utility can throttle back DERs). That method is a bit slow though, for things like sunlight backup Enphase microinverters use a direct communication link, ethernet over powerline.

Wind?
I suspect what you've heard is that Wind Turbines need to divert excess power (typically into a heating coil). The reason why wind turbines need this isn't because electricity must flow, it's because it makes the turbines less expensive to build. Wind turbines use the electrical load as mechanical resistance to prevent the blades from spinning too fast.

So, Solar panels are safe to just disconnect and turn on/off like a light bulb. That doesn't mean something won't have frayed and you have a short somewhere. With electricity, it's always good to not make assumptions and test with a meter.

I hoping to make this one and done!

Hope that helps! Best of luck and please keep us in the loop!
 
Back
Top