Comparing ‘AI’ For Basic Plant Care With Human Brown Thumbs

The future of healthy indoor plants, courtesy of AI. (Credit: [Liam])
The future of healthy indoor plants, courtesy of AI. (Credit: [Liam])
Like so many of us, [Liam] has a big problem. Whether it’s the curse of Brown Thumbs or something else, those darn houseplants just keep dying despite guides always telling you how incredibly easy it is to keep them from wilting with a modicum of care each day, even without opting for succulents or cactuses. In a fit of despair [Liam] decided to pin his hopes on what we have come to accept as the Savior of Humankind, namely ‘AI’, which can stand for a lot of things, but it’s definitely really smart and can even generate pretty pictures, which is something that the average human can not. Hence it’s time to let an LLM do all the smart plant caring stuff with ‘PlantMom’.

Since LLMs so far don’t come with physical appendages by default, some hardware had to be plugged together to measure parameters like light, temperature and soil moisture. Add to this a grow light and a water pump and all that remained was to tell the LMM using an extensive prompt, containing Python code, what it should do (keep the the plant alive), and what Python methods are available. All that was left now was to let the Google’s Gemma 3 handle it.

To say that this resulted in a dramatic failure along with what reads like an emotional breakdown on the part of the LLM would be an understatement. The LLM insisted on turning the grow light on when it should be off and had the most erratic watering responses imaginable based on absolutely incorrect interpretations of the ADC data, flipping dry and wet. After this episode the poor chili plant’s soil was absolutely saturated and is still trying to dry out, while the ongoing LLM experiment, with an empty water tank, has the grow light blasting more often than a weed farm.

So far it seems like that the humble state machine’s job is still safe from being taken over by ‘AI’, and not even brown thumb folk can kill plants this efficiently.

Hydrogen Trains: Not The Success Germany Hoped They Would Be

As transport infrastructure in Europe moves toward a zero-carbon future, there remain a number of railway lines which have not been electrified. The question of replacing their diesel traction with greener alternatives, and there are a few different options for a forward looking railway company to choose from. In Germany the Rhine-Main railway took delivery of a fleet of 27 Alstom hydrogen-powered multiple units for local passenger services, but as it turns out they have not been a success (German language, Google translation.). For anyone enthused as we are about alternative power, this bears some investigation.

It seems that this time the reliability of the units and the supply of spare parts was the issue, rather than the difficulty of fuel transport as seen in other failed hydrogen transport problems, but whatever the reason it seems we’re more often writing about hydrogen’s failures than its successes. We really want to believe in a hydrogen future in which ultra clean trains and busses zip around on hydrogen derived from wind power, but sadly that has never seemed so far away. Instead trains seem inevitably to be following cars, and more successful trials using battery units point the way towards their being the future.

We’re sure that more hydrogen transport projects will come and go before either the technological problems are overcome, or they fade away as impractical as the atmospheric railway. Meanwhile we’d suggest hydrogen transport as the example when making value judgements about technology.

Black and white photo of Evertop computer on desk

The Evertop: A Low-Power, Off-Grid Solar Gem

When was the last time you saw a computer actually outlast your weekend trip – and then some? Enter the Evertop, a portable IBM XT emulator powered by an ESP32 that doesn’t just flirt with low power; it basically lives off the grid. Designed by [ericjenott], hacker with a love for old-school computing and survivalist flair, this machine emulates 1980s PCs, runs DOS, Windows 3.0, and even MINIX, and stays powered for hundreds of hours. It has a built-in solar panel and 20,000mAh of battery, basically making it an old-school dream in a new-school shell.

What makes this build truly outstanding – besides the specs – is how it survives with no access to external power. It sports a 5.83-inch e-ink display that consumes zilch when static, hardware switches to cut off unused peripherals (because why waste power on a serial port you’re not using?), and a solar panel that pulls 700mA in full sun. And you guessed it – yes, it can hibernate to disk and resume where you left off. The Evertop is a tribute to 1980s computing, and a serious tool to gain some traction at remote hacker camps.

For the full breakdown, the original post has everything from firmware details to hibernation circuitry. Whether you’re a retro purist or an off-grid prepper, the Evertop deserves a place on your bench. Check out [ericjenott]’s project on Github here.

Turning Old Cellphones Into SBCs

[David] sent us a tip about a company in Belgium, Citronics, that is looking to turn old cellphones into single-board computers for embedded Linux applications. We think it’s a great idea, and have long lamented how many pocket supercomputers simply get tossed in the recycling stream, when they could be put to use in hacker projects. So far, it looks like Citronics only has a prototyping breakout board for the Fairphone 2, but it’s a promising idea.

