Hacker News new | past | comments | ask | show | jobs | submit login
A 2-Axis, Multihead Light Positioner (bunniestudios.com)
93 points by etiam 10 months ago | hide | past | favorite | 15 comments



BTW, if anyone has questions about the project I will be giving an NLnet webinar Q&A on May 2, 11.00 - 12.00 CEST. Details at: https://nlnet.nl/webinars/

(you should also check out Charles Papon's webinar tomorrow, also listed at the link above, if you're into RISC-V or open source hardware!)


Ever heard of Checkpoint technologies? They sell commercial infra-red in-situ inspection systems that seem superficially similar to what you're developing.

https://checkpointtechnologies.com/


fwiw the videos aren’t loading for me on chrome or safari on ios.


Looks like I had a power outage, but the site should be back.


Very impressive, as always from bunnie. The background/introduction post [1] is informative for people like me who haven't kept up. Basically he has invented (as far as I understand?) the "IRIS" technique to inspect chips that are still packaged.

This should make it easier to look "under the hood" of system-on-chips and such, to see if there perhaps are units that are unaccounted for. If someone knows more about the motivation/goals with the research that would be interesting to have further explained than the post(s).

This post is about the lighting needs the imaging requires, and the very detailed design of a mechanized light source positioning system.

Also, there seems to be an actual paper about the technique over on arXiv [2] if you want more academics.

[1]: https://www.bunniestudios.com/blog/2023/infra-red-in-situ-ir...

[2]: https://arxiv.org/ftp/arxiv/papers/2303/2303.07406.pdf


>> If someone knows more about the motivation/goals with the research that would be interesting to have further explained

The organizations that need to ensure the chips they receive are 'as expected' don't know how those chips may differ or why (or even that they will.) They just need to prove that they received what they ordered, on every unit.


> If someone knows more about the motivation/goals with the research that would be interesting to have further explained than the post(s).

I work in medical devices. One of the changes in the new FDA guidance for cybersecurity is a greater focus on verifying the integrity of your supply chain, both physical and software. This type of inspection could be a potential mitigation of some of those verification issues.


Bunnie is developing (has developed?) a portable high security app platform of sorts, the Precursor. https://www.crowdsupply.com/sutajio-kosagi/precursor

This post is tagged with Precursor, so I assume this project is to add another level of assurance about supply chain integrity for the components going into the precursor.


Wonder if there was consideration given to mechanically moving mirrors to facilitate the light positioning rather than moving the light source itself. It may lend to modularity as well if you need to switch out light sources (LED to Laser) and could allow for more complex configurations like notch filtering to remove unwanted reflective wavelengths. My first guess is it may have been cost as mirrors and filters are expensive.


You guessed right -- mirrors are expensive, and also heavy; plus it's yet another piece to align (in infrared, so you can't use unaided vision to align the optics; yes, you can use a coaxial visible beam to help with this but that's also expensive and hard for various reasons). If you're using an incoherent light source there's also the problem of collecting all that light and focusing it onto the sample, and each optical element incurs some loss of light.

In the end, the actual LEDs themselves are tiny; a PCB containing the LEDs + driver ICs is smaller, lighter, and brighter than any mirror/lens assembly I could easily source with the requisite coatings to operate efficiently in this frequency band.

For a laser source, a fiber optic source can make a lot of sense, but fiber optics optimized for this frequency of light turn out to be pretty expensive, and the termination of the fiber optic still needs a collimator lens. At the end of the day, the net weight and cost budget still works out in favor of mounting a laser diode directly on the microscope head, so long as the power requirement is under a watt. Beyond that, the weight of the heat sink starts to be a factor, and an off-board laser starts to make more sense.


> Initially, I tried to do the angular positioning entirely using an electronically addressable strip of LEDs, but the pixel density was insufficient.

It's also possible to steer light electromagnetically using a spatial light modulator, but the angular range would be very narrow.


Have you come across AOM too - https://en.wikipedia.org/wiki/Acousto-optic_modulator, they appear to be able to control the angle of light also.


I had no idea that regular black plastic IC packages were transparent enough in infrared to be able to image the underlying silicon.

Is this by design? I wonder if this transparency causes problems with photo-sensitivity in some circuits.


The technique only works on CSP and flip-chip packaged devices. These are chips that are basically bare silicon die that are bonded onto PCBs or substrates.

This package type is fairly popular in mobile phones and many types of CPUs.

The technique does not see through regular black plastic IC packages.





Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: