diy solar

diy solar

My big-bang-for-the-buck setup

plympton

I make things and fix stuff
Joined
Nov 2, 2023
Messages
137
Location
Portland
TL;DR: For a bit over $3k (after tax break) + my time, I'll offset half (or more!) of my electric use, and this should pay off in less than 5 years - even in rainy urban Portland!

I've been looking into solar for 20+ years, and never had the opportunity to dive in. With the Biden 30% off sale on panels & storage, I decided to take a dip this spring. I'm glad I wasn't a super-early-adopter - compared to what I saw in the 90's, today's gear is practically plug-n-play!

My plan is modest: To cut my electric bill by $1/day & get 'free' air conditioning this summer. After watching hours of videos & consulting the extensive knowledge pool here at Solar DIY on Reddit, I concluded that a small DIY system was the way to go. I'm an engineer, stubborn DIY'er, and the $50k quote I got from a local installer made my head explode, so...

1712072591245.pngIMG_0686.jpg
 
System Parts & Cost

On the advice of others, I picked up an Emporia Vue and hooked it up to my main + essential circuits and watched them for a few months to get an idea of what I needed. Since I'm not going whole house, I determined that I could get away with 3 kW of power - that would cover my furnace (250w), washing machine (500w), freezer (100w), fridge (80w), room AC (500w) -or- small heater (900w), server (150w), and office/wifi/internet (250w). The Vue & a Kill-A-Watt were really helpful in figuring out my "true needs".

Here's a breakdown of my expenses. I went through a few variations - 240v vs 120v was a big one. I chose 120v for the lower complexity, and a big driver was the backup-generator sizing. I almost went with a 6000xp + LifePower4 battery combo, but it's really overkill for what I need now. Really, if that took 120v generator feed, I probably would have gone for it. It would have been marginally more expensive, but way more capable and expandable.

In the end, I went with a 3 kW Firman dual fuel generator, 2 Bluetti AC300/B300 kits, 6 Hyperion 400w panels. Based on my calculations, I could fill the batteries with solar, and run my essentials (about 500w avg) overnight - they would be close to drained when the sun was about to start filling them. Rinse, repeat. My only real regret was going with the 6 circuit Transfer Switch and not the 10 - I realize that there are days I could power, say, the kitchen on solar and have the sun toast my bread, or add another window AC upstairs.

Spreadsheet.png
 

Attachments

  • PV_Panel_Size.zip
    11.7 KB · Views: 1
Solar
I don't have a south facing roof, and I can't do ground panels on my urban 50'x100' city lot. I do have a detached garage with a south facing roof. It has a less-than-optimal angle (16 degrees vs 31 optimal). In practice, it hasn't been a huge hit to generation.

To start, I used a 100w portable panel that I tested on good days to make sure I could get enough sun. It performed amazingly - going above 100w often at solar noon.
IMG_0622.jpg

Since my garage not on a "real roof", I could take liberties with mounting. I chose to use Renogy mounts directly to my roof. Lots of roof sealant, and at peace with possible leaks, these kept a low profile. They seem to do the trick. All the panels are accessible if I need to clean or maintain them. I worry that there's not enough gap under for cooling, but Portland is a moderate climate. If it's an issue I could make a rack later without much hassle.
IMG_0662.jpgIMG_0686.jpgIMG_0767.jpg
Since I'm using low-voltage MPPT's on the Bluetti, I had to run 2 strands of PV wire in a conduit from the garage to the house 3 x 37.07 Voc = 111v + ~ 12v for worst-cast Portland sun + cold << 150v max. Not too bad. Even the trench wasn't awful to dig once I committed to it. All the panels are tied together with grounding clips and I drove a nice long grounding rod nearby.
IMG_0766.jpgIMG_0765.jpgIMG_0764.jpg
 
Inverter & Battery

I went with Bluetti AC300's & 2 B300. They seemed reliable, refurbs were cheap, and they have 2 1200w MPPT's. They are a good size & stack (unlike EcoFlow). Technically they're "portable" so I could move them to a cabin in the future if I decide to go "whole hog" later. The whole thing could be removed in an hour or 2 if necessary.

From the solar runs, I have a 2-string disconnect, feeding into a box that goes into my basement. I also have a generator inlet with a 50-amp capable wire if I want to upgrade later. The solar goes into my um... less than optimal "everything room". I had to make-do. 🤷‍♂️.
IMG_0763.jpgIMG_0762.jpgIMG_0760.jpg
I built a dolly for the Bluetti stuff, mounted and connected the transfer switch, and can power the AC300 via panel or generator by moving the cord to the different outlet. I use an ESP32 with Bluetooth Proxy to connect the Bluetti to Home Assistant. It's AWESOME!

Yes, that is a laundry basket to the left. There's a chute above that. The water to the right is off. Not ideal, please don't judge!

