• 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

I want to build a system to power a Tiny Home, and man it's a lot to wrap my head around...

Cooljerk

New Member
Joined
Feb 28, 2024
Messages
2
Location
Houston, Tx
I'm sorry if this is a little too broad of a topic, but I've read through all the FAQ and have been doing homework here for the past several weeks and I just want to lay out my entire plans to make sure i'm not missing anything before I start throwing money into this project. A bit of background -- I've built myself a 120 sqft tiny home outside of Houston, Texas on a plot of land I own. I want to do an off-the-grid setup and bring my cost of living down, and a big component of that is Solar Energy. I work from home as a computer programmer, so my energy needs are a bit diverse.

Long story short, I've calculated that, with the equipment I have to do my work, I need a budget of about 4800w per day to run my development computer and some other necessities. A friend of mine in the area converted his garage into a home office and runs it off of solar power (and he also warned me how deep of rabbit hole this can all be, haha), and he gets away with a 4800w system, so that's what I've targeted.

What I'd like is a system where I can use solar panels to charge batteries to provide this 4800w daily. Ideally, I'd like to start with something at 4800w initially, then grow it with redundant batteries for prolonged periods of rain, but for right now, I'm concentrating on having a single battery system. My friend recommended I look into server batteries as they can be had at good prices here in Texas because, while they're nearly impossible to ship cheaply, we aren't afraid of driving 200 miles in this state to pick up something. And, low behold, there's a place in Dallas selling server batteries that seem like they're what I need: https://signaturesolar.com/all-products/batteries/server-rack/eg4-lifepower4/

My plan is to get one of those 24V 200aH batteries, which would provide me a capacity of 4800w if I've done my calculations correctly. That would run me $1250.

My worst-case peak solar hours by month according to a calculator site are:
Jan: 4.08
Feb: 4.27
Mar: 4.89
Apr: 5.48
May: 5.7
Jun: 5.67
Jul: 5.63
Aug: 5.82
Sep: 5.6
Oct: 5.33
Nov: 4.57
Dec: 3.74

So I'm doing my calculations based on an average of 4 hours of peak sunlight per day (since I can skirt by with less power in the winter).

I've looking into buying used Solar Panels from this place in Dallas: https://beenebrothers.com/product/ja-solar-315w-solar-panel-used/

That's a 315w solar panel. As I understand it, that means the solar panel can produce 315w per hour at peak sunlight, so at 4 hours of peak sunlight per day worst-case scenario, a single panel is producing 1260w. So I'd need 4 of those panels to produce the 4800w I want my battery to hold in one day? Is that correct? 4 of those panels at $60 each would be $240.

Assuming I get the number of panels I need, I'm next looking at an inverter. I need it to be the size of my solar array, so the inverter would need to be 4800w as well, is that correct? I've been pricing them, and they look to run at about $500, does that sound correct?

Finally, I'm looking at a charge controller. What I've read is they need to be the size of my solar array wattage divided by the voltage of my battery. So if my battery is 24V, and my Solar Array is 4800W (technically 5040), I do 4800/24 = 200a. The solar panels I'm looking at are 72 cell, so unless i'm mistaken I can go with a PWM charge controller to save a little bit of money? I've been looking around and I think a 200 amp PWM controller tends to go for around $1000?

Do these numbers sound correct? I'm looking at a rough cost, without wiring or mounts, of about $3000? Is there anything obvious I'm missing? Am I getting myself into trouble here or is this a good starting point for more research? Further, assuming in the future I wanted to add a second battery, what pitfalls should I avoid? And if I wanted to add more solar panels, that would mean I'd need a second inverter, correct? Would it be smarter to size my inverter upfront for any future expansion?

Thanks in advance for any response, I'm sorry if this was rambling. It's just simply soooo much to take in.
 
Good for you for getting started on this! You need to go back and study the difference between watts and watt-hours, and probably brush up on power, amp, volts, and then ask your question again.
 
