This idea has a similar vibe to @janamarie deciding to use an STM32 as an IO expander in a thread last weekend. It would be a neat challenge to take a schematic and just see if you can replace every chip with a microcontroller somehow 🤣
=> More informations about this toot | View the thread
I know that other people have done this before and that there are manufacturer application notes on this because it has legitimate applications in high-power converters that need complex, custom control loops. I still kind of want to try it myself.
=> More informations about this toot | View the thread
With @sad_electronics and others discussing funky switching regulator circuits in another thread, I was reminded of something I’ve been wanting to try for a while now: Take a small microcontroller such as STM32G030 and make it a switching regulator with as few external components as possible.
=> More informations about this toot | View the thread
I guess I could probably merge a couple of the design’s five analog supplies, but then, a really nice, brand-name LDO is less than a dollar so I’m not going to bother figuring out which.
=> More informations about this toot | View the thread
is eight power rails too much?
=> More informations about this toot | View the thread
That's some expensive tape right there. This is anisotropic conductive film that cures under heat and pressure, resulting in a permanent bond. I think I'll pass and go for the cheaper (~100 USD/roll) stuff that works just like normal sticky tape and doesn't require heat, but requires larger pad size (~1x2mm vs. ~0.5x0.75mm IIRC).
For added fun, these ACF tapes have a shelf life that's only between six months and two years, and the heat-curing type even needs to be refrigerated.
[#]electronics
=> More informations about this toot | View the thread
You may not like it, but this is what peak performance looks like.
=> More informations about this toot | View the thread
I just found out that LCSC has started stocking whitelabel devboards of the kind you can find on aliexpress under their house brand. Prices seem competitive with aliexpress, and you get a real invoice and no order quantity limits.
Down below are links to an STM32 bluepill clone [1], a little neat-looking Gowin FPGA board [2] and a very pink arduino uno clone [0].
[0] https://www.lcsc.com/product-detail/Development-Boards-Kits_LCSC-boards-LCKFB-ColorEasyDuino-Pink_C22397129.html
[1] https://www.lcsc.com/product-detail/Development-Boards-Kits_LCSC-boards-LCKFB-DKX-STM32F103C8T6_C22396880.html
[2] https://www.lcsc.com/product-detail/Development-Boards-Kits_LCSC-boards-LCKFB-LJPI-FPGA-G1_C41421753.html
[#]electronics
=> More informations about this toot | View the thread
I’ve just finished watching Pantheon, a US-made animated sci-fi series, and it was great. I’d highly recommend it if you’re into sci-fi. The writing was very good, it builds intelligently and respectfully on previous works in the genre and the animation’s visuals are great with some highlights that would not look out of place in a work by Satoshi Kon.
[#]scifi #series #anime
=> More informations about this toot | View the thread
TIL that you can just watch cosmic rays hit the Superkamiokande experiment's giant (30m diameter) water tank 1000m below the Japanese alps in real time online:
https://www-sk.icrr.u-tokyo.ac.jp/realtimemonitor/
[#]science #physics
=> More informations about this toot | View the thread
TIL that the Zotero (academic citation / bibliography manager) browser addon actually has decent support for generating citations of youtube videos. It automatcially includes all the metadata like description and upload date in the citation even.
=> More informations about this toot | View the thread
The way they spin that into basically "tomorrow you can buy this!" in the press release is pretty gross. I think press releases like this, and the resulting journalistic articles that regurgitate the overhyped claims, failing to notice the bullshit, are really harmful to science communication in general. This shit erodes trust, because it leads to an endless stream of hypey press releases with very little real-world results to show in the end.
=> More informations about this toot | View the thread
This is cool research that they are doing, and the paper does not contain anything outright wrong. In the paper, the glucose monitoring stuff is presented as a purely hypothetical motivation, a "maybe one day we could use it for that" kind of thing.
=> More informations about this toot | View the thread
Here's the plot from their paper. Note that the horizontal axis, showing glucose concentration, is labelled in mlligram per milliliter, not per deciliter as you'd use in medicine. 500 milligram per milliliter is halfway to a saturated solution.
Also note how whilie it improves total SNR, their metasurface tech actually worsens the system's sensitivity to glucose level. It's funny-sad to read how they argue around that in the paper.
=> More informations about this toot | View the thread
Note that nowhere in their paper do they claim that they are able to measure glucose concentration in a blood sample, much less in living tissue. Instead, their only actual claim is that in a lab setting, where there are no other variables, they can distinguish between pure water and a thicc glucose solution.
=> More informations about this toot | View the thread
Since there's no reason to assume that the output from the off-the-shelf infineon mmWave radar chipset that they use interacts with glucose in particular in any meaningful way, from what I can tell they are effectively just measuring density. A 500mg/ml (~50%) solution of glucose in water has about 20% higher density than pure water.
=> More informations about this toot | View the thread
Today in irresponsible #science marketing bullshit: #UWaterloo puts out a real turd of a press release, linked below.
The press release: "[our stuff] senses glucose levels for diabetics more accurately than ever before"
The actual research: basically a better antenna that can distinguish between distilled water and a ~25000 mg/dL glucose solution absent any other factors. For context: At around 500 mg/dL you're dead.
Link to the press release:
https://uwaterloo.ca/news/media/no-more-needles-tracking-blood-sugar-your-wrist
=> More informations about this toot | View the thread
One of my pet peeves is software that runs unit tests as part of release builds or packaging.
[#]linux
=> More informations about this toot | View the thread
Currently at #38C3 about to watch a talk by stacksmashing on Apple’s USB controllers and some dude with a Cellebrite-branded backpack and hoodie sits down in the second row. Cellebrite is a manufacturer that’s well-known for their smartphone “forensic” imaging devices. Kinda weird for a chaos event where traditionally the makers of surveillance software weren’t very welcome?
=> More informations about this toot | View the thread
My #chipofthemonth for December is this little analog XYZ color sensor chippy, AMS/Osram AS73210, just mostly because it looks really neat. The thing is basically just an array of parallel-connected photodiodes with thin-film interference filters on top. The XYZ color space is used professionally because it approximates the receptor response in human color perception better than e.g. sRGB.
Alas, this chip is obsolete in favor of newer offerings of AMS that have digital readout.
[#]electronics
=> More informations about this toot | View the thread
=> This profile with reblog | Go to jaseg@chaos.social account This content has been proxied by September (ba2dc).Proxy Information
text/gemini