IMG_0761.jpgIMG_0759.jpg
 

Attachments

  • IMG_0759.jpg
    IMG_0759.jpg
    56.5 KB · Views: 19
  • IMG_0538.jpg
    IMG_0538.jpg
    121.6 KB · Views: 18
Conclusion

I'm REALLY please with how this came together. I've done a ton of projects that went off the rails, but this came together nicely. The Bluetti? Marginally reliable, and their support stinks - until you "find a way". The software works - just barely, but good enough. It's the right system for me, for now.

I learned a ton, and (unlike my cabin build) didn't suffer in the process and would be willing to do it again.

I still have clean up to do - the wires from the panel to my Bluetti need some conduit. I know it's not "to code", but I believe it's at least safe. If there's anything obvious 😱😨please let me know!

Thanks to all the info and support - this was fun! My kids are SICK of me talking about solar and saving $2/day, but at least I have a friendly audience here. 😁
 
Good job on the work and thank you for sharing your experience with lots of details.
 
Interesting setup. Those Hyperion panels almost look square in the pictures, are they truly 66 x 45” ? I’m still trying to figure out a similar setup and what I want to do. I’m really wanting DIY to keep the costs down, but for the life of me can’t get anyone to tell me how close my panels can be from a wall or peak without trying to sell me a whole system. I just need the code! lol.
 
System Parts & Cost

On the advice of others, I picked up an Emporia Vue and hooked it up to my main + essential circuits and watched them for a few months to get an idea of what I needed. Since I'm not going whole house, I determined that I could get away with 3 kW of power - that would cover my furnace (250w), washing machine (500w), freezer (100w), fridge (80w), room AC (500w) -or- small heater (900w), server (150w), and office/wifi/internet (250w). The Vue & a Kill-A-Watt were really helpful in figuring out my "true needs".

Here's a breakdown of my expenses. I went through a few variations - 240v vs 120v was a big one. I chose 120v for the lower complexity, and a big driver was the backup-generator sizing. I almost went with a 6000xp + LifePower4 battery combo, but it's really overkill for what I need now. Really, if that took 120v generator feed, I probably would have gone for it. It would have been marginally more expensive, but way more capable and expandable.

In the end, I went with a 3 kW Firman dual fuel generator, 2 Bluetti AC300/B300 kits, 6 Hyperion 400w panels. Based on my calculations, I could fill the batteries with solar, and run my essentials (about 500w avg) overnight - they would be close to drained when the sun was about to start filling them. Rinse, repeat. My only real regret was going with the 6 circuit Transfer Switch and not the 10 - I realize that there are days I could power, say, the kitchen on solar and have the sun toast my bread, or add another window AC upstairs.

View attachment 206536
If I'm reading your cost chart correctly, it looks like you are getting an 80 % tax credit on the panels.
About 42 % on the AC300/B300.
 
Interesting setup. Those Hyperion panels almost look square in the pictures, are they truly 66 x 45” ? I’m still trying to figure out a similar setup and what I want to do. I’m really wanting DIY to keep the costs down, but for the life of me can’t get anyone to tell me how close my panels can be from a wall or peak without trying to sell me a whole system. I just need the code! lol.
Good luck getting anyone to tell you what's good without submitting full drawings & paying for a permit. That was my experience when I was looking at buying a parcel of land & trying to determine whether it was buildable.

I didn't do any permits or inspections. Since it was my garage I could a) see if it leaked easily, and b) decide if I cared or not. I'm pretty sure I used enough sealant to prevent leaks for at least a few years. In my case I had to choose whether to go below the gutter, or cantilever over the roof ridge, or just go right into the ridge (which I'm sure isn't good practice).

I believe they're almost exactly 68x45. Add 2-3" to the width for the Renogy mounts I used.
 
Please share! We all love reading ridiculous quotes 😂
I'll see if I can dig it up. It wasn't so ridiculous as "didn't pencil out". Sol-Ark 15k inverter, too many panels, all the grid-tie stuff to make the interconnect, etc. And I'd have to get a new roof. My startup is on fumes, so trying to squirrel cash away for the fall, so it was a no-go (I don't finance anything except my mortgage, so I do without instead of going into debt).
 
If I'm reading your cost chart correctly, it looks like you are getting an 80 % tax credit on the panels.
About 42 % on the AC300/B300.
There's some hand-wavy math in there. Bluetti partially refunded me after delivery because both of their inverters were defective & DOA. So there's the invoice-vs-actual, and not sure how to explain that to the IRS.
 
I'll see if I can dig it up. It wasn't so ridiculous as "didn't pencil out". Sol-Ark 15k inverter, too many panels, all the grid-tie stuff to make the interconnect, etc. And I'd have to get a new roof. My startup is on fumes, so trying to squirrel cash away for the fall, so it was a no-go (I don't finance anything except my mortgage, so I do without instead of going into debt).
It's not a bad quote, just way way overkill - about $21k in hardware before all the cables, mounts, cutoffs, etc. I'm curious how my 2.4 kW south facing setup fares compared to the 8 kW East/West estimate he proposed. He was a nice guy, small business, flexible in some ways, but the delta was just insurmountable.
 