It's just simply soooo much to take in.
Yes, it is. I've been studying up on this stuff for three or four months now, and I feel like I'm still at the bottom of the learning curve, but at least I'm headed upwards.

Your question shows some weak spots in understanding that you will definitely want to learn before completing your design. First, you need to understand the difference between a PWM and an MPPT charge controller. Yes, the PWM is cheaper, possibly more reliable, but can be 30% or more less efficient. But that's not the most important part: you need to understand that the solar panels' voltage must be more carefully matched to the battery voltage if using PWM. It's also crucial to understand the difference between running solar panels in series versus in parallel, and how to match the voltage output of the panels (VoC: open-circuit voltage) to the charge controller to prevent damaging the latter. The VoC of the panels, which increases with cold temperatures, must never exceed the maximum voltage limit of the charge controller and/or inverter...not even by a little bit. So one must carefully calculate, at the lowest possible temperature for his or her location, what that VoC might be: and make sure this is still a suitable margin below the maximum limit of the controller. Amperage and/or wattage is a separate matter. For most controllers/inverters, the PV wattage can exceed the specified "maximum" by as much as 50% without harm--it will simply be wasted (not used).

Also, be aware that used panels will not output their full potential. Panels can decrease in output by as much as 3% per year. How many years have they been in service? Even new panels may not see their full wattage, even in full sun. That wattage is calculated in laboratory conditions, given specific temperatures, angle of light, etc., without the limitations of atmospheric haze, above-average temperatures, or a layer of dust on the panel.

Dust is a real problem over here where I am. While they say we have three seasons (officially), I like to say we have two seasons: mud season and dust season! Dust on the panels must be cleared off nearly every couple of days in some locations during the "dry" season--or risk depletion of the batteries and having no usable power.
 
First thing you need to do is get the description of what you need correct.

Batteries are storage of energy, and are explained inAh and Wh, never just W…
It sounds pedantic, but it will help differentiate what you actually need.

A 4800W system is a far cry from a 4800Wh system.

First, what is the max W output you require? Computers frequently have a W power requirement, big stuff exceeds 700W. Running that 8 hours would be 5600Wh

Total accumulated Wh is the figure your battery will need to have.
 
A couple more points to think about,

1)A 100w panel doesn't usually output @ 100%. I'd figure 80% (80w) at best.

2)Did you figure anything for ac?

3)Electronics don't run at 100% efficiency. Various efficiencies for various types.

It sounds like you are on the right path, and have a decent idea of your needs. I'd suggest that you post your usage figures and calculations and then we can better help you out.

Ed
 
48V for the win. I'm another noob, and while I'm happy with the 24V I'm playing with for my specific application (my boat), I quickly realized that 48V is better in every other application I wanted to use it in. 24V is great for small marine and RV applications, but 48V is better in almost every circumstance to 24V for a stationary application were you might need more than 1500wt at a time. Had to completely pause and rethink my plans once I started running the numbers of cost and efficiency as well as looking at selection and availability of reputable equipment.

Based on your description, if you don't have a specific NEED for 24v DC, (like running a 24v DC motor) as a 24v user I CAN"T recommend it beyond being an improvement over 12v.

2cents.
 
Thanks for the replies so far. To address some concerns people brought up, i do have access to an ac running off of the electrical grid, so i don't need to calculate it into my solar capacity. My goal is to basically run everything except ac off of solar.

Regarding my development computer, i actually do development on a raspberry pi, which ive calculated has a max power consumption of 40w at full load. Raspberry pi runs off of a 3 volt 5 amp wall wart.

The most pressing issue i think is my confusion between watts and watt hours - that is something i keep getting confused with when reading up on this stuff. Without asking too much of anybody, could someone please go through my post and correct me on where im mixing them up specifically? For example, i see solar panels listed at 315w... is that to mean it produces 315w in an hour at peak sunlight? Or am i confused?

