solar powered laptop project

Thread Starter

Teljkon

Joined Jan 24, 2019
267
I have an old laptop that I am going to try to power purely from solar power I was curious if anyone here could turn me on to the formula's and math. I will need to figure its power demands. I have the original charging brick that I could cannibalize,. However I also have a universal charger made to work off a cars cigarette lighter an I have found a way to plug that directly into a small solar charge controller. Totally a learning project the long term goal is to power it 24/7 from solar. In short a miniature system that I will hopefully be able to scale into bigger stuff in the future.
 

Ya’akov

Joined Jan 27, 2019
9,079
PV panels are cool, and solar projects can be fun. I feel you may have bitten off more than you wanted to chew with this project, though.

An order of magnitude sanity check suggests that your universal adapter is rated ~100W at 12V. A 12V 100W PV panel is about about 1m x .5m, and that rating is best case, full sunlight. If you are in a good area, you might have 75% output during daylight. But, again, that's just wishful thinking. You will probably need two or three such panels for reliable operation.

Even then, you say your goal is "24x7" operation. That means your laptop's battery would have to be able to handle the non-daylight hours, which is very unlikely unless it is doing almost nothing. So, that means you need a larger battery to accumulate the power while it is available.

Your first step is to actually measure the power requirements of your system. Without actual information about what the system will need to run, you can't calculate anything useful.

I'd suggest, as a start, you do a small MCU project, which you could then scale up. You can use a smaller battery and panel but it is still the same process with much more chance of success.

Have fun!

[EDIT: typo repair]
 

Ya’akov

Joined Jan 27, 2019
9,079
MCU? I am thinking you meant Micro control unit? are you talking about powering an Arduino/Ras pie?
Yes. An RPi would be particularly well suited. It runs a real OS, and you could do generalized things with it. There is going to be a lot of information about using solar power with one as well.

It would be possible, if you wanted, to include a monitor, and use the RPi as a desktop computer with a graphical interface. Of course, if you want a network connection, it would have to be using equipment not on commercial power, if you care about independence from that.

The thing is, by scaling back the power requirements, your inevitable mistakes will be of smaller magnitude and cost will be lower. The RPi is an amazingly useful device on its own, though.
 

MisterBill2

Joined Jan 23, 2018
18,179
Fortunate to have the original power module. From that you can read the voltage and current requirements and know what the demand from a solar panel will be. Many of the laptops require about 19 volts at a bit less than 2 amps, while a few much older ones may use 12 volts.
And certainly some of the laptops do require a specific power module, primarily to force you to buy an overpriced replacement from them. Probably if we know the brand and model you can get a response telling you if it demands an OEM module or not.
 

Thread Starter

Teljkon

Joined Jan 24, 2019
267
I am not apposed to the suggestion of down grading to a pie I actually have one sitting idle. The goal of this adventure has nothing to do with the compute unit. and I am not a stickler on using outside power for things like a internet router.

The goal is simply to learn about solar enough to have real knowledge about powering electronics with it. Understand the ratios if you will how much one needs to push power of X amount of gear. I am still going to start with the laptop just because why not.

Again though do you guys know a place to look at formula in the raw or maybe search terms I should use. Yaakov can spit out an estimate from base knowledge that I will fail to drive a laptop. I am trying to get from here to there.

Thanks
 

MisterBill2

Joined Jan 23, 2018
18,179
OK, if the solar cell system is going to do any more than directly power a load,yes, it gets a bit complicated because batteries are involved and batteries do not like abuse. Fortunately there are many good published circuits and also a lot of products available.
 

MisterBill2

Joined Jan 23, 2018
18,179
Probably, if the computer will actually be used for anything important, you will need to have a battery to provide enough time to do an "emergency save and close" in case there is an interruption.
Otherwise, just an efficient regulator to avoid any over-voltage problems.
 

Ya’akov

Joined Jan 27, 2019
9,079
Probably, if the computer will actually be used for anything important, you will need to have a battery to provide enough time to do an "emergency save and close" in case there is an interruption.
Otherwise, just an efficient regulator to avoid any over-voltage problems.
One of the aspirations for the project in the TS first post is 24x7 operation so the battery is absolutely essential.
 

Thread Starter

Teljkon

Joined Jan 24, 2019
267
One of the aspirations for the project in the TS first post is 24x7 operation so the battery is absolutely essential.
since it is a laptop it makes sense that the on board bms should be able to handle allot of it. I do have an old Gel lead acid battery from a broken jumper that could be added.
 

Thread Starter

Teljkon

Joined Jan 24, 2019
267
I'll suggest first configuring an Arduino to measure and record the energy produced by the solar panel, and the energy consumed by a load.
I think this makes perfect sense as well as a phase two then for phase three add a batttery and or more solar depending on data collected. Ill probably set this up this morning. thanks guys.
 

Thread Starter

Teljkon

Joined Jan 24, 2019
267
Interesting. So I am getting about 30 ma from both panels in parallel. These panels advertise 12v at 8w and 5w peak respectively. So even at peak power I am only capable of pushing 1.084 amps. even a raspberry pie needs 2.5 amps at 5v so I am a ways away from that. I mean it is winter and it was afternoon with indirect sunlight but I expected a little more.

So my math says 12v, .03 A, .36 Watts, 400 Ohms

I know I can stick it on a buck converter to easily to drop the volts. What I still don't know, or understand how to estimate, is will I get more amps out of it? To this day I don't understand why you cant find a calculator online that has all four parts of the formula set up on the same quad? There all related so why wouldn't it work?

OHMS-LAW-300.jpg


I under stand Ohms is a measure of resistance but cant I assume a static environment? Also the charge controller didn't want to show up at all. I am just figuring not enough power. Any advice is always appreciated I am going to stick the multimeter on it again with better sun. Another consideration is I am suctioned to a home window there may be a UV filter tanking my draw. So that could be the death stroke.
 
Last edited:

Ya’akov

Joined Jan 27, 2019
9,079
1. RPi power supply recommendations at power requirements are two different things. This chart:

40EF2FC1-9C97-4F01-A68B-4DD251AAC755.jpeg

taken from here: https://www.raspberrypi.com/documentation/computers/raspberry-pi.html#power-supply shows the actual current draw expected during RPi operation.

2. UV light is not an issue. PV panels do not use UV light which is too energetic. The largest majority of energy in solar radiation is in the visible light range so the semiconductors in PV panels is chosen so the bandgap matches the energy in that part of the spectrum (1.1 to 1.7 eV) for the most efficiency.

Your problem is either bad PV panels or not enough visible light. Don’t sweat the UV.
 

Audioguru again

Joined Oct 21, 2019
6,674
Name-brand solar panels produce their rated current at their rated output voltage when outdoors pointing directly at the sun at noon on a sunny day. Your solar panels are almost in the dark.
 
Top