09/12/2015 at 13:10 •
Despite my prior budget claims, things worked out for me, and I was able to get all the electronics ordered for the laptop. Since I went with the primarily module-based design. Not much work was needed to make it work. All I had to do was solder the switches to the boost converter modules, change the plug on the solar panel, solder wires between the solar management module and the boost converter, and rig up the usb A to dual uB cable. Took maybe an hour.
Since I am going to the Hackaday DC meetup tonight, I wanted to bring it with me. However, it was just a pile of wires and stuff, not exactly transportable. Luckily, the box that the screen came in was large enough to attach the solar panel and screen to the lid and contain everything else inside. It looks a bit janky; everything is attached with twisted wire, but it will work until I can design and 3D print a proper case for it.
I went to test the AVR flashing, and found that one of the pins I had setup as part of the programmer on the pi was previously occupied. Fortunately, all I had to do was change the pin until I found one not occupied and it works great.
09/07/2015 at 16:23 •
I acquired the screen for the laptop. I also got a usb current meter so I can determine how much current the whole setup is pulling. After changing the max usb output current an cutting a couple jumpers on the screen, I was able to power the screen through the pi. It turned out that the whole setup, including keyboard and wifi dongle, pulled less than 750 mA, or 3.75W. The battery that I am looking to get, and hoping will be as large as claimed, has about 57Wh, so the battery life ought to be around 13 hours, which exceeds my original goal.
I found a nice writeup from a fellow who was setting up a CLI only linux box (http://lcorg.blogspot.com/2009/02/project-building-all-text-linux.html) and used some of his suggestions, most notably the "screen" function, which is great because I often need multiple command lines open.
Now begins designing the case for the laptop. I can at least work on the front panel, which will need to hold the screen, a breadboard, and a socket for programming.
09/01/2015 at 03:02 •
I found a source for a much larger, single Lipo battery, http://www.all-battery.com/li-ion3-7v15600mahrechargeablebattery31059.aspx. Assuming this is legit, which I need to check. This battery would be one 57.72 Wh battery, replacing two 24.42 Wh batteries. This would give a better battery life and eliminates one boost converter module. I think that I will order one of these as one of Adafruit solar lipo management modules and make sure that they work without dying before I put things together.
08/31/2015 at 02:15 •
Due to, lets call it budget issues, the hardware will be acquired slowly over time. Fortunately, this gives me time to get the code up and running on the Pi prior to getting the hardware. So, I set up avr gcc on the pi. It compiles code with no issue. Haven't tested it actually flashing code because I was trying figure out wiringPi software PWM. The screen that I plan to use has a PWM input to control screen brightness, so I wanted to have a simple program to control brightness from the Pi. The code currently works. At least it can dim an LED as expected. Have not hooked it up to a scope. According to the wiringPi website, it will only operate at ~100Hz. I do not know if this is enough and wont until I can order the screen. Code is in the attached github. One thing I did notice was initializing wiringPi in the program. According to the PWM page on the website, I can use either wiringPiSetup() or wiringPiSetupGpio(). I found that, while using either one compiles, only wiringPiSetup() actually does anything. This might be operator error.