compromising emanations, 2005.11.29 notes

that notes are not finished yet; come back in a week or so; last changed: 2005.12.15

2005.11.29 I made some experiments with the compromising emanation. Here are my notes.

• the scene
• probes
  • probe 1
  • probe 2
  • probe 3
  • probe 4
  • probe 5
  • probe 6
• recorded data on floppy

the scene

The equipment I used was:

• the monitor,
• the antenna,
• the oscilloscope,
• the camera.

The monitor had a vertical refresh rate of 60 Hz and the horizontal one of 75 kHz. It was the one that emitted the emanations I looked at.

The antenna was made of two wires fixed (with a sticking tape) to the ruler. Each wire was 20 cm long.

The antenna was connected to the oscilloskope.

I used the camera to record everything.

1

I put some black image on the monitor's screen. I set the oscilloscope to 20 μs per one div of the scale. Such was its settings:

I put the antenna on the monitor and found such place for it that I could see something on the oscilloscope. I saw such image:

avi

I guessed that the ticks I saw was the horizontal refresh ticks (I mean: the tikcs that occur when the beam of electrons come back from the right to the left).

2

I changed the horizontal resolution to 5 μs per one div of the scale to see everything better. The vertical resolution was 2mV per div. That was the settings:

I saw something like that:

avi

The distance between ticks is 2.5 divs = 12.5 μs. So the frequency of that ticks is 80 Hz. So it is quite possible that that ticks are the ticks of the horizontal refresh.

3

Then I changed horizontal resolution to 2 μs per div. What I saw on the oscilloscope was that:

avi

4

Then I switched on averaging - every 16 screens was averaged. I saw something like that:

avi

5

Then, with still the same settings, I compared the emanation of the black screen against the white one.

The black screen:

The white screen:

I really could not see any difference between the two.

6

Then I found out that when I took away the metal box which stood by the antenna-oscilloscope cable, the image on the oscilloscope disappeared. Well, seems that the box worked as the antenna and not antenna I made.

avi

So then I removed the box and put the antenna in the new place, where it could really detect something. Then once again I tried to compare the black screen against the white one, but the effect was exactly the same.

recorded data on floppy

Then I recorded some measured data on floppy. Sampling rate was 500 MHz, data was averaged every 16 passes, vertical resolution was 1mV/div, horizontal one was 2 μs/div. I put the white vertical band on the screen (I used this image) and recorded some measures. Here they are.

For example: one of that data files looks like that in gnuplot: