Reverse engineering of microchips by photo
Bright vertical stripes are metal conductors; horizontal stripes from which only dark borders are visible are the conductors from polysilicon ; an irregularly shaped region with a bright black border is doped part of the silicon substrate; yellow circles are the connections between the layers of the chip.
Each transistor is formed by a polysilicon conductor crossing the doped silicon region:
By tradition, such transistors are called MOSFETs (“metal oxide semiconductors”), even when the gate is not metallic, but polysilicon. The author half-wittedly suggested that no polysilicon transistor manufacturer wanted to call them POS .
The operation of the transistor is that when a positive voltage is applied to the gate, the doped region, including the source and drain, becomes conductive; when the voltage is removed from the shutter, the source and drain open.
Here is the same photo with marked transistors and conductors: (I added the designation of the connections between the layers in the author’s illustration)
Of the vertical metal conductors, only three are involved (connected to the semiconductor layer); the rest just pass by. The red conductor outside the frame is connected to the power supply, the blue one to the ground. The polysilicon conductors included in the frame are signed with letters; transistors whose gates are formed by these conductors are signed with numbers. At first I was confused by the fact that - unlike the circuit diagram - transistors do not have clear boundaries on the microcircuit: the source of one transistor is simultaneously the drain of the next transistor.
The blue (doped) region above the conductor A is connected at the point with a grounded metal conductor. When there is voltage on A, then transistors 1 and 6 open, and the entire blue region above the conductor B becomes conductive, i.e. grounded. The yellow metal conductor connected to the blue region at the point .
When there is voltage on conductor B, transistors 2 and 7 open, and the yellow conductor is grounded through the connections and .Finally, when there is voltage on C, 3 and 8 open, and the yellow conductor is grounded through and . As a result, the yellow conductor is grounded when .
The Out explorer is connected to the blue region at the point . This point is connected via ground when all three transistors 6, 7, 8 are open, i.e. when .
Finally, the point is connected to yellow conductor when one of the transistors 4 or 5 is open, i.e. when . This means that will is grounded through the yellow conductor when . If the point is not grounded neither through the yellow conductor nor through the upper right part of the blue region, then the connection is with a red conductor pulls it to a positive voltage: a long narrow strip of doped silicon gives the desired resistance. The author does not mention that the Out conductor, crossing a narrow blue strip, forms another transistor: when Out is grounded, this additional transistor closes and the suspender turns off.
I drew a schematic diagram, generally maintaining the layout of the elements:
As a result, a positive voltage is obtained on Out when :
This scheme forms a single-bit block ALU : inputs A, B and C are operands, D and E select the operation (AND or OR). In the four-bit ALU of the Z80 processor, this pattern is repeated four times:
In this photo, T34BM1 (Soviet clone Z80, taken apart and photographed BarsMonster ) it can be seen that the remaining blocks in the Z80 ALU are repeated four times.
That’s how, marking the transistor behind the transistor from the photograph of the microcircuit, an online simulator Visual 6502 was created.