Ancestors

Written by DJ🌞:donor: on 2025-01-10 at 13:10

Today's #nakeddiefriday entry is a TK14583 by Toko America, a FM IC for "high speed" comms coming from [redacted]. A short 🧵

Full-res map etc: https://siliconpr0n.org/archive/doku.php?id=infosecdj:toko-america:tk14583

=> View attached media

=> More informations about this toot | More toots from infosecdj@infosec.exchange

Written by DJ🌞:donor: on 2025-01-10 at 13:21

The die is fabbed with two metal layers and built with bipolar devices. Interestingly, the die has 28 pads while the device itself only 24 pins. There are two smaller pads on the right likely used for factory testing, but I am not sure which other two were not bonded out.

We should be able to spot these units: an oscillator+mixer, a limiter with RSSI circuit, a detector, a comparator with OC output, and an amplifier.

=> More informations about this toot | More toots from infosecdj@infosec.exchange

Toot

Written by DJ🌞:donor: on 2025-01-10 at 13:46

The top is taken by the oscillator and its mixer, converting RF input to IF.

=> View attached media

=> More informations about this toot | More toots from infosecdj@infosec.exchange

Descendants

Written by DJ🌞:donor: on 2025-01-10 at 13:47

In the bottom left we got the comparator. The output transistor is placed right between pis 8 and 9.

=> View attached media

=> More informations about this toot | More toots from infosecdj@infosec.exchange

Written by DJ🌞:donor: on 2025-01-10 at 13:48

Right next to it is the amplifier circuit.

Note the large chunk of uncommitted devices beneath a metal pour. I wonder what this was to be used for?

=> View attached media

=> More informations about this toot | More toots from infosecdj@infosec.exchange

Written by DJ🌞:donor: on 2025-01-10 at 13:53

And in the middle of the chip sits the limiter+detector+RSSI combo.

I struggled to figure out which pins correspond to which pads on the left there, as there are 5 pads and 4 of them are clearly bonded out but only 3 pins on the device are allocated for the limiter.

=> View attached media

=> More informations about this toot | More toots from infosecdj@infosec.exchange

Written by DJ🌞:donor: on 2025-01-10 at 14:03

Interesting to see how the designers adjusted bipolar transistor characteristics by effectively putting multiple in parallel? instead of changing the geometry. I wonder why.

/🧵

=> View attached media

=> More informations about this toot | More toots from infosecdj@infosec.exchange

Written by tnt on 2025-01-10 at 14:43

@infosecdj If they depend on those to have a given ratio, it's better to copy the geometry than to change it.

=> More informations about this toot | More toots from tnt@chaos.social

Written by DJ🌞:donor: on 2025-01-10 at 19:34

@tnt they have different geometries elsewhere, so that doesn't seem to be the main factor?

=> More informations about this toot | More toots from infosecdj@infosec.exchange

Written by tnt on 2025-01-10 at 22:26

@infosecdj I guess maybe in that part they needed some precise ratio and not in other ? I'm not sure what part of the circuit that is TBH.

=> More informations about this toot | More toots from tnt@chaos.social

Written by DJ🌞:donor: on 2025-01-10 at 22:27

@tnt Ah, this is from the very top right corner, sorry.

=> More informations about this toot | More toots from infosecdj@infosec.exchange