I dont want to make it sound like i want someone to do all my homework for me, i definitely cut my teeth in a world of "RTFM" coming from comp sci, but theres a point where rubber meets road where i have to talk to real people to complete my knowledge gap. Text is too unclear without being able to ask questions.
 
The most pressing issue i think is my confusion between watts and watt hours
Watthours are how the power company measures energy delivered to your meter. A thousand Watthours is a kiloWatthour. It is also a measure of the capacity of a battery. Watts are a measure of instantaneous power. One Watt consumed or produced for an hour is a Watthour.
 
Watts is power ... or watts is how much energy a thing can use at any given moment. Watt hours are a measure of power over time.

So, a thing that uses 100 watts for 1 hour will use 100 Wh (Watt-hours) ... and in 10 hours it would use 1kWh (kiloWatt-hour).
 
Last edited:
...

The most pressing issue i think is my confusion between watts and watt hours - that is something i keep getting confused with when reading up on this stuff. Without asking too much of anybody, could someone please go through my post and correct me on where im mixing them up specifically? For example, i see solar panels listed at 315w... is that to mean it produces 315w in an hour at peak sunlight? Or am i confused?

...
W=VA measurement of electrical power. If you have 1 volt at 1 amp it equals 1 watt. Watt-hour is a measurement of electrical watts delivered over an hours time period. Say your device is a 20W light bulb. You leave it on for 2.5 hours. Thus 2.5h X 20W= 50Wh

Your PV panels are rated in the ability to make watts at a certain level of sun intensity. (STC) If they get enough sun to develop the rated full 315w for 1 hour you would have 1h X 315W = 315Wh

Whether watts are delivered for an hour, or fraction of it, it can all be expressed as Wh.
 
Watts is power ... or watts is how much energy a thing can use at any given moment. Watt hours are a measure of power over time.

So, a thing that uses 100 watts for 1 hour will use 100 Wh (Watt-hours) ... and in 10 hours it would use 1kWh (kiloWatt-hour). Note the lower and upper case "W" ... lower case for watts and upper case for Watt-hours.
I was under the impression the word watt was not capitalized, but any abbreviation using watt is capatalized.
 
W=VA measurement of electrical power. If you have 1 volt at 1 amp it equals 1 watt. Watt-hour is a measurement of electrical watts delivered over an hours time period. Say your device is a 20W light bulb. You leave it on for 2.5 hours. Thus 2.5h X 20W= 50Wh

Your PV panels are rated in the ability to make watts at a certain level of sun intensity. (STC) If they get enough sun to develop the rated full 315w for 1 hour you would have 1h X 315W = 315Wh

Whether watts are delivered for an hour, or fraction of it, it can all be expressed as Wh.
W=VA ONLY for DC or Pure resistive loads.
AC inductive loads W is far different from VA
 
Choose 48v, not 24v, and don't buy those used panels when Signature has new 370's for $90 and you're already driving there.
This. Don't spend money on those used ones

"Panels were removed from a commercial ground mount due to a grass fire. They’ve been sorted out so that the worst you’ll see on any of these is some smoke stains on the back sheet and the junction box being wavy from heat.

They are ready for their second life!"
 
Theoretical math is one thing; putting it in practice is often another. Here are some practical examples.

You have a 100Ah battery. You do not wish to discharge it below 20% of its capacity (0.2C), i.e. its DoD (depth of discharge) should be no greater than 80%. Assuming it starts fully charged, 0.8C (80%) of 100Ah is 80Ah. Before we can calculate watts, we must know the battery voltage as well. Let's say the voltage is 24v: 24 volts x 80 ampere-hours = 1920 watt-hours (Wh), following the W=VA formula and adding the hours.

If the tiny home consumes electricity at the rate of 120 watts over a period of 16 hours, this will equal the 1920 Wh. The watts by themselves are the rate of usage, not connected to time. But 120 watts for a full hour is 120 Wh (watt-hours).

If our 100Ah battery is at 48 volts instead, we have: 48 x 80 = 3840 Wh. If this is consumed at the same rate, it could last for 32 hours (double the time). Otherwise, it could provide power at twice the wattage for the same period of time.

