The Ware for June 2022 is shown below.

Thanks to an anonymous benefactor for donating a few of these for this months’ Ware. The board itself is a bit sparse, but, there are some hefty clues regardless. I think there’s a good chance someone will guess it from this image alone. However, I’ve got a few other images in my back pocket in case it turns out to be too hard to guess. Either way, I’ll add them to this post once some guesses are in!

Because the board is so sparse, I thought maybe it would be fun to also dump the contents of the one chip that is on it. Not that it gives any particularly useful hint about what it does, but because it was fairly easy to do; just an SOIC test clip and a Raspberry Pi does the trick:

sudo i2cdump 1 0x50
I will probe file /dev/i2c-1, address 0x50, mode byte
(sample 1)
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: 00 00 94 4f 00 9e eb 2e c6 0d 12 bf ee 5b 49 2f    ..?O.??.?????[I/
10: 2e 9d 1e 34 f6 30 dd 1a 05 19 df 35 ab 74 df 75    .??4?0?????5?t?u
20: 06 bc 3d e4 f5 fe 7f 2d e6 8b 5b a2 0f 83 6b b5    ??=????-??[???k?
30: 04 7a 3a ae 68 96 5f f8 55 8a ce 3c 91 be 5b c3    ?z:?h?_?U??<??[?
40: e1 07 00 00 00 00 2e 00 0a 19 08 c9 d9 83 50 10    ??......??????P?
50: 13 20 a3 82 01 30 80 9a fd 92 06 3a 06 31 36 35    ? ???0?????:?165
60: 39 34 4a 12 11 9a 01 0e 08 02 15 00 80 88 c5 20    94J????????.???
70: 01 2d 00 00 c8 c3 00 00 00 00 00 00 00 00 00 00    ?-..??..........
80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
****
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................

(sample 2)
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: 00 00 1c 44 fc 2b 6d 07 02 55 9a fe 0d ed 91 98    ..?D?+m??U??????
10: ab 6b 94 51 db bd 2f cb 93 cc e3 b8 e1 17 14 85    ?k?Q??/?????????
20: 9b 5e 0d fd 6b 18 c2 da 67 a6 73 98 99 cb f4 40    ?^??k???g?s????@
30: 3e ab 40 b4 48 eb aa c2 94 94 49 29 12 93 da 3e    >?@?H?????I)???>
40: f0 08 00 00 00 00 2e 00 0a 19 08 95 e2 83 50 10    ??......??????P?
50: 13 20 a3 82 01 30 80 9a fd 92 06 3a 06 31 36 35    ? ???0?????:?165
60: 39 34 4a 12 11 9a 01 0e 08 02 15 00 80 88 c5 20    94J????????.???
70: 01 2d 00 00 c8 c3 00 00 00 00 00 00 00 00 00 00    ?-..??..........
80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................
****
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00    ................

It’s always instructive to dump a couple of samples. Without doing any numerical analysis, eyeballing the two dumps side-by-side makes me think whatever drives this is little-endian (given the formatting of some constants in address 0x40 and above), and the data from 0x04-0x40 is probably cryptographic in nature; assuming the implementation didn’t roll their own cipher, it’s probably either an AEAD, or an HMAC. I say this because the first 2-4 bytes from 0x00-0x04 are likely not ciphertext. However, the block size of AES is 16 bytes, so, it’s not any simple block-based encryption scheme, due to the odd 12 bytes or so that are present. However, the format could make sense if 12 bytes served as the nonce for AES-GCM-SIV, and then maybe the last 16 bytes are the authentication tag; that would yield 32 bytes of encrypted, authenticated data, which would be enough for…

…I’ll stop talking there, before I totally give it away!

Edit: looks like someone has already guessed what it is, so here’s some more photos of it!

The small bit of research I did on the device indicates it uses LAMP for amplification, so the device runs at an elevated but constant temperature, and the results are read out using an electrochemical method. Basically, the reagents are mixed with the sample, and they are pumped through a small channel (lined with some kind of blue film) that goes over gold-plated electrodes on the circuit board. The reaction presumably changes some sort of electrically measurable parameter as it evolves — conductivity, pH, impedance, something like that. I thought the round black object molded into the clear plastic assembly would be a heating element, but it also seems to be an electrode of some sort, as it measures an open circuit before the reagent mix is punctured, but has a high DC impedance once liquids are introduced to the chamber.

The I2C ROM likely encodes per-device calibration parameters, as well as (presumably encrypted and authenticated) traceability data such as expiration dates, serial numbers and the like. The encryption would enforce the expiration of the reagents (hence the disposal of the cartridges to yours truly), and also foil the ability of third parties to make interoperable cartridges.

The Ware for May 2022 is a Lenovo Thinkpad Minidock, Type 4338 from back when I had a T520 Thinkpad — circa 2011, about a decade ago. It’s slightly unusual for its time period, because it was probably one of the last brand-name OEM pieces of hardware that featured a parallel port. As a hardware hacker I bemoaned the parallel port’s obsolescence: it was the closest thing we had to standardized GPIOs on a “full sized OS” until the Raspberry Pi. Anyways, I was cleaning out some old hardware and thought it’d be interesting to see what’s inside. Congrats to Matthew for nailing it, email me for your prize!

The Ware for May 2022 is shown below:

If you’re like me, you’re wondering where the month of May went. I guess that’s what you get for spending too much time writing software.

This one is a quickie that I grabbed out of my scrap pile. The photo is cropped to try and make it slightly more challenging, but, there should also be enough details left to unequivocally identify the original equipment from which this was extracted.

The ware for April 2022 is part of a tx/rx module for putting video and audio over a single optical fiber. As noted by Don Straney, the contributor of the ware:

This was being used to remotely feed a video signal to a projector in an MRI setup, for neuroscience experiments (although these likely had many more uses than that). Usually the projector has to sit in an awkward corner somewhere, just outside the shielded room, to get the right angle to project straight down the scanner’s bore and off a screen or mirror in front of the subject’s face, while the researchers are sitting in the control room 40 feet or so away, with the entire shielded room between them and the projector – this makes for some long cable runs.

It was actually interesting to see the system that they were part of, because it looked like a very “cottage industry” low-volume sort of thing; seemed pretty professional from the outside, but the internals were a hacked-together combo of 3rd-party boards like these ones, with wires soldered on to tap into power and the any pushbutton controls, and a little bit of custom stuff to power all the 3rd-party devices from a common power supply and “press” their onboard buttons as needed to make it a clean self-contained system.

Interestingly, the name on the board identifies the manufacturer as “Communication Specialties, Inc.”:

However, Black Box seems like the type of company that would OEM many of its products, so maybe that’s just the name of the OEM or a company they acquired. So, I’ll give the prize to Matt for — somehow — figuring out where this design came from. Congrats, email me for your prize!

The Ware for April 2022 is shown below.

This is once again another fine Don Straney ware; thank you for the contribution!