I know there are all kinds of lists on github.
Is there a list of companies that are inherently good (I know it's subjective) and are hiring developers?
Would you mind elaborating on that? Some days I feel like I’d rather work for a non-profit than where I am.
Lots of possible roles, take a look!
2 years ago build somebody a simple ultrasonic scanner with a Raspberry Pi.
Anyway it's great. A lot of such tech is just expensive .. because it is since ever and customer pay the price.
These days I think almost all resources should be thrown into inventing and developing carbon capture technologies.
I don't mean to belittle the achievement or sound harsh, but when thinking of priorities one has to consider what will happen if something is not done.
[1] Hint we can't do this via the raw growth and exploitation used to lift the first billion people out of poverty.
There's so much that may or may not work: the USB port must support OTG/Host on the hardware side, which is already a hit-and-miss. The port must be able to supply the power required for the ultrasound. Then the Android stack must support USB OTG - and that must have been properly tested during manufacturing. And then the vendor must have compiled the Linux kernel below Android with support for uvcvideo (which is not standard enabled in cheap phones). Then the CPU and video unit must be powerful enough to run the software with video playback.
This could be a revolution, but the Ultrasound Scanners need to be in the hands of low income farmers then imagine how quickly the dam ecosystem would develop across to humans.
I've been looking to get a cheap one from China, but you can't, I'm not sure why, but perhaps it needs a certain chip that is not made cheaply yet. There are DIY's around but seem immature.
If you want to change the world, get the price down for use on animals.
This is a good first step, but while the medical community is in change it'll move at a snail pace which means a lot of people will die.
I'm not sure why the Africans need a $2000+ butterfly scanner rather than this kind of thing https://www.aliexpress.com/item/Veterinary-ultrasound-machin... ?
I presume the Chinese one isn't as good. The images do look a bit blurry.
Regarding the price of piezoelectric transducers, the typical buzzers (i.e. the flat round piezo discs) are very cheap.
===
Idea 1: Using piezo buzzer elements (the piezo disks):
The piezo discs are ridiculously cheap, but large monolithic transducers. There is a trick to partition them into multiple transducers:
The piezo disc transducers are essentially: a steel membrane, onto which a thin layer of piezoelectric material is deposited, onto which a very thin and weak layer of metal is vaporized, onto which leads can be soldered (at low temperature).
The steel membrane defines the resonance frequency in its monolithic configuration, and can be thickened to make it less flexible, say by electrodeposition of copper in a copper sulfate bath, in order to make the steel membrane a more rigid base (since we are interested in the local pressures and not the global pressure which deforms the disc).
The thinnest metallic layer on top can be easily patterned even mechanically, see for example:
https://web.archive.org/web/20071214032759/http://www.geocit...
Here the single disc transducer is patterned into 4 independently transducing quadrants. For context on the motivation of that page, it is for actuating a scanning tip for a Scanning Tunneling electron Microscope (STM).
Essentially they just use a ruler and a fine blade to cut the metal layer, and then clean up any remains with a pencil eraser to clean the surface for soldering and eliminate any conductive burrs after the cut to prevent shorts between the quadrants. It's pretty genius.
For context on why they want the 4 quadrants for the STM see: https://web.archive.org/web/20121114015437/http://www.geocit...
So in theory you could lay out a grid of piezo sensors this way, and then solder afterwards, or alternatively pattern the thin metal layer in such a way with traces such that between the top pads of each new subtransducer there are thin traces going to the edge of the piezo disc...
The patterning could happen mechanically or perhaps more sanely lithographically: spin coat (on a CPU fan) some photoresist, pattern with minification (don't need 5nm resolution) then etch the metal layer away... This seems to be within financial reach of top universities of even the poorest nations...
The cheapest buzzer element on DigiKey (helpfully without leads presoldered): https://www.digikey.com/product-detail/en/murata-electronics...
$ 0.19827 @ 8k pieces
(Cheapest in the naive sense of sorting by price, ideally you'd go through the list of 107 active products and inspect the price as a function of quantity)
The largest buzzer element on DigiKey (larger to allow more sub transducers to be defined):
https://www.digikey.com/product-detail/en/pui-audio-inc/AB65...
$ 2.67 @ single piece $ 1.4 @ 1000 pieces
(Largest in the naive sense of sorting by diameter, ideally you'd again go through the list of 107 active parts, and check the datasheets or images to estimate the actual piezo area diameter as opposed to the metal base diameter)
For the full list of active (currently buyable) DigiKey piezo elements: https://www.digikey.com/products/en/audio-products/buzzer-el...
For patterning one could use an analog camera, and focus it on a high resolution monitor, try and develop a couple of times to find the best setting for final resolution. then place the piezo with photoresist in place of the film...(probably irreversibly modify the analog camera a bit). Photoresist should be relatively cheap if its consumption is properly planned.
===
Idea 2: Using pressure sensitive paints, and optical readout.
Wikipedia https://en.wikipedia.org/wiki/Pressure-sensitive_paint
Pressure sensitive paints work by sensing the local oxygen concentration, which may be problematic in aquous conditions, perhaps it could still work on dissolved oxygen? Perhaps variations on oxygen-based PSP could be designed for liquids (better impedance matching with human body).
Perhaps a layer of PSP can be cured/dried in a flexible foamed state (with trapped air bubbles). Such that it can be brought in contact with water but still retain trapped air?
===
Idea 3: Using "schlieren photography": I am unable to find it back, but years ago I saw a video (I believe from some israeli university/college) where they somehow used Schlieren photography in water (which is rather incompressible) to visualize the beam quality and spot size of the transducer & accoustic lens / ...
If Schlieren imaging was used to characterize the quality of the ultrasonic beam, it suggests that perhaps Schlieren imaging could be used directly, if properly miniaturized etc...
One can easily find papers on the subject of using Schlieren imaging for characterizing hydrophones etc...
Technically there is a serial cable you can buy for hobbyist/personal/internal use but "Under current Apple policy this cable may not be used with apps sold on the App Store." https://redpark.myshopify.com/products/serial-cable-l2-db9v