Attachments

  • Upliftic Energy.pdf
    1.5 MB · Views: 9
Good luck getting anyone to tell you what's good without submitting full drawings & paying for a permit. That was my experience when I was looking at buying a parcel of land & trying to determine whether it was buildable.

I didn't do any permits or inspections. Since it was my garage I could a) see if it leaked easily, and b) decide if I cared or not. I'm pretty sure I used enough sealant to prevent leaks for at least a few years. In my case I had to choose whether to go below the gutter, or cantilever over the roof ridge, or just go right into the ridge (which I'm sure isn't good practice).

I believe they're almost exactly 68x45. Add 2-3" to the width for the Renogy mounts I used.
Yeah, it pretty much makes me want to build a “solar shed” in the backyard so that it’s basically a ground-mount with storage, and I don’t have to mess with all that stuff. Although to net-meter gotta get permits.
 
I'll see if I can dig it up. It wasn't so ridiculous as "didn't pencil out". Sol-Ark 15k inverter, too many panels, all the grid-tie stuff to make the interconnect, etc. And I'd have to get a new roof. My startup is on fumes, so trying to squirrel cash away for the fall, so it was a no-go (I don't finance anything except my mortgage, so I do without instead of going into debt).
At least in Portland with PGE though, you can switch to a time of use program and the solar ROI can be under 6 years for a grid tied system. In Vancouver, WA our power is under 9 cents/kwh, no TOU option, so there just isn’t ROI on anything unless you DIY it. I mean there it, but it’s like 15-25 years depending on how much you pay. The PUD said if you pay $3500/kw installed the ROI is 25 years.
 
It's not a bad quote, just way way overkill - about $21k in hardware before all the cables, mounts, cutoffs, etc. I'm curious how my 2.4 kW south facing setup fares compared to the 8 kW East/West estimate he proposed. He was a nice guy, small business, flexible in some ways, but the delta was just insurmountable.
Yikes!
I'm not a fan of quotes that itemize out tax credits, but not the parts you are buying. No detail about what battery or racking system you're getting but they make sure to include how many trees you save... Maybe it's just me, but that says "look over here, not over there".
 
At least in Portland with PGE though, you can switch to a time of use program and the solar ROI can be under 6 years for a grid tied system. In Vancouver, WA our power is under 9 cents/kwh, no TOU option, so there just isn’t ROI on anything unless you DIY it. I mean there it, but it’s like 15-25 years depending on how much you pay. The PUD said if you pay $3500/kw installed the ROI is 25 years.
Holy cow that's cheap! Before PGE stopped TOD option (like Jan 1 this year), I was considering that - it was pretty reasonable, and you could get an ROI for just load-shifting with batteries, no solar. The new TOU option is much less viable - the off-peak rate is higher, and on-peak is just punitive. I have a max 6000-watt electric oven and 2 kids so can't time-shift dinner to either end of peak (.41 $/kWh, 5p-9p). When that dies I (ironically?) can get gas and reconsider it.
 
Yikes!
I'm not a fan of quotes that itemize out tax credits, but not the parts you are buying. No detail about what battery or racking system you're getting but they make sure to include how many trees you save... Maybe it's just me, but that says "look over here, not over there".
He was using a 3rd party quote site, so I don't hold that against him. I don't think "normal" people care enough to ask for the details, sadly. Leads to a lot of confusion when promises != what's delivered.
 
Holy cow that's cheap! Before PGE stopped TOD option (like Jan 1 this year), I was considering that - it was pretty reasonable, and you could get an ROI for just load-shifting with batteries, no solar. The new TOU option is much less viable - the off-peak rate is higher, and on-peak is just punitive. I have a max 6000-watt electric oven and 2 kids so can't time-shift dinner to either end of peak (.41 $/kWh, 5p-9p). When that dies I (ironically?) can get gas and reconsider it.
I’d be surprised if Portland doesn’t ban gas stoves and ovens by that time lol.

You know I was looking at the GE heat pump wash and dryer and realizing that might have a better ROI than solar lol. Using like 100-200W vs 5000W.
 
I’d be surprised if Portland doesn’t ban gas stoves and ovens by that time lol.

You know I was looking at the GE heat pump wash and dryer and realizing that might have a better ROI than solar lol. Using like 100-200W vs 5000W.
I'm pretty sure nobody is banning gas stoves anytime soon - too many connected folks like to cook with gas. Those electric dryers, however, are THIRSTY. Electric stoves, too (when everything on mine is on, it will go north of 6000w!). I haven't heard that heat-pump-anything has all the bugs worked out - yet. They'll get there, but all the gear isn't rock-solid - small compressors, fans, etc.
 
Back
Top