One of the things that’s stopping us from re-using old phones, of course, is the lack of easy access to the peripherals. On the average phone, you’ve got one USB port and that’s it. The Citronics dev kit provides all sorts of connectivity: 4x USB 2.0, 1x Ethernet 10/100M, and a Raspberry Pi Header (UART, SPI, I2C, GPIO). At the same time, for better or worse, they’ve done away with the screen and its touch interface, and the camera too, but they seem to be keeping all of the RF capabilities.

The whole thing runs Linux, which means that this won’t work with every phone out there, but projects like PostmarketOS and others will certainly broaden the range of usable devices. And stripping off the camera and screen has the secondary advantages of removing the parts that get most easily broken and have the least support from custom Linux distros.

We wish we had more details about the specifics of the break-out boards, but we like the idea. How long before we see an open-source implementation of something similar? There are so many cheap used and broken cellphones out there that it’s certainly a worthwhile project!

An Inexpensive Way To Break Down Plastic

Plastic has been a revolutionary material over the past century, with an uncountable number of uses and an incredibly low price to boot. Unfortunately, this low cost has led to its use in many places where other materials might be better suited, and when this huge amount of material breaks down in the environment it can be incredibly persistent and harmful. This has led to many attempts to recycle it, and one of the more promising efforts recently came out of a lab at Northwestern University.

Plastics exist as polymers, long chains of monomers that have been joined together chemically. The holy grail of plastic recycling would be to convert the polymers back to monomers and then use them to re-make the plastics from scratch. This method uses a catalyst to break down polyethylene terephthalate (PET), one of the more common plastics. Once broken down, the PET is exposed to moist air which converts it into its constituent monomers which can then be used to make more PET for other uses.

Of course, the other thing that any “holy grail” of plastic recycling needs is to actually be cheaper and easier than making new plastic from crude oil, and since this method is still confined to the lab it remains to be seen if it will one day achieve this milestone as well. In the meantime, PET can also be recycled fairly easily by anyone who happens to have a 3D printer around.

Producing Syngas From CO2 And Sunlight With Direct Air Capture

The prototype DACCU device for producing syngas from air. (Credit: Sayan Kar, University of Cambridge)

There is more carbon dioxide (CO2) in the atmosphere these days than ever before in human history, and while it would be marvelous to use these carbon atoms for something more useful, capturing CO2 directly from the air isn’t that easy. After capturing it would also be great if you could do something more with it than stuff it into a big hole. Something like producing syngas (CO + H2) for example, as demonstrated by researchers at the University of Cambridge.

Among the improvements claimed in the paper as published in Nature Energy for this direct air capture and utilization (DACCU) approach are that it does not require pure CO2 feedstock, but will adsorb it directly from the air passing over a bed of solid silica-amine. After adsorption, the CO2 can be released again by exposure to concentrated light. Following this the conversion to syngas is accomplished by passing it over a second bed consisting of silica/alumina-titania-cobalt bis(terpyridine), that acts as a photocatalyst.

The envisioned usage scenario would be CO2 adsorption during the night, with concentrated solar power releasing it the day with subsequent production of syngas. Inlet air would be passed only over the adsorption section before switching the inlet off during the syngas generating phase. As a lab proof-of-concept it seems to work well, with outlet air stripped from virtually all CO2 and very high conversion ratio from CO2 to syngas.

Syngas has historically been used as a replacement for gasoline, but is also used as a source of hydrogen (e.g. steam reformation (SMR) of natural gas) where it’s used for reduction of iron ore, as well as the production of methanol as a precursor to many industrial processes. Whether this DACCU approach provides a viable alternative to SMR and other existing technologies will become clear once this technology moves from the lab into the real world.

Thanks to [Dan] for the tip.

Ultra-Low Power Soil Moisture Sensor

Electricity can be a pretty handy tool when it stays within the bounds of its wiring. It’s largely responsible for our modern world and its applications are endless. When it’s not running in wires or electronics though, things can get much more complicated even for things that seem simple on the surface. For example, measuring moisture in soil seems straightforward, but corrosion presents immediate problems. To combat the problems with measuring things in the natural world with electricity, [David] built this capacitive soil moisture sensor which also has the benefit of using an extremely small amount of energy to operate.

The sensor is based on an STM32 microcontroller, in this case one specifically optimized for low-power applications. The other low-power key to this build is the small seven-segment e-ink display. The segments are oriented as horizontal lines, making this a great indicator for measuring a varying gradient of any type. The microcontroller only wakes up every 15 minutes, takes a measurement, and then updates the display before going back to sleep.

To solve the problem resistive moisture sensors have where they’re directly in contact with damp conditions and rapidly corrode, [David] is using a capacitive sensor instead which measures a changing capacitance as moisture changes. This allows the contacts to be much more isolated from the environment. The sensor has been up and running for a few months now with the coin cell driving the system still going strong and the house plants still alive and properly watered. Of course if you’re looking to take your houseplant game to the next level you could always build a hydroponics system which automates not only the watering of plants but everything else as well.