Note that in these calculations, the watts calculated for the output must already include those consumed by the inefficiencies of the inverter, so additional equipment (lights, fans, computers, refrigerator, etc.) get to use what is leftover after the inversion process has used its portion.

Solar panels are a different story. Unlike a battery, whose capacity and charge level may be known, panels have a maximum output rating which may quite seldom ever be reached, and can fluctuate wildly in output due to many factors: angle of incidence of the sunlight, haziness/clarity of the atmosphere, temperature of the panels, shading (clouds, trees, chimneys, dust on the panels, or perhaps even the passing flock of birds), etc. The MPPT of the charge controller may not instantly respond to changing conditions, and some efficiency is lost while it updates to track the maximum power point. All this to say that panels are notorious for not producing their full rated output. Once you have a particular figure in mind as your production goal, it is highly probable that you will want to have panels rated in excess of that goal by a certain margin. Rain, for example, may cut solar output down to a mere 10% or less of that which one would achieve on a sunny day. If one had ten times the panels, one might maintain the desired output even in this adverse condition. Of course, when the clouds vanish again, one would then likely have way more power than could be put to use. This is part of the fun of designing the system--getting things balanced to where one is satisfied with the result. But the panels' output must be carefully matched to the charge controllers' and/or the inverters' power limitations, especially voltage, so panels, inverters, etc. all have a dominoes effect--change one, and the others need to be reconsidered as well.

The details get even stickier: for example learning that some inverters will be damaged if the battery is disconnected from them while the inverter is still connected to the solar panels--and that some batteries (those with a BMS) may automatically disconnect themselves for cell voltages that are too high or too low, or for temperatures too high or low. To mitigate against disasters, one must either design relays into the system that would disconnect the PV array in the event of a battery disconnection, or else use an inverter model that is not sensitive to battery interruptions and can operate without a battery. Some inverters can be "overpaneled" and some cannot. Some DC fuses will protect a system from extreme overload and some will just let the surge pass through! There are so many ways to cobble together a system which in the end may use incompatible components! So many ways to muck it up, even over what might be considered a small and nearly insignificant part! Every part of the installation seems important, and it takes considerable time to research each part and know its characteristics.
 
Panels were removed from a commercial ground mount
One thing to consider is that some commercial panels are 96 cell panels which may operate at a higher voltage. That is not as important in a string inverter because it is the sum of the cells in series. In the case of matching high voltage cells to a micro inverter, it may require a micro which has a higher voltage rating. In my case with some used commercial Sun Power panels I used IQ7-X micros, which cost a little bit more. I do not know what the equivalent IQ8 model nmber is.
 
Will your loads be daytime or night time?

For daytime loads, consider AGM battery and overpanel with SE and SW facing PV strings. Your loads will be mostly powered PV direct.

Is that 4800 Wh of load over an 8 hour day? 600W mostly steady?
In that case a 12V system could be fine.

You have grid AC available to power A/C?
Then why a PV system at all? Just plug into same circuit.

A window A/C for 120sf building might run on 500W, draw 2500W for a second to start. A small inverter could do that, and about 2000W of PV panels ($500 worth) would power it most days.

So if you wanted to be disconnected from the grid, that is doable. Except for heating in winter, need to burn something.
 
Your math for panel production came to a reasonably accurate guesstimate answer. Just a more creative way of doing it.
4x315=1260
1260x5hrx.7efficiency=4410wh per day as a rough average. (5040 @80% efficiency)
Instead of buying a separate pwm or Mppt you could get an all in one inverter for under your $1000 estimate of a controller.
 
It sounds like the Tiny Home (TH) has a grid connection available; if so, just use an AIO inverter appropriate to your chosen system voltage (24v?). Feed the grid to the inverter.

