Hi, I am interested in the Artemis module, after seeing Robin_Hodgson’s Artemis PCB and low power optimizations and Freethegameboy.info’s ability to use an Apollo3 to make a solar powered Gameboy, a solar powered laptop sounds like it would be a very interesting project.
It’d be awesome to build an old-school linux OS with a minimal set office and light web apps (like nanolinux or tinycore linux) running on Artemis module and powered by a solar panel around the keyboards and an integrated power manager like E-peas AEM10941 or TI BQ25570 on the same PCB.
My hackaday page is also here: https://hackaday.io/project/177716-the- … kic-laptop Feel free to contact me there or my via my twitter handle @techrecount (or DM). Thanks and looking forward to seeing more projects here!
The applications this could be useful for would be a sensor where outlets are not available. There are definitely more efficient ways to integrate the solar charging circuitry, such as https://www.tindie.com/products/jaspers … n-battery/ but I am only skilled with connectors that I can plug and play.
I don’t think you’re going to find a bunch of plug and play parts that you can just connect to make your idea reality, You’d probably need to custom engineer something that uses some existing parts in it.
You might start with a pi zero. That and an e-ink screen would probably consume less than 2 watts of power and should work fine on a 5 watt solar power source. Still not plug and play, but would be a lot easier route than to build a bunch of stuff from scratch.
I’ve moved on from the Pi Zero after already determining the average energy consumption: https://www.pidramble.com/wiki/benchmar … onsumption (b/t 0.4W-0.7W for the Zero+ 1-2W for the Display). It is also due to the fact that the Rpi Zero is on a 40nm process node. Some application processors use as little as 5nm, and could support solar power, but few SBCs use less than 30nm- I know one- the Khadas VIM3, which is 12nm, but it has a lot of other peripherals unnecessary for a basic typewriter) The problem is that I’d like this laptop to work indoors and outdoors, and ambient lighting levels isn’t likely to change in the long term (i.e. there isn’t a future need for brighter indoor design, as it may already be at 500lux or more.
The Artemis/Apollo4 could theoretically run indoors 24/7 w/o needing a large battery- if the PCB was designed for a limited function, like a Pomera typewriter: https://goodereader.com/blog/reviews/di … ter-review it wouldn’t need a large kernel for drivers that aren’t used. Additionally, having the kernel modules swappable per app could allow the OS to run in the limited SRAM of the MCU, and could be loaded like a multi-boot Live linux USB when a different app were needed, to minimize the power consumption of the RAM (i.e little to no SPI PSRAM/FRAM would be needed).
There are many proof-of-concepts showing solar powered Raspberry Pi’s- mainly outdoors, but I’d like to see laptops more like solar powered calculators- to do that, it wouldn’t be a very performance-geared laptop, and wouldn’t serve a large market segment. but I do like the idea of an integrated circuit, which would make it more efficient. E-peas has one with an MCU: https://twitter.com/epeas_news/status/1 … 6652998663 At the 0:22 mark, It mentions the AM-1522 solar cell. This video shows an eink display connected to the same MCU https://www.youtube.com/watch?v=_Wh1qm_Cemg Having efficient solar cells and the power manager on the same PCB as the MCU would enhance the efficiency as well as lowering the cost.
There already also integrated solar modules, like https://hackaday.io/project/159139-tiny … odule-tsem but I agree a custom engineering solution would be needed. I’d like to scale it up to power at least 3-5mW. That said, that may not be enough to power an e-paper display. But then again, the number of applications would be limited to text editing, and maybe pdf reading. A crowd-funded path to market seems possible after polling interest. Eventually the MCU could use more RAM and storage that could be powered by the solar without using the battery at a faster rate than the harvesting rate. But starting out, I’d like an RTOS that could be like the ESP32- able to run a number of apps with GUIS.
It would be very cool if you can build this and make it work. If you can beat the $250 price the Pomera has and have the web functionality you’re wanting to add, I think you may have something people will pay for.