Start your system with just:
- 24v battery bank: 12v100ah x 4 ($300/ea?), set up for 24v
- 24v aio inverter ($600 or so? spend lots of time reviewing the features)

For phase 1, get just this part up and running, with all the appropriate cables (use busbars), fuses, and such, and tie it in to the TH electrical system (say another few hundred for all of these). You can vary the battery-bank design with larger Ah choices, at more cost. Once the batteries & cabling & such are in, then you know how much budget can be put towards the AIO; overspend on the AIO to get more features important to you (and bank of solar panels down the line).

You'll need to figure out how to get all of this into the TH footprint. Don't know if your TH is on wheels, or stationary (like a shed) on the property; if on wheels, everything has to support occasional traveling.

Note that the battery-bank is built such that if one battery gives you grief, you can reconfigure the battery-bank to keep going, while pulling out and troubleshooting the problem battery. Can't do this if the single server battery gives grief; or, get to 2 server batteries at least. Plenty of threads here where a "single thing" stops the whole system.

If you have night-time lower power rates, then the grid charges the battery-bank at night, and you use the banked power during the day. This is also your fallback to any loss of production from panels, for whatever reason. If you will run the AirCond from this setup, then oversize AIO even more for that function.

This is how I build power systems into our TH's, and the panels always get added last; we have stable power from the start. As we are off-grid, a generator steps in for the "grid" connection at our property.

Now you can focus on phase 2, adding panels; sorting out mounting and cabling, and feeding everything to the AIO. Perhaps they sit on the ground, or are tied into the TH (mobile) structure. AIO has the mppt built-in.

Hope this helps ...
 
I've built myself a 120 sqft tiny home outside of Houston, Texas on a plot of land I own. I want to do an off-the-grid setup and bring my cost of living down, and a big component of that is Solar Energy. I work from home as a computer programmer, so my energy needs are a bit diverse.
I hope you understand that solar/inverter/batteries is a capital investment, thus has a pay-off period. However, I believe it is wise to be more self-sufficient. That capability may not need an ROI justification. Houston certainly has it's share of storms, from a hard-hitting tropical depression to hurricanes, and now winter weather blasts. We lived on the outskirts of Houston for decades and have been exposed to as much as 11 days without power - during AC season.

I think one of the first things you need to do is get a handle on your electrical wants. If you are currently using your place, turn to your electric bill to determine monthly kWh for starters. Go back as far as possible if you want to compare year-vs-year for each month. Next, see if your provider will give you a breakdown per day. I keep mine to a spreadsheet. For example for me, I have seen days below 10kWh, and the highest 138, but the typical day seems to be half way between these. Because kWh is a quantity, you then know how to address your goals. Live off-grid? Survive 4 days without any grid? Cut your bill in half? You also need to know the peak kW your inverter needs to support your devices, for it must be big enough to meet or exceed an instantaneous load demand. You mentioned a computer system - how about lights, fan, TV, microwave, air fryer, hot plate, coffee maker, desktop oven, electric blanket, etc.? If you want to know the load of a plug-in object, I would suggest a product named kill-a-watt usage meter. If you want to know the total electric load of your home, I would look for a meter that can clamp to your incoming electric line to measure amps. Amps X volts = watts.

Suggestions:
1. Determine Inverter size (watts) to meet load requirements for all devices that can possibly run at once. Costly and challenging to add capacity to this.
2. Determine battery size (kWh) to get you through power outages, or low-light multiple day rains or low sun intensity winter days. Easy to add more capacity.
3. Determine solar panel (watts) production needs. Must be able to develop enough to cover each day's kWh needs, plus fill batteries. Easy to add more capacity - if you have space.

[edit:] For inverter, I'm thinking of an all-in-one type that has connections for Grid or Generator, PV (solar), and batteries. Some of the latest have all of the switches and amperage protections built in. For measuring amperage of your home's input, I'm thinking of a hand-held meter that can clamp around a load wire: black if 120V, black and red (separately) if 240V, but for 240V each contributes at 120V.
 
Last edited:

diy solar

diy solar
Back
Top